“Where the product is a prescription drug, however, it is widely accepted that the manufacturer’s duty to warn runs to the physician rather than the patient.”
How later courts described this case
- “Where the product is a prescription drug, however, it is widely accepted that the manufacturer’s duty to warn runs to the physician rather than the patient.”
Written by the judges who cited it.
The opinion
United States District Court
District of Massachusetts
IN RE: FRESENIUS GRANUFLO/ ) MDL No. 13-2428
NAUTRALYTE DIALYSATE PRODUCTS )
LIABILITY LITIGATION )
)
This document relates to: )
)
Gloria Cothern Dunaway )
Case No. 13-11714 )
)
Mervin Boyd
)
Case No. 13-11717
)
Michael McNulty )
Case No. 13-12403 )
)
Charles Cameron )
Case No. 13-12446 )
)
Daniel Carter )
Case No. 13-12459 )
)
Joyce Marie Clark )
Case No. 13-12460 )
)
Kathy Dennis )
Case No. 13-12467 )
)
Kimberly Ross
)
Case No. 13-12478
)
Beulah Williams )
Case No. 13-12486 )
)
Sophia Walker )
Case No. 13-12487 )
)
Janice McGhee )
Case No. 13-13172 )
)
Max Riben )
Case No. 15-11134 )
)
Josephine Gallardo Hernandez )
Case No. 18-11224 )
MEMORANDUM & ORDER
GORTON, J.
This Multi-District Litigation arises from the use of acid
concentrates in the treatment of dialysis patients who died
following the procedures. The acid concentrates at issue,
NaturaLyte and GranuFlo, are manufactured by the defendants
Fresenius Medical Care Holdings, Inc. d/b/a Fresenius Medical
Care North America; Fresenius USA, Inc.; Fresenius USA
Manufacturing, Inc.; and Fresenius USA Marketing, Inc.
(collectively “Fresenius”). All of the defendants move for
summary judgment on the claims of certain opt-out plaintiffs.
Here, the Court addresses Fresenius’s motions for summary
judgment based on 1) lack of evidence of elevated serum
bicarbonate levels, 2) lack of evidence of causation 3) claims
involving Naturalyte and 4) the learned intermediary doctrine.
Subject to the motion related to elevated serum bicarbonate
levels are the following plaintiffs: Gloria Cothern Dunaway,
Mervin Boyd, Michael McNulty, Daniel Carter, Joyce Marie Clark,
Kimberly Ross, Beulah Williams, Sophia Walker, Janice McGhee and
Max Riben. Subject to the motion related to the lack of
evidence of causation are the following plaintiffs: Gloria
Cothern Dunaway, Mervin Boyd, Michael McNulty, Daniel Carter,
Joyce Marie Clark, Kathy Dennis, Kimberly Ross, Sophia Walker,
Janice McGhee, Max Riben and Josephine Gallardo Hernandez.
Subject to the motion involving Naturalyte are the following
plaintiffs: Charles Cameron, Daniel Carter, Sophia Walker, Max
Riben and Josephine Gallardo Hernandez. All remaining 13
plaintiffs are subject to the learned intermediary doctrine
motion.
Because there are no outstanding genuine issues of material
fact, the Court will allow the motions for summary judgment as
against all pertinent plaintiffs.
I. Background
A. Factual Background
1. The Second Amended Complaint & Plaintiffs’
General Causation Theory
Plaintiffs’ complaint is premised on the theory that
Fresenius failed to warn doctors about how to use GranuFlo and
NaturaLyte safely with their hemodialysis patients. According
to plaintiffs, the acetate in GranuFlo and NaturaLyte leads to a
“dangerous increase” in serum bicarbonate levels in patients
undergoing hemodialysis which results in metabolic alkalosis
triggering cardiac arrest and sudden cardiac death. In
particular, plaintiffs allege that alkalosis
is caused by too much bicarbonate in the blood [and
that it is those patients with] elevated bicarbonate
levels in their blood
who are at an increased risk of sudden cardiac arrest.
Plaintiffs contend that Fresenius should therefore have advised
doctors to
pay attention to the increase in serum bicarbonate
levels [and to] reduce the amount of bicarbonates
being delivered . . . during dialysis to take into
account the additional bicarbonate from NaturaLyte
and/or GranuFlo.
2. Facts Applicable to All Serum Bicarbonate
Plaintiffs
Ray Hakim, MD, former Chief Medical Officer for Fresenius
Medical Services, authored a memorandum dated November 4, 2011
(“the Hakim Memo”) that was addressed to medical directors and
attending physicians regarding the subject of “Dialysate
Bicarbonate, Alkalosis and Patient Safety.” The Hakim Memo
discussed the results of a “case-control study” that
evaluated risk factors in [hemodialysis] patients who
suffered from [cardiopulmonary] arrest in the facility
. . . compared to other [hemodialysis] patients . . .
within the same facilities between January 1, and
December 31, 2010.
The data in the Hakim Memo depicted no statistically significant
increased risk of in-center cardiopulmonary arrest for patients
with pre-dialysis serum bicarbonate levels in the mid to low
20s. When focusing on bicarbonate levels alone, patients with a
pre-dialysis serum bicarbonate level of 28 milliequivalents
(“mEq/L”) or more were depicted as having the greatest relative
risk for cardiopulmonary arrest during dialysis as compared to
other groups and no other group was marked with a statistically
significant increased risk. When pre-dialysis potassium lab
values were included in the analysis, patients with a pre-
dialysis serum bicarbonate value under 28 mEq/L and potassium
greater than or equal to four mEq/L had no increased risk. Dr.
Hakim testified that an earlier draft of the Hakim Memo defined
alkalosis as “pre-dialysis bicarbonate of greater than or equal
to 28 milliequivalents,” but that language was not included in
the final version of the memo.
Plaintiffs retained Dr. Derek Fine, as an expert witness on
general and specific causation in this litigation. He is an
Associate Professor of Medicine at Johns Hopkins University
School of Medicine and has a clinical practice that includes
treating dialysis patients at a DaVita outpatient dialysis unit
in Baltimore, Maryland. Dr. Fine testified during his
deposition that a “normal” range for pre-dialysis serum
bicarbonate is subject to “varying opinion” but that he would
like to see the [serum] bicarb[onate] somewhere
between, in most cases, 20 and 24 [mEq/L] [and that he
would tell his fellows and nurse practitioners that
the] K/DOQI guidelines say greater than 22 [mEq/L] is
a reasonable target.
Dr. Fine also testified that if he were asked to place an “upper
limit” for pre-dialysis serum bicarbonate, that number would be
27 mEq/L. Dr. Fine testified that, in general, it is
unnecessary to adjust a bicarbonate prescription because most
patients are not alkalotic, so the key is to ensure his
nephrology physician fellows are aware “that alkalosis is bad.”
He further testified that studies show that “high [serum]
bicarb[onate] is bad” and “associated with mortality” and
“sudden cardiac arrest.” His expert report notes that “normal”
serum bicarbonate levels are 22 to 26 mEq/L for arterial blood
and 23 to 27 mEq/L for venous blood.
Plaintiffs also retained Dr. Sushrut Waikar, as an expert
witness on general and specific causation in this litigation.
