AngioDynamics, Inc. v. C.R. Bard, Inc. et al

Filing 189

REDACTED MEMORANDUM-DECISION AND ORDER: RE 188 . It is hereby ORDERED that Bard's motion in limine (Dkt. No. 132 ) is GRANTED. It is further ORDERED that Bard's motion for summary judgment (Dkt. No. 133 ) is DENIED in its entirety. It is further ORDERED that AngioDynamics' motion for summary judgment (Dkt. No. 134 ) is DENIED in its entirety. Signed by Judge Brenda K. Sannes on 5/5/2021. (nmk)

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 1 of 105

UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF NEW YORK
ANGIODYNAMICS, INC.,
Plaintiff,
v.
C.R. BARD, INC. and BARD ACCESS SYSTEMS, INC.,
Defendants.
Appearances:
For Plaintiff:
Philip J. Iovieno
Nicholas A. Gravante, Jr.
Mark A. Singer
Cadwalader, Wickersham & Taft LLP
200 Liberty Street
New York, NY 10281
Adam R. Shaw
Anne M. Nardacci
Boies Schiller Flexner LLP
30 South Pearl Street, 11th Floor
Albany, NY 12207
For Defendants:
Andrew J. Frackman
Edward N. Moss
O’Melveny & Myers LLP
7 Times Square
New York, NY 10036
James P. Nonkes
Philip G. Spellane
Harris Beach PLLC
99 Garnsey Road
Pittsford, NY 14534

1:17-cv-00598 (BKS/CFH)

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 2 of 105

Hon. Brenda K. Sannes, United States District Judge:
MEMORANDUM-DECISION AND ORDER
I.

INTRODUCTION
Plaintiff AngioDynamics, Inc. (“AngioDynamics”) brings this antitrust action against

Defendants C.R. Bard, Inc. and Bard Access Systems, Inc. (collectively, “Bard”), asserting a
claim of illegal tying in violation of section 1 of the Sherman Act (codified at 15 U.S.C. § 1)
under “per se” and “rule of reason” theories of liability. (Dkt. No. 1). AngioDynamics seeks
treble damages, a permanent injunction, and declaratory relief. (Id. at 29). Presently before the
Court are: (1) AngioDynamics’ motion for partial summary judgment on liability and antitrust
injury, (Dkt. No. 134); (2) Bard’s motion for summary judgment seeking dismissal of the
complaint, (Dkt. No. 133); and (3) Bard’s motion in limine to preclude the trial testimony of
AngioDynamics’ causation and damages expert, Dr. Alan Frankel, (Dkt. No. 132). The Court
heard oral argument on the motions on April 6, 2021. For the reasons below, both parties’
motions for summary judgment are denied, and Bard’s motion in limine is granted.
II.

FACTS 1
This case centers on AngioDynamics’ claim that Bard’s policy of only selling the

proprietary stylet for its Tip Location System (“TLS”) preloaded into its own peripherally
inserted central catheters (“PICCs”), and refusing to sell its TLS stylet separately for use with its
competitors’ PICCs, constitutes an illegal tie in violation of the Sherman Act. The facts and
evidence relevant to the Court’s resolution of the pending motions are summarized below.
The following facts are drawn from the parties’ statements of undisputed material facts and responses pursuant to
Local Rule 56.1 (formerly Local Rule 7.1(a)(3)), (Dkt. Nos. 133-2, 134-2, 144-1, 146), to the extent those facts are
well-supported by pinpoint citations to the record, as well as the exhibits attached thereto and cited therein to the
extent they could “be presented in a form that would be admissible in evidence” at trial. Fed. R. Civ. P. 56(c)(2). In
considering the parties’ cross-motions for summary judgment, the Court “in each case constru[es] the evidence in the
light most favorable to the non-moving party.” Krauss v. Oxford Health Plans, Inc., 517 F.3d 614, 621-22 (2d Cir.
2008).
1

2

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 3 of 105

A.

Background on PICCs and TLSs
1.

PICC and TLS Technology

Bard and AngioDynamics compete to develop, manufacture, market, and sell vascular
access medical devices, including PICCs, to hospitals and other medical care providers. (Dkt.
No. 133-2, ¶¶ 1-2; Dkt. No. 134-2, ¶ 1; Dkt. No. 144-1, ¶¶ 1-2; Dkt. No. 146, ¶ 1). PICCs are
long, thin, soft, flexible catheters inserted into the body through a vein, most commonly the
basilica vein in the upper arm, and navigated to the distal superior vena cava, the large vein
leading to the right atrium of the heart. (Dkt. No. 133-2, ¶¶ 3-4; Dkt. No. 134-2, ¶ 2; Dkt. No.
144-1, ¶¶ 3-4; Dkt. No. 146, ¶ 2). Clinicians use PICCs to deliver medications, fluids, and
nutrients into a patient’s body, sample blood, and power-inject contrast media. (Dkt. No. 133-2, ¶
3; Dkt. No. 134-2, ¶ 3; Dkt. No. 144-1, ¶ 3; Dkt. No. 146, ¶ 3). PICCs are generally suited for
patients requiring long-term intravenous medical treatment. (Dkt. No. 134-2, ¶ 3; Dkt. No. 146, ¶
3). PICCs can be placed either at a patient’s bedside by a nurse or in an interventional radiology
(“IR”) suite, usually by a physician. (Dkt. No. 134-2, ¶ 14; Dkt. No. 146, ¶ 14).
During placement of a PICC, clinicians often use a guidewire (also known as a “stylet”)
inside the PICC to stiffen it so that it can be threaded through the patient’s veins. (Dkt. No. 1332, ¶ 5; Dkt. No. 144-1, ¶ 5). After completing the PICC placement procedure, the clinician will
remove the stylet from the PICC and discard it. (Dkt. No. 133-2, ¶ 5; Dkt. No. 144-1, ¶ 5).
Because there are several places where a patient’s veins branch before reaching the superior vena
cava, clinicians sometimes route PICCs incorrectly. (Dkt. No. 133-2, ¶ 6; Dkt. No. 144-1, ¶ 6). In
addition, sometimes clinicians get the final placement incorrect. (Dkt. No. 133-2, ¶ 7; Dkt. No.
144-1, ¶ 7). Historically, clinicians used a chest x-ray or fluoroscopy (a medical imaging
technique that uses x-rays) to confirm that a PICCs’ final placement was correct. (Dkt. No. 1332, ¶¶ 7, 12; Dkt. No. 134-2, ¶ 10; Dkt. No. 144-1, ¶¶ 7, 12; Dkt. No. 146, ¶ 10).
3

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 4 of 105

To assist with the PICC navigation process and minimize the complications associated
with incorrect PICC placement, certain companies developed TLSs. (Dkt. No. 133-2, ¶ 8; Dkt.
No. 144-1, ¶ 8). TLSs can offer two key functions: pinpointing the location of the stylet as it
moves through the body (“navigation”) and confirming the PICC’s location once it has been
placed (“confirmation”). (Dkt. No. 133-2, ¶ 8; Dkt. No. 144-1, ¶ 8). A TLS may feature
navigation functionality, confirmation functionality, or both. (Dkt. No. 133-2, ¶ 9; Dkt. No. 1441, ¶ 9). Navigation technology uses magnetic tracking or Doppler technology to provide
information regarding directionality of the PICC as it moves through the patient’s veins, assisting
clinicians in threading the PICC. (Dkt. No. 133-2, ¶ 10; Dkt. No. 134-2, ¶ 13; Dkt. No. 144-1, ¶
10; Dkt. No. 146, ¶ 13). Confirmation technology enables a clinician to confirm the final location
of the PICC within the superior vena cava using a patient’s electrocardiographic (“ECG”)
waveform. (Dkt. No. 133-2, ¶ 11; Dkt. No. 134-2, ¶ 11; Dkt. No. 144-1, ¶ 11; Dkt. No. 146, ¶
11).
While many clinicians now use TLSs to place PICCs because doing so is less expensive,
less time consuming, and more accurate than placing PICCs without TLSs, not all clinicians use
TLSs. (Dkt. No. 133-2, ¶ 12; Dkt. No. 134-2, ¶ 10; Dkt. No. 144-1, ¶ 12; Dkt. No. 146, ¶ 10).
For example, physicians placing PICCs in an IR suite still typically use fluoroscopy, rather than
a TLS, to confirm PICC placement. (Dkt. No. 133-2, ¶ 12; Dkt. No. 134-2, ¶ 16; Dkt. No. 144-1,
¶ 12; Dkt. No. 146, ¶ 16). There are also some hospitals in which nurses continue to place PICCs
without navigation assistance or use chest x-rays rather than TLSs for confirmation. (Dkt. No.
133-2, ¶ 12; Dkt. No. 134-2, ¶ 17; Dkt. No. 144-1, ¶ 12; Dkt. No. 146, ¶ 17). However, a
majority of PICCs placed by nurses at a patient’s bedside use TLSs with navigation capabilities.
(Dkt. No. 133-2, ¶ 12; Dkt. No. 134-2, ¶ 17; Dkt. No. 144-1, ¶ 12; Dkt. No. 146, ¶ 17).

4

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 5 of 105

2.

PICC Purchasing Decisions

PICCs differ from each other in a variety of ways, including with respect to the material
they are made from, the number of lumens (tubes or channels), the outside diameter, whether the
PICC is valved (which can help prevent the backflow of blood into the PICC) or non-valved, and
whether the PICC is preloaded with a TLS stylet. (Dkt. No. 133-2, ¶ 13; Dkt. No. 144-1, ¶ 13).
Manufacturers typically sell PICCs in a kit that also contains various accessories, which vary
depending on, among other things, whether the PICC will be placed by a nurse at the patient’s
bedside or by a physician in an IR suite. (Dkt. No. 133-2, ¶¶ 14-15; Dkt. No. 144-1, ¶¶ 14-15).
The process by which hospitals decide whether to purchase a particular manufacturer’s
PICC is “complex,” “not monolithic” and “varies by hospital.” (Dkt. No. 133-2, ¶¶ 15-17; Dkt.
No. 144-1, ¶¶ 15-17). Depending on the hospital, various constituencies may be involved in the
purchasing decision, including doctors, nurses and representatives from the supply chain, risk
management and infection control departments. (Dkt. No. 133-2, ¶ 15). In some cases, hospitals
have “value analysis committees,” or “VACs,” which play a role in the procurement process and
consist of representatives from various hospital departments. (Id.). Some hospitals are part of
Group Purchasing Organizations (“GPOs”) or Integrated Delivery Networks (“IDNs”), which
negotiate pricing for their member hospitals and have played an increasingly larger role in
influencing the purchasing decisions of their member hospitals. (Dkt. No. 133-2, ¶ 16).
Hospitals consider a variety of factors when deciding whether to purchase PICCs, or
particular types of PICCs, from a given supplier, including, among other things, price, quality of
the PICCs, clinical outcomes, safety, PICC functionality and features (including whether they are
preloaded with a TLS stylet or not, whether they are valved or not, and whether they have a
flare-tip or small diameter, among other factors), the components of the kit in which the PICCs
come (or the potential kit options), the breadth of a manufacturer’s product portfolio, the benefits
5

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 6 of 105

of dealing with a single vendor for multiple products, the manufacturer’s clinical team and
training and customer support, and the requirements of GPO and IDN contracts. (Dkt. No. 133-2,
¶ 17). Different hospitals may weigh and prioritize these factors differently, depending on their
particular needs and concerns. (Dkt. No. 144-1, ¶¶ 15-17).
B.

Relevant Competitors and their Key Products

Several companies compete in the sale of PICCs. The three largest competitors, in terms
of market share, are Bard, AngioDynamics, and Teleflex Incorporated (“Teleflex”). (Dkt. No.
133-2, ¶ 21; Dkt. No. 144-1, ¶ 21). These competitors’ products that are most relevant to this
litigation are described below.
1.

AngioDynamics’ BioFlo PICCs

In 2012, AngioDynamics began selling PICCs known as BioFlo PICCs after acquiring
the technology from another company. (Dkt. No. 133-2, ¶ 33; Dkt. No. 144-1, ¶ 33). BioFlo
PICCs are the first PICCs in the market to use Endexo Technology, a technology designed to
reduce thrombus accumulation (blood clots). (Dkt. No. 134-2, ¶ 4; Dkt. No. 146, ¶ 6). Thrombus
accumulation is a common problem associated with PICC placement that can lead to serious
complications such as deep vein thrombosis (“DVT”) and pulmonary embolism (“PE”). (Dkt.
No. 134-2, ¶¶ 5-6; Dkt. No. 146, ¶¶ 5-6). The parties dispute whether the clinical evidence and
published literature establishes that BioFlo is actually effective at reducing thrombus
accumulation and its resulting complications, with both parties citing record evidence in support
of their respective positions. (Dkt. No. 133-2, ¶¶ 34-36; Dkt. No. 134-2, ¶¶ 4, 7; Dkt. No. 144-1,
¶¶ 34-36; Dkt. No. 146, ¶¶ 4, 7).
Bard has spent significant time, money and energy working to develop a PICC with an
anti-thrombus coating, including with Endexo Technology, but has been unsuccessful. (Dkt. No.
134-2, ¶ 8; Dkt. No. 146, ¶ 8). Bard claims, however, that its FT PICC, which has a taper with a
6

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 7 of 105

smaller diameter at the axillary arch in the arm vein where the majority of DVTs occur, has
proven effective at reducing DVT rates. (Dkt. No. 146, ¶ 8).
2.

Bard’s TLS

Bard sells PICCs under several brand names, has been selling PICCs for decades, and
sells the majority of PICCs in the U.S. (Dkt. No. 133-2, ¶ 22; Dkt. No. 144-1, ¶ 22). In 2006,
Bard became the first company to develop a TLS with navigation technology, which it called the
Sherlock tip navigation system. (Dkt. No. 133-2, ¶ 24; Dkt. No. 144-1, ¶ 24). In 2011, Bard
received FDA approval for a tip confirmation system called Sapiens. (Dkt. No. 133-2, ¶ 24; Dkt.
No. 144-1, ¶ 24). Then, in 2012, Bard launched the Sherlock 3CG TLS, which included both
navigation and confirmation capabilities. (Dkt. No. 133-2, ¶ 24; Dkt. No. 144-1, ¶ 24). Bard’s
Sherlock 3CG system, which only works with Bard’s special, patented proprietary stylet, is the
first and only TLS to combine three technologies: (i) ultrasound technology to identify a suitable
vein for inserting the PICC; 2 (ii) magnetic tracking navigation technology to monitor and guide
the PICC through the venous system; and (iii) ECG technology to confirm the final location of
the PICC’s tip in the superior vena cava. (Dkt. No. 133-2, ¶¶ 25-26; Dkt. No. 134-2, ¶ 21; Dkt.
No. 144-1, ¶¶ 25-26; Dkt. No. 146, ¶ 21).
Bard only sells its proprietary stylet pre-loaded into its PICCs; it does not sell the stylet
“single sterile,” i.e. as a standalone product that can be loaded into a different company’s PICCs
by a nurse or other medical professional at the patient’s bedside. (Dkt. No. 133-2, ¶¶ 26-27, 30;
Dkt. No. 134-2, ¶¶ 22-23; Dkt. No. 144-1, ¶¶ 26-27, 30; Dkt. No. 146, ¶¶ 22-23). As a result, a
customer who wishes to use Bard’s TLS can only do so by purchasing the stylet preloaded into a

While Bard’s TLS system combines ultrasound with navigation and confirmation technology, “ultrasound is
universally available,” “there are many [ultrasound systems] in the market,” and a hospital can use any ultrasound
system, including Bard’s or another company’s, in conjunction with Bard’s TLS. (Dkt. No. 146, ¶ 21).
2

7

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 8 of 105

Bard PICC; there is no option to purchase a Bard TLS separately and combine it with a
competitor’s PICC. (Dkt. No. 133-2, ¶ 30; Dkt. No. 134-2, ¶ 22; Dkt. No. 144-1, ¶ 30; Dkt. No.
146, ¶ 22). To date, Bard has only made a single exception to its TLS policy: sales of a
standalone stylet for a limited time to Cleveland Clinic, the circumstances of which are discussed
further below in Section II.E.5.a infra. (Dkt. No. 133-2, ¶ 30; Dkt. No. 134-2, ¶ 23; Dkt. No.
144-1, ¶ 30; Dkt. No. 146, ¶ 23). While Bard does not sell its TLS stylets separately from its
PICCs, it does sell its PICCs separately from its stylets and acknowledges that certain customers
combine its PICCs with other companies’ TLSs, though it does not expressly “condone the use
of an alternative stylet with [its PICCs].” (Dkt. No. 134-2, ¶ 24; Dkt. No. 146, ¶ 24).
Bard executives’ stated justification for its TLS policy is that loading Bard’s TLS stylet
into a PICC at a patient’s bedside increases the risk of stylet breakage, serious patient injury,
contamination of the sterile field and resulting patient infection. (Dkt. No. 133-2, ¶ 28).
However, AngioDynamics challenges this justification, citing to analyses by several of its
experts disputing that there is a clinical benefit to preloading a TLS stylet or an increased risk
from loading the stylet bedside. (Dkt. No. 144-1, ¶ 28). Bard also claims that it designed its TLS
stylet to be preloaded into its PICCs by trained operators in its manufacturing facilities and not
by nurses at a patient’s bedside, but AngioDynamics disputes this, noting that Bard has FDA
approval to sell its TLS stylet single-sterile for bedside insertion, which it obtained in connection
with its decision to sell Cleveland Clinic a single-sterile stylet. (Dkt. No. 133-2, ¶ 29; Dkt. No.
144-1, ¶ 29).
According to AngioDynamics, aside from Bard’s sales of a single-sterile stylet to
Cleveland Clinic, Bard “took steps to hide the fact that it had FDA approval to sell its stylet
single sterile and attempted to block customers from ordering the stylet single sterile”; Bard

8

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 9 of 105

disagrees with this characterization, and justifies its actions by claiming that Bard “did not want
to release false information to the market by suggesting that the product was available to the
market broadly when in fact it was not.” (Dkt. No. 133-2, ¶ 39; Dkt. No. 146, ¶ 39).
AngioDynamics cites to the following record evidence:
•

A Bard executive wrote to team members, “Fucking A. Keep on down low!” when
referring to Bard’s application to the FDA to sell its TLS stylet as a standalone product.
(Dkt. No. 138-12, at 2).

•

A Bard vice president told others at Bard that Bard’s decision to make the standalone
stylet available to Cleveland Clinic was “top secret[.]” (Dkt. No. 138-13, at 2).

•

A Bard product manager instructed Bard employees to keep any product information
about the standalone TLS stylet “private as much as possible,” explaining that “[w]e
don’t want to publicize that we offer a standalone Sherlock 3CG option to the masses.”
(Dkt. No. 138-14, at 3).

•

When Bard’s corporate representative was asked why Bard did not want to publicize the
standalone option to the masses, he testified that Bard “would not have wanted to create
demand for it.” (Dkt. No. 138-15, at 8). He also testified that Bard did not want to
publicize the “one-off experiment with the Cleveland Clinic” because “the only reason it
worked for Cleveland was because we fully understood that risk and we had a complete
connectivity with the clinical management in a hospital that satisfied us we could onboard this safely,” and that Bard did not want to “promot[e] something when it was a
limited market release.” (Dkt. No. 147-16, at 22-23).

•

The ordering information for Bard’s standalone TLS stylet was not available on Bard’s
computer ordering system. (Dkt. No. 138-2, at 4).

•

A Bard PICC Product Manager instructed that the product code for Bard’s standalone
TLS stylet “should remain ‘invisible’ to all of our customers” except for Cleveland
Clinic and “should not be published in any pricing catalogs, online, etc.” (Dkt. No. 13818, at 2).
3.

AngioDynamics’ Attempts to Acquire a TLS

Since at least 2011, AngioDynamics has attempted to develop or acquire a TLS. (Dkt.
No. 133-2, ¶ 48; Dkt. No. 134-2, ¶ 52; Dkt. No. 144-1, ¶ 48; Dkt. No. 146, ¶ 52). Since then,
there have been times during which AngioDynamics offered a TLS. (Dkt. No. 133-2, ¶¶ 49-50;
Dkt. No. 144-1, ¶¶ 49-50). For example, from 2013 through 2016, pursuant to a license from
9

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 10 of 105

another company, AngioDynamics sold a confirmation-only TLS called Celerity, for which it
attempted to develop a navigation component. (Dkt. No. 133-2, ¶ 51; Dkt. No. 134-2, ¶ 53; Dkt.
No. 144-1, ¶ 51; Dkt. No. 146, ¶ 53). However, in January 2016, when the Celerity technology
was up for sale, Teleflex outbid AngioDynamics and acquired the technology, at which point
AngioDynamics ceased its efforts to obtain FDA approval for navigation-enabled Celerity and
discontinued selling Celerity aside from a small number of units from its remaining inventory.
(Dkt. No. 133-2, ¶ 52; Dkt. No. 134-2, ¶ 54; Dkt. No. 144-1, ¶ 52; Dkt. No. 146, ¶ 54). 3 In
addition to Celerity, AngioDynamics has offered other TLSs at times in the past, but it has never
offered a TLS with both navigation and confirmation capabilities. (Dkt. No. 133-2, ¶¶ 49-50;
Dkt. No. 144-1, ¶¶ 49-50). In late 2019, AngioDynamics acquired a TLS called C3 Wave, which
has confirmation, but not navigation, capabilities. (Dkt. No. 133-2, ¶ 56; Dkt. No. 144-1, ¶ 56).
Several AngioDynamics executives have publicly stated their belief that C3 Wave will help
AngioDynamics compete in the PICC market, though deposition testimony from
AngioDynamics’ General Manager of Vascular Access clarified that these statements refer to the
more limited market for PICCs that use confirmation-only TLS technology, not the much larger
market for PICCs that use navigation-enabled TLSs like Bard’s. (Dkt. No. 133-2, ¶¶ 58-59; Dkt.
No. 144-1, ¶¶ 58-59).
4.

Teleflex’s PICC and TLS Technology

Like Bard and AngioDynamics, Teleflex markets and sells PICCs. (Dkt. No. 133-2, ¶ 63;
Dkt. No. 144-1, ¶ 63). Like Bard, but unlike AngioDynamics, it also markets and sells TLSs that

In January 2017, Teleflex received FDA approval for a TLS that was based off the same technology that
AngioDynamics would have used to develop Celerity with navigation. (Dkt. No. 144-1, ¶ 52). Several AngioDynamics
executives have testified as to their frustration and disappointment that Teleflex outbid AngioDynamics for the
Celerity technology, and their belief that navigation-enabled Celerity would have helped AngioDynamics compete in
the PICC market. (Dkt. No. 133-2, ¶¶ 53-54; Dkt. No. 144-1, ¶¶ 53-54).

3

10

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 11 of 105

provide both navigation and confirmation capabilities. (Dkt. No. 133-2, ¶ 63; Dkt. No. 144-1, ¶
63). Unlike Bard, Teleflex historically has sold, and continues to sell, its TLSs both preloaded
into its PICCs and as separate, standalone products that are compatible with third-party PICCs.
(Dkt. No. 133-2, ¶ 64; Dkt. No. 134-2, ¶ 42; Dkt. No. 144-1, ¶ 64; Dkt. No. 146, ¶ 42).
Therefore, hospitals that want to pair AngioDynamics’ BioFlo PICCs with a navigation-enabled
TLS are able to do so with Teleflex’s TLSs. (Dkt. No. 133-2, ¶ 65; Dkt. No. 144-1, ¶ 65).
AngioDynamics has, at times, marketed the option of pairing BioFlo with Teleflex’s TLS
technology to its customers, and some hospitals do utilize this option. (Dkt. No. 133-2, ¶¶ 66-67;
Dkt. No. 144-1, ¶¶ 66-67).
C.

Relevant Markets and the Competitors’ Market Shares

The parties dispute the relevant product market for purposes of the Court’s antitrust
analysis. The parties appear to agree that PICCs constitute a distinct market from other types of
vascular access devices; that all PICCs compete in the same market, regardless of the various
distinctions between particular PICCs; and that the relevant geographic market is the United
States. (Dkt. No. 144-1, ¶ 13; Dkt. No. 136-9, at 26-27, 36-37; Dkt. No. 147-13, at 13-14, 16).
However, Bard contends that the only relevant product market is the market for “differentiated
PICCs”—i.e. PICCs that differ from each other on a number of characteristics such as size,
material, and others—and that Bard’s preloaded TLS stylet is merely one way in which Bard’s
PICCs are differentiated, rather than a distinct product with its own market. (Dkt. No. 145, at 715, 17-19). AngioDynamics contends that PICCs and TLSs comprise two separate, distinct
product markets. (Dkt. No. 134-1, at 14-19; Dkt. No. 152, at 7-9, 10-11).
Bard has historically captured the majority of the PICC market, with AngioDynamics and
Teleflex capturing significantly smaller market shares. (Dkt. No. 133-2, ¶ 21; Dkt. No. 144-1, ¶
21; Dkt. No. 138-78, at 2-3).
11

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 12 of 105

(Dkt. No. 133-2, ¶ 21; Dkt. No. 144-1, ¶ 21).
PICCs used with TLSs constitute by far the largest segment of the PICC market. (Dkt.
No. 133-2, ¶ 57; Dkt. No. 146, ¶ 57). The evidence cited by AngioDynamics suggests that,
between 2013 and 2018, Bard sold the overwhelming majority of TLSs on the market—more
than

percent—and Bard has not offered evidence contesting that assertion. (Dkt. No. 134-2, ¶

47; Dkt. No. 146, ¶ 47). Teleflex, the only other seller of navigation-enabled TLS stylets, sold a
much smaller percentage of TLSs—only

percent—over the same period. (Dkt. No. 134-2, ¶

49; Dkt. No. 146, ¶ 49). Other TLSs sold today do not include navigation components and make
up a negligible percentage of TLS sales. (Dkt. No. 138-20, at 25 n.107). Because Bard, which
sells the majority of the TLSs, does not sell its TLS stylets single-sterile as a matter of policy,
only a small percentage of all TLS stylets on the market are sold single-sterile. (Dkt. No. 146, ¶
43). However, the percentage of stylets sold single-sterile by Teleflex, Bard’s only meaningful
competitor in the TLS space, is significant: between 2013 and 2018,

of Teleflex’s

TLS stylets were sold separately from its PICCs. (Dkt. No. 134-2, ¶ 43; Dkt. No. 146, ¶ 43).
The parties dispute whether, assuming TLS stylets comprise a separate market from the
PICC market, there are high barriers to entry in the TLS market that enable Bard to sustain such
a high market share. AngioDynamics cites “high development costs, the requirement for FDA
clearance of medical devices, and Bard’s ownership of patented TLS technology” as such
barriers to entry. (Dkt. No. 134-2, ¶ 50). The evidence cited by the parties on each of these issues
may be summarized as follows:
•

High Development Costs: The parties agree that both AngioDynamics and Bard have
spent significant time, effort and money developing navigation-based TLSs (Bard

12

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 13 of 105

successfully, and AngioDynamics unsuccessfully). (Dkt. No. 134-2, ¶¶ 51-52; Dkt. No.
146, ¶¶ 51-52). Bard nonetheless contests AngioDynamics’ assertion that high
development costs pose a significant barrier to entry, noting that AngioDynamics’ own
expert has identified seven different competing TLS products that have been developed
by Bard’s rivals in the past. (Dkt. No.146, ¶ 50).
•

FDA Approval: AngioDynamics cites the need for FDA approval of medical products
as a barrier to entry for developing a TLS with navigation. (Dkt. No. 134-2, ¶ 50). Bard
points out, however, that AngioDynamics’ own expert economist opined in his
deposition that FDA clearance is “not [an] insurmountable” barrier, and that “[i]f you
have a successful technology . . . simply getting the FDA approval shouldn’t be that
difficult.” (Dkt. No. 146, ¶ 50).

