AKI Risk With SGLT2 Inibitors? Maybe Not, Clinic Data Show

Miriam E Tucker

August 28, 2017

Sodium glucose cotransporter-2 (SGLT2) inhibitors don't appear to raise the risk for acute kidney injury (AKI), data from two large health systems suggest. The authors conclude that the drug class may, in fact, have a protective effect, lowering the risk of AKI.

The analysis, from the Mount Sinai chronic kidney disease registry and the Geisinger Health System, was published online August 21, 2017 in Diabetes Care by Steven Coca, DO, with the Icahn School of Medicine at Mt Sinai, New York, and colleagues.

Although the SLGT2 inhibitor class provide significant benefits in type 2 diabetes, in 2016 the U.S. Food and Drug Administration strengthened a prior warning about a possible increased risk for AKI with the SGLT2 inhibitors canagliflozin (Invokana, Invokamet, Janssen) and dapagliflozin (Farxiga, Xigduo XR, AstraZeneca), based on 101 cases reported between March 2013 and October 2015. The warning did not include empagliflozin (Jardiance, Boehringer Ingelheim).

"There is the possibility that concerns for AKI might discourage use of these medications. The reports of AKI did not have a comparator group; thus, the risk attributable to SGLT2 inhibitors was unclear, because it is well-known that patients with type 2 diabetes are at already at high baseline risk of developing AKI," Dr Coca and senior author Girish N Nadkami, MD, said in an email to Medscape Medical News. Dr Nadkami is also from the Icahn School of Medicine.

In a propensity-matched analysis of nearly 1600 SGLT2-inhibitor users and the same number of nonusers with type 2 diabetes, the researchers actually found a somewhat lower risk for AKI among those using SGLT2 inhibitors over a 14-month follow up.

Moreover, the frequency and severity of AKI was lower for SGLT2-inhibitor users than nonusers, with the findings holding true in subgroup analyses.

"Thus, the take-home message is that there does not seem to be an increased risk of AKI in patients taking SGLT2 inhibitors compared with similar patients with type 2 diabetes," Dr Coca and Dr Nadkami said.

At the same time, of course, they urge common sense. "Overall, the SGLT2 inhibitors seem to be nephroprotective. But, due to their mechanism of action, there is the possibility for patients to get volume depleted, particularly when [the drugs are] used concurrently with diuretics or when taken during periods of illness with gastrointestinal fluid losses or lack of oral intake. However, the same risks are present for thiazide-type and loop diuretics. They can and do cause volume depletion and can lead to prerenal forms of AKI. Thus, caution is necessary when there are multiple stressors on volume status."

Propensity-Matching Analysis Shows No Increased AKI Risk

The study populations, all with type 2 diabetes, included 377 SGLT2-inhibitor users and 377 nonusers from Mount Sinai and 1207 of each from Geisinger, both groups propensity matched for potential confounders including age, gender, HbA1c, and comorbidities. In both databases, canagliflozin was the most commonly used SGLT2 inhibitor (72% at Mt Sinai, 61% at Geisinger). Use of empagliflozin was much more common at Geisinger, 29% vs 9% at Mt Sinai.

Acute kidney events were defined as an increase in serum creatinine by ≥0.3 mg/dL within 48 hours or increase in serum creatinine to ≥1.5 baseline known or presumed to have happened in the prior 7 days.

After propensity matching, the proportions with AKI among SGLT2 users and nonusers were 3.8% and 9.7%, respectively, at Mt Sinai (P = .002), with incidence rates three vs eight per 100 patient-years, respectively. At Geisinger, those proportions were 2.2% among SGLT2-inhibitor users and 4.6% among nonusers (< .01), with incidence rates 1.7 vs 3.8 per 100 patient-years, respectively.

Sensitivity analyses also yielded lower AKI rates among users than nonusers. No differences were seen in peak serum creatinine (1.7 vs 1.6 mg/dL, = .9) or change in creatinine (0.6 vs. 0.6 mg/dL, = .8) for users vs nonusers. Overall, the unadjusted hazard ratios for AKI were 60% lower for SGLT2-inhibitor users vs nonusers, and this didn't change after multiple confounder adjustments.

Of the 40 SGLT2-inhibitor users combined from both centers who had AKI, the medication was discontinued and not restarted in 30.

In both unadjusted and adjusted analyses in data from the two centers, there were no differences in AKI risk between dapagliflozin and canagliflozin. Only Geisinger had sufficient data on empagliflozin, and the results between users and nonusers were similar there as well (adjusted hazard ratio, 0.96).

Based on these and prior data, "there seems to be a class effect for the improvement in cardiorenal end points; thus, the [FDA's} distinction between the agents does not seem appropriate," Dr Coca and Dr Nadkami told Medscape Medical News.

What Is the Best Clinical Approach?

In its 2016 warning, the FDA advised the following for patients taking SGLT2 inhibitors:

Healthcare professionals should consider factors that may predispose patients to acute kidney injury prior to starting them on canagliflozin or dapagliflozin. These include decreased blood volume; chronic kidney insufficiency; congestive heart failure; and taking other medications such as diuretics, blood-pressure medicines called angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs), and nonsteroidal anti-inflammatory drugs (NSAIDs). Assess kidney function prior to starting canagliflozin or dapagliflozin and monitor periodically thereafter. If acute kidney injury occurs, promptly discontinue the drug and treat the kidney impairment.

Asked to comment on that, Dr Coca and Dr Nadkarni noted that more than 70% of patients in the current cohorts — and about 80% in both the EMPA-REG and CANVAS trials — were on ACE inhibitors/ARBs with no increased risk for AKI in the current study and evidence of long-term renoprotection in both of the trials.

"Thus, while the theoretic issue of decreased efferent arteriolar tone from the ACE inhibitor/ARB in conjunction with the afferent vasoconstriction in SGLT2 inhibitors due to induction of tubuloglomerular feedback is present, it does not seem to manifest in more AKI or harm to the kidneys."

The authors agree, of course, that kidney function should be closely monitored and that if AKI does occur, the etiology should be determined. However, they also addressed the issue of restarting the medications, which the FDA warning did not:  

"While a temporary holding of meds during acute illness or hospitalization is prudent, rational and thoughtful decision-making should ensue about whether to restart the medication. If the AKI episode is attributable to acute illness, then there is no reason the SGLT2 inhibitor cannot be restarted to afford the long-term cardiovascular and renal benefits. Even if the AKI is deemed to be due to in part to SGLT2 inhibitor–induced volume depletion, adjustment of other diuretics, or instruction to withhold during illness, or change in dose might be the prudent step, rather than complete abandonment."

In response to a Medscape Medical News query, an FDA spokesperson confirmed that the 2016 warning about AKI with dapagliflozin and canagliflozin still stands but added, "FDA is aware of this paper and will review the information contained within; however, we cannot comment on this at this time."

The study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health. Dr Coca serves as a consultant on SGLT2 inhibitors for Janssen Pharmaceuticals. The coauthors have no additional disclosures.

Diabetes Care. Published online August 21, 2017. Abstract  

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.