Dr. Waikar is an Associate Professor of Medicine at Harvard
Medical School and he treats nephrology patients, including some
who are on dialysis, at Brigham & Women’s Hospital in Boston,
Massachusetts. At his deposition, Dr. Waikar testified that the
typical serum bicarbonate range is 20 to 26 mEq/L and the range
he targets for his own patients’ pre-dialysis serum bicarbonate
levels is “22 to 26 [mEq/L], around there, would be reasonable,
maybe 22 to 24 [mEq/L].” Dr. Waikar also testified that he
would adjust the bicarbonate prescription for a patient based on
[t]he presence or absence of chronic obstructive
pulmonary disease, the presence or absence of severe
metabolic alkalosis or acidosis.
When asked to explain what he meant by a patient presenting with
metabolic alkalosis, Dr. Waikar gave the example of a patient
with a serum bicarbonate concentration level of 35 mEq/L when he
comes into the dialysis unit. He was also asked to explain what
he meant when he referred to “significant alkalosis,” and he
gave examples of a patient with serum bicarbonate levels of 30
or 35 mEq/L.
The third expert witness on general causation retained by
plaintiffs is Dr. David Goldfarb. He is a professor at New York
University and treats dialysis patients at a Veterans Affairs
unit in the New York Harbor Healthcare System. In discussing
bicarbonate levels that would be a potential cause for concern,
Dr. Goldfarb agreed that levels below 22 mEq/L are “associated
with adverse outcomes,” as are bicarbonate levels of 28 or 30 or
35 mEq/L. He also testified that in his dialysis practice, he
gives his patients 35 mEq/L of bicarbonate and that none of the
treating nephrologists in the chronic dialysis unit prescribes
different levels of bicarbonate. In fact, Dr. Goldfarb
indicated that 1) his recommendation would be not to change the
bicarbonate prescription for any of the chronic kidney disease
patients at the New York Harbor clinic and 2) he never adjusts
the prescription for treatment based on pre or post-dialysis
serum bicarbonate values.
Finally, plaintiffs retained Dr. Steven C. Borkan, as a
fourth expert witness on general and specific causation. Dr.
Borkan is a professor at Boston University and maintains an
active clinical nephrology practice in facilities affiliated
with DaVita. During his June, 2015 deposition, Dr. Borkan
testified that his “target” pre-dialysis serum bicarbonate range
for his own patients is between 22 and 24 mEq/L. At his
October, 2015 deposition, he added that he does not
dial back the bicarbonate delivered in prescription
[to his own patients unless the patient] has a
predialysis bicarbonate level that’s above 24 [mEq/L].
Dr. Borkan testified as a general and case-specific
nephrology expert witness in the bellwether trial, Fiorella Dial
v. Fresenius Medical Care Holdings, Inc., et al., in February,
2017 before United States District Judge Douglas P. Woodlock.
At that time he assured the jury that the “normal” pre-dialysis
serum bicarbonate level for a dialysis patient is about 22 to 24
mEq/L and confirmed that such a level is his “target” range for
his patients. On cross-examination at that trial, however, Dr.
Borkan acknowledged that the Dial decedent’s pre-dialysis serum
bicarbonate laboratory value was 26 mEq/L before four of his
monthly dialyses but that on three of those occasions, his
bicarbonate level decreased to a reading of 22 to 24 mEq/L
without any change in prescription. Dr. Borkan’s expert report
on general causation defines “elevated” serum bicarbonate levels
as greater than 26 mEq/L.
Drs. Fine, Waikar, Goldfarb and Borkan all rely on the data
discussed in the Hakim Memo to support their opinions on general
causation in their expert reports. Dr. Fine also relied on the
Hakim Memo at his deposition for his opinion that alkalosis is a
trigger when
someone has an event on dialysis, a cardiopulmonary
arrest or cardiac arrest, [and] in patients who are
having cardiac events, they’re more likely to have a
high bicarbonate.
During his deposition, Dr. Goldfarb identified data in the Hakim
Memo as “the data that’s important” to support his opinion that
“the increase in serum bicarbonate . . . was associated with an
increase in sudden death.” Dr. Borkan also testified that the
Hakim Memo is the basis for some of his opinions.
3. Facts Applicable to All Causation Plaintiffs
As discussed above, the Hakim Memo sets forth Dr. Hakim’s
findings regarding the relative risk to dialysis patients of
cardiopulmonary arrest and sudden cardiac death based on their
pre-dialysis serum bicarbonate levels.
On December 1, 2014, Dr. Hakim testified that he would want
a patient to sit for between 90 minutes and two hours before
doing a post-dialysis bicarb draw because
the ability . . . to metabolize bi-acetates is
different in different patients. Some take longer;
some take less time. But the data that I’ve seen is
that it goes up up to 90 minutes after termination,
and then it starts coming down again.
Plaintiffs’ cardiology experts, Drs. Joseph G. Akar, Julian
M. Aroesty, Zayd A. Eldadah, Joseph Shawn Miles, Arthur Z.
Schwartzbard and Douglas Zipes all rely, at least in part, on
the Hakim Memo to support their expert opinions on general
causation. Dr. Miles, when asked what research he relied on to
support his ultimate opinion, testified that he relied on “[t]he
November 4, 2011 memo and prior Fresenius documents.”
Dr. Akar’s report opines that 1) “[c]omplex arrhythmias”
require a trigger and an “underlying substrate that allows its
perpetuation,” 2) dialysis patients “are highly vulnerable to
the development of arrhythmias in the setting of the alkalotic
process and hypokalemia” which involves an “intracellular
shifting of potassium,” and 3) NaturaLyte and GranuFlo
“produc[e] a process of alkalosis” that exposes patients to “an
increased risk of cardiac arrest and death.” The report further
notes that rapid shifts in potassium levels during dialysis have
been associated with sudden death. Thus,
the more rapid and the greater the changes are in pH,
the higher the gradients that are created, and the
more rapid and steeper shifts in potassium levels.
During his June, 2015 deposition, Dr. Akar testified that
GranuFlo and NaturaLyte
have the potential to provide excess acetate, and this
excess acetate has the potential to cause significant
alkalosis, and alkalosis has the potential
to . . . have a significant effect on ionic channels
which has a potential to produce sudden cardiac death.
Dr. Akar noted that looking at the statistics with respect to
sudden cardiac death around the time of dialysis, “0 to 12 hours
is a particularly high period in which sudden cardiac death due
to arrhythmias occurs.”
Dr. Lucius M. Lampton, who submitted expert reports on
behalf of plaintiffs Boyd, Carter, Clark, Dunaway, Dennis,
McGhee, McNulty, Ross and Walker, attached and incorporated Dr.
Akar’s report by reference in his case-specific expert reports.
Dr. Aroesty’s report opines that end-stage renal disease
(“ESRD”)
patients have high comorbidity (e.g. diabetes,
hypertension, atherosclerosis) making them
particularly vulnerable to SCA/SCD [sudden cardiac
arrest/sudden cardiac death] triggers
and that hypokalemia and alkalosis can be triggers for sudden
cardiac arrest and sudden cardiac death. It further states that
the
change in dialysate formulation to include diacetate
[in GranuFlo] was accompanied by a progressive
increase in pre [hemodialysis] blood pH (i.e.
alkalosis) [and] the incremental increase in pH
(alkalosis) resulted in a shift of potassium (K) ions
across the cell membrane with consequent increased
risk of VT/VF [ventricular tachycardia/ventricular
fibrillation] and SCA/SCD.
His report also notes that a rapid change in a patient’s
electrolyte and acid/base balance during hemodialysis is an
additional risk factor for development of ventricular
tachycardia/ventricular fibrillation and sudden cardiac
arrest/sudden cardiac death.