•

Patents: The parties agree that Bard has multiple patents on its TLS stylets. (Dkt. No.
134-2, ¶ 55; Dkt. No. 146, ¶ 55). AngioDynamics’ liability expert, Professor George
Alan Hay, opined that entry into the alleged TLS market is “complicated by the fact that
both Teleflex’s and Bard’s navigation systems are patented,” as this patent protection
requires an entrant into the TLS market to “invent around” these companies’ existing
intellectual property, and “entry with a similar technology could lead to costly patent
litigation, even if there is no infringement.” (Dkt. No. 134-2, ¶ 56; Dkt. No. 146, ¶ 56).
During his deposition, however, Dr. Hay admitted that he “ha[s] not studied the patents,”
does not know “which patents Bard has on tip location,” and could only make general
assumptions about “which specific technological features of the Bard tip location system
are covered by patents.” (Dkt. No. 146, ¶¶ 50, 56). As noted above, Dr. Hay also
identified seven different TLS products developed by Bard’s competitors,
notwithstanding Bard’s patent protection. (Dkt. No. 146, ¶ 56). Bard’s corporate
representative testified that “[a competitor] could easily have reverse engineered [Bard’s
3CG] stylet, and they would have to change an approach to creating a magnetic field to
avoid our intellectual property. But they could do that and . . . that has been done.” (Dkt.
No. 138-26, at 4).
D.

General Explanations for AngioDynamics’ Alleged Lost Sales

Citing record evidence, Bard contends that AngioDynamics has lost existing PICC
business or been unable to win new PICC business because of a variety of factors other than
Bard’s TLS policy, including “price, contracts, quality issues, product recalls, competing
technologies, incomplete product lines, lack of data, customer relationships, and product
backorders.” (Dkt. No. 133-2, ¶ 38). AngioDynamics does not appear to meaningfully dispute
that these factors have impacted its PICC business, but it contends that Bard’s TLS policy is at
least one materially contributing factor to its loss of PICC sales to Bard and/or its failure to win

13

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 14 of 105

PICC business from Bard. (Dkt. No. 144-1, ¶ 38). Some of the most significant factors Bard
discusses with respect to the pending motions are discussed below.
1.

Pricing

AngioDynamics’ BioFlo PICCs are sometimes priced at a premium to other PICCs in the
market at the point of sale. (Dkt. No. 133-2, ¶ 39; Dkt. No. 144-1, ¶ 39). According to Tom
Aldrich, a former sales executive at AngioDynamics, AngioDynamics’ former CEO believed
that “customers should pay more for BioFlo than the other PICCs on the market.” (Dkt. No. 1332, ¶ 40; Dkt. No. 144-1, ¶ 40). AngioDynamics also believes that any price premium on BioFlo
PICCs is justified by evidence that BioFlo PICCs’ thrombus-accumulation-reducing properties
result in long-term cost savings to the hospitals using them. (Dkt. No. 144-1, ¶¶ 39, 41-42).
Nonetheless, AngioDynamics concedes that some hospitals are unwilling to pay a premium for
BioFlo PICCs. (Dkt. No. 133-2, ¶ 43; Dkt. No. 144-1, ¶ 43).
AngioDynamics does not dispute that, as a result of this price premium, the combination
of BioFlo PICCs with Teleflex’s TLS stylets is “sometimes” more expensive than Bard’s PICCs
preloaded with its stylets. (Dkt. No. 133-2, ¶ 41; Dkt. No. 144-1, ¶ 41). However,
AngioDynamics asserts that because AngioDynamics competes with Bard and other competitors
on price, its pricing of BioFlo is specific to each customer, and that therefore, its pricing of
BioFlo is sometimes higher and sometimes lower than Bard’s pricing of its PICCs. (Dkt. No.
144-1, ¶ 41; see also Dkt. No. 138-61, at 2).
2.

Quality/Reputational Issues

AngioDynamics has suffered from “multiple quality issues across multiple product lines
and areas.” (Dkt. No. 133-2, ¶ 45; Dkt. No. 144-1, ¶ 45). As relevant to its PICC business, in
2015, AngioDynamics recalled its entire line of Morpheus PICCs due to quality issues. (Dkt. No.
133-2, ¶ 46; Dkt. No. 144-1, ¶ 46). There is evidence in the record suggesting that the Morpheus
14

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 15 of 105

recall damaged AngioDynamics’ relationship with at least some of its customers, and that these
issues negatively impacted AngioDynamics’ PICC sales. (Dkt. No. 133-2, ¶¶ 46-47). In internal
correspondence, General Manager of Vascular Access Chad Campbell referred to a “Morpheus
hangover”; Campbell testified that this merely referred to the fact that AngioDynamics had some
“slow orders as [it] converted [its customers] to [its] other products,” though he acknowledged
that AngioDynamics also lost business entirely as a result of the Morpheus recall. (Dkt. No. 1332, ¶ 46; Dkt. No. 144-1, ¶ 46; Dkt. No. 144-47, at 4). Unlike AngioDynamics, Bard has never
recalled an entire line of its PICCs. (Dkt. No. 133-2, ¶ 23; Dkt. No. 144-1, ¶ 23).
3.

AngioDynamics’ Lack of a TLS

Because some customers value the convenience of preloaded PICCs and dealing with a
single vendor, and thus prefer to purchase PICCs and TLSs from the same supplier even if they
have the option to purchase them separately, (Dkt. No. 146, ¶¶ 65-66), Bard cites the fact that
AngioDynamics does not offer its own TLS with both navigation and confirmation capabilities
as a factor negatively impacting its PICC sales, independent of Bard’s TLS policy. (Dkt. No.
133-2, ¶¶ 61-62). Bard cites to AngioDynamics’ thus-far unsuccessful efforts to develop its own
navigation-enabled TLS, as well as AngioDynamics’ internal documents reflecting that it
believes its lack of navigation-enabled TLS to be a reason for its loss of PICC market share and
its difficulties competing with Bard’s PICCs. (Dkt. No. 133-2, ¶¶ 48-62).
4.

FDA Approval

Bard also cites regulatory issues as a factor impacting AngioDynamics’ ability to win
sales of PICCs against Bard independent of Bard’s TLS policy. As discussed further below, see
Section E.5.a infra, Bard first obtained FDA approval to sell its TLS stylet single-sterile in
November 2014 in order to satisfy a request by Cleveland Clinic, which wanted to trial Bard’s
TLS with AngioDynamics’ BioFlo PICCs. (Dkt. No. 133-2, ¶¶ 68-71; Dkt. No. 134-2, ¶¶ 28-29;
15

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 16 of 105

Dkt. No. 144-1, ¶¶ 68-71; Dkt. No. 146, ¶¶ 28-29). Bard obtained FDA clearance to sell its stylet
as a standalone product that could be combined with any “open-ended, nonvalved, polyurethane
[PICCs],” including those sold by Bard’s competitors. (Dkt. No. 133-2, ¶¶ 70-71; Dkt. No. 1342, ¶ 29; Dkt. No. 144-1, ¶¶ 70-71; Dkt. No. 146, ¶ 29). While Bard’s purpose for obtaining this
FDA clearance was to accommodate Cleveland Clinic’s request, the clearance it received did not
limit Bard to selling its single-sterile stylet only to Cleveland Clinic. (Dkt. No. 134-2, ¶ 30; Dkt.
No. 146, ¶ 30).
An important limitation on Bard’s FDA clearance for its single-sterile stylet is that it is
only approved to sell its stylet for use with competitors’ non-valved PICCs; Bard has never
sought nor obtained FDA clearance to sell its stylet single-sterile for use with competitors’
valved PICCs. (Dkt. No. 133-2, ¶ 71; Dkt. No. 144-1, ¶ 71). Therefore, a hospital’s use of Bard’s
stylet with AngioDynamics’ valved PICCs would be considered an “off-label use,” i.e. a use that
is not included in the FDA’s approved indication for the product. (Dkt. No. 133-2, ¶ 71; Dkt. No.
144-1, ¶ 71). Bard argues that this constitutes an independent barrier to AngioDynamics’ ability
to compete for customers who prefer valved PICCs, which comprise a significant percentage of
both Bard’s and AngioDynamics’ PICC sales. (Dkt. No. 133-2, ¶¶ 72-74; Dkt. No. 144-1, ¶¶ 7274). AngioDynamics rejects this contention, arguing that AngioDynamics or another competitor
could seek FDA approval to market Bard’s single-sterile stylet to such customers; that, based on
its own testing of Bard’s stylet with valved PICCs, AngioDynamics believes such approval could
be obtained; and that in any event, the law does not prohibit medical professionals from choosing
to use medical products in ways that are “off-label.” (Dkt. No. 144-1, ¶ 71).

16

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 17 of 105

E.

Customer-Specific Evidence of AngioDynamics’ Alleged Lost Sales
1.

Overview

A central dispute between the parties is whether there is evidence that specific customers
would have purchased AngioDynamics’ PICCs for pairing with Bard’s TLS but for Bard’s TLS
policy. Bard concedes that “the topic of whether Bard sells Stylets separately from its PICCs
and/or whether Bard would do so has arisen” in its discussions with 17 hospitals, though it
claims that it has never received a formal Request for Proposal regarding a stand-alone stylet.
(Dkt. No. 133-2, ¶¶ 84, 86; Dkt. No. 134-2, ¶ 40; Dkt. No. 144-1, ¶¶ 84, 86; Dkt. No. 146, ¶ 40). 4
AngioDynamics has its own list of at least 22 institutions that it believes have requested that
Bard sell its stylets on a standalone basis and been refused. (Dkt. No. 133-2, ¶ 83; Dkt. No. 1441, ¶ 83). AngioDynamics also has provided what it contends is a “non-exhaustive” list of 31
customers—including hospitals, IDNs, and GPOs—from which it claims that it “potentially” lost
sales as a result of Bard’s TLS policy. (Dkt. No. 133-2, ¶ 87; Dkt No. 144-1, ¶ 87).
2.

Customer Requests to Bard Regarding Standalone Stylet

AngioDynamics points to a number of documents that reflect specific examples of
customer inquiries to Bard regarding the possibility of purchasing a standalone TLS stylet to pair
with AngioDynamics’ or another competitor’s PICCs, and of Bard preparing for such inquiries.
The Court summarizes the relevant portions of these communications as follows:
•

In February 2015, in an internal email chain, a Bard employee wrote: “We need an
official ‘talk track’ on the following subject so that everyone in Sales/Marketing/Nursing
speaks with one voice on this issue. Not to make a big deal but it actually becomes more
strategic and relevant now that Angio has a wire.” (Dkt. No. 138-3, at 2). The employee
was referring to the following passage from a District Manager’s monthly report: “The
AngioDynamics rep in Cincy is telling accounts ‘[Bard] got the indication to sell their

Bard contends that “[m]ost of the hospital communications to Bard regarding standalone stylets came from nurses,
not personnel making decisions about PICC purchases,” but AngioDynamics disputes this characterization, citing
record evidence that the decision-making process and the personnel involved varies by hospital. (Dkt. No. 133-2, ¶
85; Dkt. No. 144-1, ¶ 85).

4

17

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 18 of 105

wire with our BloFlo PICCs.’ . . . I’ve coached [sales representatives] specifically how to
handle with their customers if it’s brought up to them. In short, our response is, ‘That is
not an option available at this time.’” (Id.).
•

In February 2015, another Bard employee described a phone call from a customer to a
Bard clinical nurse “asking if they could get the 3CG standalone wire for there [sic] at
risk patients so they could use a coated PICC.” (Dkt. No. 138-5, at 2). The employee
went on to write: “[The customer] said they had been told that the Cleveland Clinic is
using a different PICC with 3CG. I will be able to squash this, but I think we will be
hearing a lot of people bringing this up.” (Id.).

•

In March 2015, Bard executives discussed the fact that a customer was “getting pressure .
. . to evaluate bioflo” and “contacted [Bard personnel] about pulling a separate 3CG wire
with the Bioflo PICC line.” (Dkt. No. 138-75, at 2).

•

In March 2015, in an internal email chain, Bard personnel discussed the “talk track . . . so
that [Bard personnel are] armed with the correct verbiage in case someone comes
calling.” (Dkt. No. 138-2, at 3). One Bard employee wrote, “If you receive any inquiries
from other hospitals/customers about [the standalone stylet], please tell them that it is not
currently available for purchase. If they mention Cleveland Clinic specifically . . . you
can tell them that Cleveland Clinic is doing an evaluation, but there are no plans to make
the kit available at other facilities at the current time.” (Id. at 2).

•

In March 2015, a Value Access Coordinator from a customer reached out to a Bard
Territory Manager, writing: “I did contact Supply Chain at Cleveland Clinic. They are
using the Bard 3CG technology and separately ordered/stocked Bard 3CG stylets with 2
DIFFERENT PICC vendors (both NOT bard). So . . . there are facilities getting this
single sterile 3CG stylet? Can you send me pricing on the separately stocked 3CG stylet
for my comparison.” (Dkt. No. 138-76, at 4).

•

In March 2015, a representative of HealthTrust, a GPO, wrote to a Bard Vice President
that “[i]f [Bard’s single-sterile TLS stylet] is being sold to HealthTrust members, as I
have been informed it is, then it needs to be added to contract.” (Dkt. No. 138-7, at 2).

•

In an April 2015 email, a Bard representative described a “very agitated” customer who
“want[ed] something in writing stating [that the stand-alone stylet was unavailable] and
the reason why,” because “she has been told otherwise.” (Dkt. No. 138-4, at 2).

•

In April 2015, a customer contacted a Bard sales representative about “the BARD guide
wire being made available separate from the PICC kit so it can be used to place any
PICC,” stating that the customer “would still like to try the antimicrobial PICC if
possible.” (Dkt. No. 138-77, at 3). In a subsequent internal email, the Bard employee
wrote that “[w]e explained that there is not a coated PICC that comes with 3CG and the
director stated that Cleveland Clinic utilizes our wire with [Teleflex] PICCs. I understand
that this isn’t going to happen, but I wanted to make sure you are aware they are taking
[sic] about it.” (Id. at 2-3). Another Bard employee wrote: “This came up today as well at
18

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 19 of 105

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 20 of 105

4.

Deposition Testimony of AngioDynamics Employees

The parties dispute whether the deposition testimony of AngioDynamics officers and
employees demonstrates that AngioDynamics lost any sales to Bard as a result of Bard’s TLS
policy. That testimony is summarized below.
a.

Tom Aldrich

Tom Aldrich worked for AngioDynamics from 2001 to 2017, and was the head
salesperson in AngioDynamics’ Vascular Access Group (the business segment that includes
PICCs) from 2013 to 2017. (Dkt. No. 136-16, at 5-11). Aldrich testified that he “[knew] that
[AngioDynamics] lost accounts as a result of our PICC line not being compatible with Bard’s tip
location,” but did not know how many, and was only able to name one example of a customer
that “might have been one that was interested in using our PICC line with the Bard tip location
but [was] unable to.” (Id. at 27-31). Aldrich also generally lacked knowledge regarding the
entities on AngioDynamics’ list of customers that it “potentially lost” as a result of Bard’s tying
policy. (Id. at 47-55). Aldrich did recall that AngioDynamics lost business from HealthTrust, a
GPO, because some of the hospitals “wanted a navigation system” and AngioDynamics was “not
able to convert them because they were using Bard,” but he did not recall details about which
hospitals within the GPO this applied to, or which individuals at those hospitals conveyed the
information to him. (Id. at 52-54).
b.

Chad Campbell

Chad Campbell is the General Manager of AngioDynamics’ Vascular Access Group.
(Dkt. No. 136-25, at 3-5). Campbell testified that “the number one reason [AngioDynamics loses
business to Bard] is because customers prefer [Bard’s] navigation and tip location” product, and
are forced to use Bard’s PICCs in order to gain access to that product. (Dkt. No. 136-25, at 1923). However, he also testified that he lacked “firsthand personal information” regarding
20

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 21 of 105

business AngioDynamics lost as a result of Bard’s TLS policy, and that he has not directly
communicated with any hospital regarding their interest in combining AngioDynamics’ BioFlo
PICCs with Bard’s TLS. (Dkt. No. 144-48, at 8). His only knowledge of such lost business came
“through reps, managers, heads of sales, marketing, clinical, my teammates on my team.” (Id.).
Campbell demurred when asked to name specific examples of a customer whose business
AngioDynamics lost as a result of Bard’s TLS Policy, responding that “whatever the team has
produced is who” and that he could not name “anyone specifically that I’ve talked to from a
customer standpoint.” (Id.).
c.

Scott Centea

Scott Centea is AngioDynamics’ Senior Director of Global Marketing. When shown
AngioDynamics’ non-exhaustive list of 31 entities from which it believed it “potentially lost
sales,” Centea identified 21 of them that he knew AngioDynamics lost sales from “due to not
having a tip location system or our PICC being able to work with the Bard tip location system.”
(Dkt. No. 136-30, at 6-8). When asked about the basis for his knowledge of these lost sales, he
testified as follows:
So as a corporate account manager during this time, I was responsible for helping set up
contracts as well as remaining very close to the sales representatives. I was also
responsible for managing some of the sales. And we would have discussions, as well as
. . . quarterly business reviews and progress reports. It’s never fun when we lose an
account. And so they were always typically brought to my attention, or someone within
the organization’s attention, to better understand what we could have done differently to
potentially save the business.
(Id. at 7-8).
Centea testified that AngioDynamics’ sales representatives “were being asked daily, if
[AngioDynamics’] BioFlo could be used with Sherlock and 3CG.” (Dkt. No. 144-37, at 5).
When asked whether he had ever had “an actual conversation where any of these accounts ever
told [him] that they would rather buy [AngioDynamics’] PICC with Bard’s single sterile stylet
21

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 22 of 105

for more than Bard’s PICC with Bard’s Sherlock stylet,” Centea testified that he has “had so
many of those conversations, they all seem to run together.” (Dkt. No. 136-30, at 13-14).
However, because purchasing Bard’s stylet single-sterile to combine with AngioDynamics’
PICCs was not an option for customers, AngioDynamics was usually unable to reach the point
where it could engage in discussions about the price of that option, and so Centea could not
recall a situation in which a customer said that “we will, no matter what the price is, we will buy
their single stylet, because it was never offered.” (Id. at 16).
Centea specifically identified Community Health Systems and Florida Hospital as two
customers with whom he was “pretty confident” that he “personally had a conversation” about
pairing AngioDynamics’ PICCs with Bard’s TLS stylet. (Id. at 14). As to Community Health
Systems, he could not “recall the exact conversation,” but recounted the following details:
[The conversation was with] either Sheila Jarrett or Marsha Hodge . . . I believe it was in
person. We would hold monthly business reviews, updating them on . . . the progress of
the conversion. And I believe Nicolas Abella was there as well . . . I remember Nicolas,
who was their head of clinical or nursing I believe at the time, was saying that, based on
some of the results, that if there was an option to receive the Bard single sterile stylet
separately, that they would definitely take a look at . . . the pricing and run an economic
analysis by knowing that that was what’s going to be best for their accounts.
(Id. at 14-15). When pressed on whether the Community Health Systems representatives told
him that they “would buy the Angio combined with Bard product if it was more expensive than
the Bard product preloaded,” Centea testified that “we never got to those conversations, because
we knew that Bard was never willing to sell the single sterile stylet.” (Id. at 16).
Similarly, as to Florida Hospital, Centea testified that his conversation “was one of those,
well, if they do somehow decide to sell it to us, we . . . would take a look at, again, the BioFlo
and the efficacy of the BioFlo and whatever price Bard was to offer, because we didn’t have a
price. It was really difficult to understand what the economic impact would [sic]. But if they said

22

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 23 of 105

it was reasonable, that it would make sense for them to convert.” (Id. at 16-17). Centea
acknowledged that he “did not have conversations with specifics to the cost and/or price of what
that stylet would be” and therefore did not know the price premium Florida Hospital would be
willing to pay to combine AngioDynamics’ PICCs with Bard’s TLS. (Id. at 17-19).
d.

William Millar

William Millar is AngioDynamics’ Southeast Region Business Manager. Millar testified
that, based on “a general conversation with PICC nurses and [AngioDynamics’] sales team,” he
understood that “there is interest out there for our technology if it can be made available with
[Bard’s] stylet,” and that there were “definitely hypotheses going around in conversations in my
field every day out there of when we’re trying to offer our technologies to our solutions.” (Dkt.
No. 144-38, at 3). As one example, he testified that the Director of Supply Chain at St.
Dominic’s Hospital in Mississippi “had [AngioDynamics] come in to do a presentation of
BioFlo, believed in the technology and we were provided the feedback that . . . we could not
proceed with the trial because [St. Dominic’s had] Bard 3CG in the facility and the nurses prefer
to keep or require that technology to move forward” but that “[i]f it was made available, [St.
Dominic’s] would definitely consider the BioFlo technology but at this point cannot proceed . . .
based on the inability of not being able to marry the Sherlock 3CG system with BioFlo PICCs.”
(Id. at 3-4). Millar also testified that facilities in the Community Health System Family told
AngioDynamics that they would purchase AngioDynamics’ PICCs if they had the option to pair
them with Bard’s 3CG TLS, but could not recall which specific facilities did so. (Dkt. No. 13619, at 16-19).
When asked for additional examples of hospitals that would have purchased
AngioDynamics’ PICCs had Bard made its standalone stylet available, Millar provided the
names of hospitals, which he testified was non-exhaustive, including, among others: JFK,
23

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 24 of 105

Singing River, Biloxi Regional, and Christus Cabrini. (Dkt. No. 144-38, at 5-8). Millar testified
that the information he had about AngioDynamics’ potential lost sales came from his
conversations with other AngioDynamics employees or, in some cases, hospital employees
themselves. (Dkt. No. 133-2, ¶ 96; Dkt. No. 144-1, ¶ 96).
As to JFK, Millar testified that his knowledge that AngioDynamics lost business as a
result of Bard’s TLS policy came from conversations with either JFK’s local clinical resource
director or the AngioDynamics sales representative responsible for JFK; he could not recall
which one. (Dkt. No. 154-4, at 16-17). He testified that AngioDynamics was the “incumbent
supplier” and “the need for navigation tip location was a new initiative for [JFK],” which
“changed” the “business landscape” and caused AngioDynamics to lose its PICC business when
JFK opted to use Bard’s TLS. (Id. at 16). He did not know whether JFK considered Teleflex’s
TLS a viable alternative solution, did not know how AngioDynamics’ pricing compared to
Bard’s, and was not aware of whether the hospital realized any savings by switching to Bard
PICCs, though he recalled that “pricing was not, to my knowledge, a decision-making factor into
it.” (Id. at 13-16).
As to Singing River Hospital, Millar testified that AngioDynamics was unable to win
their PICC business because “they were using Bard’s Sherlock device at the time and they
required that stylet to place PICCs,” and that hospital representatives told him that “if the
technology was available to them to purchase separately . . . in order to buy our PICCs, they
would consider a trial.” (Id. at 27-28). When pressed on whether Singing River had committed to
having a trial of the combined technologies (as opposed to merely considering it), Millar testified
that the option “wasn’t available so we couldn’t go down the pathway of saying they would
because Bard wasn’t providing that opportunity for the customer to choose.” (Id. at 28). He was

24

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 25 of 105

unaware as to whether Singing River ever asked Bard about the possibility of a standalone stylet.
(Id. at 28-29).
As to Biloxi Regional, Millar testified that Biloxi was an existing Bard PICC customer
whose business AngioDynamics was unable to win, and that Biloxi’s Director of Cardiology told
him that if Bard’s standalone stylet became available, Biloxi “would conduct a trial to move to
our business.” (Id. at 29-30). He testified that he had “numerous” conversations with the Director
of Cardiology and several PICC nurses in which they conveyed that “if [the option to combine
the technology] was made available and you could use that system, ‘We would consider your
PICCs.’” (Id. at 37-38). When asked whether he had reason to believe that Teleflex’s TLS would
not be a viable alternative for Biloxi, Millar testified that “[t]hey were already invested into the
3CG system – Sherlock system, and their interest was if they could maintain the – the current use
of the 3CG Sherlock device and incorporate the BioFlo PICC, they would take interest in it.” (Id.
at 36).
As to Christus Cabrini, Millar testified that in the 2014-2016 timeframe, Christus was
“interested in the BioFlo technology” and was “using it . . . on a couple of other platforms we
have,” and that AngioDynamics was “down the pathway of moving toward trial potentially based
off the interest in the BioFlo.” (Id. at 40-43). However, the nurses on the PICC team were
“comfortable” with Bard’s TLS, “were using that as a reliability tool for the Sherlock device, a
navigation component,” and “wanted that Sherlock device to continue to be used” because
“that’s how they were trained”; thus, because Bard’s system could not be combined with
AngioDynamics’ PICCs, AngioDynamics “didn’t have an offering . . . to continue down the
pathway of trialing our PICCs.” (Id.). As a result, although AngioDynamics “had a lot of
meetings with them showing our technology [and] educating them on that” and “they had interest

25

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 26 of 105

[in AngioDynamics’] technology,” “that opportunity didn’t come to fruition because we didn’t
have the Sherlock stylet available to them.” (Id. at 42-43). Millar testified that his knowledge
was based on direct, in-person conversations with the PICC team at Christus Cabrini, though he
could not recall specifically who. (Id. at 43-44).
5.

Other Evidence Regarding Specific Customers

Both parties also cite to documentary evidence—including internal correspondence
among AngioDynamics and Bard employees, as well as correspondence between the parties’
employees and customers—that purportedly demonstrate examples of customers switching from
AngioDynamics’ PICCs to Bard’s PICCs or foregoing AngioDynamics’ PICCs in favor of
Bard’s PICCs, and the reasons for those decisions. The Court summarizes this evidence below.
a.

