Since the introduction of omeprazole in 1989, proton pump inhibitors (PPIs)
have steadily become the mainstay in the treatment of acid-related disorders.
The general clinical uses of PPIs include – healing of peptic ulcer disease,
peptic ulcer-related gastrointestinal bleeding, prevention of NSAID (Motrin-type
medications) induced gastroduodenal ulcers, treatment of Zollinger-Ellison
syndrome, treatment of erosive gastritis, treatment of functional dyspepsia,
and eradication of H. pylori infection.
There are currently six FDA approved PPIs – omeprazole (Prilosec),
esomeprazole (Nexium), lansoprazole (Prevacid), dexlansoprazole (Dexilant),
pantoprazole (Protonix), and rabeprazole (AciPHex).
Despite an excellent safety profile throughout their first two decades of
use, the nearly universal popularity of PPIs has prompted several concerns
about both their short- and long-term effects
Since their inception, long-term use of PPIs have been known to have
potential side effects, including: decreased absorption of certain vitamins,
increased susceptibility to a multitude of enteric infections such as C.
difficile. However, most of the published evidence is inadequate to establish a
definite association between PPI use and the risk for development of serious
After 25 years of use, it has become increasingly apparent that PPIs cause
certain types of kidney disease – namely acute interstitial nephritis (AIN) and
more recently, chronic kidney disease (CKD).
More recently, a series of class actions have been brought against the drug
manufacturers based on strict liability. The basis of the suit is that PPIs
cause certain types of kidney problems, that the manufacturer was aware of this
and failed to warn consumers. Failure to warn is a marketing defect cause of
action. Manufacturers are required to issue warnings about any serious side
effects associated with pharmaceuticals and provide appropriate instructions.
Failure to warn may be based on strict liability or negligence. If a negligence
cause of action is asserted, plaintiffs will need to establish that the
manufacturer failed to use reasonable care to provide appropriate information
about a foreseeable risk significant enough to justify the cost of giving the
information. However, in a strict liability cause of action, plaintiffs need to
prove that there was a defective warning and that it caused the injuries
suffered by the plaintiff. The primary issue in either case is whether the
warning was adequate.
Currently, all claims are pending in court. There are more than 4200. The
federal cases are being consolidated. Other lawsuits are pending in state and
local courts, but I found none that have been set for trial.
These cases appear to me to be most likely than not destined for ultimate
failure. Some background first.
One of the alleged types of injury – acute interstitial nephritis (AIN) is a
type of acute kidney damage that is very specific. It is characterized by the
presence of inflammatory cells, including eosinophils (a type of white blood
cell) in the urine and blood. It is readily identifiable on kidney biopsy.
It is caused predominantly by certain drugs (the classic of which is
methicillin), but many other drugs are implicated. In fact, virtually any drug
can cause AIN. A few known drugs and classes include: 1,2
- Nonsteroidal anti-inflammatory agents (NSAIDs),
including selective cyclooxygenase (COX)-2 inhibitors (i.e. motrin,
- Penicillins and cephalosporins (related antibiotics)
- Rifampin (anti-tubercular drug)
- Ciprofloxacin (a common antibiotic used for urinary
- Cimetidine (another type of antacid)
AIN can also has other causes:
- Allopurinol (used to treat gout)
- Infections: multiple organisms
have been associated with AIN, some of which are indolent (you never know
you have it). Some are Legionella, Leptospira, cytomegalovirus,
Streptococcus, tuberculosis, Epstein-Barr and others.
- Systemic diseases: these primarily
include SLE, sarcoidosis, and Sjögren’s syndrome.
AIN is a very specific type of kidney injury. It’s pathogenesis (how it
progresses) has been well described, as is it’s histologic characterization. 4
This would appear to favor plaintiffs.