Dr. Eldadah’s report concludes that sudden “derangement” in
serum potassium levels can cause “abnormal heart rhythms” that
can be fatal and that
sudden cardiac death or injury occurred in dialysis
patients who received GranuFlo or NaturaLyte because
these compounds cause: (a) an increased load of
acetate in the body, which caused (b) an increased
load of serum bicarbonate in the body (due to the
conversion in the liver of acetate to bicarbonate),
which caused (c) an acute drop in serum potassium
concentrate, which caused (d) lethal cardiac
arrhythmias.
Dr. Eldadah’s report also states that “ventricular tachycardia
and/or ventricular fibrillation” and sudden cardiac death can
ensue from “rapid” changes in blood pH that “derange the orderly
flow of electricity through the heart muscle.”
Dr. Miles’s report determines that higher concentrations of
dialysate bicarbonate cause metabolic alkalosis, which causes
hypokalemia, hypocalcemia and hypoxia and can result in
shifts in potassium, calcium and oxygen [which] can
cause sudden cardiac arrest and death, myocardial
infarction and stroke.
Moreover, Dr. Miles’s report states that “exposure to
bicarbonate and acetate in the dialysate” during dialysis
subjects patients to “rapid potassium shifting resulting in
hypokalemia, which is a well-known cause of sudden cardiac
death.”
Dr. Schwartzbard’s report postulates that elevated
bicarbonate levels and low potassium concentrations can cause
life threatening ventricular arrhythmias “in the susceptible
ESRD population, leading to sudden cardiac arrest (SCA) and
death (SCD).” The shift in bicarbonate leads to electrolyte
disorders, such as hypokalemia and hypocalcemia, which cause an
increased risk of cardiac arrhythmia. Dr. Schwartzbard explains
that “[w]hen a patient who was previously well within an hour
prior to his demise dies suddenly, the event is termed sudden
cardiac death (SCD).”
Dr. Zipes’s report theorizes that
alkalosis due to elevated serum bicarbonate
concentrate, . . . a low serum potassium
concentration, . . . a combination of elevated serum
bicarbonate concentration and low serum potassium
concentrate, . . . rapid electrolyte shifts following
administration of Granuflo [sic] and Naturalyte [sic],
and . . . acidosis each can trigger life threatening
ventricular arrhythmias in the susceptible ESRD
dialysis patients, [which] can lead to cardiopulmonary
arrest and death.
Dr. Zipes’s report also notes that a “catastrophic arrhythmic
event . . . is called sudden cardiac arrest (SCA) and leads to
death unless reversed promptly.” Dr. Zipes testified during his
June, 2015 deposition that his understanding of this case is
that there was a problem with the dialysate, here GranuFlo
and/or NaturaLyte, that created a metabolic electrolyte
imbalance resulting in sudden cardiac arrest and ultimately
death.
4. Facts Applicable to NaturaLyte Plaintiffs
NaturaLyte and GranuFlo are acid concentrates used in the
creation of dialysate, the dialysis solution. GranuFlo is a dry
powder acid concentrate that contains various electrolytes, four
mEq/L of sodium acetate and four mEq/L of acetic acid.
Together, those two solutions form sodium diacetate. When
combined with a bicarbonate concentrate and water, GranuFlo
provides eight mEq/L of acetate to the dialysis solution.
NaturaLyte, the subject of one of the four motions before the
Court, is a liquid acid concentrate that contains various
electrolytes and four mEq/L of acetic acid. When combined with
a bicarbonate concentrate and water, NaturaLyte provides four
mEq/L of acetate to the dialysis solution.
Fresenius facilities use those products in their dialysis
procedures. It also sells and markets its products to other
dialysis facilities, including many clinics that compete with
Fresenius facilities, such as DaVita Dialysis Centers, Dialysis
Clinics Inc. and Renal Ventures Management LLC.
GranuFlo and NaturaLyte have been on the market for many
years. The United States Food and Drug Administration (“FDA”)
cleared NaturaLyte for marketing in 1981 and, a decade later,
cleared GranuFlo for marketing.1 In the years since NaturaLyte
was cleared for sale, other manufacturers of acid concentrates
for hemodialysis have also offered a liquid product with four
mEq/L of acetate and they continue to do so. Notably, the
labels for other liquid acid concentrate products with four
mEq/L of acetate identify the acetate contents in the same
manner that the NaturaLyte label identified its acetate
contents.
Fresenius sold over 305 million gallons of NaturaLyte in
the United States between 2000 and 2012. NaturaLyte has been
used in clinical settings since the early 1980s and has been
used in hundreds of millions of hemodialysis treatments.
Several of plaintiffs’ experts, including Dr. Fine, Dr. Paul
Miller and Dr. Waikar testified that they used NaturaLyte to
treat their patients in a safe and effective manner.
1 NaturaLyte and GranuFlo are regulated as medical devices by the
FDA and are subject to FDA clearance rather than FDA approval.
5. Facts Applicable to Learned Intermediary Doctrine
Plaintiffs
Fresenius’s Chief Medical Office issued several memoranda,
including the Hakim Memo discussed above, between 2000 and 2011
which discuss GranuFlo and NaturaLyte, acetate, acid/base
balance, serum bicarbonate levels, alkalosis, the concept of
“total buffer” and potential mortality and cardiac risks. Those
memoranda were distributed to physicians and faculty staff as
well as posted to the Fresenius Intranet website. The Hakim
Memo was also sent to DaVita the same day that it was
distributed to Fresenius physicians.
In March, 2012, Fresenius also issued an “Important
Prescribing Information” notification to all known customers
that had purchased GranuFlo or NaturaLyte. That notification
stated, in part that
NaturaLyte Liquid contributes 4.0 mEq/L of acetate and
GranuFlo contributes 8.0 mEq/L of acetate to the final
dialysate; which in addition to bicarbonate, combine
to the total buffer that the patient receives from the
dialysate. Acetate is also contained in the dialysis
acid concentrates produced by other manufacturers.
Since acetate is rapidly converted into bicarbonate by
the liver, the bicarbonate prescription entered into
the dialysis machine underestimates the total buffer
that the patient receives from the dialysate by ~8
mEq/L with dialysate prepared from Granuflo (powder)
or by ~4 mEq/L with dialysate prepared from NaturaLyte
(liquid).
The notification recommended that physicians individualize
prescriptions and “review[] them monthly with consideration of
patient’s pre-dialysis bicarbonate and dialysate total buffer.”
Furthermore, plaintiffs’ experts Drs. Goldfarb and Fine
both testified that nephrology fellows know from medical school
that acetate metabolizes into bicarbonate in the liver. Dr.
Miller agreed, stating that
most nephrologists who have been through high school
and then college and then medical school, would
understand that acetate converts in the body to
bicarbonate.
Plaintiffs’ experts also testified that NaturaLyte containers
are labeled with the contents in the acid concentrate, including
the acetate concentration.
B. Procedural Background
As relevant here, the plaintiffs against whom summary
judgment is sought opted out of the global settlement agreement.
Those plaintiffs are subject to the Lone Pine Order entered by
United States District Judge Douglas P. Woodlock on January 26,
2017. Under that order, the opt-out plaintiffs were required to
provide: 1) an affidavit by counsel that attested to the fact
that counsel had reviewed documents or data supporting the
contention that GranuFlo or NaturaLyte acid concentrate was used
during the last dialysis treatment of the subject patient prior
to the alleged injury and 2) an affidavit executed by a
qualified physician or other medical expert setting forth an
opinion about specific causation. The plaintiffs were given
until July 28, 2017 to decide whether to opt in to the
settlement or comply with the Lone Pine Order.