Cleveland Clinic

To date, Cleveland Clinic is the only customer to whom Bard has offered its TLS stylet
as a stand-alone product. (Dkt. No. 133-2, ¶ 30; Dkt. No. 144-1, ¶ 30). In 2014, Cleveland Clinic
began using AngioDynamics’ BioFlo PICCs in an effort to lower its DVT rates, an effort which
was successful. (Dkt. No. 134-2, ¶¶ 25-26; Dkt. No. 146, ¶¶ 25-26). Later that year, Cleveland
Clinic made a request to Bard to purchase Bard’s TLS stylet single sterile so that it could trial it
with AngioDynamics’ BioFlo PICCs as an alternative to Teleflex’s TLS, which it was using to
place the BioFlo PICCs. (Dkt. No. 133-2, ¶ 68; Dkt. No. 134-2, ¶¶ 27-28; Dkt. No. 144-1, ¶ 68;
Dkt. No. 146, ¶¶ 27-28). At that time, Bard did not have FDA approval to market standalone
Bard TLS stylets for pairing with other companies’ PICCs, but it sought and obtained FDA
approval to do so in order to accommodate Cleveland Clinic’s request. (Dkt. No. 133-2, ¶¶ 7071; Dkt. No. 134-2, ¶ 29; Dkt. No. 144-1, ¶¶ 70-71; Dkt. No. 146, ¶ 29).
Ultimately, Bard agreed to sell the standalone stylet to Cleveland Clinic for $101. (Dkt.
No. 133-2, ¶ 69; Dkt. No. 134-2, ¶ 33; Dkt. No. 144-1, ¶ 69; Dkt. No. 146, ¶ 33). A Bard
26

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 27 of 105

executive testified that Bard felt comfortable offering its standalone stylet to Cleveland Clinic,
despite its general safety concerns regarding bedside loading of the product, because Bard felt
that it “had a complete connectivity with the clinical management in a hospital that satisfied us
we could on-board this safely.” (Dkt. No. 146, ¶ 35 (quoting Dkt. No. 147-16, at 23)). A Bard
internal weekly report indicates that Bard earned a fairly high margin on these sales, though Bard
cites to evidence suggesting that the margin calculation does not include certain relevant costs
associated with a standalone stylet (including liability, training and reputational risk costs), and
that the price at which Bard sold its standalone stylet to the Celveland Clinic was a promotional
price that was discounted from the price at which Bard would be willing to sell the standalone
stylet to the market generally. (Dkt. No. 134-2, ¶ 48; Dkt. No. 146, ¶ 48).
Cleveland Clinic began its trial of Bard’s TLS with AngioDynamics’ BioFlo PICCs in
early 2015, and found that, with the use of Bard’s TLS, it was successfully able to place the
BioFlo PICCs at the patient’s bedside and confirm placement without the need for a chest x-ray.
(Dkt. No. 133-2, ¶ 75; Dkt. No. 134-2, ¶ 36; Dkt. No. 144-1, ¶ 75; Dkt. No. 146, ¶ 36). However,
beginning in October 2015, Cleveland Clinic conducted a three-month trial of Bard’s FT PICC
preloaded with Bard’s TLS stylet, after which it chose to begin using that integrated product,
rather than BioFlo, as its primary PICC product. (Dkt. No. 133-2, ¶¶ 76-77; Dkt. No. 134-2, ¶ 37;
Dkt. No. 144-1, ¶¶ 76-77; Dkt. No. 146, ¶ 37). The parties dispute the reasons for Cleveland
Clinic’s decision, but the evidence they cite suggests that the following factors were relevant: (i)
Bard agreed to sell Cleveland Clinic its integrated product at the same price it had been
previously selling the stand-alone stylet, and thus the integrated product was less expensive than
the combination of a BioFlo PICC and a Bard TLS stylet; (ii) Bard’s PICC kit contained
accessories that Cleveland Clinic’s PICC team preferred; (iii) according to the head of Cleveland

27

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 28 of 105

Clinic’s PICC team, the team observed lower DVT rates using Bard’s FT PICCs than with
BioFlo (though AngioDynamics disputes the credibility and significance of this testimony, see
Dkt. No. 144-1, ¶ 77); (iv) Cleveland Clinic’s PICC team believed that the fact that Bard’s
PICCs were preloaded, which eliminated the need to load the stylet at the beside, reduced the
risk of patient infection, a factor which “weighed heavy in the decision”; and (v) the relationship
between AngioDynamics and Cleveland Clinic may have deteriorated after AngioDynamics
released data concerning the results of Cleveland Clinic’s use of BioFlo prematurely and without
Cleveland Clinic’s permission. (Dkt. No. 133-2, ¶ 77; Dkt. No. 134-2, ¶ 37; Dkt. No. 144-1, ¶
77; Dkt. No. 146, ¶ 37).
b.

University of Colorado

On April 24, 2015, a PICC team manager from the University of Colorado reached out to
Bard to inquire about a standalone TLS stylet. (Dkt. No. 134-2, ¶ 78; Dkt. No. 144-1, ¶ 78).
According to an email from a Bard representative, after Bard indicated a standalone stylet was
not available, the customer became “very agitated” and “want[ed] something in writing stating”
that the stylet was “not available as a standalone” and the reason why, because “she has been told
otherwise.” (Dkt. No. 136-57, at 2). Ultimately, the University of Colorado proceeded to
combine AngioDynamics’ PICCs with Teleflex’s TLS stylets. (Dkt. No. 134-2, ¶ 79; Dkt. No.
144-1, ¶ 79).
In August 2017, the University of Colorado switched from AngioDynamics’ BioFlo
PICCs to Bard’s preloaded PICCs. (Dkt. No. 133-2, ¶ 80; Dkt. No. 144-1, ¶ 80). According to an
email written by an AngioDynamics employee, the decision was “based on contract savings and
a corporate rebate agreement.” (Dkt. No. 136-58, at 4). The email also noted that the “clinical
champion for BioFlo” at the University of Colorado “has been disappointed in the [upper
extremity DVT] results [from BioFlo] for the past few years and her data collection doesn’t
28

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 29 of 105

allow her to argue/justify the contract savings to supply chain anymore.” (Id.). The email further
referenced the fact that “[o]ther [h]ospitals within” the University’s medical system were
“HUGE Bard supporters,” and that “[t]hey were told if University switched then [sic] the other
teams could get their new equipment, if she didn’t then [sic] no funding would be available for
future requests this year.” (Id. at 5).
c.

Torrance Hospital

In August 2017, AngioDynamics lost its PICC business with Torrance Hospital to Bard.
(Dkt. No. 133-2, ¶ 81; Dkt. No. 144-1, ¶ 81). In an email reporting on the loss, an
AngioDynamics employee attributed the loss to “the discontinuance of Navigator [an audio-only
tip navigation system AngioDynamics had offered] and the hospitals [sic] need for a Navigation
piece for” its PICC placement. (Dkt. No. 136-59, at 3). The email explained that the decision
“came down to ICU nurses, Administration and Radiology finding a safe and acceptable system
from Bard, Teleflex or Medcomp that would allow them to place Piccs [sic] in the ICU for their
patients . . . We have tried to influence the decision but in the end they chose Bard.” (Id.). The
email cited, as an additional reason for the change, the fact that Torrance Hospital received “a
tremendous savings of over 15% with the change to Bard,” and noted that the hospital was
“never open to paying any type of premium for [AngioDynamics’ BioFlo PICC] and remained
steadfastly concerned about reducing acquisition costs.” (Id.).
d.

Hillcrest Medical Center

According to a 2013 internal Bard memo, in that year, AngioDynamics won the PICC
business of Hillcrest Medical Center in Tulsa, Oklahoma from Bard “based on [BioFlo PICCs’]
superior results during long evaluation period.” (Dkt. No. 138-31, at 3). The memo noted that, at
the time, Hillcrest did not use tip location to place PICCs, and that such non-tip-location
accounts were “by far [Bard’s] most vulnerable.” (Id.).
29

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 30 of 105

Scott Centea testified that Hillcrest was a customer that had “collected data” and prepared
a “poster presentation” for nursing conferences “show[ing] the efficacy of the BioFlo PICC and
how it reduced . . . DVTs.” (Dkt. No. 144-37, at 12-13). He also testified that he learned from
other AngioDynamics employees that Hillcrest had “inquired [to Bard] about the ability or the
opportunity . . . to potentially purchase a stylet from Bard” for use with AngioDynamics’ PICCs
after AngioDynamics had informed them that this “was an option in the market” based on Bard’s
selling of the stylet to Cleveland Clinic. (Dkt. No. 144-37, at 13-15).
In a February 2015 email, summarizing a three-month BioFlo trial being conducted at
Hillcrest, a Bard employee wrote that Hillcrest had experienced “hemolysis and poor flow rates”
with AngioDynamics’ valved BioFlo PICC and “cracked/leaking hubs” with their non-valved
BioFlo PICCs, and that Hillcrest ultimately “made the decision to stay with Bard Solo after
having multiple issues with hemolysis.” (Dkt. No. 144-29, at 2).
Later in 2015, Hillcrest decided to implement tip location and considered TLSs from
Bard, Teleflex and AngioDynamics; an internal Bard memo speculated that the reason for
evaluating all three vendors’ technology, rather than merely using AngioDynamics’ Celerity TLS
that was available at the time, was dissatisfaction with the BioFlo PICCs’ results. (Dkt. No. 13832, at 3, 5).
In October 2015, the Clinical Resource Management Director at Ardent Health Services
wrote to AngioDynamics representatives: “Following an extensive trial at Hillcrest Medical
Center, to include a multitude of vendor product combinations, we will be standardizing to
BARD PICC Lines and 3CG technology . . . I know you cannot be surprised by the decision, as
currently AngioDynamics does not have similar technology.” (Dkt. No. 144-12, at 2-3). In a
subsequent follow-up email to Hillcrest’s corporate parent, Scott Centea wrote, “This entire

30

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 31 of 105

situation could have been avoided if Bard was willing to sell you their single sterile 3CG stylet. .
. . If they were willing to offer you this product, like they have at other facilities, [then] your
PICC nurses could have used the Sherlock/3CG system with our BioFlo.” (Dkt. No. 138-35, at
2).
In November 2015, a Bard employee wrote, “Big win at Hillcrest to finally win PICCs
back after 2 year[s] of Bioflo. Total growth should be about $700k . . . [T]hey did attest that they
actually had used more Cathflo[, a product used to dissolve thrombus formations in or around the
PICC,] while using Bioflo. They declined to evaluate Celerity, based on the fact that they didn’t
have an algorithm aspect like 3CG Diamon[d] and VPS did.” (Dkt. No. 147-40, at 2).
e.

Christus St. Patrick

In October 2014, Bard’s Louisiana Territory Manager, Shane Beaudean, wrote to other
Bard employees that a “Bio-Flo trial [at Christus St. Patrick] has concluded and the decision has
been made to stick with Bard.” (Dkt. No. 144-14, at 2). When asked for more details regarding
why the hospital chose to stick with Bard, Beaudean responded:
The PICC nurse really disliked Celerity. She said it felt like she was using 10 year old
technology. She was required to screenshot a baseline P-wave, initial deflection P-wave,
and max P-wave for each patient. She was instructed by the rep to do this. She really felt
like this slowed her down and she missed our real-time navigation. On one patient she
achieved max P and the catheter was up the U. She was instructed how to check for
malpositions using our Site-Rite 6 which she really disliked due to sterility issues. The
nurses hated the triple lumen PICC compared to ours. Nurses had problems with blood
draws so much so that the Angio rep showed them “a new technique for drawing blood.”
They had issues with the flow rates of the PICC’s [sic] to the point that the rep talked to
BioMed about changing IV pump settings which they refused to do. At the end of the
day, the PICC nurse felt that Bard’s technology was far superior on all fronts.
(Id. at 2). A November 2014 email from an AngioDynamics territory manager provides a
somewhat different explanation based on his own discussions with Christus St. Patrick
representatives. The email says that the PICC nurse “gave us outstanding reviews on the overall
performance of BioFlo” and “stated the data should reflect a switch to BioFlo,” but that “one of
31

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 32 of 105

the Executives brought up the fact that we do not have Navigation with Celerity” and “stated that
Christus should look into converting once we have Navigation.” (Dkt. No. 138-37, at 6). After a
subsequent meeting, a different AngioDynamics employee confirmed that the PICC nurse’s
“feedback has been terrific regarding the performance of BioFlo,” but that “as expected, [the
nurse] (and Christus as a whole) enjoys the time saver and clinical benefit of eliminating [chest
x-rays]” through TLS technology, and that Christus “advised that when we can eliminate [chest
x-rays], then we can move to the next step with St Patrick as well as with Christus as a whole.”
(Id. at 4).
f.

McLaren Oakland

In a March 2014 memo, a Bard employee wrote: “We have finalized a $120,000 PICC
conversion [with McLaren Oakland] starting May 5th. This conversion will knock Angio D and
Bioflo out of its only account in this system and gives us full compliance with Bard and 3CG.
They are planning on eliminating X-Ray [with Bard TLS] on day one of placements.” (Dkt. No.
144-15, at 3). In a subsequent May 2014 email describing the conversion, a Bard employee wrote
that “[t]he biggest reasons for this conversion are clinical buy-in and McLaren’s Corporate
initiative to eliminate Chest X-Ray.” (Dkt. No. 144-16, at 2). A Bard Regional Manager
responded to that email, saying, “This [Bard] win brings AD’s total lost business, in Michigan
alone, to a staggering $880,000!!! Trust me when I say, Bioflo is a BioFLOP.” (Id.). A
spreadsheet prepared by Bard’s sales team several months later reports that McLaren Oakland
“has had great results going to solo from bioflo.” (Dkt. No. 147-41, at 2).
g.

Summerlin

In September 2015, an AngioDynamics territory manager wrote that the PICC Registered
Nurse at Summerlin was “[o]pen to BioFlo, BUT SKEPTICAL,” and listed concerns about,
among other things, whether BioFlo would “do what we say it will,” BioFlo’s larger lumens size
32

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 33 of 105

compared to the PICCs Summerlin was currently using, and the fact that the TLS
AngioDynamics offered at the time, Celerity, did not yet have a navigation component. (Dkt. No.
147-42, at 3). In October 2015, in connection with an upcoming trial of BioFlo PICCs with
Celerity, a representative of Summerlin hospital wrote to AngioDynamics employees explaining
that the PICC Nurse “has many concerns about the AngioDynamic catheters,” but that “[h]is
main concern is the inability to do tip navigation to prevent malpositions and the lack of some
size catheters he uses.” (Dkt. No. 138-42, at 3).
In his deposition, Scott Centea testified, based on his understanding from another
AngioDynamics employee involved in the conversation, that Summerlin told AngioDynamics
that it was “having difficulties . . . placing our BioFlo PICC without [Bard’s] 3CG system,” and
that he believed Summerlin unsuccessfully requested a standalone stylet from Bard to pair with
the BioFlo PICCs. (Dkt. No. 152-5, at 3-5).
In November 2015, a Bard Sales Trainer and Territory Manager wrote to other Bard
employees describing his takeaways from a discussion with a PICC nurse at Summerlin Hospital
who was “currently evaluating the BioFlo PICC and Celerity.” (Dkt. No. 144-17, at 3). One of
the “main takeaways” was that Bard’s Sherlock TLS “is irreplaceable - by far this is our #1
competitive advantage - Celerity doesn’t have a way to address malpositions.” (Id.). However,
the Summerlin team also expressed various other concerns about BioFlo and Celerity, including
stylet stiffness, lumen sizes, cumbersomeness, and the fact that Celerity “doesn’t come preloaded.” (Id.). Approximately a month later, when Summerlin opted to discontinue the BioFlo
evaluation, the customer provided feedback to Scott Centea in an email reflecting that “[t]he
overriding concern was the lack of a navigation system” which led to “concern about inaccurate
placement as well as the increased time to place and assure correct placement”; however, the

33

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 34 of 105

customer also listed various other issues with the BioFlo/Celerity system, including TLS
equipment failures, cumbersomeness, oozing and cracking of the PICCs (including one PICC
cracking while inside a patient’s body), and concerns about patient safety. (Dkt. No. 144-18, at
4). Reflecting on this feedback, another AngioDynamics employee wrote that, “[f]rom the
sounds of it the biggest factor in them ending their trial mid-term was CELERITY . . . Definitely
further supports the notion that tip location is often the deciding factor in many of our trials, and
more often than not seems to trump the PICC performance and outcomes.” (Id. at 2).
h.

UHS Delaware

In May 2014, in discussing Universal Health Systems of Delaware’s (“UHS Delaware”)
decision to purchase Bard’s PICCs, a Bard National Accounts Manager wrote:
“AngioDynamics/Bioflop was once again turned away as they could not compete against our
clinically superior products. This is a perfect example of why we need all of our customers using
3CG technology. The proprietary technology including 3CG, HF, FT and PowerGlide help
lockdown our business as our competition cannot offer these products.” (Dkt. No. 138-44, at 2).
In December 2015, following a “multi site trial of AngioDynamics BioFlo and Celerity,”
the Value Analysis Program Director at UHS of Delaware, Patricia Tyrrell, wrote to a Bard
National Accounts Manager: “I wanted to let you all know that the decision was made today by
the PICC work group, that we will be staying with Bard. The biggest issue in moving forward
was lack of navigation.” (Dkt. No. 144-19, at 3). After receiving this email, the National Account
Manager forwarded it to other Bard employees, writing: “All of you did a great job protecting
what is yours with

in PICC and Midline spend. Once again the significance of getting our

facilities to 3CG played an important role here.” (Id. at 2).
In subsequent correspondence between Tyrrell and Scott Centea, Centea wrote that “[t]he
initial objective with this evaluation was to compare the merits of BioFlo against a non-BioFLo
34

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 35 of 105

PICC (Bard)” but “[w]hat ended up happening is that our Celerity was being compared against
3CG Sherlock. Unfortunately, the primary objective of the evaluation was silenced rather
quickly.” (Dkt. No. 144-21, at 2). Tyrrell responded: “[W]hen we look [sic] PICCs again, we
will again include Bioflow [sic] knowing that your navigation will be available at that time.”
(Id.).
i.

Baycare

Bard emails reflect that, in 2012, Bard was able to capture “close to $1 million” in new
PICC business from Navilyst (BioFlo’s previous owner) by converting Baycare Health Systems
to Bard’s TLS with “help from dozens of our Clinical Specialists,” which allowed the system to
eliminate the use of chest x-rays for tip confirmation. (Dkt. No. 144-22, at 2). In an August 2013
email describing the conversion, a Bard employee wrote that, during Baycare’s trial of Bard’s
TLS, Bard “strongly stressed the trial was not about [comparing PICC technology], it was about
catheter tip position confirmation and eliminating X-ray,” but that Bard also “just happened to
have a proximal valved Picc as they were used to and if they function equally then no
comparison needed.” (Dkt. No. 144-23, at 2). The email also noted that Baycare “actually loved
the [Bard] kit component upgrades . . . over current Navilyst” kits. (Id.).
Years later, in a January 2017 email from William Millar to Tom Aldrich listing “key
losses and risks for Q2,” Millar listed Baycare as a “$850,000 Loss due to no Tip
Location/Confirmation.” (Dkt. No. 144-20, at 2). In his deposition, Millar expanded on this,
testifying that, based on his
conversations with [Baycare personnel] . . . BayCare was interested in moving to a
navigation tip location system. We conducted a trial head-to-head against Bard, our
midlines versus their midlines, because at the time that was also an interest . . . they had
clinical data outcomes proving that we had better outcomes with the BioFlo midline
versus the Bard device, yet . . . they wanted tip location 3CG and they could not use our
PICC lines with the Bard system, so they made the business decision to move [to] Bard

35

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 36 of 105

PICCs and midlines from AngioDynamics because of the requirement of not being able
to get provided that solution.
(Dkt. No. 138-46, at 3-4). Millar further testified that Baycare switched from AngioDynamics’
midline catheters to Bard’s midline catheters, even though better patient outcomes were observed
with AngioDynamics’ catheters, and that their stated reason was the need for Bard’s TLS stylet
and the inability to pair it with AngioDynamics’ PICCs. (Id. at 5-20).
j.

St. Vincent’s

In an April 2016 email listing updates on PICC opportunities, Daniel Tighe, a Bard
District Manager, wrote: “St. Vincent – Billings – Angio D competitive conversion . . .
Scheduled for June 13th.” (Dkt. No. 144-25, at 2). In a subsequent November 2016 memo, Tighe
wrote to a sales representative: “Excellent job of converting [the St. Vincent’s] account from
Angio D to 3CG in the first quarter. The team seems very satisfied with 3CG.” (Dkt. No. 138-40,
at 2).
k.

UH Cleveland

Internal Bard documents reflect that, in 2015, Bard was able to convert UH Cleveland to
its TLS system, putting Bard “one step closer to OWNING Cleveland with both PICCs and Tip
Confirmation.” (Dkt. No. 144-26, at 3; Dkt. No. 144-27, at 3). Scott Centea testified that, after
experiencing multiple complications with a particular patient using Bard’s PICC lines, UH
Cleveland decided to try a BioFlo PICC (which needed to be placed by a physician in the IR
suite rather than at the patient’s bedside, since it could not be paired with Bard’s TLS), and the
patient had “zero complications” with the BioFlo PICC. (Dkt. No. 144-37, at 6-11). Centea
testified that a UH Cleveland representative informed an AngioDynamics representative that,
following this “fantastic patient success story,” UH Cleveland asked Bard whether it would be
possible to buy Bard’s stylet single-sterile, but Bard “said no,” and therefore UH Cleveland

36

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 37 of 105

continued purchasing Bard’s PICCs with its preloaded stylet rather than AngioDynamics’ BioFlo
PICCs. (Id. at 7-8). Centea could not recall the specific AngioDynamics representative who
relayed this information to him or the specific UH Cleveland representatives who were involved,
but testified that the story was “pretty well-known because of how we worked closely with this
patient,” that the story became “pretty high profile testimony” regarding BioFlo’s benefits, and
that “a lot of people were asking within AngioDynamics, as well as other institutions,” whether
UH Cleveland had switched to BioFlo. (Id. at 10-11).
l.

VA Little Rock

In a May 2014 email, in response to a question regarding whether AngioDynamics had
lost or was going to lose VA Little Rock’s business, Millar responded: “Lost. Bard 3CG was
implemented as tip location/navigation was decided to be the requirement.” (Dkt. No. 144-28, at
2). In February 2015, a Bard employee wrote that VA Little Rock “was an Angiodynamics
account who used Bioflo and Navigator. To eliminate X-ray, they trialed 3CG and the Solo,
changed to Bard with no increase in cathflo and less placement complications.” (Dkt. No. 14429, at 3).
III.

STANDARD OF REVIEW
Under Rule 56(a), summary judgment may be granted only if all the submissions taken

together “show that there is no genuine issue as to any material fact and that the moving party is
entitled to judgment as a matter of law.” Celotex Corp. v. Catrett, 477 U.S. 317, 322 (1986); see
also Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247-48 (1986). The moving party bears the
initial burden of demonstrating “the absence of a genuine issue of material fact.” Celotex, 477
U.S. at 323. A fact is “material” if it “might affect the outcome of the suit under the governing
law,” and is genuinely in dispute “if the evidence is such that a reasonable jury could return a
verdict for the nonmoving party.” Anderson, 477 U.S. at 248; see also Jeffreys v. City of New
37

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 38 of 105

York, 426 F.3d 549, 553 (2d Cir. 2005) (citing Anderson, 477 U.S. at 248). The movant may
meet this burden by showing that the nonmoving party has “fail[ed] to make a showing sufficient
to establish the existence of an element essential to that party’s case, and on which that party will
bear the burden of proof at trial.” Celotex, 477 U.S. at 322; see also Selevan v. N.Y. Thruway
Auth., 711 F.3d 253, 256 (2d Cir. 2013) (explaining that summary judgment is appropriate where
the nonmoving party fails to “‘come forth with evidence sufficient to permit a reasonable juror to
return a verdict in his or her favor on’ an essential element of a claim” (quoting In re Omnicom
Grp., Inc. Sec. Litig., 597 F.3d 501, 509 (2d Cir.2010))).
If the moving party meets this burden, the nonmoving party must “set out specific facts
showing a genuine issue for trial.” Anderson, 477 U.S. at 248, 250; see also Celotex, 477 U.S. at
323-24; Wright v. Goord, 554 F.3d 255, 266 (2d Cir. 2009). “When ruling on a summary
judgment motion, the district court must construe the facts in the light most favorable to the nonmoving party and must resolve all ambiguities and draw all reasonable inferences against the
movant.” Dallas Aerospace, Inc. v. CIS Air Corp., 352 F.3d 775, 780 (2d Cir. 2003). Still, the
nonmoving party “must do more than simply show that there is some metaphysical doubt as to
the material facts,” Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586 (1986),
and cannot rely on “mere speculation or conjecture as to the true nature of the facts to overcome
a motion for summary judgment,” Knight v. U.S. Fire Ins. Co., 804 F.2d 9, 12 (2d Cir. 1986)
(quoting Quarles v. Gen. Motors Corp., 758 F.2d 839, 840 (2d Cir. 1985)). Furthermore, “[m]ere
conclusory allegations or denials . . . cannot by themselves create a genuine issue of material fact
where none would otherwise exist.” Hicks v. Baines, 593 F.3d 159, 166 (2d Cir. 2010) (quoting
Fletcher v. Atex, Inc., 68 F.3d 1451, 1456 (2d Cir. 1995)).

38

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 39 of 105

When considering cross-motions for summary judgment, a court “must evaluate each
party’s motion on its own merits, taking care in each instance to draw all reasonable inferences
against the party whose motion is under consideration.” Hotel Emp. & Rest. Emp. Union, Local
100 of New York, N.Y. & Vicinity v. City of New York Dep’t of Parks & Recreation, 311 F.3d
534, 543 (2d Cir. 2002) (quoting Heublein v. United States, 996 F.2d 1455, 1461 (2d Cir. 1993)
(internal quotation marks omitted)). “In the context of antitrust cases . . . summary judgment is
particularly favored because of the concern that protracted litigation will chill pro-competitive
market forces.” Pepsi Co, Inc. v. Coca-Cola Co., 315 F.3d 101, 104-05 (2d Cir. 2002) (citing
Tops Mkts., Inc. v. Quality Mkts., Inc., 142 F.3d 90, 95 (2d Cir.1998)). “Although all reasonable
inferences will be drawn in favor of the non-movant, those inferences ‘must be reasonable in
light of competing inferences of acceptable conduct.’” Id. (quoting Tops Mkts., 142 F.3d at 95).
IV.