Acute interstitial nephritis (AIN) is characterized by the presence of
inflammatory infiltrates and edema within the interstitium, usually associated
with an acute deterioration in renal function.5,6
The development of drug-induced AIN is not dose dependent, and recurrence or
exacerbation can occur with a second exposure to the same or a related drug.4
This would seem to work against plaintiffs.
The actual incidence of AIN is reported to be less than 1% of the total
population. This does not indicate the actual number of those affected since 1%
represents a larger number than 0.01%. Moreover, most studies report the
incidence as a percentage of kidney biopsies taken. When the analysis was
restricted to patients with acute kidney failure, AIN accounted for 15–27% of
lesions.7,8,9. This type of reporting makes it very difficult to extrapolate to
the actual incidence amongst the general population.
There have been a number of observational studies that have linked proton
pump inhibitors (as a class) with AIN. To date, the best study comes out of
John Hopkins and was published in 2016. Based on two large collections of patient,
the authors report proton pump inhibitors (PPIs) as an independent risk factor,
not only for AIN, but also for chronic kidney disease disease (CKD). https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2481157
This study was observational. An observational study is a great way of
isolating adverse effects of drugs as they exist in the real world. On the
other hand, they are not the “gold standard” study – which is a double blind,
placebo-controlled study. Observational studies have significant limitations.
For instance, all in the cohort from Hopkins were likely to be taking multiple
medications, have obesity and hypertension. Consider that, in both groups,
there was a statistically significant increase in concomitant use of NSAIDs
(which are much more likely to cause kidney injury) among PPI users, which is a
confounding variable. They report that PPI’s were associated with a 20-50%
greater risk of developing CKD when compared to nonusers, with a 10-year
absolute risk increase of 3.3%.
The proper interpretation of this study is that it identified a small, but
significant increase in kidney injury from PPIs. In other words, there is a
small potential risk that is applied to a large number of people who take these
drugs. The association is probably real and demands that physicians and
patients be cautious.
Using the Department of Veterans Affairs national databases, Xie et al also
assessed kidney outcomes in patients using PPI’s and another antacid
medication. In this study, there was a significant association discovered
between duration of exposure and worsening outcomes. This tidbit probably
favors the defense.
The connection between AIN and CKD
The study in JAMA unveiled another variable – that PPIs are associated with
chronic kidney disease. In fact, this makes sense and is not confounding at
all. The mechanism for the development of CKD is likely the result of AIN,
which if untreated progresses to the more permanent injury.
We should note, all of the studies to date are retrospective observational
studies. Aside from intrinsic biases, are incapable of demonstrating a definite
causal relationship. With adverse effects being uncommon, the absolute risk to
patients remains quite small. The cases that I’ve looked at are misusing and
misquoting the studies. In any event, the fact that they are observational
would seem to favor the drug manufacturers.
A complaint alleging that a manufacturer failed to provide adequate warning
of a product’s dangers may be brought under at least three legal doctrines:
strict liability, implied warranty, and negligence. 10
Failure to warn of a product’s dangerous propensities may,
under appropriate circumstances, constitute a breach of duty upon which an
action for negligence might be predicated, and may also serve as the basis for
holding a defendant liable under a strict products liability theory. Lewis
v. Lead Industries Ass’n, Inc
., 342 Ill. App. 3d 95, 276;
Generally speaking, a prima facie claim that a product was defective due to
a failure to warn consists of three elements in addition to those in a state’s
defective product statute:
(1) the defendants had a duty to warn; (2) the lack of an adequate warning
made the product unreasonably dangerous, and therefore defective; and (3) the
lack of an adequate warning was
a proximate cause of the injury. Kellogg v. Wyeth
762 F. Supp. 2d 694 (D. Vt. 2010) (applying Vermont law); Surre v. Foster
, 831 F. Supp. 2d 797 (S.D. N.Y. 2011) (applying New York law).
