Heart failure is more prevalent in diabetes than in the general population. All too often, patients with such serious co-morbidities are excluded from clinical trials. As a result, there remains uncertainty around the benefits of diabetes medications in these sub-groups of patients. Often, we need to look at observational data to get the best available evidence to inform our treatment choices.
A recent systematic review in Circulation and Heart Failure addressed this question:
- In patients with diabetes and heart failure, does metformin as compared with alternative treatments affect the risk of overall mortality?
The reviewers searched for relevant studies in the most important databases. Two independent reviewers selected the studies for inclusion and extracted their data.
Data were pooled using a random effects model, to reflect the likelihood that there would be important clinical differences between the studies.
Nine observational studies were found that addressed the review question. Data from 34,504 patients were included, of whom 6,624 were using metformin. As expected, most of the studies looked at metformin in various combinations with other treatments.
Overall mortality for metformin users was 23%, compared with 37% in the comparison groups. After adjusting for other risk factors, the relative risk was 0.80 (95% CI 0.74–0.87) in favour of metformin compared with the comparison groups.
This benefit was seen consistently across several planned sub-group and sensitivity analyses.
In patients with reduced left ventricular ejection fraction (LVEF) and chronic kidney disease, no significant difference was found, possibly due to lack of data.
- This was a good quality systematic review that conformed to PRISMA guidance.
- It is likely that the reviewers found all of the relevant published evidence in the area. Although they do not report their specific search strategy, it is likely that well-indexed terms such as metformin and heart failure would be picked up. The one exception, as noted by the reviewers, is where use of metformin would be analysed as a sub-group within an analysis of a heart failure cohort.
- It is not clear what proportion of the patients had type 1 or type 2 diabetes. Most of the studies identified patients from the prescription of antidiabetic medications. From perusing this information, it seems that most had type 2 diabetes.
- The review does not explicitly describe how the studies were evaluated for methodological quality. However, a summary of the appraisals of each paper is available with the online supplement.
- Although the review adjusted the analysis for other clinical factors, it still seems likely that the comparator groups could have contained more ill patients.
Eurich DT, Weir DJ, Majumdar SR, Tsuyuki RT, Johnson JA, Tjosvold L, Vanderloo SE, McAlister FA. Comparative Safety and Effectiveness of Metformin in Patients With Diabetes Mellitus and Heart Failure : Systematic Review of Observational Studies Involving 34,000 Patients. Circ Heart Fail. 2013;6:395-402. [PubMed abtract]