- Email: [email protected]
Acute renal failure with sodium-glucose-cotransporter-2 inhibitors: Analysis of the FDA adverse event report system database
Accepted Manuscript Acute Renal Failure with Sodium-Glucose-Cotransporter-2 Inhibitors: Analysis of the FDA Adverse Event Report System Database Amichai Perlman, PharmD, Samuel N. Heyman, MD, Ilan Matok, PhD, Joshua Stokar, MD, Mordechai Muszkat, MD, Auryan Szalat, MD PII:
S0939-4753(17)30238-7
DOI:
10.1016/j.numecd.2017.10.011
Reference:
NUMECD 1793
To appear in:
Nutrition, Metabolism and Cardiovascular Diseases
Received Date: 5 May 2017 Revised Date:
3 September 2017
Accepted Date: 6 October 2017
Please cite this article as: Perlman A, Heyman SN, Matok I, Stokar J, Muszkat M, Szalat A, Acute Renal Failure with Sodium-Glucose-Cotransporter-2 Inhibitors: Analysis of the FDA Adverse Event Report System Database, Nutrition, Metabolism and Cardiovascular Diseases (2017), doi: 10.1016/ j.numecd.2017.10.011. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Acute Renal Failure with Sodium-Glucose-Cotransporter-2 Inhibitors: Analysis of the FDA Adverse Event Report System Database
1
RI PT
Amichai Perlman, PharmD1,2, Samuel N Heyman, MD1, Ilan Matok, PhD2, Joshua Stokar, MD1, Mordechai Muszkat, MD1, Auryan Szalat, MD1
Corresponding Author: Amichai Perlman
SC
Department of Internal Medicine, Hadassah University Hospital, Jerusalem, Israel. Division of Clinical Pharmacy, Institute for Drug Research, School of Pharmacy, Faculty of Medicine, The Hebrew University of Jerusalem, Israel.
2
Address: POB 12065. Postal code 9112001. Jerusaelm, Israel.
M AN U
Phone: +972-2-675-7578, +972-50-517-2768 Fax: +972-2-675-7246
AC C
EP
TE D
Email address: [email protected]
ACCEPTED MANUSCRIPT
Abstract
Background and Aims: Sodium-glucose-cotransporter-2 (SGLT2) inhibitors have
RI PT
recently been approved for the treatment of type II diabetes mellitus (T2DM). It has been proposed these agents could induce acute renal failure (ARF) under certain
conditions. This study aimed to evaluate the association between SGLT2-inhibitors and
SC
ARF in the FDA adverse event report system (FAERS) database.
Methods and Results: We analyzed adverse event cases submitted to FAERS
M AN U
between January 2013 and September 2016. ARF cases were identified using a structured medical query. Medications were identified using both brand and generic names. During the period evaluated, 18,915 reports (out of a total of 3,832,015 registered in FAERS) involved the use of SGLT2-inhibitors. SGLT2-inhibitors were
TE D
reportedly associated with ARF in 1,224 of these cases (6.4%), and were defined as the “primary” or “secondary” cause of the adverse event in 96.8% of these cases. The proportion of reports with ARF among reports with SGLT2 inhibitor was almost three-
EP
fold higher compared to reports without these drugs (ROR 2.88, 95% CI 2.71-3.05, p<0.001). The proportion of ARF reports among cases with SGLT2-inhibitors was
AC C
significantly greater than the proportion of ARF among cases with T2DM without SGLT2-inhibitors (ROR 1.68, 95% CI 1.57-1.8, p<0.001). Among the SGLT2-inhibitors, canagliflozin was associated with a higher proportion of reports of renal failure (7.3%), compared to empagliflozin and dapagliflozin (4.7% and 4.8% respectively, p<0.001).
ACCEPTED MANUSCRIPT
Conclusion: SGLT2-inhibitors are associated with an increase in the proportion of reports of ARF compared to other medications. SGLT2-inhibitor agents may differ from
RI PT
one another in their respective risk for ARF.
AC C
EP
TE D
M AN U
SC
Keywords: Diabetes; Renal failure; SGLT-2 inhibitors; Adverse effects.
ACCEPTED MANUSCRIPT
Background Sodium-glucose-cotransporter-2 (SGLT2) inhibitors, have been approved in
RI PT
recent years for use in the treatment of type II diabetes mellitus (T2DM). Currently approved agents include canagliflozin, dapagliflozin, and empagliflozin. These agents have all proven effective treating T2DM in terms of HbA1c reduction, as well as
additionally beneficial in terms of blood pressure reduction, weight loss, and improved
SC
lipid profile.[1] The large randomized control trial EMPA-REG, demonstrated that
M AN U
empagliflozin reduces cardiovascular outcomes.[2] While empagliflozin was associated with an early reduction in eGFR following initiation,[3] this study reported empagliflozin slowed progression of kidney disease and reduced rates of clinically relevant renal events[3]. Similar positive results have recently been reported with cangliflozin in the CANVAS trial.[4]
TE D
These positive results indicate SGLT-2 inhibitors constitute a significant advancement in the pharmacological management of T2DM. However, concern has
EP
been raised regarding the possibility of an increased risk for acute renal failure (ARF) in some individuals using these agents.[5] SGLT-2 inhibitors induce diuresis which can
AC C
lead to intravascular volume depletion. These agents also reduce trans-glomerular pressure, leading to reduced glomerular filtration. Lastly, SLGT-2 inhibitors increase medullary oxygen consumption, increasing the risk for hypoxia.[6] These effects might be clinically important under circumstances of volume depletion, a-priori reduced transglomerular pressure (such as with concomitant use of diuretics or agents blocking the rennin-angiotensin axis) and where there is a predisposition to medullary hypoxic injury, such as with use of nonsteroidal anti-inflammatory drugs (NSAIDs) or radiocontrast
ACCEPTED MANUSCRIPT
agent. On June 14th, 2016, the U.S. Food and Drug Administration (FDA) strengthened an existing warning about the risk of acute kidney injury for canagliflozin and dapagliflozin, following evaluation of cases reported from March 2013 through October
RI PT
2015.
Meta-analyses of data from randomized controlled trials of SGLT2-inhibitors
reported findings suggesting possible acute decline in renal function. A meta-analysis
SC
published in 2013 included 13 randomized controlled studies and reported that In
M AN U
patients with moderate renal impairment, use of dapagliflozin or high doses of canagliflozin was associated with increased incidence of renal-related adverse events.[7] A subsequent meta-analysis of randomized controlled studies published in 2016, reported ”these agents are associated with a small increase in serum creatinine,[8] while a more recent meta-analysis including 58 studies reported that
TE D
canagliflozin and dapagliflozin were significantly associated with a greater risk of composite renal events compared to control groups, while empagliflozin was associated
EP
with decreased risk.[9]
In this study, we aimed to evaluate the association between SGLT2-inhibitors
AC C
and ARF in the FDA adverse event report system (FAERS) database, and characterize reports linking these agents to acute renal failure.
Methods
We analyzed adverse event cases submitted to FAERS between January 2013 and September 2016. FAERS is a database that contains information on adverse event reports submitted to FDA. The database is designed to support the FDA's post-
ACCEPTED MANUSCRIPT
marketing safety surveillance program for medication. Reports of adverse events are submitted by healthcare professionals (such as physicians, pharmacists, nurses and others), consumers (such as patients, family members, lawyers and others), and drug
RI PT
manufacturers and marketers. Drug manufacturers, distributors, and license holders, are legally bound to actively collect and report all adverse effects related to their
SC
products.
ARF cases were identified using Medical Dictionary for Regulatory Activities
M AN U
(MedDRA) terms related to acute renal failure. We used a predefined grouping of ARF related MedDRA terms based on the broad Standardized MedDRA Query (SMQ) that has been developed by the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH).[10] Cases involving SGLT2-inhibitors were identified using the terms for the generic and brand
TE D
names of the currently approved SGLT-2 inhibitors canagliflozin, dapagliflozin, and empagliflozin. Cases using other medications for the treatment of diabetes were
EP
identified using the reported indications, and a list of relevant preferred MedDRA terms. Use of concomitant ACE-inhibitors, ARBs, diuretics, and NSAIDs, was identified using
AC C
the terms for the generic and brand names of these medications. For a more detailed list of terms used for the analysis see appendix.