In area of advertising of pharmaceuticals, rebuttable presumption exists
that when manufacturer complies with Food and Drug Administration (FDA)
advertising, labeling, and warning requirements, the manufacturer has satisfied
its duty to warn consumer about potentially harmful side effects of its
product. Perez v. Wyeth Laboratories Inc
., 161 N.J. 1, 734 A.2d
I located a complaint against PPI manufacturer, Astra Zeneca, recently filed
in federal court. My purpose was to examine and evaluate their evidence.
Evidence – in the form of medical studies – would have to exist and establish
that Astra Zeneca was aware of the risk. For the plaintiffs to prevail, there
had to exist the predicate medical knowledge that PPIs caused kidney damage. If
the science exists, this would establish a duty to warn.
I was able to locate one case that had been filed. The original complaint is
telling, in that it spells out all of the evidence that the plaintiffs have.
What I found was surprising.
In Re: Proton-Pump Inhibitor Products Liability Litigation (No. II),
Clarice Armstrong v. Astra Zeneca Pharmaceuticals LP –
complaint was filed in New Jersey District Court on September 22, 2017.
In the section entitled “Factual Allegations,” the plaintiffs allege that
the manufacturer was aware that PPIs cause renal injury. In support of this,
they cite and discuss a number of publications. In fact, as I reviewed each of
the studies, I found their evidence to be largely insufficient to support their
claim. Below is a synopsis of their evidence. I noticed that there were three
occasions where the plaintiffs double quoted the same study, referencing them
differently. In other words, the same study was re-packaged and presented
separately. This gave the appearance (at least to me) of there being more data
than there actually was.
- Simpson, I, et al., Proton pump inhibitors and
acute interstitial nephritis: Report and analysis of 15 cases,
Nephrology (2006)11: 381-85
- This publication was a case report, based out of the
largest referral center in Austria.
- Note that in all cases, withdrawal of the PPI resulted
- This particular case report was limited in being a
small sample size without comparator groups.
- As a result, no estimate of the effect size was
- Determining if an association between PPI use and AIN
requires studies using larger sizes and control groups.
- At best, this study is inconclusive.
2. Geevasinga, N, Coleman PL, Webster AC. Proton pump inhibitors and
acute interstitial nephritis, Clin Gastroenterol Hepatol.
2006 May, 4(5):
- This was a retrospective case review based on thousands
of potential cases at multiple teaching hospitals extracted from the
Therapeutic Goods Administration of Australia (TGA).
- From this, 18 cases were identified.
- Withdrawal of the PPI resulted in resolution in all
- Given the volume of PPI prescribed, no credible
conclusion can be reach.
- Again, there is no comparator groups.
3. Klepser, D., Collier, D.,
Cochran, G., Proton pump inhibitors and acute kidney injury: a nested case
. BMC Nephrology 2013, 14:150
- This study was another retrospective case-control study
looking a privately insured population in a single (undisclosed)
Midwestern state of 185,000 patients which identified 195 patients who
developed AIN and were taking PPIs.
- To identify AIN, multiple ICD-9 codes (medicare
diagnostic codes) were used to capture unrecognized cases.
- PPI exposure status was obtained through prescription
- This analysis, like the first two studies, has certain
- They include misclassification bias (including
instances of renal injury not associated with PPI exposure, thus
overestimating the true relationship between PPI use and renal disease).
- On the other hand, this study did not capture OTC PPI
utilization – therefore potentially underreporting prevalence.
- Surveillance bias is another source of overestimation,
a potential erroneous estimate of relative risk and odds ratio
4. Brewster, UC, Perazella, MA, Acute kidney injury following proton
pump inhibitor therapy
, Kidney International, 2007:71, 589-93
- This publication was a single case presented by a
student followed by a discussion.
- It is not a persuasive argument.
5. Blank, ML, et al. A nationwide nested case-control study indicates an
increased risk of acute interstitial nephritis with proton pump inhibitor use
Kidney International, 2014: 86, 837-44
- According to this study out of New Zealand, “The
magnitude of the suspected increase in risk of acute interstitial
nephritis among proton pump inhibitor users is uncertain.”