Plaintiffs’ operative pleading asserts claims for: 1)
strict liability, 2) negligent failure to warn, 3) negligent
design, 4) negligence, 5) negligent misrepresentation, 6) breach
of implied warranty of merchantability, 7) breach of implied
warranty of fitness for a particular purpose, 8) breach of
express warranty, 9) fraud, 10) violation of consumer protection
laws, 11) loss of consortium, 12) wrongful death and 13)
survival actions.
In August and September, 2017, Fresenius filed four
successive summary judgment motions as to the remaining opt-out
plaintiffs. Judge Woodlock held oral argument on those summary
judgment motions in November, 2017.
Plaintiff Josephine Gallardo Hernandez filed her complaint
in January, 2018. Fresenius moved for summary judgment against
her with respect to the issues of serum bicarbonate, causation
and Naturalyte in January, 2019. Because Fresenius has
incorporated by reference the facts and arguments set forth in
its original four motions for summary judgment, the Court will
address that motion in conjunction with those affecting the
other plaintiffs.2
This Multi-District Litigation was reassigned to this
session of this district court in June, 2023. Upon
reassignment, the assigned judicial officer held a status
conference in July, 2023, to determine the status of the
remaining cases. The Court ordered the parties to file status
reports on or before August 17, 2023, to inform the Court as to
any potential resolution of those cases.
Fresenius reported at the status conference and in its
status reports that the summary judgment motions have been
fully-briefed and all of the above-captioned cases are ripe for
rulings on the merits by this Court. Counsel for plaintiffs
Dunaway, Boyd, McNulty, Cameron, Carter, Clark, Dennis, Ross,
Williams, Walker and McGhee protest only now in their status
report that such plaintiffs do not agree that the cases are ripe
for rulings on the merits. Those plaintiffs did not object at
the July, 2023, status conference to this session deciding the
fully-briefed motions for summary judgment nor have they even
yet suggested why the cases are not ripe for decision.
2 That motion for summary judgment is Docket No. 31 in Case No.
18-11224.
After careful consideration of all the briefs on file, the
transcript of oral argument before Judge Woodlock and his
extensive prior labor on the subject, the Court will allow
defendants’ pending motions for summary judgment.
II. Legal Standard
The role of summary judgment is “to pierce the pleadings
and to assess the proof in order to see whether there is a
genuine need for trial.” Mesnick v. Gen. Elec. Co., 950 F.2d
816, 822 (1st Cir. 1991) (quoting Garside v. Osco Drug, Inc.,
895 F.2d 46, 50 (1st Cir. 1990)). The burden is on the moving
party to show, through the pleadings, discovery and affidavits,
“that there is no genuine dispute as to any material fact and
the movant is entitled to judgment as a matter of law.” Fed. R.
Civ. P. 56(a).
A fact is material if it “might affect the outcome of the
suit under the governing law . . . .” Anderson v. Liberty Lobby,
Inc., 477 U.S. 242, 248 (1986). A genuine issue of material
fact exists where the evidence with respect to the material fact
in dispute “is such that a reasonable jury could return a
verdict for the nonmoving party.” Id.
If the moving party satisfies its burden, the burden
shifts to the non-moving party to set forth specific facts
showing that there is a genuine, triable issue. Celotex Corp. v.
Catrett, 477 U.S. 317, 324 (1986). The Court must view the
entire record in the light most favorable to the non-moving
party and make all reasonable inferences in that party’s favor.
O’Connor v. Steeves, 994 F.2d 905, 907 (1st Cir. 1993). Summary
judgment is warranted if, after viewing the record in the non-
moving party’s favor, the Court determines that no genuine issue
of material fact exists and that the moving party is entitled to
judgment as a matter of law.
III. Analysis
The Court addresses separately below the four successive
summary judgment motions filed by Fresenius.
A. Serum Bicarbonate
Fresenius moves for summary judgment on the claims of ten
plaintiffs with respect to the serum bicarbonate levels of the
decedents. Fresenius asserts that it is entitled to summary
judgment because 1) patients with bicarbonate levels below 28
mEq/L are outside the Hakim Memo’s risk range and therefore
there is a lack of evidence of medical causation and 2) the
testimony of plaintiffs’ experts shows that the alleged failure
of Fresenius to warn could not proximately cause injury to
patients with bicarbonate levels of 26 or lower.
Causation is an essential element for each of plaintiffs’
claims. Plaintiffs are required to establish two kinds of
causation: general and specific. See In re Neurontin Mktg.,
Sales Practices, & Prods. Liab. Litig., 612 F. Supp. 2d 116, 123
(D. Mass. 2009). Specifically,
[g]eneral causation is established by demonstrating,
often through a review of scientific and medical
literature, that exposure to a substance can cause a
particular disease, [and specific causation] is
established by demonstrating that a given exposure is
the cause of an individual’s disease.
Id. (citations omitted). Thus, plaintiffs must establish
general causation by showing the drug’s capacity to cause the
injury generally and specific causation by showing “that the
drug did cause the injury in this case.” Kerlinsky v. Sandoz
Inc., 783 F. Supp. 2d 236, 240 (D. Mass. 2011). Here, Fresenius
argues that plaintiffs have failed to adduce any evidence to
meet their burden of establishing both general causation and
specific causation.
1. General Causation Theory
Fresenius’s motion rests in part on the proposition that
the Hakim Memo, on which all of plaintiffs’ nephrology experts
rely, demonstrates that plaintiffs’ general causation theory is
unavailing. That theory is that GranuFlo and NaturaLyte
increase patients’ serum bicarbonate to dangerous levels, which
leads to alkalosis, which leads to cardiac arrest. The Hakim
Memo, however, indicates that the heightened risk of
cardiopulmonary arrest applies only to patients whose last pre-
dialysis lab results indicate a serum bicarbonate level of 28
mEq/L or greater or a potassium level of less than four mEq/L.
Because none of the patients at issue here had a last pre-
dialysis serum bicarbonate level that was 28 mEq/L or higher,
Fresenius contends that no individual plaintiff is able to
establish specific causation by virtue of the fact that none of
them qualifies under the general causation theory. In fact, the
range of levels recorded for plaintiffs’ decedents is from 19
mEq/L to 26 mEq/L, all below the threshold established by the
Hakim Memo.
Moreover, Fresenius notes that plaintiffs’ general
causation nephrology experts 1) conducted no independent studies
on the alleged association between serum bicarbonate levels and
the risk of cardiac arrest and 2) explicitly relied on the Hakim
Memo in rendering their own opinions with respect to whether
heightened serum bicarbonate levels cause alkalosis and, in
turn, cardiac arrest.
The plaintiffs here, whose decedents’ final pre-dialysis
serum bicarbonate levels all fell below 28 mEq/L, do not qualify
under their own core liability theory and they have adduced no
evidence to support general causation.
a. The Plaintiffs’ General Opposition
Plaintiffs cite portions of expert testimony selectively to
support their theory but in essence they allege that dialysis
can cause a spike in serum bicarbonate levels leading to
alkalosis in any patient and alkalosis can cause sudden cardiac
arrest and death. Plaintiffs cannot create an issue of fact by
contorting or expanding their theory of general causation.