DISCUSSION
AngioDynamics moves for summary judgment on all liability elements of its tying claim,

as well as on the issue of antitrust injury, leaving the amount of damages as the only triable issue.
(Dkt. No. 134). Bard moves for summary judgment seeking dismissal of the complaint, asserting
that AngioDynamics has failed to raise a triable issue of fact as to antitrust injury and damages.
(Dkt. No. 133). In conjunction with its motion for summary judgment, Bard moves to exclude
the testimony of AngioDynamics’ causation and damages expert. (Dkt. No. 132). Bard contends
that if its motion is successful, AngioDynamics would have “no evidence of any damages” with
which it could “create a triable issue as to damages.” (Dkt. No. 133-1, at 36). The Court
addresses each element in turn, and addresses Bard’s motion in limine in connection with its
analysis of damages.

39

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 40 of 105

A.

Liability Elements

“A tying arrangement is ‘an agreement by a party to sell one product but only on the
condition that the buyer also purchases a different (or tied) product, or at least agrees that he will
not purchase that product from any other supplier.’” Smugglers Notch Homeowners’ Ass’n, Inc.
v. Smugglers’ Notch Mgmt. Co., Ltd., 414 F. App’x 372, 374 (2d Cir. 2011) (quoting Eastman
Kodak Co. v. Image Tech. Servs., Inc., 504 U.S. 451, 464 (1992)). To state a tying claim under
the Sherman Act, a plaintiff must prove that: (1) “the sale of one product (the tying product) is
conditioned on the purchase of a separate product (the tied product)”; (2) “the seller uses actual
coercion to force buyers to purchase the tied product”; (3) “the seller has sufficient economic
power in the tying product market to coerce purchasers into buying the tied product”; (4) “the
tie-in has anticompetitive effects in the tied market”; and (5) “a not insubstantial amount of
interstate commerce is involved in the tied market.” Kaufman v. Time Warner, 836 F.3d 137, 141
(2d Cir. 2016) (citing E&L Consulting, Ltd. v. Doman Indus. Ltd., 472 F.3d 23, 31 (2d Cir.
2006)).
AngioDynamics’ complaint alleges that Bard has violated section 1 of the Sherman Act
by unlawfully tying the purchase of Bard TLS stylets to the purchase of Bard PICCs. (Dkt. No.
1, ¶ 95). The “tying products”—the products over which Bard allegedly has market power—are
Bard’s TLSs, and the “tied products”—the products that Bard allegedly coerces buyers to
purchase—are Bard’s PICCs. (Id. ¶¶ 96–96, 104). AngioDynamics argues that based on the
record evidence, there is no genuine dispute of material fact as to any of the five elements. (Dkt.
No. 134).
1.

Separate Products

“The ‘separate product’ element requires that the alleged tying product and tied product
be separate, i.e., they must exist in separate and distinct product markets.” Kaufman, 836 F.3d at
40

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 41 of 105

141. “This is because if there is no separate market for the allegedly tied product, there can be no
fear of leveraging a monopoly in one market to harm competition in a second market.” Id. at 142.
Courts apply a “consumer demand test” to determine whether two products are separate for
antitrust law purposes. Id. Two products are separate only if “there is a sufficient demand for the
purchase of [the tied product] separate from [the tying product] to identify a distinct product
market in which it is efficient to offer [the former] separately from [the latter].” Jefferson Par.
Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 21-22 (1984), abrogated on other grounds by Ill. Tool
Works Inc. v. Indep. Ink, Inc., 547 U.S. 28, 31 (2006). Separateness “turns not on the functional
relation between [the products], but rather on the character of the demand for the two items.” Id.
at 19. Factors relevant to whether there is “separate and distinct consumer demand” for the two
products include, among others, “the history of the products being, or not being, sold separately”
and “the sale of the products separately in similar markets.” Kaufman, 836 F.3d at 142.
AngioDynamics contends that there is no genuine dispute of material fact as to whether
Bard’s PICCs and its TLSs constitute separate products. AngioDynamics cites the following
evidence in support of its position that they are separate products: (1) documentary evidence and
Bard’s own admissions establishing that some customers have inquired about the possibility of
purchasing a standalone TLS stylet, and that one (Cleveland Clinic) actually did for a time; (2)
documentary evidence reflecting Bard’s efforts to avoid publicizing to customers the fact that it
had obtained FDA approval to sell its TLS stylet single-sterile, and was doing so with respect to
Cleveland Clinic, at least partly out of a concern that customers might demand such a product;
and (3) the fact that Teleflex, Bard’s only competitor that sells navigation-enabled TLSs, “has
always sold its TLS Stylets separately from its PICCs and continues to offer this option to

41

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 42 of 105

customers today,” and a significant number of its TLS customers elect that option. (Dkt No. 1341, at 16-19; Dkt. No. 152, at 8).
In arguing that there is a genuine dispute of material fact as to the separate products
element, Bard relies on the analysis of its expert, Professor Fiona Scott Morton, who opines,
among other things, that: (1) “the total demand for a separate 3CG stylet represented by [the
hospitals the parties have identified as having inquired about a standalone stylet] is at most
relatively small”; (2) based on this low demand and the costs of offering a standalone stylet,
“Bard would reasonably conclude that there would likely be little demand for a separate 3CG
stylet at a price that would make it profitable for Bard to sell the stylet separately”; and (3) “[t]he
fact that Teleflex sells a relatively small number of . . . stylets separately is not evidence that
there would be enough demand for a separate 3CG stylet to induce Bard to sell it separately,”
including because Teleflex and Bard Stylets “are based on different technology” and the two
companies “face different costs for selling them separately,” rendering it “not at all unreasonable
or surprising that the two firms would come to different conclusions about selling their stylets
separately.” (Dkt. No. 145, at 8 (quoting Dkt. No. 147-13, at 33-34); id. at 10-11 (quoting Dkt.
No. 147-13, at 31); id. at 13-14 (quoting Dkt. No. 147-13, at 35)). Bard also argues that: (1)
Bard’s witness testimony suggests that it developed its TLS stylets to be used exclusively as
components in Bard’s PICCs and that they were “not intended or designed to be loaded in the
field”; (2) AngioDynamics’ witness testimony establishes that, when it tracks its “market share at
the PICC level,” it does not separately track market share for “disposables” like TLS stylets; (3)
neither “anecdotal” examples of customer inquiries regarding a standalone stylet nor documents
reflecting Bard’s “prepar[ation] for the possibility that a customer might make an inquiry”
establish that there was significant enough demand for Bard’s TLS as a separate product to

42

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 43 of 105

satisfy the consumer demand test, particularly where none of the inquiring individuals submitted
a formal Request for Proposal, discussed the price they would be willing to pay for a standalone
stylet, or had complete decision-making authority on behalf of their hospitals; and (4)
“AngioDynamics’ own economic expert . . . agreed that one must evaluate whether there would
be sufficient demand for Bard’s standalone stylets at a price that covered all the relevant costs of
offering the product,” but “has not quantified the relevant costs Bard would have to cover if it
offered the standalone stylet, nor did he determine at what price demand for standalone stylets
would dry up.” (Dkt. No. 145, at 9-13).
Based on the parties’ arguments and the record before it, the Court finds that Bard has
raised a genuine dispute of material fact sufficient to survive summary judgment with respect to
the separate products element. AngioDynamics is correct that the record evidence of customer
inquiries to Bard regarding a standalone stylet and Bard’s efforts to conceal knowledge of the
standalone stylet in the market, as well as the fact that Bard’s only significant TLS competitor
offers its standalone stylet separately from its PICCs and many of its TLS customers choose to
purchase it that way, constitutes strong evidence that consumer demand exists for TLSs as a
separate product from PICCs. However, Bard has presented expert analysis suggesting that,
whatever demand for Bard’s standalone stylet may have existed in the market, it was not
significant enough to make it “efficient [for Bard] to offer [TLSs] separately from [PICCs],” and
thus satisfy the “consumer demand” test for purposes of antitrust law. See Jefferson Par., 466
U.S. at 21-22. Because, based on the record before the Court, both parties take positions that a
reasonable fact-finder could credit, their disputes are appropriately resolved at trial, not at the
summary judgment stage.

43

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 44 of 105

AngioDynamics relies on Park v. Thomson Corp., No. 05-cv-2931, 2007 WL 119461,
2007 U.S. Dist. LEXIS 2001 (S.D.N.Y. Jan. 11, 2007) and In re Visa Check/Mastermoney
Antitrust Litigation, No. 96-cv-5328, 2003 WL 1712568, 2003 U.S. Dist. LEXIS 4965 (E.D.N.Y.
Apr. 1, 2003) for the proposition that “[w]here the items have been sold separately, distinct
products exist,” and argues that under this case law, the fact that Teleflex sells its TLS separately
from its PICCs establishes the existence of separate products. (Dkt. No. 134-1, at 14-15; Dkt.
No. 152, at 7-8). However, the Court does not read either case to suggest that Teleflex’s sale of a
standalone TLS establishes the existence of separate products as a matter of law. In Park, the
Court found that BAR/BRI’s Multistate Bar Examination (“MBE”) and state-specific bar review
courses constituted separate products by relying on the fact that “numerous sellers offer an MBEonly or specific-state course to buyers every year,” and granted summary judgment to the
plaintiff on the separate products element. Park, 2007 WL 119461, at *3, 2007 U.S. Dist. LEXIS
2001, at *9-10. The In re Visa court found that “[o]verwhelming evidence establishes that
merchant demand for credit card services is distinct from merchant demand for debit card
services,” including that “those services are sold separately; many merchants would refuse to use
off-line debit services if given the choice to do so; and the defendants themselves have
repeatedly acknowledged in their business strategy and marketing activities the distinctive
attributes of their off-line debit services compared to their credit card services.” In re Visa, 2003
WL 1712568 at *2, 2003 U.S. Dist. LEXIS 4965, at *8-9.
The evidence here is not so clear-cut. It establishes that there are only two competitors
who sell navigation-enabled TLSs—one that sells its TLS stylet as a standalone product and one
that does not. It also suggests that there are genuine technological distinctions between the two
companies’ products that may or may not affect the price at which they would need to offer their

44

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 45 of 105

respective stylets single-sterile in order to cover their associated costs, and consequently, the
level of consumer demand for each of them as standalone products when sold at that price. (Dkt.
No. 136-6, at 33-34 (describing technological differences between Teleflex and Bard stylets);
Dkt. No. 147-44, ¶¶ 4-10 (same); Dkt. No. 147-13, at 35 (opining that technological differences
would impact the companies’ costs of selling their respective TLS stylets single-sterile, the price
at which those single-sterile stylets would be offered, and the resulting consumer demand)). 5
While the Court agrees with the In re Visa court that “[t]he proper question is not whether it was
more efficient for [Bard] to offer [its PICCs and TLS stylets] together, but whether the nature of
the demand is such that those services could be offered separately,” In re Visa, 2003 WL
1712568 at *3, 2003 U.S. Dist. LEXIS 4965, at *9, here (unlike in In re Visa) the genuine
disputes of fact do not merely relate to “whether it was more efficient for [Bard] to offer [its
PICCs and TLS stylets] together,” but to whether the “nature of the demand” for Bard’s TLS
stylet (as distinct from Teleflex’s TLS stylet) is such that Bard could efficiently sell the products
separately at all. Therefore, the Court denies AngioDynamics’ motion for summary judgment
with respect to the separate products element of its claim.
2.

Coercion

“Actual coercion by the seller that in fact forces the buyer to purchase the tied product is
an indispensable element of a tying violation.” Unijax, Inc. v. Champion Int’l, Inc., 683 F.2d
678, 685 (2d Cir. 1982). Indeed, “the essential characteristic of an invalid tying arrangement lies
in the seller’s exploitation of its control over the tying product to force the buyer into the

Moreover, the fact that Bard sold single-sterile stylets to Cleveland Clinic for a price at which Bard appeared to make
a fairly high margin, (Dkt. No. 138-79, at 3), does not definitively resolve the question of whether sufficient demand
existed to allow Bard to efficiently sell the products separately to the market at large, given the record evidence
suggesting that this price was based on factors specific to Cleveland Clinic and was discounted from the price at which
Bard would have been able to offer its single-sterile stylet to the market. (Dkt. No. 146, ¶ 48 (citing and summarizing
record evidence)).
5

45

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 46 of 105

purchase of a tied product that the buyer either did not want at all, or might have preferred to
purchase elsewhere on different terms.” Jefferson Par., 466 U.S. at 12. “When such ‘forcing’ is
present, competition on the merits in the market for the tied item is restrained and the Sherman
Act is violated.” Id. “The element of actual coercion is designed to weed out the many cases where
the bundling of separate products is due to consumer demand. If a consumer wants to purchase a
bundle of the alleged tying and tied products, the seller is simply satisfying consumer demand and
monopolization concerns are irrelevant.” Kaufman, 836 F.3d at 142. A seller’s “use of strong

persuasion, encouragement, or cajolery to the point of obnoxiousness to induce [a buyer] to buy
its full line of products does not . . . amount to actual coercion,” which is present only if the
seller “goes beyond persuasion and conditions its [buyer’s] purchase of one product on the
purchase of another product.” Unijax, 683 F.2d at 685 (internal quotation marks omitted).
An “unremitting policy of tie-in, if accompanied by sufficient market power in the tying
product to appreciably restrain competition in the market for the tied product constitutes the
requisite coercion . . . given foreclosure of a not insubstantial volume of interstate commerce.”
Hill v. A-T-O, Inc., 535 F.2d 1349, 1355 (2d Cir. 1976). Thus, when the seller has “a policy of
never offering the [tying product] separately from the [tied product],” id., allegations of
individual coercion are unnecessary. See Park, 2007 WL 119461, at *4, 2007 U.S. Dist. LEXIS
2001, at *11-12 (“When a policy of conditioned sales is demonstrated, proof of coercion on an
individual basis is unnecessary.”); cf. Reisner v. Gen. Motors Corp., 511 F. Supp. 1167, 1177
n.21 (S.D.N.Y. 1981) (“The ‘unremitting policy of tie-in,’ which under some circumstances can
substitute for coercion, is appropriate only in a situation . . . where the policy was applied to all
buyers.” (citation omitted)), aff’d, 671 F.2d 91 (2d Cir. 1982).
AngioDynamics argues that the element of coercion is met here because “Bard admits
that it has a policy of only selling its TLS Stylets preloaded in its PICCs—forcing those who
46

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 47 of 105

want to purchase Bard’s TLS to also purchase its PICCs—and that it has only made one
exception to that policy, for the Cleveland Clinic.” (Dkt. No. 134-1, at 20-21; see also Dkt. No.
152, at 10). AngioDynamics also argues that “the record establishes, and Bard has admitted, that
many institutions have requested to purchase Bard’s TLS Stylets separately from its PICCs,” and
that “Bard refused these requests and has not sold its PICCs to any of these institutions—only to
the Cleveland Clinic.” (Id. at 21).
Bard responds that AngioDynamics “has a complete failure of proof” on the coercion
element because “not a single hospital will testify that it ‘either did not want [the Bard PICC] at
all, or might have preferred to purchase [PICCs] elsewhere on different terms’”; to the contrary,
“the only hospital that the jury will hear, Cleveland Clinic, will testify [that] it had the choice to
purchase Bard’s standalone stylets and use them with AngioDynamics’ PICCs, but quickly
decided that it preferred Bard’s preloaded PICCs.” (Dkt. No. 145, at 15-16 (quoting Jefferson
Par., 466 U.S. at 12)). Bard also argues that AngioDynamics cannot satisfy the coercion element
“by showing only that Bard does not offer its stylets separately from its PICCs,” without any
evidence that particular customers were actually coerced into purchasing the products together.
(Id. at 16-17).
Here, it is undisputed that, with the sole exception of Cleveland Clinic for a limited
period in the past, Bard has never, and does not, allow its customers to purchase its TLS stylets
separately from its PICCs. In other words, as a matter of Bard’s policy, customers who wish to
purchase Bard’s TLS stylet are forced to purchase Bard’s PICCs along with it. (Dkt. No. 133-2,
¶¶ 26-27, 30; Dkt. No. 134-2, ¶¶ 22-23; Dkt. No. 144-1, ¶¶ 26-27, 30; Dkt. No. 146, ¶¶ 22-23).
Thus, Bard’s policy is precisely the type of “unremitting policy of a tie,” i.e. the “policy of never
offering the [tying product] separately from the [tied product],” that the Second Circuit has long

47

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 48 of 105

held “constitutes the requisite coercion,” assuming the other elements of market power and
foreclosure of a not insubstantial amount of commerce are satisfied. Hill, 535 F.2d at 1355.
Bard’s argument that AngioDynamics cannot prevail on the coercion element without
proof that individual customers were coerced into purchasing the tied products together misses
the mark. While Bard correctly states that the Second Circuit requires a showing of actual
coercion to prevail on a tying claim, it has offered no rebuttal to the case law in this Circuit
establishing that a “policy of never offering the [tying product] separately from the [tied
product],” Hill, 535 F.2d at 1355, which “[is] applied to all buyers,” Reisner, 511 F. Supp. at
1177 n.21, in itself constitutes such “actual coercion,” since any customer who wants the tying
product is, by definition, forced to purchase it with the tied product or not at all. See Park, 2007
WL 119461, at *4, 2007 U.S. Dist. LEXIS 2001, at *11-12 (“When a policy of conditioned sales
is demonstrated, proof of coercion on an individual basis is unnecessary.”); In re Visa, 2003 WL
1712568, at *2, 2003 U.S. Dist. LEXIS 4965, at *7-8 (holding, on a plaintiff’s motion for
summary judgment, that the defendants “indisputably” engaged in a tie because the defendants’
“rules require merchants who accept [the tying product] to also accept [the tied product]” and
“Defendants do not claim otherwise”).
Bard argues that “AngioDynamics cannot meet the Hill standard,” because Hill
establishes that, “in addition to a policy of tying, a plaintiff must also show (i) market power in
the tying market, and (ii) foreclosure of a not insubstantial amount of commerce.” (Dkt. No. 145,
at 17). Bard is correct that, under Hill, an “unremitting policy of a tie” only transforms into
“actual coercion” where the defendant has sufficient market power such that its policy can
“appreciably restrain competition in the market for the tied product” and “foreclos[e] a not
insubstantial volume of interstate commerce.” Hill, 535 F.2d at 1355. Indeed, without such

48

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 49 of 105

market power, a defendant cannot actually coerce customers into purchasing its tied products,
since those customers can simply purchase the desired products separately elsewhere. In its most
recent discussion of the “actual coercion” element, while it did not address the continued validity
of Hill, the Second Circuit took a consistent approach, noting that “the ‘separate product’ and
‘market power’ requirements are usually essential to the coercion element,” since those separate
elements are necessary to distinguish tying policies that result from the defendant’s market
power from situations where “the bundling of separate products is due to consumer demand.”
Kaufman, 836 F.3d at 141-43.
The Court reads these cases to hold that an “unremitting policy of a tie” constitutes actual
coercion if, and only if, the other elements of a tying claim—market power, separate products,
and the foreclosure of a not insubstantial amount of commerce—are met. Therefore, if a plaintiff
proves that a defendant has an unremitting policy of a tie that applies to all customers, the
plaintiff need not present proof of individualized coercion, but must still prove the other
elements of its tying claim in order to establish that that “unremitting policy” in itself constitutes
“actual coercion.” Here, the fact that Bard engages in an “unremitting policy of a tie” is not in
dispute; given that, proof that Bard’s TLS policy coerced individual customers who did not want
Bard’s PICCs into buying them is unnecessary to prove the “actual coercion” element of
AngioDynamics’ claim. However, the Court cannot conclude that Bard’s “unremitting policy of
a tie,” in itself, satisfies that element unless it determines that Bard has sufficient market power,
and forecloses a sufficiently “not insubstantial” amount of commerce, to transform that
“unremitting policy” into “actual coercion” within the meaning of the antitrust laws. Because, as
discussed below, genuine factual disputes remain as to these distinct elements of

49

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 50 of 105

AngioDynamics’ tying claim, the Court must, at this stage, deny AngioDynamics’ motion for
summary judgment on the actual coercion element.
3.

Market Power in the Tying Product Market

With respect to the third element of AngioDynamics’ tying claim—market power in the
tying product market—the Second Circuit has explained:
The third element at issue here—market power in the tying product—is essential to a
would-be monopolist’s coercion via tie-in. Without the leverage of market power, a
seller’s inefficient tie-in will fail because a rational consumer will buy the tying product
from the seller’s competitor. “As a simple example, if one of a dozen food stores in a
community were to refuse to sell flour unless the buyer also took sugar it would hardly
tend to restrain competition in sugar if its competitors were ready and able to sell flour by
itself.” N. Pac. Ry. Co., 356 U.S. at 6-7, 78 S.Ct. 514. Hence, without market power,
there is little risk of anticompetitive harm from the seller’s tie-in.
Market power is “the ability of a single seller to raise price and restrict output.” Eastman
Kodak, 504 U.S. at 464, 112 S.Ct. 2072 (quoting Fortner Enters., Inc. v. U.S. Steel Corp.,
394 U.S. 495, 503, 89 S.Ct. 1252, 22 L.Ed.2d 495 (1969)). It can be shown by specific
evidence of a seller’s ability to control prices or exclude competitors from the market.
See K.M.B. Warehouse Distribs., Inc. v. Walker Mfg. Co., 61 F.3d 123, 129 (2d Cir.
1995). Market share is a proxy for market power. See id.; Eastman Kodak, 504 U.S. at
464, 112 S.Ct. 2072. A high market share alone, however, is insufficient to infer a seller’s
market power if other characteristics of the product market, such as low barriers to entry,
high cross elasticity of demand, or technological developments in the industry, interfere
with the seller’s control of prices. See Tops Mkts., Inc. v. Quality Mkts., Inc., 142 F.3d 90,
98-99 (2d Cir. 1998) (“A court will draw an inference of monopoly power only after full
consideration of the relationship between market share and other relevant market
characteristics.”). Indeed, in a tying case, “the best way” to plead market power is to
allege facts that, if proven, “establish directly that the price of the tied package is higher
than the price of components sold in competitive markets.” Will v. Comprehensive
Accounting Corp., 776 F.2d 665, 671-72 (7th Cir. 1985) (Easterbrook, J.).
Kaufman, 836 F.3d at 143.
AngioDynamics argues that Bard’s overwhelming share of the market for TLSs
establishes its market power. (Dkt. No. 134-1, at 22-23; Dkt. No. 152, at 11-12). AngioDynamics
also contends that Bard’s significant market share is protected by high barriers to entry in the
TLS market, including high development costs, the requirement for FDA clearance of medical
devices, and Bard’s ownership of patented TLS technology. (Dkt. No. 134-1, at 22-23; Dkt. No.
50

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 51 of 105

152, at 12). Bard argues that genuine factual disputes exist as to the market power element
because: (1) the parties disagree as to whether the “TLS market” is a separate and distinct
relevant antitrust market, or whether the relevant market is the market for “differentiated
PICCs,” and AngioDynamics has not presented evidence regarding Bard’s market power in the
“differentiated PICC” market; and (2) even assuming the market for TLSs constitutes its own
antitrust market, there are factual disputes regarding the significance of barriers to entry in the
TLS market, and whether “Bard has maintained high share because it has offered a good product
at a cheaper price rather than because of high entry barriers.” (Dkt. No. 145, at 17-21).
Bard does not meaningfully dispute that it sold over the overwhelming majority of all
TLSs between 2013 and 2018. (Dkt. No. 134-2, ¶ 47; Dkt. No. 146, ¶ 47). Assuming the TLS
market constitutes its own antitrust market, this extremely high market share alone would
constitute significant evidence of market power. See, e.g., In re Visa, 2003 WL 1712568, at *4,
2003 U.S. Dist. LEXIS 4965, at *11-12 (finding that market share of “nearly 60 percent” “easily
qualifies as ‘appreciable economic power’ for purposes of the per se [tying] rule”); Park, 2007
WL 119461, at *8, 2007 U.S. Dist. LEXIS 2001, at *21-22 (noting that “[h]ow much market
power is necessary to find a per se illegal tying arrangement is an unsettled issue of antitrust
law” but that the fact that “Defendants possess an 80-90% market share might, standing alone,
permit an inference of market power”); Ortho Diagnostic Sys., Inc. v. Abbott Labs., Inc., 920 F.
Supp. 455, 473 (S.D.N.Y. 1996), as corrected (Mar. 15, 1996) (holding that “a market share
above 30 percent is necessary to establish the requisite [market] power” for purposes of a tying
claim under Sherman Act § 1). However, the Court nonetheless finds that there are factual
disputes that preclude summary judgment for AngioDynamics on this element.

51

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 52 of 105

First, for the same reasons the Court found that a genuine dispute of material fact existed
as to whether TLSs and PICCs are separate products, the Court finds that the contrasting expert
analyses by Dr. Hay and Professor Scott Morton as to whether these products constitute separate
relevant antitrust markets, (Dkt. No. 136-9, at 26-35; Dkt. No. 147-13, at 13-16; Dkt. No. 13822, at 4-6), viewed together with the record evidence as a whole, raise a genuine factual dispute
as to this question. 6 Because an analysis of market power “only makes sense with reference to a
particular market,” Heerwagen v. Clear Channel Commc’ns, 435 F.3d 219, 229 (2d Cir. 2006),
the question of whether Bard has sufficient market power for purposes of AngioDynamics’ tying
claim must survive summary judgment as well. 7
Second, even assuming the TLS market constitutes a distinct market, there are factual
questions regarding the barriers to entry AngioDynamics identifies and their impact on Bard’s
market power. From the record evidence, there is little doubt that high development costs, the
need for FDA approval, and patents held by Bard pose barriers for market entrants seeking to
develop TLSs. (Dkt. No. 134-2, ¶¶ 50-56 (citing record evidence discussing these barriers to
entry)). However, there are factual questions as to whether those barriers are significant enough

Indeed, at oral argument, AngioDynamics’ counsel agreed that, should the Court find a genuine dispute of material
fact as to whether Bard’s PICCs constitute a separate product from its TLSs, it should also find a factual dispute as to
the proper market definition for purposes of the “market power” element.