We evaluated the proportion of cases reporting renal failure with SGLT2 -
inhibitors in three ways. First, we compared the proportion of cases reporting ARF with SGLT2-inhibitors to the proportion of cases reporting ARF with all other medications. Second, we compared the proportion of cases reporting acute renal failure with SGLT2-
ACCEPTED MANUSCRIPT
inhibitors to the proportion of cases reporting acute renal failure among cases using other medications for the treatment of diabetes. Lastly, we compared the proportion of
RI PT
cases reporting acute renal failure among the specific SGLT-2 inhibitor agents.
We collected descriptive data on the characteristics of cases reporting acute renal failure with SGLT2-inhibitors, including age, weight, level of suspicion of SGLT2-
SC
inhibitor induced adverse effect (“primary suspect”, “secondary suspect”, “interacting”, or “concomitant"), concomitant medications (angiotensin-converting-enzyme inhibitors
M AN U
(ACE-I), angiotensin II receptor blockers (ARBs), diuretics, non-steroidal antiinflammatory drugs (NSAIDS), and outcome. For outcome we used the data available in FAERS. Adverse event case reports to the FDA are required to include classification of outcome according to U.S. reporting regulations. Serious adverse event cases are classified with one or more of the following possible outcome outcomes: “Death”, “Life-
TE D
Threatening”, “Hospitalization (Initial or Prolonged)”, “Disability”, “Congenital Anomaly”, “Required Intervention to Prevent Permanent Impairment/Damage”, “Other Serious
EP
Important Medical Event”. The outcome from the latest version of a case was used for this study. Characteristics of cases reporting renal adverse effects with SGLT2-
AC C
inhibitors were compared to characteristics of cases reporting all other adverse effects with SGLT2-inhibitors.
In order to evaluate the possibility that the FDA warning issued on June 2016
impacted the relative reporting rate of renal failure with SGLT2-inhibitors, we evaluated the association between SGLT2-inhibitors and renal failure using a multiple logistic regression model including terms for use of SGLT2-inhibitors as well as the date the
ACCEPTED MANUSCRIPT
adverse event was reported. Reporting dates were dichotomized as “before” or “after” the FDA warning. We evaluated whether the association between SGLT2-inhibitors and
SGLT2-inhibitors and report date status.
RI PT
acute renal failure was modified by this warning by including an interaction term for
Difference in reports of acute renal failure was reported using the reporting odds
SC
ratio (ROR) as a measure of disproportionality, and statistical significance was tested using chi-square and logistic regression.[11–15] Data processing and analysis were
M AN U
conducted using R version 3.3.2. Before the analysis, duplicate reports were removed according to the FDA’s recommendation for adopting the most recent case version.[14]
Results
During the period evaluated, 3,832,015 unique adverse event cases were
TE D
reported in FAERS. Of these, 18,915 reports involved the use of SGLT2-inhibitors. Individuals reported with SGLT2-inhibitor associated adverse events were on average
EP
58 years of age, and weighed 95 kg. SGLT2-inhibitors were reportedly associated with ARF in 1,224 cases (6.5%), and were judged to be the “primary” or “secondary” cause
AC C
of the adverse event in almost all cases (96.8%). Many of the cases reporting SGLT2inhibitor use and acute renal failure reportedly led to hospitalization, or to prolongation of hospitalization (n=521, 42.6%), and 16 cases reportedly ended with death (Table 1).
The number and proportion of cases reporting acute renal failure using three
analyses are detailed in Figure 1. The first analysis evaluated the proportion of cases reporting acute renal failure with SGLT2 -inhibitors compared to the proportion of cases
ACCEPTED MANUSCRIPT
reporting acute renal failure with all other medications. In this analysis, cases with SGLT2 -inhibitors were associated with a significantly greater proportion of reports of acute renal failure compared to cases with all other medications (ROR 2.88, 95% CI
RI PT
2.71-3.05, p<0.001). When restricting analysis to reports prior to the FDA warning, a similar signal was evident (ROR 2.85, 95% CI 2.67-3.03, p<0.001). However, there was a significant interaction between the FDA warning and the rate of report, such that the
M AN U
3.39-4.46, p<0.001; p for interaction <0.001).
SC
relative reporting rate of renal failure was greater following the report (ROR 3.9, 95% CI
SGLT2-inhibitors are used for the treatment of T2DM, itself a major risk factor for development of kidney disease, possibly introducing confounding by indication. We therefore further evaluated the proportion of cases reporting acute renal failure by restricting the comparison group to cases using other medications for treatment of
TE D
diabetes mellitus. In this analysis, we found SGLT-2 inhibitors were associated with a more moderate, but still significant, increase in the proportion of reports with acute renal
EP
failure (ROR 1.68, 95% CI 1.57-1.8, p<0.001). When examining the proportion of reports of acute renal failure among the individual SGLT-2 inhibitor agents, we found
AC C
canagliflozin was associated with a greater proportion of reports of acute renal failure (7.3%), compared to empagliflozin and dapagliflozin (4.7% and 4.8% respectively, p<0.001, Figure 2).
The characteristics of cases with SGLT2-inhibitors reporting acute renal failure,
differed from cases with SGLT2-inhibitors not reporting acute renal failure. Cases reporting ARF with SGLT2-inhibitors were more likely to be male, tended to be
ACCEPTED MANUSCRIPT
overweight, and much more likely to report use of concomitant diuretics, or ACEinhibitors or ARBs, compared to cases reporting other adverse events with SGLT2-
RI PT
inhibitors. Rates of reported use of NSAID agents were similar in both groups (Table 3).
Discussion
The results of this study show that in the FDA adverse events database, SGLT2-
SC
inhibitors are associated with increased odds of reporting of acute renal failure. The strength of this association (ROR 2.88, 95% CI 2.71-3.05, p<0.001) met widely
M AN U
accepted and validated criteria indicating a likely signal between these medications and acute renal failure, and the outcome of a large proportion of cases was deemed serious or led to new or prolonged hospitalization. The association between SGLT2-inhibitors and acute renal failure remained significant when comparing cases with SGLT2-
TE D
inhibitors to cases using other medications for the treatment of T2DM. Further analysis revealed that this signal was driven primarily by a disproportionate number of cases
EP
reporting acute renal failure with canagliflozin.
The association we identified could be the result of biases inherent to analyses of
AC C
spontaneous adverse event reporting databases, as well as to observational studies at large. While one may assume differences in the incidence of acute renal failure could drive disproportionate reporting of this adverse event, this signal could also be the result of other factors driving spontaneous adverse event reporting. These include several documented phenomena, such as the “Weber effect” (the tendency of reports to peak within the first years following approval of a drug)[16, 17] and “stimulated reporting” (the tendency of reports to peak following the publishing of regulatory warnings or media
ACCEPTED MANUSCRIPT
reports)[18]. Some studies have reported these phenomena are no longer prevalent in the modern-day FAERS database, as greater care is taken in the maintenance of the quality of the database, and stakeholders are legally bound to actively and
RI PT
programmatically collect and submit all reports of adverse events.[19, 20] Our analysis does suggest an increase in the rate of reporting of ARF with SGLT2-inhibitors following the FDA warning. However, the signal of a significantly increased reporting of ARF with
SC
SGLT-inhibitors is clearly present prior to the issue of this warning.
M AN U
The link we observed between SGLT2-inhibitors and ARF, could also be the result of confounding by indication, as these medications are prescribed for T2DM, itself strongly associated with an increased risk for the development of renal failure. To address this possibility, we examined the proportion of reports of acute renal failure with SGLT2-inhibitors compared with other cases with medications prescribed for T2DM. We
TE D
found SGLT2-inhibitors were associated with an increased proportion of reports of acute renal failure also when compared to other cases with medication prescribed for T2DM,
EP
however the effect was attenuated suggesting a significant component of the observed signal arose might be attributed to confounding by indication.