- The cohort identified 572,661 patients retrospectively.
- From this, 72 validated cases of which 46 were definite.
- Aside from the potential sources of error described
previously for case control studies, the conclusion was “absolute risks
were very low.”
Finally, the previously mentioned JAMA publication was referenced in the
case. Recall the JAMA article was published in 2016 and this case was published
PPIs were clinically introduced more than 25 years ago. They have since
proven invaluable, safe and highly effective.9 Given the prevalence of acid
related GI disorders, the PPIs are considered life-saving by the medical
community. They came into existence at a time when little else outside of
surgery to remove affected portions of the upper GI system were available.
Consider that by 2009, more than 119 million PPI prescriptions were filled in
the United States, accounting for nearly $14 billion dollars in prescription
PPI sales, in addition to billions more in over-the-counter (OTC) sales. This
according to IMS health, US Sales and Prescription Information.http://www.imshealth.com/portal/site/ims/menuitem.d248e29c86589c9c30e81c033208c22a/?vgnextoid=9c61ba440c900310VgnVCM10000071812ca2RCRD&cpsextcurrchannel=1
The case that I analyze above is based on no evidence based data. In short,
the scientific articles did not raise an objectively reasonable safety concern.
In my view, they did not trigger a duty to warn. The most influential research
to date which makes a causal connection is from JAMA. I realize that my
inexperience in the law renders me naïve. Yet the above case lacked evidence.
Double quoting and re-identifying the same data differently gives the
impression that there is more data than there actually is. Using the study from
Yale – which was a mere “poster presentation” by a student – appears deceptive.
In short, without a tobacco industry type smoking gun, there does not appear to
be adequate evidence that the manufacturers knew of a defective drug. Even
against a preponderance standard, the evidence will not survive. It makes more
sense that after an extended period of time – after acquiring data based on the
consumption of billions of tablets, we have only recently learned of the
potential for kidney damage.
The evidential value of the epidemiological studies is difficult to
determine. Over a period of years, there was little evidence that PPIs cause
renal injury. Knowledge of harm has occurred temporally. Twenty years ago, the
evidence was sparse. Over time, the evidence began to accrue in the form of
case studies, followed by observational studies. Only recently – due to the
JAMA report, do we have a clearer picture, sufficient to create a causal link.
Even with the JAMA report, causation is not necessarily established.
The primary issue in these cases is whether the epidemiological evidence
submitted by the plaintiffs clearly proves the causal relationship between PPI
use and disease.
Epidemiology studies the distribution and determinants of diseases in a
defined population group. Hence, epidemiological evidence tends to have
statistical and stochastic characteristics. The plaintiffs use epidemiological
evidence to demonstrate a causal relationship between a risk factor and a
disease by showing that there is a substantial probability of a certain disease
developing in a certain group of people exposed to a risk factor, and that the
plaintiffs themselves belong to that group and have developed the disease after
being exposed to the risk factor.
In the present case, defendants will successfully argue that statistical and
stochastic epidemiological evidence is not sufficient to prove proximate or
Proving “but for” causation is difficult. Renal disease can be caused by
many different levels of insult. Even AIN, which is a more specific disease has
hundreds of causes. It is tenuous at best to make the causal connection at the
individual level. This was seen in tobacco lawsuits – factors other than
smoking are involved in the development of a disease, and also because of the
lapse of time between smoking and the manifestation of the disease. On the
other hand, the epidemiological evidence in the tobacco lawsuits was absolute
and rock solid. The epidemiological association between the disease and the
risk factor was proven to be significant. However, even in tobacco litigation,
the epidemiological evidence was group-level and not individual-level, a hurdle
that was overcome on a theory of “probability of causation.”