6

AngioDynamics argues that Professor Scott Morton’s opinion is based on the premise that there is no separate use
for TLSs independent of their use with PICCs, and that this premise is contrary to law because “[t]he Supreme Court
and courts in this Circuit have repeatedly held . . . that it does not matter whether ‘there is no separate use’ for the
tying product.” (Dkt. No. 152, at 11 (citing Park, 2007 WL 119461, at *4, 2007 U.S. Dist. LEXIS 2001, at 10-11;
Kodak, 504 U.S. at 463)). AngioDynamics is correct that the mere fact that two products are interdependent does not
preclude a finding that they are separate products or exist in separate antitrust markets. However, Professor Scott
Morton’s opinion is not based merely on the fact that Bard’s TLSs have no use separate from Bard’s PICCs. Rather,
she opines that consumers “do not evaluate the price of [a TLS] stylet in comparison with the price of other stylets,”
but rather “evaluate the price of that stylet paired with a PICC in comparison with other combinations of stylets and
PICCs,” and “choose the differentiated PICC system that best suits its needs based on the system’s overall features
and price.” (Dkt. No. 147-13, at 15). Professor Scott Morton concludes that, “[b]ecause the relevant competition
between manufacturers occurs at the differentiated PICC level and not at the TLS or TLS stylet level . . . there is not
a separate relevant product market for TLSs.” (Id. at 16).
7

52

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 53 of 105

to protect Bard’s ability to raise price and restrict output in the TLS market, particularly given
record evidence that Bard’s competitors have developed TLS products in the past
notwithstanding the associated development costs and Bard’s patents, (Dkt. No. 146, ¶¶ 50, 56),
that the price of Bard’s PICCs preloaded with its TLS stylet has sometimes been lower than the
combination of AngioDynamics’ PICCs and Teleflex’s TLS, (Dkt. No. 133-2, ¶¶ 39-41; Dkt. No.
144-1, ¶¶ 39-41), that
, (Dkt. No. 133-2, ¶ 21; Dkt. No. 1441, ¶ 21), and that there is no record evidence suggesting that Bard has, in fact, raised prices or
restricted output in the TLS market. These questions are best resolved by a fact-finder after
development of a complete record at trial, rather than the Court on summary judgment.
Therefore, the Court denies AngioDynamics’ motion for summary judgment on the market
power element of its claim.
4.

“Not Insubstantial” Amount of Interstate Commerce

As to the fourth element of AngioDynamics’ claim, the Supreme Court has stated that
there must be a substantial potential for impact on competition in the tied market in order to
justify condemnation of an alleged tying arrangement. Jefferson Par., 466 U.S. at 16. “If only a
single purchaser were ‘forced’ with respect to the purchase of a tied item, the resultant impact on
competition would not be sufficient to warrant the concern of antitrust law.” Id. The Supreme
Court has defined “substantial” in this context as “substantial enough in terms of dollar-volume
so as not to be merely de minimis.” Fortner Enters., Inc. v. United States Steel Corp., 394 U.S.
495, 501 (1969) (“Fortner I”). “There is no magic number that definitively establishes whether a
plaintiff has foreclosed a ‘not insubstantial’ amount of potential sales for the tied product.”
Gonzalez v. St. Margaret’s House Housing Dev. Fund Corp., 880 F.2d 1514, 1518 (2d Cir.
1989). In Fortner I, the Supreme Court held that a sum of almost $200,000 is not “paltry or
53

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 54 of 105

‘insubstantial,’” while the Second Circuit has “said that $600,000 of commerce clearly meets
‘any test of substantiality.’” Id. (quoting Fortner, 394 U.S. at 502; Yentsch v. Texaco, Inc., 630
F.2d 46, 58 (2d Cir. 1980)).
AngioDynamics argues that, for purposes of this element, the Court should look “solely
to the amount of sales processed by the tied product.” (Dkt. No. 134-1, at 24; Dkt. No. 152, at
12-14). It asserts that “the data produced by Bard in the litigation shows that Bard sold over
PICCs pre-loaded with TLS, valued at over

during the period 2013-2018.”

(Dkt. No. 134-1, at 24; Dkt. No. 152, at 12). Bard argues that the relevant measure is not the total
volume of tied sales, but the amount of foreclosed sales, i.e. the amount of sales that would have
occurred but for the challenged tie (subtracting those sales in which the customer would have
voluntarily chose to purchase the tied package even if they had a different option). (Dkt. No. 145,
at 21-22).
The Supreme Court has held that, “[f]or purposes of determining whether the amount of
commerce foreclosed is too insubstantial to warrant prohibition of the practice . . . the relevant
figure is the total volume of sales tied by the sales policy under challenge, not the portion of this
total accounted for by the particular plaintiff who brings suit.” Fortner I, 394 U.S. at 502.
Clearly, then, the relevant metric is the amount of PICC sales Bard’s TLS policy foreclosed with
the respect to the entire market, not just AngioDynamics, and no party contends otherwise. The
more difficult, and relevant, question is whether this language means that the Court should look
simply to the total volume of Bard’s tied sales, rather than the amount of sales that Bard’s
competitors actually were foreclosed from as a result of Bard’s conduct. AngioDynamics cites
several district court cases in support of its argument that the court should look solely to the total
volume of tied sales, rather than the volume of sales that were actually foreclosed. See In re Data

54

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 55 of 105

Gen. Corp. Antitrust Litig., 490 F. Supp. 1089, 1117 (N.D. Cal. 1980) (“In Fortner I and the
earlier cases relied on therein, the Supreme Court looked solely at the total volume of the
defendant’s tied sales . . . Proof of actual foreclosure has never been required in order to satisfy
the effect on commerce test.” (citation omitted)); In re Visa, 2003 WL 1712568, at *2, 2003 U.S.
Dist. LEXIS 4965, at *7 (“In 1999 alone, merchants processed over one hundred and fifty billion
dollars in sales through the defendants’ off-line debit cards. This figure is ‘not insubstantial.’ . . .
To the extent that the defendants contend that this element incorporates the need for a showing of
‘foreclosure’ or ‘anticompetitive effect’ in the tied product market, I disagree.”).
However, the Second Circuit has noted that while “one might read Fortner I to require
the district court to consider the [total amount of tied sales], we believe that the Supreme Court
in Fortner I was primarily concerned with ascertaining the total sales lost to potential
competitors due to the tying policy.” Gonzalez, 880 F.2d at 1519 (citation omitted). The
Gonzalez court found that using the total amount of tied sales (in that case, mandatory meal
charges collected from residents of a housing facility) “overstates the amount of commerce
foreclosed because the record already indicates that many residents would continue to use the
meal plan by choice,” and that using only the amount of tied sales to the specific plaintiffs in the
litigation “may also overstate the amount of commerce foreclosed by the mandatory meal policy
because some plaintiffs would choose not to buy the product (the single, prepared meal) at all . . .
for a variety of reasons; e.g., they might cook in their rooms at a lower cost.” Id. at 1518. The
Second Circuit remanded to the district court to “determine whether the impact of the alleged
tying arrangement is substantial enough to warrant the protection of the antitrust laws” by
“tak[ing] into consideration the actual number of people affected by the disputed policy,” and

55

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 56 of 105

noted that “[i]f the district court can ascertain [the total sales lost to potential competitors due to
the tying policy], it may use that as the relevant figure.” Id. at 1519.
Notwithstanding the somewhat ambiguous language in Fortner I, this Court is bound by
the Second Circuit law in Gonzalez. Here, to estimate the amount of commerce foreclosed by
Bard’s TLS policy, AngioDynamics has simply provided its expert’s valuation of the total
volume of Bard’s sales of PICCs preloaded with its TLS. (Dkt. No. 138-20, at 53-54). This,
however, likely “overstates the amount of commerce foreclosed” by Bard’s TLS policy, as it
does not account for the near-certainty that at least some customers would have voluntarily
purchased Bard’s PICCs preloaded with its TLS stylet even if they had the option not to.
Gonzalez, 880 F.2d at 1518; cf., e.g., Park, 2007 WL 119461, at *9, 2007 U.S. Dist. LEXIS
2001, at *28 (finding that using a defendant’s “total revenues” to measure the volume of
commerce affected “would overstate the effect [of the alleged anticompetitive activity] because
many consumers purchase the integrated BAR/BRI course by choice”). Of course, given that
amounts as low as a few hundred thousand dollars are sufficient to find a “not insubstantial”
amount of foreclosed commerce, if a fact-finder were to conclude that even a relatively small
fraction of Bard’s total PICC sales were actually foreclosed to competitors by its TLS policy, the
fact-finder would likely have a basis to conclude that the amount of foreclosed commerce was
“not insubstantial.” However, as discussed below in connection with the “anticompetitive
effects” element of AngioDynamics’ claim, there are outstanding factual disputes as to whether
Bard’s TLS policy actually foreclosed any PICC sales at all. Therefore, the Court must deny
AngioDynamics’ motion for summary judgment on this element. 8

The Court recognizes that AngioDynamics’ damages expert, Dr. Frankel, puts forth a number of estimates of
AngioDynamics’ lost profits that resulted from Bard’s TLS policy. If proven, any one of these estimates would far
exceed the low threshold to allow this Court to find that a “not insubstantial” amount of commerce has been foreclosed,
without even accounting for other PICC competitors’ lost sales. (Dkt. No. 132-5, at 45). However, these lost profits

8

56

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 57 of 105

5.

Anticompetitive Effects 9

To prove anticompetitive effects, a plaintiff must prove that “the defendants’ challenged
behavior had an actual adverse effect on competition as a whole in the tied product market.”
Geneva Pharms. Tech. Corp. v. Barr Labs., 386 F.3d 485, 506-07 (2d Cir. 2004).
Anticompetitive effects may be proven directly by control of prices or exclusion of competition.
See Virgin Atl. Airways Ltd. v. British Airways PLC, 257 F.3d 256, 264 (2d Cir. 2001); see also
Yentsch, 630 F.2d at 57 (explaining that anticompetitive effects may be shown if “competitors
were foreclosed from selling to [buyers] because of [the defendant’s] policies”). If plaintiffs
establish anticompetitive effects, the burden shifts to defendants to justify their tying
arrangement. Geneva, 386 F.3d at 507. “Assuming defendants can provide such proof, the
burden shifts back to the plaintiffs to prove that any legitimate competitive benefits offered by
defendants could have been achieved through less restrictive means.” Id.
AngioDynamics argues that Bard’s TLS policy caused “substantial foreclosure in the
PICC market, which harmed not only competitors but also the competitive environment in the
PICC market.” (Dkt. No. 134-1, at 25). AngioDynamics argues that, “[a]s a result of Bard’s tie
and its overwhelming market share in the TLS market . . . Bard has almost entirely foreclosed
competition in the largest segment of the PICC market (PICCs used with TLS).” (Dkt. No. 152,
at 15). AngioDynamics also argues that Bard’s policy “suppressed innovation and investment in

estimates are obviously in dispute, and given that (as discussed below) there are genuine disputes of material fact as
to whether any of AngioDynamics’ lost profits may be fairly attributed to Bard’s TLS policy, these estimates cannot,
at this stage, serve as a basis for granting summary judgment in AngioDynamics’ favor.
As they did at the motion to dismiss stage, the parties disagree on whether AngioDynamics must prove
anticompetitive effects as an element of its per se claim, as distinct from its rule of reason claim. (Dkt. No. 134-1, at
24-25; Dkt. No. 145, at 22 n.26; Dkt. No. 152, at 14 & n.8). Because the Court has found that genuine disputes of
material fact exist as to the other elements of AngioDynamics’ claim under either a per se or rule of reason theory, a
decision on this issue would not affect the Court’s ultimate decision on either party’s motion for summary judgment.
Therefore, the Court does not address this issue.
9

57

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 58 of 105

the PICC market and frustrated customer choice.” (Dkt. No. 134-1, at 26). Bard argues that: (1)
customers who wish to use non-Bard PICCs with a TLS have the option to pair those PICCs with
Teleflex’s TLS, so Bard’s PICC competitors are not entirely foreclosed from competing for such
customers; (2)
(3)
Bard’s actions have benefitted consumers in the form of lower PICC prices and higher product
quality; (4) AngioDynamics has failed to establish BioFlo’s technological superiority to Bard’s
PICCs (a basis for AngioDynamics’ belief that many customers prefer BioFlo to Bard’s PICCs
and that Bard’s TLS policy is therefore thwarting customer access to a desired product); and (4)
the record belies any argument that Bard’s TLS policy “suppressed innovation and investment in
the PICC market,” as both AngioDynamics and Teleflex have made significant investments in
the development of innovative antithrombogenic and antimicrobial PICC technology. (Dkt. No.
145, at 22-25).
AngioDynamics’ argument is facially compelling. Because Bard’s policy forces any
customer who wishes to use Bard’s TLS stylet to pair it with Bard’s PICCs, Bard’s PICC
competitors are effectively unable to compete for the PICC business of any customer who opts to
use Bard’s TLS. And because the vast majority of customers within the market segment for TLSpaired PICCs use Bard’s TLS, Bard’s TLS policy arguably precludes Bard’s competitors from
competing with Bard in by far the largest segment of the PICC market.
Nonetheless, as Bard has demonstrated, factual disputes exist as to whether Bard’s TLS
policy has actually foreclosed any PICC sales by Bard’s competitors, given the combination of
the following factors: (1) customers who wish to use a TLS stylet with a non-Bard PICC have
the option to pair the non-Bard PICC with Teleflex’s TLS, meaning that Bard’s PICC

58

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 59 of 105

competitors are not entirely foreclosed from competing with Bard even in the market segment for
PICCs paired with TLSs; (2) record evidence suggests that customers who use Bard’s PICCs
preloaded with its TLS may prefer Bard’s integrated product for various reasons independent of
its TLS policy, including price, quality, ease of use, concerns about heightened risks to patient
safety from loading a TLS stylet at the patient’s bedside, the convenience of dealing with a
single supplier, and preferences for Bard’s PICC technology over competitors’ PICC technology,
including BioFlo; and (3) as discussed more fully with respect to the Court’s analysis of antitrust
injury, see Section IV.B.2. supra, AngioDynamics has not adduced evidence establishing beyond
dispute that Bard’s TLS policy caused AngioDynamics or Teleflex to lose PICC sales to Bard
that they otherwise would have made.
Given these factors, a reasonable jury could find that Bard’s dominance of the market
segment for PICCs paired with TLSs has resulted from customers’ overwhelming preference for
Bard’s combined PICC-TLS product over the other PICC and TLS combinations currently
available, and not because Bard’s TLS policy prevented Bard’s PICC competitors from
competing in that market segment.

Moreover, AngioDynamics cites no evidence that Bard’s conduct has increased prices,
reduced output or reduced product quality in the market. Cf. MacDermid Printing Solutions LLC

59

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 60 of 105

v. Cortron Corp., 833 F.3d 172, 182-84 (2d Cir. 2016) (explaining that “[o]ur cases suggest that
it is possible, at least in theory, to prove that a challenged action harmed competition without
offering evidence of higher prices, reduced output, or reduced quality,” but “in no precedential
opinion in this Circuit has a plaintiff successfully proved an adverse effect on competition
without offering” such evidence, though “[w]e have suggested that actions that reduce consumer
choice are inherently anticompetitive”). And the Court agrees with Bard that AngioDynamics’
claim that Bard’s TLS policy suppressed innovation in the PICC market is conclusory and
without support in the record.
Therefore, the Court finds that genuine disputes of material fact exist as to the
anticompetitive effects element of AngioDynamics’ claim, and denies AngioDynamics’ motion
for summary judgment on this element.
6.

Liability Summary

For the foregoing reasons, the Court finds that genuine disputes of material fact exist as
to each of the liability elements of AngioDynamics’ claim, for purposes of both its per se and
rule of reason theories. Therefore, the Court denies AngioDynamics’ motion for summary
judgment as to liability.
B.

Antitrust Injury

Both parties move for summary judgment on the antitrust injury element of
AngioDynamics’ claim. (Dkt. No. 133-1; Dkt. No. 134-1, at 26-34). The notion of “antitrust
injury” grew from the recognition that a competitor may be injured not only by prohibited
anticompetitive activity, but also by competition itself, and that the antitrust laws were not
intended to afford the latter injuries a remedy. See Balaklaw v. Lovell, 14 F.3d 793, 797 (2d Cir.
1994). Antitrust injury, then, simply means “injury of the type the antitrust laws were intended to
prevent and that flows from that which makes defendants’ acts unlawful.” Brunswick Corp. v.
60

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 61 of 105

Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977). “To establish antitrust injury, a private
litigant must ‘prove . . . “injur[y] in its business or property” by reason of the violation . . . [and
that] the violation was at least a material cause of the plaintiff’s injury.’” Blue Tree Hotels Inv.,
Ltd. v. Starwood Hotels & Resorts Worldwide, Inc., 369 F.3d 212, 220 (2d Cir. 2004) (citations
omitted, alterations in original). In other words, “a plaintiff must show (1) an injury-in-fact; (2)
that has been caused by the violation; and (3) that is the type of injury contemplated by the
[antitrust] statute.” Id. The antitrust injury requirement thus ensures that a “plaintiff can recover
only if the loss stems from a competition-reducing aspect or effect of the defendant’s behavior.”
Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 344 (1990).
1.

Preliminary Evidentiary Issues

Before turning to the merits of the parties’ arguments, the Court must address a
preliminary evidentiary issue. Throughout its briefing and its response to AngioDynamics’
statement of material facts, Bard objects to certain evidence AngioDynamics relies on as
inadmissible hearsay. Bard’s overall argument is that, to the extent AngioDynamics cites
statements from hospital employees (directly or as recounted second-hand by AngioDynamics
employees), those statements are hearsay (and, in the latter case, double hearsay) and thus
inadmissible. (Dkt. No. 145, at 26-30; Dkt. No. 146, ¶¶ 40(a)-(e), 59, 63-65 & “Statement of
Additional Material Facts” ¶¶ 22(d), 23(c); Dkt. No. 154, at 10 & n.5, 12-13). AngioDynamics
argues that this evidence is admissible under the “state of mind” exception to the hearsay rule to
“show customer motive, such as a customer’s reason for ceasing to do business with
AngioDynamics.” (Dkt. No. 144, at 29-30; Dkt. No. 152, at 19-20).
The Court “may only consider admissible evidence” in ruling on a motion for summary
judgment. Humphrey v. Diamant Boart, Inc., 556 F. Supp. 2d 167, 173-74 (E.D.N.Y. 2008)
(citations omitted). At the same time, “material relied on at summary judgment need not be
61

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 62 of 105

admissible in the form presented to the district court. Rather, so long as the evidence in question
will be presented in admissible form at trial, it may be considered on summary judgment.” Smith
v. City of New York, 697 F. App’x 88, 89 (2d Cir. 2017) (internal quotation marks omitted).
Thus, under Rule 56(c)(2) of the Federal Rules of Civil Procedure, “[a] party may object that the
material cited to support or dispute a fact cannot be presented in a form that would be admissible
in evidence.”
As the Second Circuit has long held, under Fed. R. Evid. 803(3), “[s]tatements of a
customer as to his reasons for not dealing with a supplier are admissible for [the] limited
purpose” of demonstrating the customer’s motive, “although not as evidence of the facts recited
as furnishing the motives.” Herman Schwabe, Inc. v. United Shoe Machinery Corp., 297 F.2d
906, 914 (2d Cir. 1962) (citations and internal quotation marks omitted); see also Hydrolevel
Corp. v. Am. Soc. Of Mech. Eng’rs, Inc., 635 F.2d 118, 128 (2d Cir. 1980) (same); Sleepy’s LLC
v. Select Comfort Wholesale Corp., 779 F.3d 191, 204 (2d Cir. 2015) (same, outside the antitrust
context); Discover Fin. Servs. v. Visa U.S.A. Inc., No. 04-cv-7844, 2008 WL 4560707, at *1,
2008 U.S. Dist. LEXIS 80801, at *4 (S.D.N.Y. Oct. 9, 2008) (“Defendants argue that this Court
should preclude [plaintiff’s] current and former executives from testifying at trial about what
Citibank’s executives purportedly told them . . . However, testimony concerning the motivation
of customers for ceasing to deal with a business is admissible under the ‘state of mind’ exception
to the hearsay rule, Rule 803(3) of the Federal Rules of Evidence, provided that there is
otherwise admissible proof that business was lost.”).
While from a different Circuit, Callahan v. A.E.V., 182 F.3d 237 (3d Cir. 1999), which
AngioDynamics cites, is instructive. There, the court found that the plaintiff beer retailers’
deposition testimony recounting statements by their customers as to the customers’ reasons for

62

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 63 of 105

no longer purchasing beer from the plaintiffs was admissible for the purpose of proving the
customers’ motives, but not for the fact of lost sales; that the plaintiffs’ own testimony that “they
knew of customers who used to purchase beer from them, but no longer did,” was admissible as
“direct evidence of an actual loss of customers,” but not as evidence of “the reason therefore”;
and that the “plaintiffs’ testimony that certain customers no longer purchased beer from them,
coupled with their testimony concerning the customers’ statements of their motive, which is
admissible hearsay under Rule 803(3), are together evidence of the fact of damage.” Id. at 25253.
AngioDynamics argues that “show[ing] customer motive, such as a customer’s reason for
ceasing to do business with AngioDynamics,” is “exactly how AngioDynamics relies on” the
evidence Bard challenges, and thus that this evidence is admissible under Rule 803(3). (Dkt. No.
144, at 29). Bard offers two responses.
First, Bard argues that “PICC purchases are complicated decisions made by hospital
committees comprised of many stakeholders,” and that “[a]t most, AngioDynamics’ evidence [of
statements by particular hospital employees regarding the hospitals’ motivations] constitutes
speculation by hospital personnel about what drove past decisions of other decision-makers.” In
support of this argument, Bard primarily relies on Amerisource Corp. v. RxUSA Int’l Inc., No.
02-cv-2514, 2009 WL 235648, 2009 U.S. Dist. LEXIS 6864 (E.D.N.Y. Jan. 30, 2009). In that
case, the court declined to apply the state-of-mind exception to admit deposition testimony and
written notes from the plaintiffs’ executives reflecting that “employees of various non-party
distributors told [them] (i) that their companies would not deal with [the plaintiff]; (ii) that their
companies would not deal with [the plaintiff] until [the defendant] gave its approval; and (iii)
that their companies would not deal with [the plaintiff] because [the defendant] said that [the

63

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 64 of 105

plaintiff] was a bad credit risk.” Id. at *1, 2009 U.S. Dist. LEXIS 6864, at *2. As relevant to
Bard’s argument, the court reasoned:
Here, the declarants are salespersons for various non-party pharmaceutical distributors
with whom [the plaintiff] sought to do business. However, credit managers rather than
salespersons decided whether or not to do business with prospective customers and the
declarants in this case had no authority or involvement in the decision to decline
RxUSA’s business. Consequently, the salespersons’ states of mind and intentions are not
relevant. Indeed, their statements reveal not their own intention but rather their belief or
memory as to the intention of the actual decision-maker, which is explicitly excluded by
Rule 803(3).
Id. at *2, 2009 U.S. Dist. LEXIS 6864, at *4. 10 This ruling may have some applicability to
certain documents or testimony AngioDynamics relies on. The parties agree that the personnel
involved in PICC purchasing decisions varies by hospital and can include nurses, doctors and
other professionals. (Dkt. No. 133-2, ¶¶ 15-17; Dkt. No. 144-1, ¶¶ 15-17). Given this
complexity, there are factual questions as to the roles the hospital employee speakers in the
evidence AngioDynamics relies on played in the decision-making processes at their hospitals,
and whether their statements can be fairly construed as reflections of their hospitals’ “thenexisting state of mind” so as to warrant admission under Rule 803(3). However, at this stage of
the proceedings, without further factual development on these questions, the Court declines to
find that the evidence of their statements AngioDynamics presents could not be presented in
admissible form at trial.

10
The court also found that the third-party salespersons’ “statements explaining why the company did not intend to
do business with [the plaintiff] [were] expressly outside the state-of-mind exception” because they constituted
“statement[s] of memory or belief to prove the fact remembered or believed.” Id., 2009 WL 235648, at *2, 2009 U.S.
Dist. LEXIS 6864, at *4-5. In Amerisource, there was “no dispute” that the hearsay statements were being offered not
only to prove the third-party company’s motivations, but also to prove the facts underlying those motives, i.e. “that
[the defendant] said [plaintiff] was a bad credit risk.” Id. at *1, 2009 U.S. Dist. LEXIS 6864, at *3. Here, to the extent
AngioDynamics offers evidence of customers’ then-existing states of mind solely to show its customers’ motivations
for switching away from or declining to purchase AngioDynamics’ PICCs, and not for the fact of lost sales itself or
the truth of the underlying facts asserted in those customers’ statements, the evidence would be admissible under Rule
803(3) for that purpose.

64

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 65 of 105

Second, Bard argues that much of AngioDynamics’ evidence “does not even quote
customers directly but rather involves AngioDynamics’ employees’ (oftentimes, hazy)
recollection of what either a hospital employee or another AngioDynamics sales representative
told them about the hospital’s PICC purchases.” (Dkt. No. 154, at 13). As Bard correctly points
out, “[e]ach hearsay statement within multiple hearsay statements must have a hearsay exception
in order to be admissible.” United States v. Cruz, 894 F.2d 41, 44 (2d Cir. 1990).
AngioDynamics does not respond to Bard’s argument regarding “double hearsay.”
Applying the foregoing principles, it appears to the Court that certain of the documents
and deposition testimony that AngioDynamics relies on would be admissible at trial under Rule
803(3) for the limited purpose of showing AngioDynamics’ customers’ motivations for declining
to purchase AngioDynamics’ PICCs, or for switching from AngioDynamics’ PICCs to Bard’s
PICCs. Other evidence is of more questionable admissibility, given the multiple layers of
hearsay embedded in certain documents and testimony, as well as questions over whether certain
documents truly reflect a hospital customer’s “then-existing state of mind” as opposed to merely
a statement of the speaker’s “memory or belief to prove the fact remembered or believed.” Fed.
R. Evid. 803(3). In the absence of fully developed briefing with respect to each piece of evidence
Bard raises objections to, the Court cannot issue a formal ruling on the admissibility of each such
piece of evidence at this time. Rather, applying the foregoing principles, the Court will determine
whether there appears to be enough record evidence that could be presented in admissible form
to raise a genuine dispute of material fact for trial with respect to antitrust injury.
2.

Injury-in-Fact and Causation
a.

Legal Standard

To establish antitrust injury, “[t]here must be a causal connection between an antitrust
violation and an injury sufficient for the trier of fact to establish that the violation was a ‘material
65

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 66 of 105

cause’ of or a ‘substantial factor’ in the occurrence of damage.” Billy Baxter, Inc. v. Coca-Cola
Co., 431 F.2d 183, 187 (2d Cir. 1970). “The fact of injury must be demonstrated with reasonable
certainty and may not be speculative,” and “failure to satisfactorily prove that a plaintiff is
injured by reason of the defendant’s alleged antitrust violations is sufficient to defeat the antitrust
claims.” Erie Conduit Corp. v. MAPA, 102 F.R.D. 877, 879 (E.D.N.Y. 1984) (citations omitted).
While antitrust plaintiffs are entitled to a “relaxed standard of proof” with respect to the
amount of damages they suffered as a result of a defendant’s conduct, “this relaxed standard of
proof applies to the amount of damages, not to whether the violation caused damage.” Hygrade
Milk & Cream Co. v. Tropicana Products, Inc., No. 88-cv-2861, 1996 WL 257581 at *16-17,
1996 U.S. Dist. LEXIS 6598, at *52-54 (S.D.N.Y. May 16, 1996) (citing MCI Comms. Corp. v.
Am. Tel. & Tele. Co., 708 F.2d 1081, 1161 (7th Cir. 1982)). That said, the amount of harm a
plaintiff needs to prove to meet its burden of demonstrating antitrust injury is not high; rather, a
plaintiff need only demonstrate “proof of some damage flowing from the unlawful [conduct];
inquiry beyond this minimum point goes only to the amount and not the fact of damage.” Zenith
Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 114 n.9 (1969). Despite this relatively
low burden, “[a]lthough there appear to be few precedents on point, the relevant decisions do
suggest that a trivial effect on a claimant’s sales is insufficient to demonstrate antitrust injury.”
Drug Mart Pharm. Corp. v. Am. Home Prod. Corp., No. 93-cv-5148, 2012 WL 3544771, at *14,
2012 U.S. Dist. LEXIS 115882, at *47-48 (E.D.N.Y. Aug. 16, 2012) (finding a de minimus
amount of lost sales insufficient to establish antitrust injury in the context of a Robinson-Patman
Act claim (discussing Allen Pen Co. v. Springfield Photo Mount Co., 653 F.2d 17, 23 (1st Cir.