AC C
From a causal perspective, this link has considerable biologic plausibility, as
SGLT-2 inhibitors induce diuresis which can cause intravascular volume depletion. They also exert a reduction in trans-glomerular pressure, an inherent mechanism believed to promote renal protection in the long run. Additionally, these agents increase medullary oxygen consumption,[6] thereby increasing the risk for hypoxic damage. Indeed, patients on SGLT2 inhibitiors exhibited enhanced plasma erythropoietin and
ACCEPTED MANUSCRIPT
reticulocytosis ,[21] likely reflecting intensified hypoxia at the deep cortex and outer medulla[22].
RI PT
Our results are in agreement with several of the currently available clinical
studies, which have suggested an acute reduction in renal function following initiation of SGLT2-inhibitors, as well as an increased risk for renal adverse events.[7–9] Recently, large randomized studies have reported benefit in renal outcomes with SGLT2-
SC
inhibitors. While the initial report of renal safety from the EMPA-REG trial indicated a
M AN U
numerically greater reduction in estimated glomerular filtration rate (eGFR) in the empagliflozin groups,[23] the subsequent analysis of efficacy reported slower decline in renal function with empagliflozin and reduced risk for renal outcomes [3]. Likewise, the CANVAS trial results suggested significant benefit in renal outcomes (though these were not designated statistically significant).[4] While these results are very promising, it
TE D
is important to note that randomized clinical trials are conducted in selected populations, with specialized follow-up, and for limited durations of time, and often mask adverse effects which become apparent only several years post marketing.[24] Of note, both
EP
studies included few patients with significant renal dysfunction, as they excluded
AC C
patients using metformin with moderate renal failure due to regulatory considerations. In addition, differences in the definition of renal outcomes could also explain conflicting findings regarding the absolute and relative risk of ARF with SGLT2-inhibitors, as studies utilized various definitions and methods to evaluate the risk of ARF with SGLT2inhibitors. For instance, the CANVAS study protocol required the collection of “all (serious and non-serious) cases” of several adverse effects, such photosensitivity or
ACCEPTED MANUSCRIPT
male genital infections, however the protocol stipulated that cases of acute kidney injury should be collected only “if serious or if adverse event causes discontinuation”. [25]
RI PT
The characteristics of cases with SGLT2-inhibitors and renal failure, suggest that the link between these agents and renal failure may be stronger in cases with certain concomitant medications, such as ACE-inhibitors, ARBs, and especially diuretics.
Cases with SGLT2-inhibitors reporting renal failure were nearly four times more likely to
SC
report concomitant use of diuretics, compared to cases with SGLT2-ihibitors reporting
M AN U
other adverse events. This observation could be the result of confounding. However, this observation can also plausibly be viewed as similar to the widely recognized phenomenon, whereby the risk for acute renal failure with ACE-inhibitors is accentuated when prescribed in combination with other medications which reduce renal perfusion, including diuretics.[26–28]
TE D
Our results indicated SGLT2-inhibitors may differ in their respective association with acute renal failure. This link with acute renal failure was strongest with
EP
canagliflozin, and weakest with empagliflozin. While the reason for this observation is unclear, these results are largely in agreement with currently available indirect
AC C
comparisons between these agents. Meta-analyses of data from randomized controlled trials of SGLT2-inhibitors have reported that these agents are associated with a small increase in serum creatinine,[8] however this effect was driven primarily by canagliflozin. Similarly, meta-analyses have reported that randomized controlled trials indicate canagliflozin and dapagliflozin are significantly associated with a greater risk of composite renal events compared to control groups, while empagliflozin is associated with decreased risk.[7, 9]
ACCEPTED MANUSCRIPT
A biological basis for the possible difference between the SGLT2-inhibitors’ renal effects is unknown, however important differences have been documented between these agents. Empagliflozin differs from dapagliflozin and canagliflozin in terms of target
RI PT
specificity to SGLT2, with studies indicating 10-fold greater specificity to SGLT2 vs SGLT1 compared to canagliflozin.[29] In addition, the agents differ in their
pharmacokinetic properties[29], and non-SGLT2 mediated effects have been
SC
documented as well[30].
M AN U
This study has a number of strengths. First, we used the FAERS database which provides rapid and relatively large amounts of data on outcomes with new medication in the “real world”. As noted above, randomized clinical trials are conducted in selected populations, with specialized follow-up, and for limited durations of time, and often mask adverse effects seen in the “real world”.[24] Second, we examined the link between
TE D
SGLT2-inhibitors and ARF as a class, as well as by individual agent. Third, we evaluated the possible contribution of stimulated reporting and confounding by indication. Fourth, we characterized the cases with SGLT2-inhibitor associated ARF and
EP
identified a potential contribution of concomitant medication.
AC C
This study has a number of limitations as well. As the study is based on analysis of the FAERS database, there is limited information regarding a wide range of patient health characteristics (i.e. baseline renal function, lab values, etc.), thus limiting the ability to control for confounding effects. Importantly, while the database has the advantage of providing a “real world” perspective, the quality of diagnosis of adverse events likely varies widely, and is not based on a validated and consistent definition of ARF. In addition, the database does not include data on the size of the exposed
ACCEPTED MANUSCRIPT
population from which the reports arose. One cannot infer the absolute risk of ARF possibly attributable to SGLT2-inhibitors without making assumptions regarding the relation between incidence rate and report rates, and without additional data regarding
RI PT
the extent of their utilization. While FAERS provides some data as to the outcomes of reported cases, the data is limited. ACE-inhibitors are known to temporarily reduce eGFR, though very seldom lead to renal replacement therapy and have renoprotective
SC
effects in the long-term. More detailed data on outcomes with SGLT2-inhibitors would therefore help in scaling the dimension of the association with ARF indicated in this
M AN U
study.
In conclusion, this study found SGLT2-inhibitors were linked with a disproportionate rate of report of acute renal failure in the FDA adverse event reporting system. In light of the recent results indicating positive cardiovascular and renal
TE D
outcomes with SGLT-inhibitors, their continued promotion seems quite justified. However, reports regarding the acute effects of these agents on renal function, and their propensity to induce ARF, are still conflicting, and the results of our study should
EP
encourage further research to understand the possible link between SGLT2-inhibitors
AC C
and ARF. In addition care is warranted especially when prescribing these agents to patients with additional risk factors for acute renal failure, such as in the setting of exposure to contrast media, and with use of concomitant medications which reduce renal perfusion.
Acknowledgements No financial support was received for the conduct of the research and/or preparation of the article, and the authors have no conflict of interest to disclose.
ACCEPTED MANUSCRIPT
Bibliography Zaccardi F, Webb DR, Htike ZZ, et al. Efficacy and safety of sodium-glucose co-
RI PT
[1]
transporter-2 inhibitors in type 2 diabetes mellitus: systematic review and network meta-analysis. Diabetes, Obes Metab 2016; 18: 783–794.
Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, Cardiovascular Outcomes,
SC
[2]
and Mortality in Type 2 Diabetes. N Engl J Med 2015; 373: 2117–2128. Wanner C, Inzucchi SE, Lachin JM, et al. Empagliflozin and Progression of
M AN U
[3]
Kidney Disease in Type 2 Diabetes. N Engl J Med 2016; 375: 323–334. [4]
Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and Cardiovascular and
[5]
TE D
Renal Events in Type 2 Diabetes. N Engl J Med 2017; 377: 644–657. Heyman SN, Khamaisi M, Rosen S, et al. Potential Hypoxic Renal Injury in Patients With Diabetes on SGLT2 Inhibitors: Caution Regarding Concomitant Use
[6]
EP
of NSAIDs and Iodinated Contrast Media. Diabetes Care 2017; 40: e40–e41. O’Neill J, Fasching A, Pihl L, et al. Acute SGLT inhibition normalizes O 2 tension
AC C
in the renal cortex but causes hypoxia in the renal medulla in anaesthetized control and diabetic rats. Am J Physiol - Ren Physiol 2015; 309: F227–F234.
[7]
Vasilakou D, Karagiannis T, Athanasiadou E, et al. Sodium–Glucose Cotransporter 2 Inhibitors for Type 2 Diabetes. Ann Intern Med 2013; 159: 262.