From tobacco litigation, we observe the presence of flawless scientific
studies. These were relentlessly attacked and withstood the challenge. The link
between science/statistics and harm could be made to a preponderance standard,
only because the science was not disputable. In fact, the data was well
established by the early 1970s. In the present case, the evidence was sparse
from beginning up until the present day. The best evidence (the VA and Hopkins
studies) only recently came about. It is unclear whether these will be
sufficient to create a causal link by a preponderance standard.
The reason the plaintiffs attempted to prove causation with epidemiological
evidence is because the types of evidence presented in other tort lawsuits were
not useful in proving the causal relationship in tobacco lawsuits. Most
compensatory damages in traditional tort lawsuits, e.g., an injury caused by a
fistfight, result from a relatively recent event and have relatively simple
causal links, usually explained by a basic understanding of physics. However,
proving the causal relationship in the case presented was much more
complicated. It appears likely that PPI users are progressively more vulnerable
to injury as a function of time – manifesting perhaps 10 or more years after
their initial exposure. Even this statement is ambiguous in that the data is
unclear. During this period, various other factors (i.e. consuming NSAIDs,
antibiotics, the presence of certain systemic diseases) may all play a part in
the development of the disease. In many cases, more than one exposure is
likely. Over a ten year period, it would be challenging to find an individual
who has not consumed a Motrin (a known nephrotoxic drug) and an antibiotic.
Of course, the issue, a complex one, has many sides – including numerous
that are competing. At one point, prescription PPIs began to include a suitable
warning in their package insert. When PPIs were made available OTC, a similar
warning on the product insert was not included. It’s hard to imagine why such a
warning was excluded. Although the OTC formulations were not identical, they
were sufficiently similar. Whatever the reason ($$), it makes the entirety of a
complex issue much simpler. It makes no sense, on one hand, to include a
similar warning on the OTC product, since I doubt that most who are in need of
the medication are able to read the small print contained on package inserts.
DM, Kelly CJ, Acute interstitial nephritis. J Am Soc
Nephrol. 1998; 9(3):506.
- Schubert C, Bates WD, Moosa MR. Acute
tubulointerstitial nephritis related to antituberculous drug therapy.
Clin Nephrol 2010; 73:413.
- Neilson EG, Pathogenesis and therapy of
interstitial nephritis. Kidney Int. 1989; 35(5):1257.
- Cameron, J.S. Allergic interstitial nephritis:
clinical features and pathogenesis.Q J Med. 1998; 66: 97–115
- Rossert JA, Fischer EA. Acute interstitial
nephritis. In: Comprehensive Clinical Nephrology, 2, Johnson RJ,
Feehally J (Eds), Elsevier Limited, Philadelphia 2003. Vol 1, p.769.
- Ten RM, Torres VE, Milliner DS, Schwab TR, Holley KE,
Gleich GJ. Acute interstitial nephritis: immunologic and clinical
aspects. Mayo Clin Proc. 1988; 63(9):921.
- Haas, M., et al., Etiologies and outcome of acute renal
insufficiency in older adults: a renal biopsy study of 259 case. Am J
Kidney Dis. 2000; 35: 433-47.
- Farrington, K., Levison, D.A. et al. Renal biopsy in
patients with unexplained renal impairment and normal kidney size. Q J
Med. 1989; 70: 221-33.
- Baker, R.J. and Pusey, C.D. The changing profile of
acute tubulointerstitial nephritis. Nephrol Dial Transplant. 2004; 19:
- See, for example, McNeal v Hi-Lo Powered
Scaffolding, Inc. (1988) 266 App DC 473, 836; Petree v Victor
Fluid Power, Inc. 831 F2d 1191; Payne v Soft Sheen Products, Inc.
(1985, Dist Col App) 486 A2d 712; Sprankle v Bower Ammonia &
Chemical Co. (1987, CA5 Miss) 824 F2d 409.
- Strand, D., Kim, D., Peura, D., 25 Years of Proton
Pump Inhibitors: A Comprehensive Review Gut Liver. 2017 Jan; 11(1):