66

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 67 of 105

1981); Hygrade Milk and Cream Co., 1996 WL 257581 at *18-19, 1996 U.S. Dist. LEXIS 6598,
at *59-60)). 11
An antitrust plaintiff need “not necessarily [prove that the defendant’s misconduct was]
the sole cause” of the plaintiff’s alleged injury, and “a plaintiff need not exhaust all possible
alternative sources of injury in fulfilling his burden of proving compensable injury.” In re Actos
End-Payor Antitrust Litig., 848 F.3d 89, 97-98 (2d Cir. 2017) (quoting Zenith Radio Corp., 395
U.S. at 114 n.9); see also In re Publ’n Paper Antitrust Litig., 690 F.3d 51, 66 (2d Cir. 2012)
(holding that “an antitrust defendant’s unlawful conduct need not be the sole cause of the
plaintiffs’ alleged injuries; to prove a ‘causal connection’ between the defendant’s unlawful
conduct and the plaintiff’s injury, the plaintiff need only ‘demonstrate that [the defendant’s]
conduct was a substantial or materially contributing factor’ in producing that injury” and that
“the injuries alleged would not have occurred but for [the defendant’s] antitrust violation”)
(citations omitted). “[R]equiring otherwise ‘would effectively deny private remedies, because
multiple causes always affect everyone.’” In re Gabapentin Patent Litig., 649 F. Supp. 2d 340,
356 (D.N.J. 2009) (quoting Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 338a at
317 (2d Ed. 2000)).

Citing case law from other Circuits, AngioDynamics argues that the relaxed standard of proof typically applied to
the amount of damages in antitrust cases also applies to the element of antitrust injury, and that an antitrust plaintiff
need only prove a “trifle of injury” resulting from the defendant’s conduct in order to satisfy this element. (Dkt. No.
134-1, at 27-29; Dkt. No. 144, at 12-13; Dkt. No. 152, at 15-16 (relying on In re Data Gen. Corp. Antitrust Litig., 490
F. Supp. at 1117-18; In re K-Dur Antitrust Litig., No. 01-cv-1652, 2008 WL 2660776, at *4, 2008 U.S. Dist. LEXIS
71772, at *21-22 (D.N.J. Feb. 21, 2008); Digidyne Corp. v. Data Gen. Corp., 734 F.2d 1336, 1339 n.1 (9th Cir.
1984))). The Court applies the Second Circuit’s standard for evaluating antitrust injury, as set forth above, and declines
to follow the cases AngioDynamics cites to the extent they are inconsistent with that standard.
11

67

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 68 of 105

b.

The Parties’ Arguments

Bard argues, in summary, that: (1) no hospital witness will testify at trial that it would
have purchased an AngioDynamics PICC to pair with Bard’s TLS but for Bard’s TLS policy; 12
(2) the evidence AngioDynamics relies on consists only of “[s]econd-hand, self-serving, and
speculative” testimony from its own executives, anecdotal documentary evidence of discussions
with specific customers, and “speculative comments” in Bard’s internal emails and documents;
and (3) none of that evidence establishes that any customer currently using Bard’s preloaded
PICCs would have purchased AngioDynamics’ PICCs but for Bard’s TLS policy or that Bard’s
TLS policy was a “material cause” of any PICC sales AngioDynamics lost, given the various
reasons some customers prefer to purchase Bard’s preloaded product independent of its TLS
policy. As to these reasons, Bard has adduced evidence that they include: (1) the lack of FDA
approval for the use of Bard’s standalone stylet with other companies’ valved PICCs; (2) hospital
preferences for certain characteristics that Bard’s PICCs have and AngioDynamics’ PICCs do
not; (3) the premium that AngioDynamics charges for BioFlo PICCs relative to other PICCs in
the market; (4) AngioDynamics’ inability to definitively prove BioFlo’s superiority to other
PICC technology through peer-reviewed clinical trials; (5) AngioDynamics’ past product recalls
and the resulting damage to its reputation in the market; (6) Bard’s more substantial training and
clinical support programs relative to AngioDynamics’; (7) Bard’s wider breadth of product lines
relative to AngioDynamics; (8) the effect of GPO and IDN contracts on purchasing decisions; (9)
possible risks to patient safety and other inconveniences of loading a TLS stylet at a patient’s
bedside rather than purchasing it preloaded into a PICC; (10) some customers’ preference for

Pursuant to a stipulation dated April 17, 2020, the parties agreed that neither party will call at trial any witnesses
from third-party hospitals, hospital groups or representatives, except for two witnesses from Cleveland Clinic who
have already been deposed. (Dkt. No. 136-27).
12

68

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 69 of 105

purchasing PICCs and TLSs from a single supplier; and (11) the fact that Bard would likely need
to charge a premium for its single-sterile stylet in order to compensate for its additional costs of
offering that product standalone, “including liability and reputational risks Bard would face in
the event of stylet breakage and patient injury.” (Dkt. No. 133-1, at 12-32; Dkt. No. 145, at 2534; Dkt. No. 154, at 6-12).
AngioDynamics argues, in summary, that: (1) the documentary evidence it relies on
demonstrates specific (and non-exhaustive) examples in which customers declined to purchase
AngioDynamics’ PICCs, or chose to switch from AngioDynamics’ PICCs to Bard’s PICCs,
because they had no option to pair AngioDynamics’ PICCs with Bard’s TLS stylet; (2) this same
evidence demonstrates that, even if multiple factors were at play in any particular customer’s
purchasing decision, the inability to pair AngioDynamics’ PICCs with Bard’s TLS stylet was at
least one “material factor” in that decision, which is all AngioDynamics needs to show in order
to establish antitrust injury; (3) economic evidence also supports a finding of injury and
causation, including that Bard sells the overwhelming majority of the TLSs on the market and
that “in those segments of the PICC market where Bard’s TLS is not a factor, or less of a factor,”
“AngioDynamics sells substantially more PICCs.” (Dkt. No. 134-1, at 26-34; Dkt. No. 144, at
12-30; Dkt. No. 152, at 17-20). 13
c.

Bard’s Motion for Summary Judgment

The record contains the following types of evidence that, viewed collectively and in the
light most favorable to AngioDynamics for purposes of Bard’s summary judgment motion, raises

The parties also present arguments as to whether the testimony of AngioDynamics’ causation and damages expert,
Dr. Frankel, is sufficient to establish injury-in-fact or causation. (Dkt. No. 133-1, at 17-24, 26-32; Dkt. No. 134-1, at
33; Dkt. No. 144, at 17, 24-26; Dkt. No. 154, at 14-16). For the reasons explained in Section IV.C.1 of this Decision,
the Court is granting Bard’s motion to exclude Dr. Frankel’s causation opinion. Therefore, the Court does not consider
Dr. Frankel’s opinion in connection with its injury-in-fact or causation analysis and, consequently, does not address
Bard’s specific critiques of his opinion or AngioDynamics’ responses here.
13

69

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 70 of 105

genuine issues of material fact as to whether AngioDynamics lost a more-than-de-minimus
amount of PICC sales to Bard and whether Bard’s TLS policy was a material cause of those lost
sales: (1) evidence that Bard received inquiries from some customers regarding the possibility of
purchasing a stand-alone TLS stylet to pair with AngioDynamics’ or another competitor’s
PICCs, which Bard consistently rebuffed; (2) internal Bard communications reflecting Bard’s
efforts to avoid disseminating knowledge of its sale of single-sterile TLS stylets to Cleveland
Clinic, at least in part out of a concern that customers would demand such an option if they knew
it was available; (3) internal Bard documents which could be read to reflect the views of some
Bard employees that Bard’s TLS policy prevented competitors like AngioDynamics from
effectively competing for some PICC customers that used Bard’s TLS, including one employee’s
observation that being forced to sell a stand-alone TLS stylet would “open the door for
[AngioDynamics]” to “pick off 5-10%” of customers currently using Bard’s PICCs preloaded
with its TLS, (Dkt. No. 136-63, at 3; Dkt. No. 136-64, at 2); (4) deposition testimony of
AngioDynamics executives identifying particular PICC customers that AngioDynamics has lost
to Bard; (5) deposition testimony from AngioDynamics executives recounting discussions with
specific customers (some of which the witnesses participated in directly, and some of which they
were informed about by sales representatives) in which those customers expressed interest in
AngioDynamics’ PICCs, but conveyed that they could not consider purchasing them because the
PICCS could not be paired with Bard’s TLS; (6) emails from customers explaining their reasons
for choosing Bard’s PICCs over AngioDynamics’ PICCs, including, in some cases, because they
wanted to use Bard’s TLS, which could only be obtained by buying Bard’s PICCs; and (7)
emails from AngioDynamics or Bard employees recounting similar discussions with customers.
See Section II.E supra (summarizing this record evidence in detail).

70

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 71 of 105

While, as Bard argues, some of AngioDynamics’ evidence regarding discussions with
customers appears to include multiple layers of hearsay or is otherwise of questionable
admissibility, much of AngioDynamics’ evidence does not suffer from these problems, and many
of the “double hearsay” problems that do exist could be corrected by, for example, calling the
Bard or AngioDynamics sales representatives who directly spoke with the relevant customers to
testify at trial. See, e.g., Smith Wholesale Co., Inc. v. R.J. Reynolds Tobacco Co., No. 03-cv-30,
2005 WL 8162569, at *5 14 (E.D. Tenn. Feb. 23, 2005) (finding that a “double hearsay” problem
with the plaintiff’s corporate representatives’ testimony about what sales representatives told
them regarding conversations with customers was corrected by calling the sales representatives
themselves as witnesses, and that the sales representatives’ testimony was admissible under Fed.
R. Evid. 803(3)). As discussed previously, see Section IV.B.1 supra, the combination of
evidence that AngioDynamics did, in fact, lose PICC sales to Bard (in the form of deposition
testimony from AngioDynamics witnesses and the parties’ email communications) and evidence
regarding customers’ statements about their motivations for not purchasing, or switching away
from, AngioDynamics PICCs (at least some of which is likely admissible under Fed. R. Evid.
803(3)) constitutes enough evidence that could be presented in admissible form to raise a
genuine dispute of material fact that precludes summary judgment. See, e.g., Callahan, 182 F.3d
at 252-54, 260; Complete Entm’t Res. LLC v. Live Nation Entm’t, Inc., No. 15-cv-9814, 2017
WL 6512223, at *4 n.9, 2017 U.S. Dist. LEXIS 183213, at *11 n.9 (C.D. Cal. Oct. 16, 2017)
(finding that “Plaintiffs’ records of customer statements regarding their reasons for not using

14

No parallel LEXIS citation available.

71

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 72 of 105

Plaintiffs’ services” was “material from which a factfinder could decide that at least some artistclients would have chosen Plaintiff if not for Defendants’ allegedly anticompetitive conduct”). 15
Independent of its evidentiary objections, Bard also argues that the deposition testimony
of AngioDynamics’ witnesses is too speculative and self-serving to constitute evidence of injury
or causation. (Dkt. No. 133-1, at 14-17; Dkt. No. 154, at 9-12). Bard’s arguments attacking
AngioDynamics’ witnesses’ testimony largely address the basis for their knowledge regarding
certain lost opportunities they testified about, their unfamiliarity with particular “potential lost
sales” AngioDynamics listed in its interrogatory responses, their failure to recall certain pertinent
details of discussions with particular customers, and the inconsistency of some witnesses’
recollections about particular customers with other record evidence regarding those customers’
purchasing decisions. (Dkt. No. 133-1, at 14-17; Dkt. No. 154, at 10-12). While some of these
arguments may have merit, they are questions of weight that are best resolved by a fact-finder at
trial, rather than at the summary judgment stage. 16 Bard is correct that AngioDynamics witnesses

Bard attempts to distinguish Callahan on the ground that Callahan “featured extensive [expert] economic evidence
on injury and causation that bolstered the plaintiffs’ testimony.” (Dkt. No. 154, at 9-10). While the Court is not
considering Dr. Frankel’s opinion on causation, and this case is thus distinguishable from Callahan in that respect,
this distinction is not dispositive. While the Callahan court considered the evidence before it collectively, nowhere
did it suggest that the evidence would have been insufficient to raise a triable issue of fact without the expert analysis,
and it is well-established that a plaintiff may prove antitrust injury without the use of expert evidence. See, e.g., Smith
Wholesale Co., Inc., 2005 WL 8162569, at *5 (“Contrary to defendant’s insistence, expert testimony is not necessarily
required to prove antitrust injury if the facts and circumstances otherwise fall within the ambit of F.R.E. 701, and here
they do.”); Cason-Merenda v. Detroit Med. Ctr., 862 F. Supp. 2d 603, 642-43 (E.D. Mich. 2012) (“Plaintiffs correctly
note that they may forge the necessary link between Defendants’ alleged antitrust violations and antitrust injury
without expert testimony.”); Valassis Comms., Inc. v. News Corp., No. 17-cv-7378, 2019 WL 802093, at *12, 2019
U.S. Dist. LEXIS 27770, at *37 (S.D.N.Y. Feb. 21, 2019) (“Any defect in [the plaintiff’s damages expert’s] opinion
on the amount of damages does not undermine [the plaintiff’s] ability to prove the fact of injury.”). Here, the Court
finds that the record as it stands is sufficient to raise a genuine dispute as to injury-in-fact and causation; whether that
evidence is enough to prove AngioDynamics’ case is a question for trial. Bard’s attempt to distinguish Complete
Entm’t Resources on the grounds that that case “concerns ‘records of customer statements’ rather than plaintiff’s own
testimony and deals with admissibility rather than sufficiency” is also inapposite. (Dkt. No. 154, at 10 n.4). Here,
AngioDynamics relies on both its own witnesses’ testimony and documentary evidence regarding its customers’
motivations, and the court in Complete Entm’t Resources was not dealing solely with admissibility, but rather was (as
here) considering whether there was sufficient admissible evidence in the record to raise a triable issue of fact as to
antitrust injury. Id.
15

The line of cases Bard cites for the proposition that the “‘isolated self-serving statements’ of a plaintiff’s corporate
officers may not provide substantial evidence upon which a jury can rely” is inapposite, as those cases involved
16

72

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 73 of 105

cannot testify with certainty as to what a customer would have done had it had the option of
pairing AngioDynamics’ PICCs with Bard’s TLS. However, as executives responsible for the
sale of AngioDynamics’ vascular access products, they can certainly testify as to their personal
knowledge of PICC sales that AngioDynamics lost to Bard and what customers told them
regarding their reasons for not purchasing AngioDynamics’ PICCs. Thus, for summary judgment
purposes, the Court finds that this testimony, when viewed collectively with the record as a
whole, constitutes competent evidence from which a genuine dispute of material fact can be
found. 17
Bard also spends a significant portion of its briefing attacking AngioDynamics’ anecdotal
examples of particular customers choosing Bard’s PICCs over AngioDynamics’ PICCs, arguing,
with respect to each customer, that the relevant documentary evidence and deposition testimony:
(1) demonstrates that the customer had numerous reasons for choosing Bard’s PICCs
independent of its TLS policy; (2) is ambiguous as to the precise reasons for the customer’s
purchasing decision; and/or (3) does not demonstrate that the customer would have chosen to
pair AngioDynamics’ PICCs with Bard’s TLS if given the option to do so, particularly when
accounting for other factors impacting the customer’s decision. (Dkt. No. 133-1, at 20-22; Dkt.
No. 145, at 26-31; Dkt. No. 154, at 6-9). Bard also more generally argues that AngioDynamics

conclusory statements by officers that found no support in, and in many cases flatly contradicted, the evidence in the
record. See Argus Inc. v. Eastman Kodak Co., 801 F.2d 38, 41-46 (2d Cir. 1986); Chrysler Credit Corp. v. J. Truett
Payne Co., 670 F.2d 575, 581-82 (5th Cir. 1982). By contrast, the witness testimony here regarding lost sales and the
reasons for those losses, as a general matter, is not inconsistent with, and finds some support in, the documentary
evidence. See Section II.E.5 supra (reviewing evidence).
17
In a similar vein, Bard also argues that customers’ statements to AngioDynamics regarding their reasons for not
purchasing AngioDynamics’ PICCs (even if admissible to show the customers’ motivations) are inherently unreliable.
(Dkt. No. 133-1, at 16-17). AngioDynamics takes issue with this characterization. (Dkt. No. 144, at 18-20). These
arguments concern the weight the relevant evidence should be given, and are best resolved by a fact-finder at trial
rather than at the summary judgment stage.

73

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 74 of 105

has not demonstrated that Bard’s TLS policy caused AngioDynamics’ injury, given the various
other independent factors affecting customers’ purchasing decisions. (Dkt. No. 133-1, at 25-32;
Dkt. No. 145, at 25-34; Dkt. No. 154, at 6-12). The Court need not address each of Bard’s
arguments about each specific anecdote and relevant factor. Viewing the evidence as a whole
and in the light most favorable to AngioDynamics, the Court finds that a reasonable fact-finder
could conclude that the desire for Bard’s TLS, which could only be purchased preloaded into
Bard’s PICCs, played a significant role in these customers’ decision to choose Bard’s PICCs
over AngioDynamics’ PICCs; that any customer that chose to use Bard’s TLS was necessarily
precluded from using a competitor’s PICCs; that even accounting for other factors at play, at
least some of the customers may have made a different PICC purchasing decision if given the
option to pair AngioDynamics’ PICCs with Bard’s TLS; and that Bard’s TLS policy was one
material cause (even if not the only cause) of customers’ decision not to purchase
AngioDynamics’ PICCs. See Section II.E.5 supra (summarizing the relevant evidence in detail).
Given this, questions about the proper inferences to be drawn from particular pieces of evidence,
whether Bard’s TLS policy was both a but-for and material cause of each particular customer’s
purchasing decision in light of other factors at play, the precise significance of confounding
factors independent of Bard’s TLS policy, and whether the evidence sufficiently demonstrates
that AngioDynamics lost more than a de minimus amount of sales as a result of Bard’s TLS
policy are all disputed factual questions to be resolved at trial. 18
18
While the Court need not address each of Bard’s arguments with respect to the various factors it cites, there is one
argument Bard makes that the Court sees fit to briefly address. Bard argues that the failure of AngioDynamics and its
damages expert to account for the fact that Bard had not obtained FDA approval to market its TLS stylet for use with
competitors’ valved PICCs is fatal to any claim it has with respect to valved PICC sales, citing case law for the
proposition that a “regulatory or legislative bar can break the chain of causation in an antitrust case.” (Dkt. No. 1331, at 31-32 (citing In re Wellbutrin XL Antitrust Litig., 868 F.3d 132, 165 (3d Cir. 2017); In re Ciprofloxacin
Hydrochloride Antitrust Litig., 261 F. Supp. 2d 188, 206 (E.D.N.Y. 2003)). Even assuming (without deciding) that
the case law Bard cites applies here, it does not defeat AngioDynamics’ claim of antitrust injury; this lack of FDA
approval is not a factor for the large number of AngioDynamics’ PICC sales that involve non-valved PICCs.

74

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 75 of 105

The Court rejects Bard’s argument that Bard’s internal emails and documents—and,
particularly, an email from a Bard employee expressing his belief that AngioDynamics’ strategy
“[wa]s to pressure the market to pressure [Bard] into launching [the standalone stylet], so it will
open the door for them to go after special patient populations in our 3CG [TLS] accounts and
pick off 5-10%, which is better than the 0% they are getting at 3CG accounts now,” (Dkt. No.
136-63, at 3; Dkt. No. 136-64, at 2)—cannot serve as evidence of injury or causation. (Dkt. No.
133-1, at 22-24). Statements by Bard’s own employees regarding their beliefs about the impact
of Bard’s allegedly anticompetitive activity are clearly relevant to the question of whether that
activity caused injury to AngioDynamics. Arguments about the proper inferences to be drawn
from those statements or the weight they should be given merely raise factual disputes that are
best reserved for trial.
The Court also rejects Bard’s argument that AngioDynamics’ failure to present any direct
testimony from a hospital representative (at trial or via deposition) regarding the hospital’s
reasons for purchasing Bard’s PICCs over AngioDynamics’ dooms its case. (Dkt. No. 133-1, at
13-14). For the reasons explained, there is sufficient evidence in the record regarding
AngioDynamics’ lost sales and customers’ statements about the reasons for those losses to at
least raise a genuine issue for trial; it is for a fact-finder to decide whether that evidence is
sufficient to meet AngioDynamics’ burden without direct testimony from AngioDynamics’
customers. 19
Bard also argues that AngioDynamics cannot prove injury or causation because it has, at
most, presented evidence that, absent Bard’s TLS policy, some customers would have been
In a similar vein, the Court need not address the parties’ arguments over whether a survey of hospitals or interviews
of hospital representatives—which neither AngioDynamics nor its experts conducted—would have been appropriate
in this case. (Dkt. No. 133-1, at 13, 19-20; Dkt. No. 144, at 24-25; Dkt. No. 154, at 15-16). The question for the Court
is whether the record as it stands raises triable issues of material fact, and the Court finds that it does.

19

75

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 76 of 105

willing to conduct a trial of AngioDynamics’ PICCs with Bard’s TLS and compare the pricing
of AngioDynamics’ PICCs with other options; AngioDynamics has not presented evidence of the
price at which any current Bard TLS customer would have been willing to purchase Bard’s
standalone TLS stylet to combine with AngioDynamics’ PICCs, nor has it shown that any
current Bard TLS customer would have been willing to pay a premium for that option. (Dkt. No.
133-1, at 14, 16, 24, 28; Dkt. No. 145, at 27, 31; Dkt. No. 154, at 6-9). The Court does not find
that Bard is entitled to summary judgment based on this argument. Viewed in the light most
favorable to AngioDynamics, the record suggests that at least some PICC customers were
interested in exploring the option of combining BioFlo PICCs with Bard’s TLS, but Bard’s TLS
policy precluded them from even considering that option, thus depriving AngioDynamics of the
opportunity to even attempt to compete with Bard for those customers on other factors relevant to
hospitals’ purchasing decisions, such as price and clinical effectiveness. See, e.g., Section
II.E.4.c-d supra (summarizing testimony describing such examples). Moreover, record evidence
suggests that PICC pricing is a customer-specific decision, and the question of whether pricing
would deter customers from pairing AngioDynamics’ PICCs with Bard’s TLS absent Bard’s
TLS policy is rife with factual disputes, including how often AngioDynamics’ BioFlo PICCs are
actually priced at a premium to Bard’s PICCs, the size of that premium, how much of a premium
Bard would charge for its single-sterile TLS stylet, the value customers place on BioFlo for its
purported clinical benefits and long-term cost-savings, and the extent to which AngioDynamics
would be able to discount its BioFlo PICCs to induce customers to switch from Bard PICCs. See
Section II.D.1 supra (summarizing the evidence and the parties’ disputes regarding pricing).
Therefore, on this record, the Court cannot conclude as a matter of law that AngioDynamics

76

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 77 of 105

cannot demonstrate a more-than-de-minimus amount of injury directly resulting from Bard’s
TLS policy.
The Court has carefully considered all of Bard’s remaining arguments for summary
judgment and finds them to be without merit. Therefore, for the foregoing reasons, the Court
denies Bard’s motion for summary judgment with respect to injury-in-fact and causation.
d.

AngioDynamics’ Motion for Summary Judgment

While AngioDynamics has raised a triable issue of fact as to injury-in-fact and causation,
it has failed to establish its entitlement to summary judgment with respect to these issues.
Relying on deposition testimony from its own executives, email communications from
customers, and the parties’ internal correspondence, AngioDynamics cites a number of specific
examples of customers who purportedly chose to purchase Bard’s PICCs over AngioDynamics’
because they desired Bard’s TLS and, due to Bard’s TLS policy, had no option to pair it with
AngioDynamics’ PICCs. (Dkt. No. 134-1, at 29-32; Dkt. No. 144, at 9-11, 16, 22-24; Dkt. No.
152, at 17-19). However, as Bard points out, much of this evidence is ambiguous as to the
precise reasons for the customers’ purchasing decisions or suggests that numerous factors were
at play in addition to Bard’s TLS policy, and none of it establishes beyond dispute that any
customer would have purchased AngioDynamics’ PICCs to pair with Bard’s TLS but for Bard’s
TLS policy. (Dkt. No. 133-1, at 20-22; Dkt. No. 145, at 26-31; Dkt. No. 154, at 6-9). While the
Court has already found that Bard’s arguments were insufficient to win its own summary
judgment motion, they are sufficient to raise genuine factual questions as to whether Bard’s TLS
policy was both a but-for and material cause of a more-than-de-minimus amount of PICC sales
that AngioDynamics lost to Bard.
AngioDynamics’ contention that AngioDynamics “sells substantially more PICCs” “in
those segments of the PICC market where Bard’s TLS is not a factor, or less of a factor,” also
77

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 78 of 105

does not entitle it to summary judgment; rather, this is, at most, an argument that
AngioDynamics may use to support its claim at trial. (Dkt. No. 134-1, at 32-33). 20 Given other
factors at play in customers’ purchasing decisions—including some customers’ preference for
purchasing TLSs and PICCs from the same supplier, and possible differences in purchasing
patterns and PICC preferences between hospitals that purchase PICCs for pairing with
navigation-enabled TLSs and those in other segments of the PICC market, (Dkt. No. 146, at ¶ 66
(making this argument and citing record evidence))—the data AngioDynamics presents does not,
in itself, establish beyond dispute that Bard’s TLS policy is a but-for or material cause of
AngioDynamics’ relatively lower success with customers that use navigation-enabled TLS.
The Court has carefully considered all of AngioDynamics’ remaining arguments for
summary judgment and finds them to be without merit. For the foregoing reasons, the Court
denies AngioDynamics’ motion for summary judgment with respect to injury-in-fact and
causation.
3.