[8]
Storgaard H, Gluud LL, Bennett C, et al. Benefits and Harms of Sodium-Glucose Co-Transporter 2 Inhibitors in Patients with Type 2 Diabetes: A Systematic
ACCEPTED MANUSCRIPT
Review and Meta-Analysis. PLoS One 2016; 11: e0166125. [9]
Tang H, Li D, Zhang J, et al. Sodium-glucose cotransporter 2 inhibitors and risk of
RI PT
adverse renal outcomes among type 2 diabetes patients: a network and cumulative meta-analysis of randomized controlled trials. Diabetes Obes Metab. Epub ahead of print 27 February 2017. DOI: 10.1111/dom.12917.
SC
[10]
http://bioportal.bioontology.org/ontologies/MEDDRA?p=classes&conceptid=2000
[11]
M AN U
0003.
Evans SJW, Waller PC, Davis S. Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports. Pharmacoepidemiol Drug Saf 2001; 10: 483–486.
Harpaz R, DuMouchel W, LePendu P, et al. Performance of pharmacovigilance
TE D
[12]
signal-detection algorithms for the FDA adverse event reporting system. Clin Pharmacol Ther 2013; 93: 539–46.
Raschi E, Poluzzi E, Koci A, et al. Liver injury with novel oral anticoagulants:
EP
[13]
AC C
assessing post-marketing reports in the US Food and Drug Administration adverse event reporting system. Br J Clin Pharmacol. Epub ahead of print 16 February 2015. DOI: 10.1111/bcp.12611.
[14]
Sakaeda T, Tamon A, Kadoyama K, et al. Data Mining of the Public Version of the FDA Adverse Event Reporting System. Int J Med Sci 2013; 10: 796–803.
[15]
Rothman KJ, Lanes S, Sacks ST. The reporting odds ratio and its advantages
ACCEPTED MANUSCRIPT
over the proportional reporting ratio. Pharmacoepidemiol Drug Saf 2004; 13: 519– 523. Weber J. Epidemiology of adverse reactions to nonsteroidal antiinflammatory
RI PT
[16]
drugs. Adv Inflamm Res 1984;61 [17]
Hartnell NR, Wilson JP. Replication of the Weber Effect Using Postmarketing
SC
Adverse Event Reports Voluntarily Submitted to the United States Food and Drug Administration. Pharmacotherapy 2004; 24: 743–749.
Poluzzi E, Raschi E, Piccinni C, et al. Data Mining Techniques in
M AN U
[18]
Pharmacovigilance: Analysis of the Publicly Accessible FDA Adverse Event Reporting System (AERS). DOI: 10.5772/50095. [19]
Hoffman KB, Demakas AR, Dimbil M, et al. Stimulated Reporting: The Impact of
TE D
US Food and Drug Administration-Issued Alerts on the Adverse Event Reporting System (FAERS). Drug Saf 2014; 37: 971–980. [20]
Hoffman KB, Dimbil M, Erdman CB, et al. The Weber effect and the United States
EP
Food and Drug Administration’s Adverse Event Reporting System (FAERS):
AC C
analysis of sixty-two drugs approved from 2006 to 2010. Drug Saf 2014; 37: 283– 94.
[21]
Lambers Heerspink HJ, de Zeeuw D, Wie L, et al. Dapagliflozin a glucose-
regulating drug with diuretic properties in subjects with type 2 diabetes. Diabetes Obes Metab 2013; 15: 853–62.
[22]
Heyman SN, Khamaisi M, Rosenberger C, et al. Increased Hematocrit During
ACCEPTED MANUSCRIPT
Sodium-Glucose Cotransporter-2 Inhibitor Therapy. J Clin Med Res 2017; 9: 176– 177. Supplement to: Zinman B, Wanner C, Lachin JM, et al. Empagliflozin,
RI PT
[23]
cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med
2015;373:2117-28. DOI: 10.1056/NEJMoa1504720. Epub ahead of print 2015.
[24]
SC
DOI: 10.1056/NEJMoa1504720.
Downing NS, Shah ND, Aminawung JA, et al. Postmarket Safety Events Among
M AN U
Novel Therapeutics Approved by the US Food and Drug Administration Between 2001 and 2010. JAMA 2017; 317: 1854. [25]
Protocol for: Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med 2017;376:644-
[26]
TE D
57. DOI: 10.1056/NEJMoa1611925. DOI: 10.1056/NEJMoa1611925. Mandal AK, Markert RJ, Saklayen MG, et al. Diuretics potentiate angiotensin
170–4.
Bakris GL, Sarafidis PA, Weir MR, et al. Renal outcomes with different fixed-dose
AC C
[27]
EP
converting enzyme inhibitor-induced acute renal failure. Clin Nephrol 1994; 42:
combination therapies in patients with hypertension at high risk for cardiovascular events (ACCOMPLISH): a prespecified secondary analysis of a randomised controlled trial. Lancet (London, England) 2010; 375: 1173–81.
[28]
Knight EL, Glynn RJ, McIntyre KM, et al. Predictors of decreased renal function in patients with heart failure during angiotensin-converting enzyme inhibitor therapy: results from the studies of left ventricular dysfunction (SOLVD). Am Heart J 1999;
ACCEPTED MANUSCRIPT
138: 849–55. [29]
Scheen AJ. Pharmacokinetic and pharmacodynamic profile of empagliflozin, a
Baartscheer A, Schumacher CA, Wüst RCI, et al. Empagliflozin decreases myocardial cytoplasmic Na(+) through inhibition of the cardiac Na(+)/H(+)
EP
TE D
M AN U
SC
exchanger in rats and rabbits. Diabetologia 2017; 60: 568–573.
AC C
[30]
RI PT
sodium glucose co-transporter 2 inhibitor. Clin Pharmacokinet 2014; 53: 213–25.
ACCEPTED MANUSCRIPT
Table 1 - Characteristics of Cases with Renal Failure associated with SGLT2 Inhibitors (n=1224) N (%) Primary suspect: 1033 (84.4%) Secondary suspect: 152 (12.4%) Concomitant: 37 (3%) Interacting: 2 (0.2%) Hospitalized: 521 (4.3%) Life threatening: 106 (8.7%) Disability: 29 (2.4%) Death: 16 (1.3%) Other serious: 551 (4.5%)
RI PT
Degree of suspicion of SGLT2 inhibtor, n (%)
SC
Outcome*
AC C
EP
TE D
M AN U
*more than one outcome is possible
ACCEPTED MANUSCRIPT
Table 3 - Characteristics of SGLT2-inhibitor Cases With and Without Acute Renal Failure
99 (26)
94.7 (25)
603 (49.3%) 246 (20%) 130 (19.6%) 35 (2.8%)
10643 (60.2%) 2688 (15.2%) 831 (4.7%) 537 (3%)
RI PT
Cases with Other Adverse Events (n=17691) 58.1 (12)
SC
Age in years, mean (SD) Weight in Kg, mean (SD) Female, n (%) ACEI or ARBs Diuretics NSAID
Cases with Renal Failure (n=1224) 59.6 (11)
M AN U
Abbreviations: ACE-I = angiotensin-converting-enzyme inhibitors; ARBs = angiotensin II receptor blockers; NSAIDs = non-steroidal anti-inflammatory drugs
Figure 1: Forest Plot of Reporting Odds Ratio of Acute Renal Failure in Relation to the FDA Warning.
EP
.
TE D
Boxes represent effect size (ROR), vertical lines represent 95% confidence interval. Reporting odds of acute renal failure increase with horizontal distance from y-axis effect size. ROR of all SGLT2-inhibitor subgroups are significantly greater than 1 (p<0.001), interaction between SGLT2-inhibitor and FDA warning significant as well (p<0.001).
AC C
Figure 2: Reporting Odds Ratio of Acute Renal Failure in Relation to Specific SGLT2-inhibitors Boxes represent effect size (ROR), vertical lines represent 95% confidence interval. Reporting odds of acute renal failure increase with horizontal distance from y-axis effect size. ROR of all SGLT2-inhibitor subgroups are significantly greater than 1 (p<0.001).
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights •
SGLT2-inhibitors are associated with acute renal failure in the FDA adverse event reporting system. Reports of acute renal failure with SGLT2-inhibitors were more likely to report concomitant diuretics and/or ACE-inhibitors.
RI PT
• •
Plausible biological mechanisms have been proposed for these findings.