Plaintiffs’ operative complaint, the Second Amended Master
Complaint, is an administrative device filed with the intent of
setting forth claims of the individual plaintiffs against
Fresenius and plaintiffs are therefore bound by it. According
to that complaint, acidosis is a typical occurrence for patients
in kidney failure and severe acidosis can lead to shock or
death. Dialysis attempts to correct an acidotic state by adding
bicarbonate to the patient’s blood. Acidosis is the opposite of
alkalosis, which occurs when a patient’s blood has excess base.
Alkalosis is caused by too much bicarbonate in the blood and can
cause cardiac arrhythmias and/or death. Thus, a purpose of
dialysis is to add bicarbonate to a patient’s blood to correct
acidosis.
During dialysis, blood is pumped in one direction and the
dialysate in the opposite direction. A nephrologist may order a
particular dialysate solution containing specific amounts of
potassium, sodium, magnesium and calcium depending upon the
patient’s electrolyte balance. The dialysate solution used
during dialysis is a mixture of a bicarbonate concentrate and
acid concentrate. The bicarbonate concentrate is used on all
dialysis patients, although the amount of bicarbonate can be
adjusted. Because patients experiencing renal failure
tend to become acidotic, and that problem is corrected
primarily by adding bicarbonate to their blood . . . ,
all dialysate solutions contain bicarbonate to correct
the naturally occurring acidosis in patients in renal
failure.
GranuFlo and NaturaLyte are the acid concentrate portions
of the dialysates at issue and when they are introduced into the
body, the acetate within the acid concentrate is converted into
bicarbonate by the liver, which increases bicarbonate levels in
the blood. Because GranuFlo contains sodium diacetate,
plaintiffs contend that the conversion in the liver produces an
unanticipated amount of bicarbonate that exceeds what is
normally and reasonably prescribed by the physician attending to
the patient, leading to a higher “total buffer.”
In sum, plaintiffs allege:
a significant number of dialysis patients develop an
unexpectedly rapid increase in elevated levels of
bicarbonate in their blood during dialysis, as well as
the potential for added serum bicarbonate post
dialysis as the acetate in the blood continues to
metabolize into bicarbonate. Patients with elevated
bicarbonate levels in their blood suffer from
metabolic alkalosis, the opposite of acidosis, and
high bicarbonate levels in the blood increases a
patient’s risk of cardiopulmonary arrest (“CP”) or
sudden cardiac arrest.
Moreover, plaintiffs repeat that
a dangerous increase in serum bicarbonate levels in
patients undergoing hemodialysis . . . contributes to
metabolic alkalosis, which is a significant risk
factor associated with . . . heart arrhythmia,
cardiopulmonary arrest and sudden cardiac death.
Throughout the complaint, plaintiffs focus on the allegation
that “too much bicarbonate” can lead to levels “outside the
normal or tolerated range” leading to alkalosis “(high blood
pH).” Finally, plaintiffs make clear that Fresenius was aware
that patients given GranuFlo had “higher than normal pre-
dialysis bicarbonate levels” and “an increase in cases of
metabolic alkalosis.” Because Fresenius was aware that pre-
dialysis serum bicarbonate levels that were “at or above 28
mEq/L” indicated a “20% increase in death risk,” plaintiffs
allege that Fresenius had a duty to warn and should have known
their product was defective and dangerous.
Indeed, even plaintiffs’ opposition relies on their Omnibus
Memorandum in Support of General Causation, which makes clear
that their theory of general causation is that excessive
bicarbonate “total buffer” in the dialysate causes metabolic
alkalosis, arrhythmia, sudden cardiac arrest and death.
However, while purporting to be confined to that general theory
of causation, plaintiffs attempt to expand their theory to
allege that any bicarbonate level pre-dialysis that is not
between 22 and 24 mEq/L should cause concern to experts. They
do that by relying on several experts, all of whom have slightly
different ranges of serum bicarbonate levels that they consider
“normal” or in the “target” range. Thus, plaintiffs cannot come
to a consensus via their own experts as to what level of pre-
dialysis serum bicarbonate is in a non-concerning range.
Furthermore, none of their experts purported to change a
prescription on numbers between 19 and 26 mEq/L and none
testified that he or she was concerned about numbers on the low
end because of alkalosis. Thus, plaintiffs have failed to set
forth any competent evidence in support of their claims that
essentially any pre-dialysis number comports with their theory
of general causation.
Plaintiffs cite Dr. Goldfarb’s testimony that bicarbonate
levels below 22 mEq/L or 28 mEq/L and above would be concerning
and therefore contend that decedents Boyd, Jenkins, McGhee and
Myles, all of whom had pre-dialysis serum bicarbonate levels of
19 or 20, have created a genuine issues of material fact
precluding summary judgment. At best, however, plaintiffs have
proffered evidence that acidosis is of concern to doctors.
Acidosis is the opposite of alkalosis and although it may be
disconcerting, it does not support the general causation theory
plaintiffs have advanced.
Next, plaintiffs rely on Dr. Fine’s testimony that he likes
to “see the bicarb[onate] somewhere between, in most cases, 20
and 24 mEq/L.” on that basis, they contend that decedents
McNulty, Ross and Hughes, all of whom had pre-dialysis serum
bicarbonate levels of 25 or 26 mEq/L, have adduced sufficient
evidence to preclude summary judgment. Plaintiffs refer to Dr.
Waikar’s testimony that his target range is 22 mEq/L to 24 or 26
mEq/L and that the presence of chronic obstructive pulmonary
disease (“COPD”) might lead him to adjust the bicarbonate
prescription for a patient. They claim, on that basis, that Ms.
Cothern and Ms. Carter, both of whom had COPD, have raised a
genuine issue of material fact. Dr. Waikar, however, testified
clearly that COPD might lead him to increase the amount of
bicarbonate in the dialysate, i.e. he would seek to elevate
their serum bicarbonate levels. That treatment would have
caused more serious electrolyte shifts and alkalosis in patients
and, accordingly, the Court perceives no genuine issue of
material fact created thereby and the argument that Ms. Cothern
and Ms. Carter had COPD is irrelevant to plaintiffs’ theory.
On behalf of Ms. Boyd, Mr. Jenkins, Mr. McGhee and Ms.
Myles, plaintiffs contend that, because their pre-dialysis serum
bicarbonate levels were all 19 mEq/L, they are outside the range
indicated as Dr. Waikar’s target. But, pursuant to Dr.
Goldfarb’s opinion that low numbers are of concern, such levels,
which reflect acidosis, are inconsistent with plaintiffs’ theory
of general causation.
Finally, plaintiffs cite Dr. Borkan’s testimony that he
would not dial back the bicarbonate level unless the patient had
a pre-dialysis reading that is above 24 mEq/L. They interpret
that to mean that Dr. Borkan would necessarily reduce the
bicarbonate level if the reading were above 24 mEq/L. But Dr.
Borkan’s prior testimony does not support that conclusion. He
testified during the Dial trial that it was the number 28 mEq/L
that constituted a “trigger” for alkalosis. He did not discuss
26 mEq/L, 19 mEq/L or any number in between in articulating his
opinion during Dial. Indeed, he has testified that serum
bicarbonate levels of 26 mEq/L can adjust on their own, without
a change in prescription. Furthermore, despite evidence that
the Dial plaintiff had multiple readings of 26 mEq/L while she
was on dialysis, Dr. Borkan did not testify that such a reading
was of concern.