Type of Injury the Antitrust Laws Were Designed to Prevent

In addition to establishing injury-in-fact and causation, to establish antitrust injury, an
antitrust plaintiff must prove that its injury is “of the type the antitrust laws were intended to
prevent and that flows from that which makes defendants’ acts unlawful.” Brunswick Corp., 429
U.S. at 489. “Thus, allegations of an injury to a competitor are insufficient unless accompanied
by allegations of injury to competition as well.” See Xerox Corp. v. Media Scis. Intern., Inc., 511

Bard objects to AngioDynamics’ reliance on a chart purportedly showing AngioDynamics’ share of particular PICC
market segments, arguing that it “appears nowhere in Frankel’s reports,” was compiled by “AngioDynamics’ lawyers
pull[ing] numbers from three different appendices to Frankel’s reports and perform[ing] additional calculations,” is
based on unreliable data, and was not timely disclosed as an expert opinion. (Dkt. No. 153, at 7-8 & n.4). For purposes
of the parties’ summary judgment motions, the Court need not decide whether or not consideration of this
demonstrative is appropriate, as it finds that neither party is entitled to summary judgment whether or not the
demonstrative is considered.

20

78

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 79 of 105

F. Supp. 2d 372, 380 (S.D.N.Y. 2007); see also Capital Imaging Assocs., P.C. v. Mohawk Valley
Med. Assocs., Inc., 996 F.2d 537, 547 (2d Cir. 1993) (“[Plaintiff’s] position is simply that it has
been harmed as an individual competitor. It has not shown that defendants’ activities have had
any adverse impact on price, quality, or output of medical services offered to consumers in the
relevant market.”); Daniel v. Am. Bd. of Emergency Med., 428 F.3d 408, 442-43 (2d Cir. 2005)
(holding that plaintiffs failed to demonstrate antitrust injury where their “economic expert
conceded that he had performed no analysis of the consumer effect of defendants’ purportedly
anticompetitive conduct”).
“A competitor’s ‘[e]xclusion from a market is a conventional form of antitrust injury’
because it is ‘exactly the type [of injury] that antitrust laws were designed to prevent and flows
from the competition-reducing aspect of [defendant’s] conduct.’” Valassis Comms., Inc., 2019
WL 802093, at *11, 2019 U.S. Dist. LEXIS 27770, at *32 (collecting cases (quoting Higgins v.
New York Stock Exch., 755 F. Supp. 113, 116 (S.D.N.Y. 1991), aff’d sub nom., Higgins v. New
York Stock Exch., Inc., 942 F.2d 829 (2d Cir. 1991))); Xerox Corp, 511 F. Supp. 2d at 381-82
(“This threatened injury (of direct exclusion from the marketplace) is exactly the type that
antitrust laws were designed to prevent and flows from the competition-reducing aspect of [the
defendant’s] conduct.” (collecting cases)). Moreover, a competitor’s loss of sales constitutes an
injury of the sort protected by the antitrust laws if “it stems from conduct that prevents potential
customers from obtaining a desired product.” Id. at 382 (citing National Assoc. of
Pharmaceutical Mfrs., Inc. v. Ayerst Labs., 850 F.2d 904, 913 (2d Cir. 1988)).
Bard argues that, even if AngioDynamics can demonstrate that it lost sales as a result of
Bard’s TLS policy, AngioDynamics has failed to show that Bard’s actions harmed competition
in the PICC market as a whole, such as by increasing prices or reducing output. (Dkt. No. 133-1,

79

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 80 of 105

at 32-35; Dkt. No. 145, at 34-35; Dkt. No. 154, at 17-18). AngioDynamics responds that Bard’s
TLS policy has led to substantial foreclosure of competition in the PICC market, the threat of
increased concentration in that market, and the restriction of consumer choice, all of which are
the types of injuries the antitrust laws are aimed at. (Dkt. No. 144, at 30-34).
The Court finds that there are genuine disputes of material fact precluding summary
judgment for either party on this issue. AngioDynamics’ claim is that, by forcing customers who
purchase its TLS to also purchase its PICCs, Bard effectively precludes AngioDynamics and
other PICC competitors from competing in the largest segment of the PICC market. If proven
true, this would be “exactly the type [of injury] that antitrust laws were designed to prevent,”
Valassis Comms., Inc., 2019 WL 802093, at *11, 2019 U.S. Dist. LEXIS 27770, at *32; Xerox
Corp, 511 F. Supp. 2d at 381-82, and AngioDynamics “need not wait to make this claim until it
is ‘actually . . . driven from the market and competition is thereby lessened.’” Id., 511 F. Supp.
2d at 381 (quoting Brunswick Corp., 429 U.S. at 489 n. 14). Moreover, there is a question as to
whether Bard’s TLS policy “prevents potential customers from obtaining a desired product,” i.e.
the combination of AngioDynamics’ PICCs and Bard’s TLS. Id., 511 F. Supp. 2d at 382. At the
same time, for the reasons discussed in connection with the Court’s analysis of the
anticompetitive effects element of AngioDynamics’ claim, see Section IV.A.5 supra, there are
genuine disputes over whether the injuries AngioDynamics complains of are, in fact, a result of a
“competition-reducing aspect” of Bard’s conduct (i.e. Bard’s purported foreclosure of
competition in the PICC marketplace through its TLS policy), or are merely the result of lawful
competition and many customers’ preference for Bard’s preloaded PICCs over the other options
available to them. Valassis Comms., Inc., 2019 WL 802093, at *11, 2019 U.S. Dist. LEXIS

80

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 81 of 105

27770, at *32. Therefore, the Court denies both parties’ motions for summary judgment on this
issue. 21
4.

Antitrust Injury Summary

For the foregoing reasons, the Court finds that there are triable issues of fact with respect
to every component of the antitrust injury element of AngioDynamics’ claim. Therefore, the
Court denies both parties’ motions for summary judgment with respect to antitrust injury.
C.

Damages/Bard’s Motion to Exclude Dr. Frankel’s Testimony

Bard moves to exclude the testimony of AngioDynamics’ causation and damages expert,
Dr. Frankel, (Dkt. No. 132), and argues that “[s]ince AngioDynamics offers no evidence of any
damages other than through Dr. Frankel, the exclusion of his testimony means that
AngioDynamics cannot create a triable issue as to damages.” (Dkt. No. 133-1, at 36).
Under Rule 702 of the Federal Rules of Evidence, the Court is charged with a
“gatekeeping” obligation with respect to expert testimony: the trial judge must ensure “that an
expert’s testimony both rests on a reliable foundation and is relevant to the task at hand.”
Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 597 (1993). Rule 702 provides:
A witness who is qualified as an expert by knowledge, skill, experience, training, or
education may testify in the form of an opinion or otherwise if: (a) the expert’s scientific,
technical, or other specialized knowledge will help the trier of fact to understand the
evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or
data; (c) the testimony is the product of reliable principles and methods; and (d) the
expert has reliably applied the principles and methods to the facts of the case.
Bard analogizes to a Tenth Circuit case, Suture Express, Inc. v. Owens & Minor Distribution, Inc., 851 F.3d 1029,
1044 (10th Cir. 2017), in which the Court observed that significant evidence of a “market that is becoming more, not
less, competitive” “raise[d] questions about . . . whether it was really competition that was harmed instead of simply
one competitor,” and found that “[t]here is simply not enough probative evidence for a jury to find that [the
defendants’] bundling practices constitute an injury of the kind the antitrust laws are intended to prevent.” Id. at 1045.
Here, while the record contains some indica of a competitive PICC marketplace, see Section IV.A.5 supra, there is
also evidence that Bard’s TLS policy prevents competitors like AngioDynamics from being able to meaningfully
compete for PICC customers that use Bard’s TLS, and that Bard dominates the market (especially with respect to
PICCs that use TLSs) to a much greater extent than any defendant in Suture Express. Thus, unlike in Suture Express,
then, the evidence of robust competition notwithstanding Bard’s TLS policy is not so overwhelming as to preclude a
finding that Bard’s actions have harmed competition in the market.
21

81

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 82 of 105

Fed. R. Evid. 702. “To determine whether a witness qualifies as an expert, courts compare the
area in which the witness has superior knowledge, education, experience, or skill with the subject
matter of the proffered testimony.” United States v. Tin Yat Chin, 371 F.3d 31, 40 (2d Cir. 2004).
“Under Daubert, factors relevant to determining reliability include the theory’s
testability, the extent to which it has been subjected to peer review and publication, the extent to
which a technique is subject to standards controlling the technique’s operation, the known or
potential rate of error, and the degree of acceptance within the relevant scientific community.”
Restivo v. Hessemann, 846 F.3d 547, 575-76 (2d Cir. 2017) (internal quotation marks omitted).
The reliability inquiry is “a flexible one,” Daubert, 509 U.S. at 594, and the factors to be
considered “depend[ ] upon the particular circumstances of the particular case at issue.” Kumho
Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 150 (1999). When applying the gatekeeping
obligation to non-scientific testimony, a district court may choose to utilize some or all of the
above Daubert factors, or it may look to other indicia of reliability. Id.
“In undertaking this flexible inquiry, the district court must focus on the principles and
methodology employed by the expert, without regard to the conclusions the expert has reached
or the district court’s belief as to the correctness of those conclusions.” Amorgianos v. Natl. R.R.
Passenger Corp., 303 F.3d 256, 266 (2d Cir. 2002). “Thus, when an expert opinion is based on
data, a methodology, or studies that are simply inadequate to support the conclusions reached,
Daubert and Rule 702 mandate the exclusion of that unreliable opinion testimony.” Id. In other
words, “[a] court may conclude that there is simply too great an analytical gap between the data
and the opinion proffered.” Gen. Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997). “Frequently,
though, ‘gaps or inconsistencies in the reasoning leading to [the expert’s] opinion . . . go to the

82

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 83 of 105

weight of the evidence, not to its admissibility.’” Restivo, 846 F.3d at 577 (quoting Campbell ex
rel. Campbell v. Metro. Prop. & Cas. Ins. Co., 239 F.3d 179, 186 (2d Cir. 2001)).
Ultimately, a district court has “the same broad latitude when it decides how to determine
reliability as it enjoys in respect to its ultimate reliability determination.” Gen Elec. Co., 522
U.S. at 142. Moreover, “[v]igorous cross-examination, presentation of contrary evidence and
careful instruction on the burden of proof are the traditional and appropriate means of attacking
shaky but admissible evidence.” Daubert, 509 U.S. at 596. With this framework in mind, the
Court addresses Bard’s arguments for exclusion of Dr. Frankel’s opinions. 22
1.

Causation
a.

Dr. Frankel’s Causation Opinion

As to causation, Dr. Frankel concludes that Bard’s TLS policy caused injury to
AngioDynamics based on the following facts (stated in Dr. Frankel’s words):
•

Bard’s dominant share of TLS devices used in the United States and the large
share of PICCs placed with the assistance of TLS devices.

•

The ability of manufacturers to sell TLS devices and navigation stylets separately
from their PICCs, and the actual use of alternative suppliers’ PICCs with a TLS
when that option is available.

•

The preference some customers would have had for AngioDynamics BioFlo
PICCs due to BioFlo’s antithrombotic 23 properties.

•

The interest expressed by customers seeking to use AngioDynamics PICCs with
Bard TLS devices.

•

Bard’s recognition that some of its TLS customers would purchase BioFlo PICCs
if Bard sold stylets separately.

22
In the following discussion, the Court summarizes only the portions of Dr. Frankel’s opinions that are relevant to
the pending motions.

While AngioDynamics disavows any claim that it markets BioFlo PICCs as having “antithrombotic” properties and
clarifies that it only claims BioFlo PICCs reduce “thrombus accumulation,” (Dkt. No. 144-1, ¶¶ 34-36), Dr. Frankel
uses the term “antithrombotic properties” throughout his report.
23

83

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 84 of 105

•

Bard’s refusal to sell its stylets separately from its PICCs.

(Dkt. No. 132-5, at 13-14). To arrive at the foregoing factual predicates, Dr. Frankel reviews,
interprets and draws inferences from documents and data in the record, much of which overlaps
with the evidence AngioDynamics relies on in its summary judgment motion. (Id. at 14-23).
Based on the evidence he reviews, Dr. Frankel concludes that “Bard’s refusal to sell stylets
separately from its PICCs harmed AngioDynamics by preventing it from selling PICCs to Bard
TLS customers that otherwise would have preferred to use AngioDynamics PICCs for some or
all of their bedside PICC placements,” and that “Bard’s tie enabled it to obtain a larger
percentage of U.S. PICC sales than otherwise, and suppressed AngioDynamics’ sales and
profits.” (Id. at 23-24).
b.

Bard’s Motion to Exclude Dr. Frankel’s Causation Opinion

Bard moves to exclude Dr. Frankel’s causation opinion, (Dkt. No. 132-5, at 13-23), on
the grounds that: (1) Dr. Frankel “did no economic (or any other) analysis to determine whether
Bard’s conduct caused hospitals to purchase Bard’s PICCs instead of AngioDynamics’ PICCs,
or whether hospitals would have purchased more AngioDynamics PICCs but for Bard’s TLS
policy,” but instead “assumes that the jury will find causation—the very opinion he claims to be
offering”; (2) Dr. Frankel’s opinion does little more than [serve] as a conduit to publish the
[record] evidence, at least some of which is likely inadmissible, to the jury, adding the patina of
an expert economist to mere fact information”; and (3) Dr. Frankel selectively relies on evidence
that supports his opinion while ignoring evidence that contradicts it, and the evidence he relies
on fails to establish that there are customers “who do not already purchase AngioDynamics’
PICCs, who would purchase them but for the challenged conduct.” (Dkt. No. 132-1, at 11-18; see
also Dkt. No. 153, at 5-8).

84

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 85 of 105

AngioDynamics responds that: (1) Dr. Frankel’s opinion does not assume a finding on
causation, as Bard claims, but on the separate liability element of coercion, which is a proper
assumption for a causation and damages report; (2) Dr. Frankel’s opinion does not merely
“‘narrate’ factual evidence,” but “analyze[s] contemporaneous documents and data and reache[s]
a conclusion based on that analysis,” which is “precisely what expert economists do in antitrust
litigation”; and (3) Dr. Frankel “review[s] the structure of the industry, sales data, and record
evidence of Bard’s conduct and applie[s] well-established economic principles to reach each of
his conclusions,” which is “precisely what FRE 702 requires,” and his opinion is well-supported
by the evidence he relies on and the record as a whole. (Dkt. No. 143, at 13-20).
As an initial matter, the Court rejects Bard’s argument that Dr. Frankel’s opinion
improperly assumes a conclusion. In his deposition, Dr. Frankel acknowledged that his analysis
“assum[es] a finding of liability,” i.e. “a conclusion by the fact-finder that Bard coerced some
hospitals to purchase its PICCs . . . when otherwise they would have purchased Angio’s PICCs,”
but he explicitly disavowed the suggestion that his analysis “assum[es] that Angio has suffered
injury as a result of Bard’s conduct,” instead confirming that his analysis both “offer[s] [an]
opinion [that AngioDynamics was injured by Bard’s conduct],” and also “quantif[ies] the
damages associated with that injury.” (Dkt. No. 132-3, at 143-46). While the language Dr.
Frankel and his examiner used in the deposition is somewhat imprecise, from a review of Dr.
Frankel’s expert report and testimony, it appears clear that his analysis presupposes a conclusion
on liability, but not on the issues on which he was asked to opine, i.e. causation and damages.
(Dkt. No. 132-5, at 13-23). Bard argues that the “supposed distinction [between coercion and
causation] is irrelevant here because the nature of this tying claim requires that there be coercion
in order for there to have been causation.” (Dkt. No. 153, at 5-6). However, as an expert on

85

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 86 of 105

causation and damages rather than liability, it was appropriate for Dr. Frankel to do what he did
here, i.e. assume a finding that Bard adopted a coercive tying policy, and then perform an
analysis of the separate question of whether that coercive policy actually caused AngioDynamics
to lose any PICC sales that it otherwise would have made.
Nonetheless, the Court agrees with Bard that Dr. Frankel’s causation opinion must be
excluded because it does little more than summarize record evidence (including sales and market
share data and documents produced in this litigation) and lend Dr. Frankel’s expert credentials to
AngioDynamics’ interpretation of that evidence. Dr. Frankel offers no specialized economic
analysis that would assist a fact-finder in interpreting the record evidence he relies on. Rather, he
recites that evidence, draws inferences that a fact-finder could glean from merely examining the
evidence itself, and concludes from those inferences that the evidence establishes causation—a
conclusion a fact-finder is perfectly capable of making based on the evidence without the aid of
expert testimony. Such an opinion, without more, does not pass muster under Rule 702. See, e.g.,
In re Namenda Direct Purchaser Antitrust Litig., 331 F. Supp. 3d 152, 170-72 (S.D.N.Y. 2018)
(barring expert testimony that “adds nothing to the direct evidence that cannot be elicited from
fact witnesses—nothing other than the use of [the expert’s] impressive credentials to bolster the
credibility of those witnesses impermissibly”); 24 Hernandez v. Leichliter, No. 14-cv-5500, 2016
WL 684038, at *2, 2016 U.S. Dist. LEXIS 19728, at *5 (S.D.N.Y. Feb. 18, 2016) (“To the extent
[the expert] merely repeats or recasts the testimony of [Plaintiff] in order to arrive at a theory of

AngioDynamics argues that Namenda is distinguishable because there “the Court excluded the testimony of an
expert who failed to analyze contemporaneous documents and data and instead relied only on her industry experience
and fact witness testimony in forming her opinion,” whereas “Dr. Frankel analyzed contemporaneous documents and
data and reached a conclusion based on that analysis.” (Dkt. No. 143, at 19). Bard points out that this distinction
ignores the fact that the Namenda court “excluded the expert’s testimony both because the expert failed to rely on
sufficient facts or data, and for a separate reason (that Bard cites)—the proposed expert testimony would simply
regurgitate information that could be introduced through fact witnesses.” (Dkt. No. 153, at 7 n.3). The Court agrees
with Bard.
24

86

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 87 of 105

causation, he is not testifying as an expert witness based upon specialized knowledge, but rather
is acting as a conduit for another witness’s testimony in the guise of an expert’s opinion.”)
(citation omitted); Palazzetti Imp./Exp., Inc. v. Morson, No. 98-cv-722, 2001 WL 793322, at *23, 2001 U.S. Dist. LEXIS 9538, at *8 (S.D.N.Y. July 13, 2001) (holding that “proposed expert
testimony regarding the elements of a franchise agreement and their alleged absence here does
not meet the standard of Rule 702” because “neither of the elements of a franchise agreement
requires knowledge beyond the ken of the average juror” and there was “nothing . . . to believe
that the jurors would be assisted (rather than improperly swayed) by the testimony of the
expert”).
This case is unlike other antitrust cases involving claims of unlawful collusion (some of
which AngioDynamics cites), in which experts have been permitted to testify as to whether
conduct or market conditions reflected in the record evidence were consistent with the existence
of an anticompetitive conspiracy, on the grounds that such an analysis requires specialized
knowledge outside the province of an ordinary juror. See, e.g., U.S. Info. Sys., Inc. v. Int’l Bhd. of
Elect. Workers Local Union No. 3, AFL-CIO, 313 F. Supp. 2d 213, 239-41 (S.D.N.Y. 2004)
(finding that the plaintiff’s expert could not draw a legal conclusion on whether or not the
defendants engaged in anticompetitive conduct, but could “point to factors that would tend to
show anticompetitive conduct in a market,” “indicate whether he believed those factors existed
here, and what the economic significance of those factors would be,” “explain how certain
conduct could affect a market through the use of hypothetical statements” and “hypothesize that
if certain conduct did occur, economists would expect the market to react in a particular way”);
Fleischman v. Albany Med. Ctr., 728 F. Supp. 2d 130, 151-52 (N.D.N.Y. 2010) (“[Plaintiffs’
expert] seeks to testify that the information exchanges, engaged in by the Defendant hospitals,

87

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 88 of 105

made it easier for Defendants to form and maintain a wage agreement. This type of opinion may
be helpful to the jury and is admissible.”). Here, by contrast, Dr. Frankel brings no specialized
economic analysis to bear in his interpretation of the record evidence. The conclusions he draws
from that evidence consist of arguments AngioDynamics’ lawyers can readily make without the
aid of expert analysis, and are all well within the province of an ordinary, intelligent juror.
Therefore, the Court grants Bard’s motion to exclude Frankel’s causation opinion. 25
2.

Damages
a.

Legal Standard for Antitrust Damages

The Court evaluates Bard’s motion to exclude Dr. Frankel’s damages opinion in light of
the legal standard for proving antitrust damages. “[T]he actual amount [of an antitrust plaintiff’s
damages] need not be proven to the same degree of certainty as proving some quantum of
damages,” given the difficulty (and, at times, impossibility) of accurately constructing a
hypothetical world untainted by the defendant’s challenged conduct. Drug Mart Pharm. Corp. v.
Am. Home Prod. Corp., 472 F. Supp. 2d 385, 424 (S.D.N.Y. 2007) (citing Story Parchment Co.
v. Paterson Parchment Paper Co., 282 U.S. 555, 562 (1931)). The “wrongdoer is not entitled to
complain that [the damages] cannot be measured with the exactness and precision that would be
possible if the case, which he alone is responsible for making, were otherwise.” Story Parchment
Co., 282 U.S. at 563. Indeed, “[t]he most elementary conceptions of justice and public policy
require that the wrongdoer shall bear the risk of the uncertainty which his own wrong has
created.” Bigelow v. RKO Radio Pictures, 327 U.S. 251, 264-65 (1946), reh’g denied, 327 U.S.
817 (1946). Therefore, an antitrust plaintiff need only “come forward with substantial evidence

Because the Court has found that Dr. Frankel’s causation opinion must be excluded on the foregoing grounds, it
need not address the parties’ remaining arguments regarding that opinion. To the extent these arguments amount to
disputes over whether the record supports a finding of causation, the Court has addressed this question in resolving
the parties’ cross-motions for summary judgment on antitrust injury. See Section IV.B supra.
25

88

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 89 of 105

from which a jury can determine the amount of damages from a ‘just and reasonable estimate of
the damage based upon relevant data.’” Drug Mart Pharm. Corp., 472 F. Supp. 2d at 424
(quoting Bigelow, 327 U.S. at 264); see also, e.g., New York v. Julius Nasso Concrete Corp., 202
F.3d 82, 88 (2d Cir. 2000) (“Where . . . there is a dearth of market information unaffected by the
collusive action of the defendants, the plaintiff’s burden of proving damages, is, to an extent,
lightened.”).
At the same time, “even where the defendant by his own wrong has prevented a more
precise computation, the jury may not render a verdict based on speculation or guesswork.” SCM
Corp. v. Xerox Corp., 645 F.2d 1195, 1212 (2d Cir. 1981) (quoting Bigelow, 327 U.S. at 264).
Moreover, a more lenient standard in proving the amount of damages should only be applied
where the plaintiff puts forth “proof of defendant’s wrongful acts and their tending to injure
plaintiffs’ business, and from evidence in the decline of prices, profits and values, not shown to
be attributable to other causes.” Bigelow, 327 U.S. at 264 (emphasis added); see also MCI
Communications Corp., 708 F.2d at 1162 (“When a plaintiff improperly attributes all losses to a
defendant’s illegal acts, despite the presence of significant other factors, the evidence does not
permit a jury to make a reasonable and principled estimate of the amount of damages. This is
precisely the type of speculation or guesswork not permitted for antitrust jury verdicts.”)
(internal marks omitted).
b.

Dr. Frankel’s Damages Opinion

Dr. Frankel calculates AngioDynamics’ purported damages resulting from Bard’s TLS
policy by estimating its lost profits, using a formula that includes, among other variables, the
estimated percentage of Bard’s TLS‐paired PICC sales that would have been won by
AngioDynamics had Bard not engaged in alleged tying. For purposes of its motion in limine,

89

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 90 of 105

Bard only challenges Dr. Frankel’s methodology for determining that variable, and that is
therefore where the Court focuses its discussion.
Dr. Frankel determines this percentage using a “benchmark” or “yardstick” analysis. As a
first step, Dr. Frankel “consider[s] a number of potential indicators of AngioDynamics’
prospective success at selling PICCs to users of Bard TLS devices” based on documents in the
record, and selects two of these “indicators” as benchmarks to use in calculating
AngioDynamics’ lost profits. (Id. at 28-29). 26
As his first benchmark (“Benchmark 1”), Dr. Frankel selects “AngioDynamics’ share of
PICCs used with Teleflex TLS devices.” (Id. at 29). Dr. Frankel derives that percentage—
percent—from an email sent by Teleflex’s outside counsel containing a rough estimate of the
percentage of its standalone stylets that are used with AngioDynamics PICCs. (Id. at 29-30; see
also id. at 42 (explaining Dr. Frankel’s calculation of Benchmark 1)). Dr. Frankel testified that
he felt this benchmark was justified because Teleflex offers a single-sterile stylet and “Angio is
able to sell PICCs that are used

percent of the time with that system,” which is an “obvious

analogue to look at as a piece of information that bears on” how successful AngioDynamics
would be if Bard offered a single-sterile stylet as Teleflex does. (Dkt. No. 132-3, at 232).
For his second benchmark (“Benchmark 2”), Dr. Frankel conducts “a comparison of
Teleflex’s share of sales of PICCs used with its own TLS devices and Teleflex’s share of
unguided nursing PICC sales,” and “use[s] the ratio of . . . non‐Teleflex shares – the non‐
Teleflex share of Teleflex TLS usage divided by the non‐Teleflex share of unguided nursing
PICCs used – as a benchmark for Bard.” (Dkt. No. 132-5, at 31). This benchmark “assumes [that,
At oral argument, AngioDynamics’ counsel suggested that Dr. Frankel’s other potential “indicators” could serve as
a basis for a damages model if the Court excludes Dr. Frankel’s benchmarking analysis; Bard’s counsel disagreed.
(Dkt. No. 187, at 64-67, 81-82). As Bard has only moved to exclude Dr. Frankel’s benchmarking analysis, the Court
only addresses the issue that has been presented and briefed by the parties.
26

90

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 91 of 105

absent the TLS policy,] Bard would have had the same relative increased success as Teleflex in
selling PICCs for use with its own TLS devices,” and that non-Bard PICC manufacturers “would
have been relatively less successful at selling PICCs to Bard TLS users to the same extent non‐
Teleflex PICC manufacturers have less relative success selling PICCs to Teleflex TLS users.”
(Id.). Based on these assumptions, Dr. Frankel determines that “[d]uring the period 2013‐2018 . .
. [absent the TLS policy,] Bard would have sold
devices,” while “[t]he remaining

percent of PICCs used with Bard TLS

percent of PICCs—i.e., Bard TLS-guided PICCs that

would not have been sold by Bard absent Bard’s tie—would have been sold by the other PICC
suppliers.” (Id. at 31-32). Dr. Frankel then “allocate[s] those sales across those suppliers,
including AngioDynamics, pro‐rata based on their share of non‐Bard unguided nursing PICCs
sold in the United States.” (Id. at 32). Using this methodology, Dr. Frankel finds that
AngioDynamics “would have accounted for

percent of U.S. PICC sales used with Bard TLS

devices.” (Id.). In his deposition testimony, Dr. Frankel described Benchmark 2 as a “more
conservative alternative” to Benchmark 1, designed to account for the fact that “Bard might do
particularly well in selling PICCs to its own customers, just as Teleflex does particularly well in
selling PICCs to its TLS customers.” (Dkt. No. 132-3, at 232).
Dr. Frankel testified that he “[doesn’t] have a strong opinion as to which of [his
benchmarks] is better.” (Id.). He stated that his benchmark analysis is designed to “suggest[] to a
jury that . . . we can’t know with precision this variable because of [Bard’s alleged
anticompetitive] conduct, and there’s a reasonable range spanning between these two
benchmarks. And I leave it to them to make a reasonable choice or their own choice based on the
best I can do, which is to present this range of outcomes.” (Id. at 231).