•
The link between SGLT2-inhibitors and acute renal failure persisted when controlling for diabetes and the FDA warning.
SC
SGLT2-inhibitors differed from one another in the strength of their link to acute
EP
TE D
M AN U
renal failure.
AC C
•
S0939-4753(17)30238-7
DOI:
10.1016/j.numecd.2017.10.011
Reference:
NUMECD 1793
To appear in:
Nutrition, Metabolism and Cardiovascular Diseases
Received Date: 5 May 2017 Revised Date:
3 September 2017
Accepted Date: 6 October 2017
Please cite this article as: Perlman A, Heyman SN, Matok I, Stokar J, Muszkat M, Szalat A, Acute Renal Failure with Sodium-Glucose-Cotransporter-2 Inhibitors: Analysis of the FDA Adverse Event Report System Database, Nutrition, Metabolism and Cardiovascular Diseases (2017), doi: 10.1016/ j.numecd.2017.10.011. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Acute Renal Failure with Sodium-Glucose-Cotransporter-2 Inhibitors: Analysis of the FDA Adverse Event Report System Database
1
RI PT
Amichai Perlman, PharmD1,2, Samuel N Heyman, MD1, Ilan Matok, PhD2, Joshua Stokar, MD1, Mordechai Muszkat, MD1, Auryan Szalat, MD1
Corresponding Author: Amichai Perlman
SC
Department of Internal Medicine, Hadassah University Hospital, Jerusalem, Israel. Division of Clinical Pharmacy, Institute for Drug Research, School of Pharmacy, Faculty of Medicine, The Hebrew University of Jerusalem, Israel.
2
Address: POB 12065. Postal code 9112001. Jerusaelm, Israel.
M AN U
Phone: +972-2-675-7578, +972-50-517-2768 Fax: +972-2-675-7246
AC C
EP
TE D
Email address: [email protected]
ACCEPTED MANUSCRIPT
Abstract
Background and Aims: Sodium-glucose-cotransporter-2 (SGLT2) inhibitors have
RI PT
recently been approved for the treatment of type II diabetes mellitus (T2DM). It has been proposed these agents could induce acute renal failure (ARF) under certain
conditions. This study aimed to evaluate the association between SGLT2-inhibitors and
SC
ARF in the FDA adverse event report system (FAERS) database.
Methods and Results: We analyzed adverse event cases submitted to FAERS
M AN U
between January 2013 and September 2016. ARF cases were identified using a structured medical query. Medications were identified using both brand and generic names. During the period evaluated, 18,915 reports (out of a total of 3,832,015 registered in FAERS) involved the use of SGLT2-inhibitors. SGLT2-inhibitors were
TE D
reportedly associated with ARF in 1,224 of these cases (6.4%), and were defined as the “primary” or “secondary” cause of the adverse event in 96.8% of these cases. The proportion of reports with ARF among reports with SGLT2 inhibitor was almost three-
EP
fold higher compared to reports without these drugs (ROR 2.88, 95% CI 2.71-3.05, p<0.001). The proportion of ARF reports among cases with SGLT2-inhibitors was
AC C
significantly greater than the proportion of ARF among cases with T2DM without SGLT2-inhibitors (ROR 1.68, 95% CI 1.57-1.8, p<0.001). Among the SGLT2-inhibitors, canagliflozin was associated with a higher proportion of reports of renal failure (7.3%), compared to empagliflozin and dapagliflozin (4.7% and 4.8% respectively, p<0.001).
ACCEPTED MANUSCRIPT
Conclusion: SGLT2-inhibitors are associated with an increase in the proportion of reports of ARF compared to other medications. SGLT2-inhibitor agents may differ from
RI PT
one another in their respective risk for ARF.
AC C
EP
TE D
M AN U
SC
Keywords: Diabetes; Renal failure; SGLT-2 inhibitors; Adverse effects.
ACCEPTED MANUSCRIPT
Background Sodium-glucose-cotransporter-2 (SGLT2) inhibitors, have been approved in
RI PT
recent years for use in the treatment of type II diabetes mellitus (T2DM). Currently approved agents include canagliflozin, dapagliflozin, and empagliflozin. These agents have all proven effective treating T2DM in terms of HbA1c reduction, as well as
additionally beneficial in terms of blood pressure reduction, weight loss, and improved
SC
lipid profile.[1] The large randomized control trial EMPA-REG, demonstrated that
M AN U
empagliflozin reduces cardiovascular outcomes.[2] While empagliflozin was associated with an early reduction in eGFR following initiation,[3] this study reported empagliflozin slowed progression of kidney disease and reduced rates of clinically relevant renal events[3]. Similar positive results have recently been reported with cangliflozin in the CANVAS trial.[4]
TE D
These positive results indicate SGLT-2 inhibitors constitute a significant advancement in the pharmacological management of T2DM. However, concern has
EP
been raised regarding the possibility of an increased risk for acute renal failure (ARF) in some individuals using these agents.[5] SGLT-2 inhibitors induce diuresis which can
AC C
lead to intravascular volume depletion. These agents also reduce trans-glomerular pressure, leading to reduced glomerular filtration. Lastly, SLGT-2 inhibitors increase medullary oxygen consumption, increasing the risk for hypoxia.[6] These effects might be clinically important under circumstances of volume depletion, a-priori reduced transglomerular pressure (such as with concomitant use of diuretics or agents blocking the rennin-angiotensin axis) and where there is a predisposition to medullary hypoxic injury, such as with use of nonsteroidal anti-inflammatory drugs (NSAIDs) or radiocontrast
ACCEPTED MANUSCRIPT
agent. On June 14th, 2016, the U.S. Food and Drug Administration (FDA) strengthened an existing warning about the risk of acute kidney injury for canagliflozin and dapagliflozin, following evaluation of cases reported from March 2013 through October
RI PT
2015.
Meta-analyses of data from randomized controlled trials of SGLT2-inhibitors
reported findings suggesting possible acute decline in renal function. A meta-analysis
SC
published in 2013 included 13 randomized controlled studies and reported that In
M AN U
patients with moderate renal impairment, use of dapagliflozin or high doses of canagliflozin was associated with increased incidence of renal-related adverse events.[7] A subsequent meta-analysis of randomized controlled studies published in 2016, reported ”these agents are associated with a small increase in serum creatinine,[8] while a more recent meta-analysis including 58 studies reported that
TE D
canagliflozin and dapagliflozin were significantly associated with a greater risk of composite renal events compared to control groups, while empagliflozin was associated
EP
with decreased risk.[9]
In this study, we aimed to evaluate the association between SGLT2-inhibitors
AC C
and ARF in the FDA adverse event report system (FAERS) database, and characterize reports linking these agents to acute renal failure.
Methods
We analyzed adverse event cases submitted to FAERS between January 2013 and September 2016. FAERS is a database that contains information on adverse event reports submitted to FDA. The database is designed to support the FDA's post-
ACCEPTED MANUSCRIPT
marketing safety surveillance program for medication. Reports of adverse events are submitted by healthcare professionals (such as physicians, pharmacists, nurses and others), consumers (such as patients, family members, lawyers and others), and drug
RI PT
manufacturers and marketers. Drug manufacturers, distributors, and license holders, are legally bound to actively collect and report all adverse effects related to their
SC
products.
ARF cases were identified using Medical Dictionary for Regulatory Activities
M AN U
(MedDRA) terms related to acute renal failure. We used a predefined grouping of ARF related MedDRA terms based on the broad Standardized MedDRA Query (SMQ) that has been developed by the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH).[10] Cases involving SGLT2-inhibitors were identified using the terms for the generic and brand
TE D
names of the currently approved SGLT-2 inhibitors canagliflozin, dapagliflozin, and empagliflozin. Cases using other medications for the treatment of diabetes were
EP
identified using the reported indications, and a list of relevant preferred MedDRA terms. Use of concomitant ACE-inhibitors, ARBs, diuretics, and NSAIDs, was identified using
AC C
the terms for the generic and brand names of these medications. For a more detailed list of terms used for the analysis see appendix.
