COVID-19 COVID-19 Evidence Alerts from McMaster PLUS is a new service that alerts users to current best evidence for clinical care of people with threatened, suspected or confirmed COVID-19 infection.
Visit the site Sign up for alerts
Okunrintemi V, Mishriky BM, Powell JR, et al. Sodium-glucose co-transporter-2 inhibitors and atrial fibrillation in the cardiovascular and renal outcome trials. Diabetes Obes Metab. 2020 Oct 1. doi: 10.1111/dom.14211. (Original study)

Dapagliflozin is a sodium-glucose co-transporter-2 (SGLT2) inhibitor that has recently been shown to reduce the incidence of reported episodes of atrial fibrillation (AF)/atrial flutter in the DECLARE-TIMI 58 trial. This raises the question regarding whether SGLT2 inhibitors can reduce the incidence of AF in a high-risk population. We searched for trials comparing SGLT2 inhibitors to placebo in high-risk individuals with or without diabetes (ie, cardiovascular and renal outcome trials) and that reported the incidence of AF as a serious adverse event. The EMPA-REG OUTCOME trial, CANVAS, CANVAS-R, the DECLARE-TIMI 58 trial, CREDENCE, DAPA-HF, VERTIS-CV and DAPA-CKD were included. The incidence of AF, reported as a serious adverse event, was 0.9% in individuals who received an SGLT2 inhibitor compared to 1.1% in those who received placebo. Pooled results showed a significantly lower incidence of AF in individuals with and without diabetes (relative risk 0.79, 95% confidence interval 0.67,0.93). This review suggests that there is a significantly lower risk of incident AF for individuals on SGLT2 inhibitors versus placebo. While there was a statistically significant lower incidence of AF, reported as a serious adverse event, more research is needed to evaluate its clinical significance.

Discipline Area Score
Family Medicine (FM)/General Practice (GP) 6 / 7
General Internal Medicine-Primary Care(US) 6 / 7
Cardiology 5 / 7
Endocrine 5 / 7
Comments from MORE raters

Cardiology rater

As noted by the authors, "There were limited data in published reports on baseline AF in SGLT2 inhibitor versus placebo groups. This limits the ability of this report to exclude the lower incidence of AF in the SGLT2 inhibitor group being the result of a lower incidence of baseline AF in the SGLT2 inhibitor group". Presumably baseline AF was similar in both cohorts given randomization but, as noted by the authors, these results require further exploration.

Cardiology rater

This is possibly important and encouraging information for cardiologists but needs further evaluation in large scale studies.

Endocrine rater

Two issues underlie my opinion that this paper is not very helpful to clinicians caring for patients. First, as a potential side effect, only an increase in AF would be important in our decision to use SGLT-2s or not. The fact that it may decrease the incidence of AF would not affect our decision. This ties in with the 2nd issue which involves absolute vs relative risk. The fact that a 0.9% vs a 1.1% is statistically significant is only due to the large number of patients and is clinically insignificant. Finally, it would have been nice if the authors had presented the effect in diabetes vs non-diabetes patients. Perhaps if there had been a large difference, one could argue that AF is so much more common in diabetic patients that an SGLT-2 inhibitor should be favored over other drugs for its beneficial effect on preventing AF. That, of course, would require further research but that analysis might have given some impetus for pursuing it.
Comments from EvidenceAlerts subscribers

Dr. william jones (10/27/2020 8:29 AM)

Buried in the full article credits {Wiley} is the funding received from at least one of the authors from the pharmaceutical industry.