Dr. Borkan’s ambiguous testimony cannot, standing alone,
create a genuine issue of material fact. Although plaintiffs
are not necessarily bound by the Hakim Memo, it is relevant in
assessing their theory of causation, not least because their
experts all explicitly purported to rely on it in rendering
their opinions. Without any definitive expert testimony
regarding what is normal but with considerable testimony from
their own experts that all of the plaintiffs fell within an
acceptable range, they have failed to adduce competent evidence
in support of their claims.
b. Plaintiff Riben’s Opposition
The plaintiffs who joined in Riben’s opposition argue that
Fresenius has read their theory of causation too narrowly and
that any pre-dialysis bicarbonate level, low or high, could
still lead to a cardiac event solely based on the administration
of a high-bicarbonate dialysate. More directly, plaintiffs
attempt to advance a theory that it is the shift caused by a
high-bicarbonate dialysate alone, regardless of the “total
buffer,” which is important. They seek to divorce the “rapid
shift” from the alkalosis itself.
Such an expansion of their theory is not, however,
articulated in the Second Amended Master Complaint, nor is it
supported by expert testimony. Even more problematic for the
plaintiffs who purport to join the Riben opposition, they
clearly explain in their own opposition that their theory of
general causation is
that excessive bicarbonate “total buffer” . . . in the
“dialysate” . . . is capable of causing metabolic
alkalosis, arrhythmia, sudden cardiac arrest (SCA),
and death.
Moreover, as Fresenius points out, if plaintiffs’ new theory of
causation is that dialysis can dramatically increase serum
bicarbonate in a short period of time, separate from total
buffer levels, it is related to the dialysis process itself and
not the dialysates used. Thus, such an attempt to re-frame the
theory of general causation is unavailing.
2. Proximate Cause Issues
Fresenius next argues that the testimony of plaintiffs’ own
experts indicates that its purported failure to warn could not
proximately cause injury to patients who had serum bicarbonate
levels of 26 mEq/L and lower. For example, plaintiffs’ experts
testified that a physician would not be expected to make a
downward adjustment to the bicarbonate setting of a patient with
a reading of 26 mEq/L or lower. According to Fresenius, that
means plaintiffs have failed to establish a proximate cause
linking the alleged failure to warn to the alleged injuries.
To prevail on any failure to warn claim, plaintiffs must
show that the lack of warning was the proximate cause of their
decedents’ injuries. Santos-Rodriguez v. Seastar Solutions, 858
F.3d 695, 697 (1st Cir. 2017). Thus, plaintiffs must show that
the treating doctors would have done something differently had
they been forewarned as plaintiffs claim they should have been.
See, e.g., In re Neurontin Mktg. & Sales Practices & Prods.
Litig., No. 04-CV-10981-PBS, 2010 WL 3169485, at *3–4 (D. Mass.
Aug. 10, 2010) (noting that “[w]here the manufacturer fails to
provide the physician with an adequate warning, courts have held
that the manufacturer may still be shielded from liability if it
can show that the prescribing physician would not have heeded an
adequate warning”).
In this case, plaintiffs’ experts have testified that pre-
dialysis serum bicarbonate readings of 26 mEq/L and lower do not
require prescription changes. In particular, Dr. Waikar
testified that a range of 22 to 26 mEq/L was “reasonable,” Dr.
Goldfarb testified that he would be concerned about serum
bicarbonate levels that are at “28 or 30 or 35” mEq/L or below
22 mEq/L, and Dr. Fine testified that he asks his nurse
practitioners to let him know if the serum bicarbonate levels of
a patient are above 26 mEq/L. Furthermore, Dr. Fine declared
that he does not find it necessary to adjust a bicarbonate
prescription downward unless the patient is “alkalotic” or if
the patient’s serum bicarbonate levels had drastic upward swings
in a short period of time.
Although Dr. Borkan averred that he would consider making
adjustments for his own patients when their serum bicarbonate
levels exceeded 24 mEq/L, he also attested that he considered 28
mEq/L to be a “trigger” for alkalosis and that a patient’s
multiple prior readings of 26 mEq/L decreased on their own
without any change in prescription, indicating that he did not
believe a reading of 26 mEq/L necessarily required a
prescription change. Plaintiffs whose decedents had pre-
dialysis serum bicarbonate readings of 26 mEq/L or below and who
had no indications of drastic upward swings in a short period of
time have adduced no evidence to support their claims that any
of their treating doctors would have done anything differently.
On the question of proximate cause, it is clear that
plaintiffs could pick and choose among their experts one who
might say that their pre-dialysis serum bicarbonate numbers are
troubling but few of them fall into the category that would have
been of concern to any of the experts with respect to alkalotic
problems. Only plaintiffs McNulty, Ross and Williams were in a
range that any of the experts indicated would have warranted
notification. None of plaintiffs’ experts has testified that he
or she would be inclined to change the dialysate based on a pre-
dialysis serum bicarbonate level of 25 or 26 mEq/L and most of
those experts have said that those numbers fall within their
“target range.” Plaintiffs are not bound by the Hakim Memo but
they were required to produce some reliable evidence to create
genuine issues of material fact as to whether their pre-dialysis
serum bicarbonate levels were too high and would lead a doctor
to change the dialysate prescription. They have not done so.
B. Causation
Fresenius’s second motion for summary judgment contends
that it is entitled to summary judgment against 11 opt-out
plaintiffs because 1) those plaintiffs have not shown that their
decedents died as a result of arrhythmia and 2) the decedents’
injuries are not proximate in time to their last dialysis
treatments. Essentially, Fresenius contends that plaintiffs’
theory of general causation rests on the fact that the
dialysates cause alkalosis, which leads to an arrhythmia
“triggered” by an electrolyte shift. Therefore, plaintiffs must
demonstrate the “right event type” in order to elicit sufficient
evidence to meet their burden. Moreover, Fresenius argues that
an arrhythmia can only be attributed to the acid concentrate in
the dialysate if it occurs within two hours after the dialysis
treatment concludes.
As indicated with respect to the serum bicarbonate motion,
plaintiffs are required to establish causation, both general and
specific. See In re Neurontin Mktg., 612 F. Supp. 2d at 123.
Furthermore,
as is well-established under Massachusetts law,
“expert testimony is required to establish medical
causation.” This applies to both general and specific
causation.
Milward v. Rust-Oleum Corp., 820 F.3d 469, 476 (1st Cir. 2016)
(quoting Reckis v. Johnson & Johnson, 28 N.E.3d 445, 461 (Mass.
2015)). If there is no evidence regarding general causation,
then “judgment as a matter of law [is] necessarily required.”
Id.
1. Event Type
Fresenius’s first argument is based on the Hakim Memo in
which Dr. Hakim made no mention of sepsis, blood clots,
myocardial infarction or anything aside from electrolyte-related
arrhythmias. Because plaintiffs’ claims are premised almost
entirely on Fresenius’s failure to warn of the electrolyte-
related arrhythmias (which can lead to sudden cardiac arrest and
death), Fresenius contends that any other event is outside the
scope of this litigation.
Fresenius next asserts that the theory of general medical
causation espoused by plaintiffs requires proof that the
decedents’ injuries were caused by a cardiac arrhythmia
triggered by an electrolyte shift. Therefore, non-cardiac
events and cardiac events
that are not arrhythmic . . . or that involve
arrhythmias due to triggers other than electrolyte
shifts
do not fall within the scope of this litigation.