We evaluated the proportion of cases reporting renal failure with SGLT2 -
inhibitors in three ways. First, we compared the proportion of cases reporting ARF with SGLT2-inhibitors to the proportion of cases reporting ARF with all other medications. Second, we compared the proportion of cases reporting acute renal failure with SGLT2-
ACCEPTED MANUSCRIPT
inhibitors to the proportion of cases reporting acute renal failure among cases using other medications for the treatment of diabetes. Lastly, we compared the proportion of
RI PT
cases reporting acute renal failure among the specific SGLT-2 inhibitor agents.
We collected descriptive data on the characteristics of cases reporting acute renal failure with SGLT2-inhibitors, including age, weight, level of suspicion of SGLT2-
SC
inhibitor induced adverse effect (“primary suspect”, “secondary suspect”, “interacting”, or “concomitant"), concomitant medications (angiotensin-converting-enzyme inhibitors
M AN U
(ACE-I), angiotensin II receptor blockers (ARBs), diuretics, non-steroidal antiinflammatory drugs (NSAIDS), and outcome. For outcome we used the data available in FAERS. Adverse event case reports to the FDA are required to include classification of outcome according to U.S. reporting regulations. Serious adverse event cases are classified with one or more of the following possible outcome outcomes: “Death”, “Life-
TE D
Threatening”, “Hospitalization (Initial or Prolonged)”, “Disability”, “Congenital Anomaly”, “Required Intervention to Prevent Permanent Impairment/Damage”, “Other Serious
EP
Important Medical Event”. The outcome from the latest version of a case was used for this study. Characteristics of cases reporting renal adverse effects with SGLT2-
AC C
inhibitors were compared to characteristics of cases reporting all other adverse effects with SGLT2-inhibitors.
In order to evaluate the possibility that the FDA warning issued on June 2016
impacted the relative reporting rate of renal failure with SGLT2-inhibitors, we evaluated the association between SGLT2-inhibitors and renal failure using a multiple logistic regression model including terms for use of SGLT2-inhibitors as well as the date the
ACCEPTED MANUSCRIPT
adverse event was reported. Reporting dates were dichotomized as “before” or “after” the FDA warning. We evaluated whether the association between SGLT2-inhibitors and
SGLT2-inhibitors and report date status.
RI PT
acute renal failure was modified by this warning by including an interaction term for
Difference in reports of acute renal failure was reported using the reporting odds
SC
ratio (ROR) as a measure of disproportionality, and statistical significance was tested using chi-square and logistic regression.[11–15] Data processing and analysis were
M AN U
conducted using R version 3.3.2. Before the analysis, duplicate reports were removed according to the FDA’s recommendation for adopting the most recent case version.[14]
Results
During the period evaluated, 3,832,015 unique adverse event cases were
TE D
reported in FAERS. Of these, 18,915 reports involved the use of SGLT2-inhibitors. Individuals reported with SGLT2-inhibitor associated adverse events were on average
EP
58 years of age, and weighed 95 kg. SGLT2-inhibitors were reportedly associated with ARF in 1,224 cases (6.5%), and were judged to be the “primary” or “secondary” cause
AC C
of the adverse event in almost all cases (96.8%). Many of the cases reporting SGLT2inhibitor use and acute renal failure reportedly led to hospitalization, or to prolongation of hospitalization (n=521, 42.6%), and 16 cases reportedly ended with death (Table 1).
The number and proportion of cases reporting acute renal failure using three
analyses are detailed in Figure 1. The first analysis evaluated the proportion of cases reporting acute renal failure with SGLT2 -inhibitors compared to the proportion of cases
ACCEPTED MANUSCRIPT
reporting acute renal failure with all other medications. In this analysis, cases with SGLT2 -inhibitors were associated with a significantly greater proportion of reports of acute renal failure compared to cases with all other medications (ROR 2.88, 95% CI
RI PT
2.71-3.05, p<0.001). When restricting analysis to reports prior to the FDA warning, a similar signal was evident (ROR 2.85, 95% CI 2.67-3.03, p<0.001). However, there was a significant interaction between the FDA warning and the rate of report, such that the
M AN U
3.39-4.46, p<0.001; p for interaction <0.001).
SC
relative reporting rate of renal failure was greater following the report (ROR 3.9, 95% CI
SGLT2-inhibitors are used for the treatment of T2DM, itself a major risk factor for development of kidney disease, possibly introducing confounding by indication. We therefore further evaluated the proportion of cases reporting acute renal failure by restricting the comparison group to cases using other medications for treatment of
TE D
diabetes mellitus. In this analysis, we found SGLT-2 inhibitors were associated with a more moderate, but still significant, increase in the proportion of reports with acute renal
EP
failure (ROR 1.68, 95% CI 1.57-1.8, p<0.001). When examining the proportion of reports of acute renal failure among the individual SGLT-2 inhibitor agents, we found
AC C
canagliflozin was associated with a greater proportion of reports of acute renal failure (7.3%), compared to empagliflozin and dapagliflozin (4.7% and 4.8% respectively, p<0.001, Figure 2).
The characteristics of cases with SGLT2-inhibitors reporting acute renal failure,
differed from cases with SGLT2-inhibitors not reporting acute renal failure. Cases reporting ARF with SGLT2-inhibitors were more likely to be male, tended to be
ACCEPTED MANUSCRIPT
overweight, and much more likely to report use of concomitant diuretics, or ACEinhibitors or ARBs, compared to cases reporting other adverse events with SGLT2-
RI PT
inhibitors. Rates of reported use of NSAID agents were similar in both groups (Table 3).
Discussion
The results of this study show that in the FDA adverse events database, SGLT2-
SC
inhibitors are associated with increased odds of reporting of acute renal failure. The strength of this association (ROR 2.88, 95% CI 2.71-3.05, p<0.001) met widely
M AN U
accepted and validated criteria indicating a likely signal between these medications and acute renal failure, and the outcome of a large proportion of cases was deemed serious or led to new or prolonged hospitalization. The association between SGLT2-inhibitors and acute renal failure remained significant when comparing cases with SGLT2-
TE D
inhibitors to cases using other medications for the treatment of T2DM. Further analysis revealed that this signal was driven primarily by a disproportionate number of cases
EP
reporting acute renal failure with canagliflozin.
The association we identified could be the result of biases inherent to analyses of
AC C
spontaneous adverse event reporting databases, as well as to observational studies at large. While one may assume differences in the incidence of acute renal failure could drive disproportionate reporting of this adverse event, this signal could also be the result of other factors driving spontaneous adverse event reporting. These include several documented phenomena, such as the “Weber effect” (the tendency of reports to peak within the first years following approval of a drug)[16, 17] and “stimulated reporting” (the tendency of reports to peak following the publishing of regulatory warnings or media
ACCEPTED MANUSCRIPT
reports)[18]. Some studies have reported these phenomena are no longer prevalent in the modern-day FAERS database, as greater care is taken in the maintenance of the quality of the database, and stakeholders are legally bound to actively and
RI PT
programmatically collect and submit all reports of adverse events.[19, 20] Our analysis does suggest an increase in the rate of reporting of ARF with SGLT2-inhibitors following the FDA warning. However, the signal of a significantly increased reporting of ARF with
SC
SGLT-inhibitors is clearly present prior to the issue of this warning.
M AN U
The link we observed between SGLT2-inhibitors and ARF, could also be the result of confounding by indication, as these medications are prescribed for T2DM, itself strongly associated with an increased risk for the development of renal failure. To address this possibility, we examined the proportion of reports of acute renal failure with SGLT2-inhibitors compared with other cases with medications prescribed for T2DM. We
TE D
found SGLT2-inhibitors were associated with an increased proportion of reports of acute renal failure also when compared to other cases with medication prescribed for T2DM,
EP
however the effect was attenuated suggesting a significant component of the observed signal arose might be attributed to confounding by indication.
AC C
From a causal perspective, this link has considerable biologic plausibility, as
SGLT-2 inhibitors induce diuresis which can cause intravascular volume depletion. They also exert a reduction in trans-glomerular pressure, an inherent mechanism believed to promote renal protection in the long run. Additionally, these agents increase medullary oxygen consumption,[6] thereby increasing the risk for hypoxic damage. Indeed, patients on SGLT2 inhibitiors exhibited enhanced plasma erythropoietin and
ACCEPTED MANUSCRIPT
reticulocytosis ,[21] likely reflecting intensified hypoxia at the deep cortex and outer medulla[22].
