Many years on metformin linked to anemia risk

People with type 2 diabetes who take metformin for many years are more likely to develop anemia than are those who do not, according to the results of a large analysis of data from an observational, population-based study with 20 years of follow-up.

“Metformin treatment was associated with a 6 percent higher risk of anemia for every cumulative year of metformin exposure, as reported in a poster presentation at the annual meeting of the European Association for the Study of Diabetes.

In an interview,  a postdoctoral research assistant at the University of Dundee (Scotland), explained why they looked at the use of metformin and anemia risk in people with type 2 diabetes.

“The Diabetes Prevention Program (DPP) study showed that long-term metformin use in individuals with impaired glucose tolerance was associated with an increased risk of anemia, and this was independent of vitamin B 12 status.  Anemia is a common finding in people with type 2 diabetes, but the impact long-term metformin use on anemia hasn’t been studied.”

 Medical  associates obtained detailed information on metformin prescribing and hematology measures from electronic patient medical records from the Genetics of Diabetes Audit and Research in Tayside and Scotland (GoDARTS) cohort, based in Scotland. This database contains information on individuals with type 2 diabetes and matching controls and is available to researchers worldwide.

 For the analysis, the team looked for people diagnosed from 1996 onward who had a baseline haemoglobin measurement. Of 6,440 individuals with type 2 diabetes in the GoDARTS cohort, just over half had a haemoglobin measurement.

We used a definition of “moderate” anemia and we excluded patients with mild anemia or worse at diabetes diagnosis.  Anemia was considered to be a hemoglobin level of less than 12 g/dL in women and less and less than 13 g/dL in men. In all, 280 individuals with anemia were excluded from further analysis, as the aim was to follow people until they developed anemia, died, or left the area, or until the end of the follow-up period, which was set at Sept. 30, 2015. A discrete-time failure analysis was used to model the effect of cumulative metformin exposure on anemia risk.

 After a median follow-up of eight years and a median number of 11 hemoglobin measurements per patient, 2,487 study subjects (71 percent) had some exposure to metformin and 1,458 of the whole sample (41.8 percent) had become anemic. Of those who developed anemia, 745 (51 percent) were current metformin users, 194 (13 percent) were former users, and 519 (36 percent) had never taken metformin.

 “Cumulative metformin use was independently associated with an increased risk of anemia,”  it was noted (odds ratio, 1.06; 95 percent confidence interval, 1.02-1.09; P=.0006). This association was not seen when they examined the data based on sulfonylurea use (OR 1.0; 95 percent CI 0.97-1.04, P=.8.

“Anemia risk was higher with age at diagnosis, duration of diabetes, lower haemoglobin at baseline, and lower Egfr (estimated glomerular filtration rate),” she observed. Ors for first anemia event were 1.03 (95 percent CI, 0.66-0.74) per 1 g/dL of hemoglobin at diagnosis, and Egfr 0.98 (95 percent CI, 0.98-1.01) per additional 1 mL/min per 1.73  (P less than .0001 for all).

Why cumulative metformin use is associated with an increased of anemia is unclear, however, and it was  noted that this needs further investigation. “We do have data from two other clinical trials now, showing similar results, and maybe through those data we might be able to untangle it.” 

“In terms of mechanism, we can only conjecture, according to a senior author of the study and professor at the University of Dundee. It is important to stress that metformin is a great drug, and we shouldn’t stop it because of a potentially increased risk of anemia.”