The decedents of the following plaintiffs have medical
records indicating that cause of death was something other than
alkalosis leading to arrhythmia: Daniel Carter, Kathy Dennis,
Max Riben and Sophia Walker. In fact, the decedents of
plaintiffs Carter and Walker suffered from sepsis, which led to
either cardiac arrest or to septic shock and neither experienced
arrhythmia. The decedents of plaintiffs Dennis and Riben died
as a result of myocardial infarction.
2. Timing
Fresenius next submits that plaintiffs have failed to
establish that the timing of the alleged injury events were in
close proximity to the patients’ last dialysis treatments. In
particular, Fresenius challenges the claims of plaintiffs Boyd,
Carter, Clark, Dunaway, McGhee, McNulty and Ross.
Fresenius argues that the Hakim Memo relates solely to in-
center cardiac arrests and that plaintiffs’ cardiology experts
testify that it is “rapid” electrolyte shifts during dialysis
that trigger arrhythmia, which occurs suddenly once triggered.
None of the seven identified plaintiffs experienced in-center
cardiac arrest and all of the cardiac arrests occurred seven
hours or more after the conclusion of dialysis: plaintiff Boyd’s
decedent (19 hours), plaintiff Carter’s decedent (36 hours),
plaintiff Clark’s decedent (31 hours), plaintiff Dunaway’s
decedent (45 hours), plaintiff Gallardo Hernandez’s decedent (8
days), plaintiff McNulty’s decedent (30 hours), plaintiff
McGhee’s decedent (8.5 hours) and plaintiff Ross’s decedent (8.5
hours).
Plaintiffs’ general causation expert, Dr. Akar, identified
the “0 to 12 hour” interval after dialysis as “a particularly
high period in which sudden cardiac death due to arrhythmias
occur.” Despite that 0-to-12 hour interval, however, Dr. Akar
concluded that the acetate converts to bicarbonate in the body
instantaneously so that any electrolyte shifting happens “within
minutes, probably even less than minutes.” In essence, any
elevation in bicarbonate, which would “trigger” arrhythmia,
occurs almost immediately, either during dialysis or perhaps
right afterward.
The only potential issue plaintiffs raise results from the
opinion of Dr. Borkan that certain patients metabolize acetate
more slowly such that the blood bicarbonate level would “spike”
and cause acute metabolic alkalosis hours after the end of
dialysis. Based on that opinion, plaintiffs whose decedents
experienced the alleged injury event more than 40 hours after
dialysis have elicited sufficient evidence to create a genuine
issue of material fact as to whether the acetate in the
dialysate solution caused the injuries.
Dr. Borkan’s opinion is largely unsupported by the
evidence, including the studies on which he purports to rely.
The only study that arguably supports his theory regarding slow
metabolizing of acetate is the one referenced in plaintiffs’
opposition, the “Graham Study,” which Fresenius attached to its
reply brief.
The Graham Study, apparently undertaken out of concern for
minimizing the effects of acidosis in hemodialysis patients,
provides some support for the theory that a dialysate
bicarbonate solution used to elevate the level of serum
bicarbonate during dialysis might have an effect on bicarbonate
values for as long as 44 hours.
The Graham Study sought to address the problem of acidosis
in patients undergoing hemodialysis, the method of correcting it
and the consequence of failing to do so. Essentially, the study
monitored nine hemodialysis patients from just after dialysis to
just before a subsequent dialysis 44 hours later. Seven out of
the nine patients experienced “a gradual decline in bicarbonate,
whilst in two there was no change.” The study did not determine
whether it was acetate that led to the elevated serum
bicarbonate level but concluded that a slow linear decline in
bicarbonate after dialysis occurred in seven of the nine
patients and that in eight of the nine, the average of their
post- and pre-dialysis bicarbonate “accurately predicted the
time-averaged . . . bicarbonate concentration.” The study also
looked at the post- and pre-dialysis serum bicarbonate levels of
46 other patients and found that the serum bicarbonate values
were significantly lower three days after dialysis. The study
did not determine why two patients remained steady in their
bicarbonate numbers post- and pre-dialysis.
From that study Dr. Borkan concludes that
there are probably a subset of patients who we send
home from the dialysis unit with substantial metabolic
alkalosis that persists for as long as until the next
dialysis session.
But Dr. Borkan’s conclusion based on the Graham Study still
fails to support his delayed bicarbonate “spike” theory upon
which plaintiffs apparently rely to claim that NaturaLyte and/or
GranuFlo caused their decedents’ alleged injury events.
Dr. Borkan’s theory of delayed acetate to bicarbonate
conversion due to slower metabolism ultimately rests on the idea
of a “bicarbonate spike” that occurs after the completion of
dialysis. That spike, because of excess acetate, occurs at a
time attenuated from the dialysis itself and, according to Dr.
Borkan,
the timing of the spike is a key determinant of the
toxic effects of acute metabolic alkalosis including
[cardiopulmonary arrest] and death both during and
after the procedure.
That “spike” theory is, however, unsupported by any of the
evidence, including the Graham Study. At best, the Graham Study
included two patients whose serum bicarbonate levels after
dialysis remained constant. Neither of those two outliers had a
belated “spike” in their serum bicarbonate levels.
Furthermore, multiple studies cited by Dr. Borkan show that
acetate levels drop quickly soon after dialysis is completed and
he himself testified during the Dial bellwether trial that “most
patients” clear any residual acetate left in their blood from
dialysis “within 30 to 60 minutes.”
Moreover, there is no evidence in the record that any of
the decedents at issue were slow to metabolize acetate. Where a
single expert among many testifies that it is possible that some
patients, although certainly not the majority, may take more
than one hour after dialysis to clear the acetate from their
blood that is not enough to create a genuine issue of material
fact. Furthermore, there is no evidence that such was the case
for these particular plaintiffs and the Graham Study did not
even purport to be about acetate exclusively or to show the
requisite bicarbonate “spike.”
C. Naturalyte
Fresenius moves for summary judgment on the claims of five
plaintiffs involving NaturaLyte, arguing that the admissions of
Dr. Borkan and plaintiffs’ other experts and the outcome in the
bellwether trial, Dial, demonstrate that general causation is
non-existent in any NaturaLyte case.
Once again, causation is a fundamental element of
plaintiffs’ claims and “to prevail in a pharmaceutical personal
injury case,” they must proffer evidence of both general and
specific causation. See Jackson v. Johnson & Johnson & Janssen
Pharms., Inc., 330 F. Supp. 3d 616, 625 (D. Mass. 2018) (quoting
In re Neurontin Mktg., 612 F. Supp. 2d at 123). As an initial
matter, general causation is
established by demonstrating, often through a review
of scientific and medical literature, that exposure to
a substance can cause a particular disease.
In re Neurontin Mktg, 612 F. Supp. 2d at 123. As discussed
above with respect to the causation motion, in Massachusetts,
understanding medical causation is
a matter beyond the common knowledge of the ordinary
layman and proof of it must rest upon expert medical
testimony.
Jackson, 330 F. Supp. 3d at 625 (quoting Hachadourian’s Case,
340 Mass. 81, 84, 162 N.E.2d 663 (1959)).
Although Dr. Borkan endorses plaintiffs’ allegations that
NaturaLyte is dangerous and defective due to the fact that it
contains four mEq/L of acetate, his own testimony and practices
demonstrate that such a charge is unsubstantiated. On numerous
occasions, as outlined above, Dr. Borkan testified that four
mEq/L of acetate is not considered excess acetate. He testified
that his own clinic uses a NaturaLyte product and that he does
not warn his patients of that fact because the amount of acetate
contained in the solution, i.e., four mEq/L, is an acceptable
amount. Furthermore, during his trial testimony in Dial, Dr.