RI PT
Our results are in agreement with several of the currently available clinical
studies, which have suggested an acute reduction in renal function following initiation of SGLT2-inhibitors, as well as an increased risk for renal adverse events.[7–9] Recently, large randomized studies have reported benefit in renal outcomes with SGLT2-
SC
inhibitors. While the initial report of renal safety from the EMPA-REG trial indicated a
M AN U
numerically greater reduction in estimated glomerular filtration rate (eGFR) in the empagliflozin groups,[23] the subsequent analysis of efficacy reported slower decline in renal function with empagliflozin and reduced risk for renal outcomes [3]. Likewise, the CANVAS trial results suggested significant benefit in renal outcomes (though these were not designated statistically significant).[4] While these results are very promising, it
TE D
is important to note that randomized clinical trials are conducted in selected populations, with specialized follow-up, and for limited durations of time, and often mask adverse effects which become apparent only several years post marketing.[24] Of note, both
EP
studies included few patients with significant renal dysfunction, as they excluded
AC C
patients using metformin with moderate renal failure due to regulatory considerations. In addition, differences in the definition of renal outcomes could also explain conflicting findings regarding the absolute and relative risk of ARF with SGLT2-inhibitors, as studies utilized various definitions and methods to evaluate the risk of ARF with SGLT2inhibitors. For instance, the CANVAS study protocol required the collection of “all (serious and non-serious) cases” of several adverse effects, such photosensitivity or
ACCEPTED MANUSCRIPT
male genital infections, however the protocol stipulated that cases of acute kidney injury should be collected only “if serious or if adverse event causes discontinuation”. [25]
RI PT
The characteristics of cases with SGLT2-inhibitors and renal failure, suggest that the link between these agents and renal failure may be stronger in cases with certain concomitant medications, such as ACE-inhibitors, ARBs, and especially diuretics.
Cases with SGLT2-inhibitors reporting renal failure were nearly four times more likely to
SC
report concomitant use of diuretics, compared to cases with SGLT2-ihibitors reporting
M AN U
other adverse events. This observation could be the result of confounding. However, this observation can also plausibly be viewed as similar to the widely recognized phenomenon, whereby the risk for acute renal failure with ACE-inhibitors is accentuated when prescribed in combination with other medications which reduce renal perfusion, including diuretics.[26–28]
TE D
Our results indicated SGLT2-inhibitors may differ in their respective association with acute renal failure. This link with acute renal failure was strongest with
EP
canagliflozin, and weakest with empagliflozin. While the reason for this observation is unclear, these results are largely in agreement with currently available indirect
AC C
comparisons between these agents. Meta-analyses of data from randomized controlled trials of SGLT2-inhibitors have reported that these agents are associated with a small increase in serum creatinine,[8] however this effect was driven primarily by canagliflozin. Similarly, meta-analyses have reported that randomized controlled trials indicate canagliflozin and dapagliflozin are significantly associated with a greater risk of composite renal events compared to control groups, while empagliflozin is associated with decreased risk.[7, 9]
ACCEPTED MANUSCRIPT
A biological basis for the possible difference between the SGLT2-inhibitors’ renal effects is unknown, however important differences have been documented between these agents. Empagliflozin differs from dapagliflozin and canagliflozin in terms of target
RI PT
specificity to SGLT2, with studies indicating 10-fold greater specificity to SGLT2 vs SGLT1 compared to canagliflozin.[29] In addition, the agents differ in their
pharmacokinetic properties[29], and non-SGLT2 mediated effects have been
SC
documented as well[30].
M AN U
This study has a number of strengths. First, we used the FAERS database which provides rapid and relatively large amounts of data on outcomes with new medication in the “real world”. As noted above, randomized clinical trials are conducted in selected populations, with specialized follow-up, and for limited durations of time, and often mask adverse effects seen in the “real world”.[24] Second, we examined the link between
TE D
SGLT2-inhibitors and ARF as a class, as well as by individual agent. Third, we evaluated the possible contribution of stimulated reporting and confounding by indication. Fourth, we characterized the cases with SGLT2-inhibitor associated ARF and
EP
identified a potential contribution of concomitant medication.
AC C
This study has a number of limitations as well. As the study is based on analysis of the FAERS database, there is limited information regarding a wide range of patient health characteristics (i.e. baseline renal function, lab values, etc.), thus limiting the ability to control for confounding effects. Importantly, while the database has the advantage of providing a “real world” perspective, the quality of diagnosis of adverse events likely varies widely, and is not based on a validated and consistent definition of ARF. In addition, the database does not include data on the size of the exposed
ACCEPTED MANUSCRIPT
population from which the reports arose. One cannot infer the absolute risk of ARF possibly attributable to SGLT2-inhibitors without making assumptions regarding the relation between incidence rate and report rates, and without additional data regarding
RI PT
the extent of their utilization. While FAERS provides some data as to the outcomes of reported cases, the data is limited. ACE-inhibitors are known to temporarily reduce eGFR, though very seldom lead to renal replacement therapy and have renoprotective
SC
effects in the long-term. More detailed data on outcomes with SGLT2-inhibitors would therefore help in scaling the dimension of the association with ARF indicated in this
M AN U
study.
In conclusion, this study found SGLT2-inhibitors were linked with a disproportionate rate of report of acute renal failure in the FDA adverse event reporting system. In light of the recent results indicating positive cardiovascular and renal
TE D
outcomes with SGLT-inhibitors, their continued promotion seems quite justified. However, reports regarding the acute effects of these agents on renal function, and their propensity to induce ARF, are still conflicting, and the results of our study should
EP
encourage further research to understand the possible link between SGLT2-inhibitors
AC C
and ARF. In addition care is warranted especially when prescribing these agents to patients with additional risk factors for acute renal failure, such as in the setting of exposure to contrast media, and with use of concomitant medications which reduce renal perfusion.
Acknowledgements No financial support was received for the conduct of the research and/or preparation of the article, and the authors have no conflict of interest to disclose.
ACCEPTED MANUSCRIPT
Bibliography Zaccardi F, Webb DR, Htike ZZ, et al. Efficacy and safety of sodium-glucose co-
RI PT
[1]
transporter-2 inhibitors in type 2 diabetes mellitus: systematic review and network meta-analysis. Diabetes, Obes Metab 2016; 18: 783–794.
Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, Cardiovascular Outcomes,
SC
[2]
and Mortality in Type 2 Diabetes. N Engl J Med 2015; 373: 2117–2128. Wanner C, Inzucchi SE, Lachin JM, et al. Empagliflozin and Progression of
M AN U
[3]
Kidney Disease in Type 2 Diabetes. N Engl J Med 2016; 375: 323–334. [4]
Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and Cardiovascular and
[5]
TE D
Renal Events in Type 2 Diabetes. N Engl J Med 2017; 377: 644–657. Heyman SN, Khamaisi M, Rosen S, et al. Potential Hypoxic Renal Injury in Patients With Diabetes on SGLT2 Inhibitors: Caution Regarding Concomitant Use
[6]
EP
of NSAIDs and Iodinated Contrast Media. Diabetes Care 2017; 40: e40–e41. O’Neill J, Fasching A, Pihl L, et al. Acute SGLT inhibition normalizes O 2 tension
AC C
in the renal cortex but causes hypoxia in the renal medulla in anaesthetized control and diabetic rats. Am J Physiol - Ren Physiol 2015; 309: F227–F234.
[7]
Vasilakou D, Karagiannis T, Athanasiadou E, et al. Sodium–Glucose Cotransporter 2 Inhibitors for Type 2 Diabetes. Ann Intern Med 2013; 159: 262.
[8]
Storgaard H, Gluud LL, Bennett C, et al. Benefits and Harms of Sodium-Glucose Co-Transporter 2 Inhibitors in Patients with Type 2 Diabetes: A Systematic
ACCEPTED MANUSCRIPT
Review and Meta-Analysis. PLoS One 2016; 11: e0166125. [9]
Tang H, Li D, Zhang J, et al. Sodium-glucose cotransporter 2 inhibitors and risk of
RI PT
adverse renal outcomes among type 2 diabetes patients: a network and cumulative meta-analysis of randomized controlled trials. Diabetes Obes Metab. Epub ahead of print 27 February 2017. DOI: 10.1111/dom.12917.