91

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 92 of 105

c.

Bard’s Motion to Exclude Dr. Frankel’s Damages Opinion

Bard moves to exclude Dr. Frankel’s damages opinion on three grounds: (1) Dr. Frankel
bases his analysis on unreliable data; (2) Dr. Frankel’s analysis relies on benchmarks that have
been affected and distorted by Bard’s challenged conduct; and (3) Dr. Frankel failed to
sufficiently analyze the comparability of his chosen benchmarks. (Dkt. No. 132-1, at 18-28).
AngioDynamics generally responds that Bard’s arguments regarding Dr. Frankel’s
benchmarking methodology go to the methodology’s weight, not its admissibility, and do not
provide grounds for exclusion. (Dkt. No. 143, at 20-31).
i.

Reliability of Dr. Frankel’s Data

Bard argues that Dr. Frankel’s damages methodology must be excluded on the grounds
that it is based on unreliable data. Specifically, the percentage of Teleflex standalone stylets used
with AngioDynamics PICCs, which Dr. Frankel uses as the basis of both of his benchmarks,
comes from the following excerpt from an email sent by Teleflex’s outside counsel several days
before Dr. Frankel served his expert report:
Teleflex believes that the vast majority (in excess of
) of non-preloaded PICCs are
being sold for use with another company’s stylet. Based on sales personnel input in May
2019, Teleflex roughly estimates that approximately
of the standalone stylet sales is
with Bard Catheters.
[sic] used with AngioDynamics catheters and approximately
However, Teleflex’s sales data does not capture this information and accordingly
Teleflex cannot confirm these estimates.
(Dkt. No. 147-23, at 3). Bard contends that this “rough[] estimate[]” is insufficiently reliable to
pass muster under Daubert, given that the email does not indicate (and Dr. Frankel does not
know) the time frame to which it applies, how it was determined, what information it was based
on, which sales personnel contributed, or what their titles or seniority levels were. (Dkt. No. 1321, at 25-28; Dkt. No. 153, at 12-14). AngioDynamics replies that “experts are granted wide
latitude with respect to the evidence and data on which they rely”; that Dr. Frankel testified that

92

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 93 of 105

“the Teleflex estimate is precisely the type of information on which economists typically rely”;
that the estimate was “provided by Teleflex in response to subpoenas by both parties in the
normal course of discovery and was accepted in satisfaction of those subpoenas by both parties”;
that “Teleflex sales personnel are in the best position to estimate [the number of Teleflex TLS
Stylets that are paired with AngioDynamics’ PICCs] based on their daily interactions with
customers in the field”; and that “Dr. Frankel undertook independent analysis to verify the
accuracy of the estimate.” (Dkt. No. 143, at 27-31).
As a preliminary matter, AngioDynamics is correct that experts have wide latitude with
respect to the data on which they rely, and challenges concerning the reliability of an expert’s
data often present questions of weight, rather than admissibility. See, e.g., Allen v. Dairy
Farmers of Am., Inc., No. 09-cv-230, 2014 WL 266290, at *9, 2014 U.S. Dist. LEXIS 8343, at
*27 (D. Vt. Jan. 23, 2014) (“Most ‘challenge[s] to the facts or data relied upon by [an expert do]
not go to the admissibility of his testimony, but only to the weight of his testimony.’” (quoting
Aventis Envtl Sci. USA LP v. Scotts Co., 383 F. Supp. 2d 488, 514 (S.D.N.Y. 2005))); Green
Mountain Chrysler Plymouth Dodge Jeep v. Crombie, 508 F. Supp. 2d 295, 325 (D. Vt. 2007)
(“[L]imitations in some of the data on which [an expert] relied goes to the weight of [the
expert’s] testimony . . . , not its admissibility.”); Hartle v. FirstEnergy Generation Corp., No. 08cv-1019, 2014 WL 1317702, at *9, 2014 U.S. Dist. LEXIS 43033, at *29 (W.D. Pa. Mar. 31,
2014) (“Even if the data relied on by the expert is ‘imperfect, and more (or different) data might
have resulted in a ‘better’ or more ‘accurate’ estimate in the absolute sense, it is not the district
court’s role under Daubert to evaluate the correctness of facts underlying an expert’s
testimony.’” (quoting i4i Ltd. P’ship v. Microsoft Corp., 598 F.3d 831, 856 (Fed. Cir. 2010))).
Nonetheless, this general principle does not divest the Court of its gatekeeping obligation to

93

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 94 of 105

ensure that an expert’s testimony is “based on sufficient facts or data” to pass muster under Rule
702.
Bard relies heavily on ZF Meritor, LLC v. Eaton Corp., 696 F.3d 254 (3d Cir. 2012), a
Third Circuit case that illustrates the circumstances under which an expert’s data is so unreliable
as to warrant exclusion. In that case, the plaintiff’s expert based his analysis on a one-page
forecast that the company prepared to project its estimated sales volume, prices, manufacturing
costs, operating expenses, and other criteria. Id. at 291 & n.23, 292. The expert used that forecast
to “estimate[] the incremental revenues that [the plaintiff] would have earned ‘but for’ [the
defendant’s] anticompetitive conduct, and then subtracted from that figure the incremental cost
that [the plaintiff] would have had to incur to achieve such incremental sales.” Id. at 291-92. The
district court excluded the expert’s testimony, finding it unreliable for “rel[ying] on a one-page
set of profit and volume projections without knowing the circumstances under which such
projections were created or the assumptions on which they were based.” Id. at 292. The Third
Circuit affirmed, explaining that “[a]n expert’s lack of familiarity with the methods and the
reasons underlying [someone else’s] projections virtually preclude[s] any assessment of the
validity of the projections through crossexamination.” Id. at 293. Even though the expert “was
generally aware of the circumstances under which the [one-page forecast] was created and the
purposes for which it was used,” he “did not know who initially calculated the . . . figures . . . [,]
did not know whether the . . . projections were calculated by [the plaintiff’s] management, lower
level employees at [the plaintiff], or came from some outside source . . ., [and did not] know the
methodology used to create the [one-page forecast] or the assumptions on which the . . . price
and volume estimates were based.” Id. Therefore, the Third Circuit held that the district court
was justified in excluding his opinion. Id.

94

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 95 of 105

AngioDynamics argues that this case is more analogous to cases in which “ZF Meritor
has been routinely distinguished.” (Dkt. No. 143, at 30 n.15). AngioDynamics relies on Apotex,
Inc. v. Cephalon, Inc., 321 F.R.D. 220 (E.D. Pa. 2017). In Apotex, like in ZF Meritor, the
plaintiffs’ expert relied on the plaintiffs’ internal market share projections. Id. at 232. However,
the court found its case distinguishable from ZF Meritor, in part because the expert in Apotex
had also considered corroborating sources such as “a series of [the plaintiff’s] internal
projections” and “the deposition testimony of . . . [the plaintiff’s] former President” who
“reviewed and signed off on the market share forecasts relied upon by [the expert], and generally
explained how they were created.” Id. at 232-33. Thus, the court found that the Apotex expert
was “was clearly more informed about how the market share projections were created than the
expert in ZF Meritor,” and concluded that “[w]hether [the expert] relied on the best data in
forming his opinions is a question for the jury.” Id. at 233. AngioDynamics also cites other cases
in which courts similarly distinguished ZF Meritor. See e.g., In re SemCrude L.P., 648 Fed.
Appx. 205, 214 (3d Cir. 2016) (finding expert’s use of contemporaneous valuation report reliable
where the expert explained his reliance on the valuation and adjusted it based on his own
analysis); Insight Equity v. Transitions Optical, Inc., 252 F. Supp. 3d 382, 396 (D. Del. 2017)
(finding expert’s use of projections reliable where “the estimates were actually used in the course
of business and incorporated actual sales data”); In re Mushroom Direct Purchaser Antitrust
Litig., No. 06-cv-0620, 2015 WL 5767415, at *17 n.10, 2015 U.S. Dist. LEXIS 120892, at *6771 n.10 (E.D. Pa. July 29, 2015) (finding expert’s use of contemporaneous internal estimates
reliable, because the expert had “grounded his opinion on factual evidence in the record
produced by defendants”).

95

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 96 of 105

As the cases AngioDynamics cites demonstrate, ZF Meritor is not the typical case; rather,
it represents an extreme category of cases in which the data relied on by an expert is so patently
unreliable as to render it inadmissible. However, the Court finds that the data Dr. Frankel relies
on here falls into that category. In his deposition, Dr. Frankel acknowledged that he “know[s] no
further details [about the data] other than what’s in” the email, including which sales personnel
the email refers to, what their “seniority” or “experience base” was, what “products they were
responsible for selling or the geographic region that they knew about,” the “circumstances under
which the unnamed Teleflex sales personnel came up with this estimate,” what Teleflex did to
“aggregate and estimate” the figure, or the specific time period the data covered. (Dkt. No. 1323, at 260-63). Dr. Frankel’s ignorance as to the most basic facts underlying the data he relies on
is precisely the “lack of familiarity” that the Third Circuit in ZF Meritor found “preclude[s] any
assessment of the validity of the projections through crossexamination,” and thus warrants
exclusion. ZF Meritor, 696 F.3d at 293. The cases cited by AngioDynamics are inapposite, as
each of them involved experts who could point to at least some corroborating information or
record evidence demonstrating how the data was created or validating its reliability, which is
utterly lacking here.
It is no answer that “Teleflex is in the best position to estimate and report this
information,” as Dr. Frankel asserts in his rebuttal report, (Dkt. No. 132-6, at 14), or that the data
was provided by Teleflex’s counsel in response to discovery subpoenas, as AngioDynamics
argues, (Dkt. No. 143, at 28). Assuming that the data provided in Teleflex’s counsel’s email is
sufficiently reliable simply because Teleflex provided it, as Dr. Frankel did here, ignores the
email’s numerous caveats that the data constitutes a “rough estimate” and that “Teleflex cannot
confirm these estimates.” (Dkt. No. 147-23, at 3). It also ignores the fact that the email provides

96

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 97 of 105

none of the information necessary to confirm that it is suitable for use in Dr. Frankel’s analysis,
including such information as how the “rough estimate” was determined, what form of “sales
personnel input” was used to calculate it (given that “Teleflex’s sales data does not capture this
information”), and whether those estimates applied to the time period Dr. Frankel examined in
his analysis. (Id.).
AngioDynamics also attempts to distinguish ZF Meritor on the grounds that the data at
issue in ZF Meritor involved profit and volume projections, whereas the data at issue here is “an
estimate concerning actual sales” that came from “sales personnel, who have direct knowledge
on which to base such an estimate.” (Dkt. No. 143, at 29). This argument fails in light of
Teleflex’s caveats that its “sales data does not capture this information” and that it “cannot
confirm these estimates.” (Dkt. No. 147-23, at 3). Indeed, nothing in the information Dr. Frankel
utilized indicates what “sales personnel input” means, the methodology by which this “input”
was gathered from salespeople and aggregated to compute the estimates in Teleflex’s counsel’s
email, and whether that methodology was sufficiently sound and reliable to justify the use of the
estimates in a damages analysis. .
AngioDynamics further contends that the Court should accept Dr. Frankel’s data, despite
its limitations, because it was the best data available in this litigation, and because Dr. Frankel
testified that “the Teleflex estimate is precisely the type of information on which economists
typically rely.” (Dkt. No. 143, at 27-28; Dkt. No. 132-3, at 261-62, 266; Dkt. No. 132-6, at 14).
But it is well established that an expert’s “lack of access to reliable data does not justify use of
unreliable data, and militates against admission under Daubert.” Danley v. Bayer (In re Mirena
IUD Prods. Liab. Litig.), 169 F. Supp. 3d 396, 445 (S.D.N.Y. 2016). And Dr. Frankel’s
conclusory insistence that the Teleflex data is the type that an economist would normally rely on

97

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 98 of 105

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 99 of 105

computed using Teleflex’s data of questionable reliability. Dr. Frankel insists that the list of
customers AngioDynamics provided is not comprehensive since AngioDynamics personnel
“knew some of [the customers that were using Teleflex TLS stylets with AngioDynamics
PICCs], but not [all] of them,” meaning the actual percentage of such customers is greater than
the figure he was able to confirm. (Dkt. No. 132-3, at 268). Accepting this as true, this at most
establishes that the actual percentage of Teleflex TLS customers using AngioDynamics PICCs is
somewhere above the figure he was able to confirm; it says nothing about what that percentage
actually is, and it certainly does not validate the figure Dr. Frankel used in Benchmark 1. For the
foregoing reasons, the Court finds that Dr. Frankel’s benchmarking analysis is based on
unreliable data for purposes of Daubert.
ii.

Bard’s Other Arguments for Exclusion

Bard offers two other reasons why Dr. Frankel’s benchmarking analysis should be
excluded under Daubert. First, Bard argues that Dr. Frankel’s benchmarks are “tainted” by the
challenged conduct. (Dkt. No. 132-1, at 18-21; Dkt. No. 153, at 8-11). 28 Second, Bard argues
that Dr. Frankel “performed no analysis to ensure that Bard’s customers are sufficiently similar
to Teleflex’s so that it is fair and reliable to assume that they would have combined their
respective TLSs with AngioDynamics’ PICCs in the same proportion”; Bard specifically points
to Dr. Frankel’s failure to account for the fact that Teleflex TLS users have already demonstrated
“a strong preference for BioFlo, since this alternative is more expensive than a Bard preloaded

As to Benchmark 1, Bard contends that “the percentage of Teleflex’s stylets [paired with AngioDynamics’ PICCs]
would have been far lower in the but-for world because some hospitals currently using them with AngioDynamics’
PICCs would have shifted to Bard’s superior TLS,” and thus using this figure as a benchmark overstates the percentage
of Bard’s TLS stylets that would be combined with AngioDynamics’ PICCs in the but-for world. (Dkt. No. 132-1, at
18-21; Dkt. No. 153, at 8-11). As to Benchmark 2, Bard contends that “the higher the number of AngioDynamics
PICCs paired with Teleflex TLS, the smaller the relative advantage Teleflex has in selling its PICCs to its own TLS
users as compared to nurses who place PICCs without TLS, which inflates Benchmark 2.” (Dkt. No. 132-1, at 20
n.15).
28

99

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 100 of 105

PICC . . . and requires hospitals to use what Dr. Frankel alleges is an inferior TLS,” whereas
Bard TLS users “had the opportunity to make the same choice . . . but none decided that the
purported advantages of AngioDynamics’ PICCs were worth it.” (Dkt. No. 132-1, at 21-25; see
also Dkt. No. 153, at 11-12). AngioDynamics responds, in summary, that, given that: (1)
Teleflex is Bard’s only competitor to offer navigation-based TLS; (2) the two companies sell
similar products to similar types of customers (hospitals, GPOs and IDNs), and even some of the
same customers; and (3) Teleflex does not engage in an alleged tying scheme similar to Bard’s,
Teleflex’s performance in the real world is the most economically reasonable comparator to use
to estimate Bard’s performance in a world without its TLS policy, and Bard’s criticisms
regarding taint and comparability are best addressed as questions of weight rather than
admissibility. (Dkt. No. 143, at 20-27).
As a starting premise, the general rule is that arguments regarding the comparability or
appropriateness of a particular benchmark go to the weight, not the admissibility, of a
benchmarking analysis. See, e.g., In re Blood Reagents Antitrust Litig., MDL No. 09-2081, 2015
WL 6123211, at *21 n.18, 2015 U.S. Dist. LEXIS 141909, at *68 n.18 (E.D. Pa. Oct. 19, 2015)
(“Many courts . . . have held that the question whether plaintiffs have met their burden of
proving comparability should be left to the trier of fact to resolve because comparability
challenges generally involve weighing facts.”); Washington v. Kellwood Co., 105 F. Supp. 3d
293, 317 (S.D.N.Y. 2015) (“[W]hile the evidence underlying [the expert’s benchmark analysis]
may, in defendant’s view, be thin, questionable, or self-servingly selective, our role as
gatekeeper is not to divest defendant of the task of challenging the weight of such evidence
before the trier of fact.”). Moreover, a benchmark need not be perfectly comparable, so long as it
allows the jury to calculate a “reasonable estimate of damages” as required under the antitrust

100

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 101 of 105

laws. See, e.g., Fleischman, 728 F. Supp. 2d at 148 (citing McDonough v. Toys R Us, Inc., 638 F.
Supp. 2d 461, 491 (E.D. Pa. 2009)). This is particularly true where a Defendant’s anticompetitive
conduct has rendered selection of a perfectly comparable benchmark difficult or impossible. See,
e.g., Dial Corp. v. News Corp., 314 F.R.D. 108, 118-19 (S.D.N.Y. 2015) (finding arguments
regarding benchmark companies’ comparability go to weight, not admissibility, where, because
the defendant “has allegedly maintained a monopoly in the market . . . since at least 2000,”
“creating a benchmark using [the defendant’s] prices during a time of ‘robust competition’ is not
feasible,” and “the selection of perfectly comparable benchmark firms aside from [the defendant]
is impossible where [the defendant’s] alleged monopoly prevents comparable firms from
operating within its market”).
Nonetheless, there are some circumstances under which, even applying the foregoing
liberal standard, a benchmarking analysis may be so utterly deficient as to warrant its exclusion.
One such circumstance is where the expert did not merely select arguably inappropriate
benchmarks, but utterly “failed to perform any substantive analysis of those factors most relevant
to comparability.” In re Blood Reagents Antitrust Litig., 2015 WL 6123211, at *22, 2015 U.S.
Dist. LEXIS 141909, at *69 (emphasis added); see also El Aguila Food Prods., Inc. v. Gruma
Corp., 131 F. App’x 450, 453 (5th Cir. 2005) (affirming exclusion of expert testimony where the
expert “made no effort to demonstrate the reasonable similarity of the plaintiffs’ firms and the
businesses whose earnings data he relied on as a benchmark”); Loeffel Steel Prod., Inc. v. Delta
Brands, Inc., 387 F. Supp. 2d 794, 810-17 (N.D. Ill. 2005) (excluding expert testimony where
expert used eight companies in the same “industry” as the plaintiff as benchmarks but did not
“consider[] such critical factors as what services the companies provided, their customer base,
the products they sold, the geographic markets in which they operated, their prices and other

101

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 102 of 105

critical aspects of the business” to determine whether the plaintiff’s “customer base and the
services and products it offered approximated those of the eight companies”). Similarly, courts
have excluded benchmarking analyses that attribute all of plaintiff’s claimed losses to a
defendant’s misconduct, without making any effort to isolate the losses actually attributable to
that conduct from the impact of other significant differences between the plaintiff and the chosen
benchmark. See, e.g., Herman Schwabe, Inc. v. United Shoe Mach. Corp., 297 F.2d 906, 911 (2d
Cir. 1962) (affirming exclusion of analysis using the plaintiff-appellant’s share of clicking
machines in the non-shoe industry as a benchmark for sales it would have attained in the shoe
industry but for the defendant’s antitrust violations, where the expert failed to account for
relevant differences between the two industries or for “entirely lawful” factors accounting for the
defendant’s large share of the shoe machinery market); Weiner v. Snapple Beverage Corp., No.
07-cv-8742, 2010 WL 3119452, at *7, 2010 U.S. Dist. LEXIS 79647, at *23 (S.D.N.Y. Aug. 5,
2010) (finding an expert’s testimony unreliable where the expert did “not explain how his
approach would isolate the impact of [the anti-competitive conduct] from the other factors that
purportedly affect the price of [defendant’s product] and its competitors”). 29
Here, Dr. Frankel used PICC sales by Teleflex—which is Bard’s only competitor in the
navigation-enabled TLS space, and which does not engage in anticompetitive activity—as a
benchmark for the PICC sales Bard would have made had it not engaged in that activity. This

As noted, Bard also argues that courts have excluded benchmarking analyses where it found the proposed
benchmarks to be “tainted” by the alleged misconduct at issue. However, the cases Bard cites for this proposition are
distinguishable, as in contrast to Dr. Frankel, the excluded experts in those cases sought to use a defendant’s
anticompetitive profits as a benchmark for measuring the profits a competitor plaintiff would have made but for the
defendant’s conduct—in essence, seeking to reap the fruit of the defendant’s anticompetitive conduct to provide a
windfall to the plaintiff. See, e.g., Farmington Dowel Prod. Co. v. Forster Mfg. Co., 421 F.2d 61, 82 & n.48 (1969);
Admiral Theatre Corp. v. Douglas Theatre Co., 437 F. Supp. 1268, 1297-99 (D. Neb. 1977), modified, 585 F.2d 877
(8th Cir. 1978); cf. In re Urethane Antitrust Litig., No. 04-cv-1616, 2012 WL 6681783, at *6, 2012 U.S. Dist. LEXIS
181506, at *42 (D. Kan. Dec. 21, 2012) (“[C]riticisms of [an expert’s] consideration of possible taint from a conspiracy
of potential benchmark years go to the weight of his opinions and not their admissibility.”), aff’d 768 F.3d 1245 (10th
Cir. 2014).
29

102

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 103 of 105

approach at least attempts to approximate a but-for world that “[has] not been affected by
[Bard’s alleged] antitrust violations,” Schwab v. Philip Morris USA, Inc., 449 F. Supp. 2d 992,
1062 (E.D.N.Y. 2006), rev’d sub nom., McLaughlin v. Am. Tobacco Co., 522 F.3d 215 (2d Cir.
2008), and to use a benchmark company that is comparable to Bard in some important ways,
including the industry they compete in, the types of products they sell, and the customer base
they serve, i.e. “hospitals that [] place PICCs,” (Dkt. No. 132-3, at 283-94). Bard’s arguments,
however, raise valid questions about the reliability of Dr. Frankel’s analysis, especially about
whether he sufficiently considered important potential differences between the purchasing
preferences of the two companies’ customers that could impact the appropriateness of his
benchmarks. The Court need not decide whether, standing alone, these issues would warrant
exclusion of Dr. Frankel’s opinion, or would be best addressed as questions of weight at trial.
Given the Court’s finding that the data on which Dr. Frankel’s benchmarking analysis is based is
so unreliable as to require exclusion, the Court grants Bard’s motion to exclude that analysis
without reaching Bard’s remaining arguments.
3.

Remaining Theories of Relief

While the Court excludes the benchmarking analysis that is the subject of Bard’s motion
in limine, the Court declines to grant Bard summary judgment on AngioDynamics’ claim for
damages under Section 4 of the Clayton Act at this time. At oral argument, the parties disagreed
on whether the Court’s granting of Bard’s motion in limine would leave anything left of Dr.
Frankel’s testimony from which a jury could calculate a “just and reasonable” estimate of
damages. As Bard has only moved to exclude Dr. Frankel’s benchmarking analysis, and the
reliability of that specific analysis is thus the only part of Dr. Frankel’s analysis that is currently
before the Court, the Court does not reach the question of whether other aspects of Dr. Frankel’s
analysis may serve as a valid basis for calculating damages. Moreover, even assuming
103

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 104 of 105

AngioDynamics is left with no evidence from which a just and reasonable damages estimate can
be determined, this does not require dismissal of AngioDynamics’ claim for damages under
Section 4 of the Clayton Act. Where, as here, “there is insufficient proof of the amount of
damages . . . proof of an antitrust violation and the fact of damage is a sufficient basis for an
award of nominal damages” under Section 4. Valassis Comms., Inc., 2019 WL 802093, at *12,
2019 U.S. Dist. LEXIS 27770, at *34-35 (citations omitted); U.S. Football League v. Nat’l
Football League, 842 F.2d 1335, 1377 (2d Cir. 1988) (confirming district courts’ ability to award
nominal damages in antitrust cases).
Also, even if AngioDynamics does not have a viable theory by which it can quantify its
past damages, this would not preclude AngioDynamics from continuing to pursue its claim for
injunctive relief under Section 16 of the Clayton Act. “Typically, the inability to prove past
damages does not compel a finding that the plaintiff faces no threat of antitrust injury in the
future,” and thus a plaintiff that failed to prove past damages may still pursue a claim for
forward-looking injunctive relief. Cash & Henderson Drugs, Inc. v. Johnson & Johnson, 799
F.3d 202, 215 (2d Cir. 2015) (citing Zenith Radio Corp., 395 U.S. at 130). “However, in certain
situations, the lack of past injury may indicate that future injury is improbable.” Id. For example,
where the challenged conduct has been in place long enough for potential effects to manifest
themselves and there is no evidence of injury, the difference between Clayton Act Section 4’s
requirement of actual injury and Section 16’s requirement of threatened injury disappears. Drug
Mart Pharmacy Corp. v. Am. Home Prods. Corp., 2007 WL 4526618, at *10-15, 2007 U.S. Dist.
LEXIS 93493, at *45-62 (E.D.N.Y. Dec. 20, 2007). Here, the Court has found that there is
sufficient evidence in the record to create triable issues of fact as to whether AngioDynamics has

104

Case 1:17-cv-00598-BKS-CFH Document 189 Filed 05/05/21 Page 105 of 105

suffered injury as a result of an antitrust violation committed by Bard. Therefore,
AngioDynamics may continue to pursue its claim for injunctive and declaratory relief.
V.

CONCLUSION
For these reasons, it is hereby
ORDERED that Bard’s motion in limine (Dkt. No. 132) is GRANTED; and it is further
ORDERED that Bard’s motion for summary judgment (Dkt. No. 133) is DENIED in its

entirety; and it is further
ORDERED that AngioDynamics’ motion for summary judgment (Dkt. No. 134) is
DENIED in its entirety.
IT IS SO ORDERED.
Dated: _________________
May 5, 2021
Syracuse, New York

105