Borkan reaffirmed his prior admission that four mEq/L of acetate
is an “average, background” and “baseline” amount for an acid
concentrate.
Besides Dr. Borkan’s admission, several other experts
retained by plaintiffs confirmed that NaturaLyte or some other
solution containing four mEq/L of acetate is used in their
clinics. Drs. Fine and Miller testified that they treated their
dialysis patients with NaturaLyte. Dr. Miller confirmed that
NaturaLyte is “a fine product” and contained the “standard
amount” of acetate. Drs. Waikar and Goldfarb testified that the
clinics where they treat their dialysis patients use a solution
that contains four mEq/L of acetate, the same amount of acetate
contained in NaturaLyte. Taking the evidence in the light most
favorable to the nonmoving party, the Court discerns no manner
in which a reasonable jury could resolve the general causation
issue in favor of plaintiffs.
Moreover, to defeat a motion for summary judgment,
plaintiffs’ “expert opinion[s] must be more than a conclusory
assertion about ultimate legal issues.” Hayes v. Douglas
Dynamics, Inc., 8 F.3d 88, 92 (1st Cir. 1993). Plaintiffs’
experts Drs. Aroesty, Akar and Lampton opine that excess acetate
in NaturaLyte does indeed increase the risk of cardiac arrest
and death. As discussed above, Dr. Lampton’s reports for all of
the decedents are conclusory, in that they merely recount Dr.
Akar’s conclusions. Similarly, Drs. Aroesty and Akar lump
GranuFlo and NaturaLyte together when discussing the alleged
excess acetate in the solutions, suggesting that both GranuFlo
and NaturaLyte caused the alleged injuries. Such a bare
conclusion does not create a genuine issue of material fact,
particularly in the face of admissions from other experts
indicating that they use products that contain four mEq/L of
acetate with their own dialysis patients. As such, plaintiffs
have produced no competent evidence contrary to what Fresenius
has provided regarding the amount of acetate included in
NaturaLyte and thus defendants’ motion for summary judgment on
plaintiffs’ claims involving NaturaLyte will be allowed.
D. Learned Intermediary Doctrine
Finally, Fresenius argues that all 13 remaining plaintiffs’
claims are barred by the learned intermediary doctrine because
there can be no dispute that the prescribing physicians were
adequately forewarned. The Court agrees.
Plaintiffs’ claims are fundamentally grounded in the
assertion that Fresenius failed to warn them of the dangers of
NaturaLyte and/or GranuFlo. Pursuant to the learned
intermediary doctrine, however, the prescribing physician is the
relevant audience for warnings about a medical device or
prescription drug. Garside v. Osco Drug, Inc., 976 F.2d 77, 80
(1st Cir. 1992) (“Where the product is a prescription drug,
however, it is widely accepted that the manufacturer’s duty to
warn runs to the physician rather than the patient.”); see also
Plourde v. Sorin Grp. USA, Inc., 517 F. Supp. 3d 76, 89 (D.
Mass. 2021). Thus, once Fresenius adequately warned the
physicians providing the dialysis treatments, its duty to warn
was fulfilled. See Garside, 976 F.2d at 80.
Fresenius contends that for plaintiffs in at least three
sets of common circumstances, the record indisputably
demonstrates that physicians were adequately warned. Those
three groups of plaintiffs include: 1) those who received
dialysis treatments preceding their alleged injury at a
Fresenius dialysis unit at any time, 2) those who received
dialysis treatments preceding their alleged injury at a DaVita
dialysis unit after November 4, 2011 and 3) those who received
dialysis treatments preceding their alleged injury at any
dialysis unit after March 29, 2012.
As explained in the fact section above, physicians and
facility staff at Fresenius dialysis units were provided several
memoranda from the Fresenius Chief Medical Office over the
course of a decade. The memoranda are adequate warnings as a
matter of law because they specifically mention the
circumstances complained of. They repeatedly cautioned that
total buffer is the sum of the acetate and bicarbonate and that
acetate, once in contact with a patient’s blood, is
metabolically converted into bicarbonate. Specifically, the
memoranda urged physicians to
[o]bserve and monitor the patient’s serum bicarbonate
level to determine that the prescribed dialysate
bicarbonate is actually being delivered and is
appropriate for that particular patient. If not, the
physician should establish a new bicarbonate
prescription and the staff should readjust the
bicarbonate setting as is appropriate.
Because the unrefuted evidence demonstrates that Fresenius
provided adequate warnings, it has discharged its duty and
summary judgment will be allowed, regardless of how the
physicians responded to those warnings.
With respect to patients treated at DaVita clinics,
including the decedent of plaintiff Riben, there is no dispute
that DaVita clinics received the Hakim memo the same day it was
released to Fresenius physicians, November 4, 2011. Plaintiff
Riben’s decedent suffered her injuries after that date. By the
time of her injuries, DaVita clinics had already been well
informed of the warnings for NaturaLyte and GranuFlo and as
such, her attending nephrologists were also aware of the dangers
of the products.
In the third category, the decedents of plaintiffs Cameron,
Riben, Gallardo Hernandez, Walker and Williams all suffered
injuries after the distribution of the March 29, 2012 “Important
Prescribing Information” notification. That warning
specifically addressed the matters of which the plaintiffs
complain and, accordingly, it cannot be disputed that such a
warning was adequate.
Finally, it does not escape the Court that there was
testimony from nephrology experts indicating that nephrology
fellows know from early on in medical school that acetate
metabolizes into bicarbonate in the liver. Plaintiffs claim
that prescribing physicians were unaware of such information and
needed to be informed of it via warnings. Fresenius has
proffered evidence, including testimony from plaintiffs’ own
experts, that all competent nephrologists understand that
acetate converts to bicarbonate and can read the labels on the
products and see that they contain acetate. Plaintiffs, on the
other hand, have failed to produce any evidence to show that
physicians would have changed their prescribing decisions if
different disclosures had been made. Therefore, a reasonable
jury could not resolve such an issue in favor of plaintiffs and
summary judgment for defendants will be allowed on the basis of
the learned intermediary doctrine.
ORDER
For the reasons outlined above, defendants’ motions for
summary judgment on the claims of opt-out plaintiffs with
respect to:
1) NaturaLyte (Docket No. 1906) is ALLOWED;
2) elevated serum bicarbonate levels (Docket No. 1913) is
ALLOWED;
3) the learned intermediary doctrine (Docket No. 1923) is
ALLOWED; and
4) non-arrythmia events or injuries not proximate in time
to the last dialysis (Docket No. 1933), is ALLOWED.
Defendants’ motion for summary judgment in Case No. 18-
11224, with respect to plaintiff Gallardo Hernandez (Docket No.
31) is ALLOWED.
Accordingly, the cases brought by plaintiffs Dunaway, Boyd,
McNulty, Cameron, Carter, Clark, Dennis, Ross, Williams, Walker,
McGhee, Riben and Gallardo Hernandez are DISMISSED.
So ordered.
/s/ Nathaniel M. Gorton
Nathaniel M. Gorton
United States District Judge
Dated: September 7, 2023