SC
[10]
http://bioportal.bioontology.org/ontologies/MEDDRA?p=classes&conceptid=2000
[11]
M AN U
0003.
Evans SJW, Waller PC, Davis S. Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports. Pharmacoepidemiol Drug Saf 2001; 10: 483–486.
Harpaz R, DuMouchel W, LePendu P, et al. Performance of pharmacovigilance
TE D
[12]
signal-detection algorithms for the FDA adverse event reporting system. Clin Pharmacol Ther 2013; 93: 539–46.
Raschi E, Poluzzi E, Koci A, et al. Liver injury with novel oral anticoagulants:
EP
[13]
AC C
assessing post-marketing reports in the US Food and Drug Administration adverse event reporting system. Br J Clin Pharmacol. Epub ahead of print 16 February 2015. DOI: 10.1111/bcp.12611.
[14]
Sakaeda T, Tamon A, Kadoyama K, et al. Data Mining of the Public Version of the FDA Adverse Event Reporting System. Int J Med Sci 2013; 10: 796–803.
[15]
Rothman KJ, Lanes S, Sacks ST. The reporting odds ratio and its advantages
ACCEPTED MANUSCRIPT
over the proportional reporting ratio. Pharmacoepidemiol Drug Saf 2004; 13: 519– 523. Weber J. Epidemiology of adverse reactions to nonsteroidal antiinflammatory
RI PT
[16]
drugs. Adv Inflamm Res 1984;61 [17]
Hartnell NR, Wilson JP. Replication of the Weber Effect Using Postmarketing
SC
Adverse Event Reports Voluntarily Submitted to the United States Food and Drug Administration. Pharmacotherapy 2004; 24: 743–749.
Poluzzi E, Raschi E, Piccinni C, et al. Data Mining Techniques in
M AN U
[18]
Pharmacovigilance: Analysis of the Publicly Accessible FDA Adverse Event Reporting System (AERS). DOI: 10.5772/50095. [19]
Hoffman KB, Demakas AR, Dimbil M, et al. Stimulated Reporting: The Impact of
TE D
US Food and Drug Administration-Issued Alerts on the Adverse Event Reporting System (FAERS). Drug Saf 2014; 37: 971–980. [20]
Hoffman KB, Dimbil M, Erdman CB, et al. The Weber effect and the United States
EP
Food and Drug Administration’s Adverse Event Reporting System (FAERS):
AC C
analysis of sixty-two drugs approved from 2006 to 2010. Drug Saf 2014; 37: 283– 94.
[21]
Lambers Heerspink HJ, de Zeeuw D, Wie L, et al. Dapagliflozin a glucose-
regulating drug with diuretic properties in subjects with type 2 diabetes. Diabetes Obes Metab 2013; 15: 853–62.
[22]
Heyman SN, Khamaisi M, Rosenberger C, et al. Increased Hematocrit During
ACCEPTED MANUSCRIPT
Sodium-Glucose Cotransporter-2 Inhibitor Therapy. J Clin Med Res 2017; 9: 176– 177. Supplement to: Zinman B, Wanner C, Lachin JM, et al. Empagliflozin,
RI PT
[23]
cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med
2015;373:2117-28. DOI: 10.1056/NEJMoa1504720. Epub ahead of print 2015.
[24]
SC
DOI: 10.1056/NEJMoa1504720.
Downing NS, Shah ND, Aminawung JA, et al. Postmarket Safety Events Among
M AN U
Novel Therapeutics Approved by the US Food and Drug Administration Between 2001 and 2010. JAMA 2017; 317: 1854. [25]
Protocol for: Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med 2017;376:644-
[26]
TE D
57. DOI: 10.1056/NEJMoa1611925. DOI: 10.1056/NEJMoa1611925. Mandal AK, Markert RJ, Saklayen MG, et al. Diuretics potentiate angiotensin
170–4.
Bakris GL, Sarafidis PA, Weir MR, et al. Renal outcomes with different fixed-dose
AC C
[27]
EP
converting enzyme inhibitor-induced acute renal failure. Clin Nephrol 1994; 42:
combination therapies in patients with hypertension at high risk for cardiovascular events (ACCOMPLISH): a prespecified secondary analysis of a randomised controlled trial. Lancet (London, England) 2010; 375: 1173–81.
[28]
Knight EL, Glynn RJ, McIntyre KM, et al. Predictors of decreased renal function in patients with heart failure during angiotensin-converting enzyme inhibitor therapy: results from the studies of left ventricular dysfunction (SOLVD). Am Heart J 1999;
ACCEPTED MANUSCRIPT
138: 849–55. [29]
Scheen AJ. Pharmacokinetic and pharmacodynamic profile of empagliflozin, a
Baartscheer A, Schumacher CA, Wüst RCI, et al. Empagliflozin decreases myocardial cytoplasmic Na(+) through inhibition of the cardiac Na(+)/H(+)
EP
TE D
M AN U
SC
exchanger in rats and rabbits. Diabetologia 2017; 60: 568–573.
AC C
[30]
RI PT
sodium glucose co-transporter 2 inhibitor. Clin Pharmacokinet 2014; 53: 213–25.
ACCEPTED MANUSCRIPT
Table 1 - Characteristics of Cases with Renal Failure associated with SGLT2 Inhibitors (n=1224) N (%) Primary suspect: 1033 (84.4%) Secondary suspect: 152 (12.4%) Concomitant: 37 (3%) Interacting: 2 (0.2%) Hospitalized: 521 (4.3%) Life threatening: 106 (8.7%) Disability: 29 (2.4%) Death: 16 (1.3%) Other serious: 551 (4.5%)
RI PT
Degree of suspicion of SGLT2 inhibtor, n (%)
SC
Outcome*
AC C
EP
TE D
M AN U
*more than one outcome is possible
ACCEPTED MANUSCRIPT
Table 3 - Characteristics of SGLT2-inhibitor Cases With and Without Acute Renal Failure
99 (26)
94.7 (25)
603 (49.3%) 246 (20%) 130 (19.6%) 35 (2.8%)
10643 (60.2%) 2688 (15.2%) 831 (4.7%) 537 (3%)
RI PT
Cases with Other Adverse Events (n=17691) 58.1 (12)
SC
Age in years, mean (SD) Weight in Kg, mean (SD) Female, n (%) ACEI or ARBs Diuretics NSAID
Cases with Renal Failure (n=1224) 59.6 (11)
M AN U
Abbreviations: ACE-I = angiotensin-converting-enzyme inhibitors; ARBs = angiotensin II receptor blockers; NSAIDs = non-steroidal anti-inflammatory drugs
Figure 1: Forest Plot of Reporting Odds Ratio of Acute Renal Failure in Relation to the FDA Warning.
EP
.
TE D
Boxes represent effect size (ROR), vertical lines represent 95% confidence interval. Reporting odds of acute renal failure increase with horizontal distance from y-axis effect size. ROR of all SGLT2-inhibitor subgroups are significantly greater than 1 (p<0.001), interaction between SGLT2-inhibitor and FDA warning significant as well (p<0.001).
AC C
Figure 2: Reporting Odds Ratio of Acute Renal Failure in Relation to Specific SGLT2-inhibitors Boxes represent effect size (ROR), vertical lines represent 95% confidence interval. Reporting odds of acute renal failure increase with horizontal distance from y-axis effect size. ROR of all SGLT2-inhibitor subgroups are significantly greater than 1 (p<0.001).
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights •
SGLT2-inhibitors are associated with acute renal failure in the FDA adverse event reporting system. Reports of acute renal failure with SGLT2-inhibitors were more likely to report concomitant diuretics and/or ACE-inhibitors.
RI PT
• •
Plausible biological mechanisms have been proposed for these findings.
•
The link between SGLT2-inhibitors and acute renal failure persisted when controlling for diabetes and the FDA warning.
SC
SGLT2-inhibitors differed from one another in the strength of their link to acute
EP
TE D
M AN U
renal failure.
AC C
•