medwireNews: Metformin consistently causes an early reduction in hemoglobin and increases the risk for moderate anemia in patients with type 2 diabetes, a MASTERMIND consortium analysis of two randomized controlled trials (RCTs) and a real-world study shows.
The researchers report that the absolute reductions in hemoglobin were not large in these trials, at an average 0.5 g/dL at 5 years in the ADOPT study and 0.5 g/dL at 3 years in the UKPDS, but they say that this was sufficient to give a large increase in moderate anemia rates.
And this was supported by data from the observational, real-world GoDARTS study, showing the annual risk for anemia increased by 2% with every 1 g/day of metformin.
“Because the mechanisms for metformin-related moderate anemia are unknown, the effects are modest, and the benefits of metformin are proven, we would not in any way advocate avoidance or discontinuation of metformin, even in patients with anemia, but a reduction in [hemoglobin] in the first few years after initiation of metformin might be anticipated,” say Ewan Pearson (University of Dundee, UK) and fellow researchers.
In ADOPT, among the 3967 participants recently diagnosed with type 2 diabetes, 1343 were taking metformin, 1289 sulfonylureas, and 1335 thiazolidinediones. Over 5 years of follow-up, the rates of anemia were 2.8%, 1.5%, and 5.7%, respectively, meaning the odds were almost doubled for those taking metformin (odds ratio [OR]=1.93), compared with sulfonylureas, and quadrupled for those taking thiazolidinediones (OR=4.18).
Similarly, in UKPDS, among 1473 people with type 2 diabetes, there were 300 taking metformin, 461 taking sulfonylureas, 360 using insulin, and 352 being treated through diet alone, and anemia rates were a corresponding 6.3%, 1.1%, 2.5%, and 1.7% over 9 years of follow-up. This translated to adjusted ORs for anemia of 4.42 for those taking metformin, compared with diet only, 0.53 for those taking sulfonylureas, and 1.79 for those using insulin.
The researchers note in Diabetes Care that the reduction in hemoglobin from baseline was immediate in those taking metformin, occurring by the first 6-month measurement in ADOPT and the first 3-year measurement in UKPDS. There was no further decrease after 3 years in ADOPT, and in UKPDS all treatment groups had reduced hemoglobin by years 6 and 9 with no greater difference among those in the metformin treatment arm compared with others.
These findings were echoed in the GoDARTS population of 3485 individuals with type 2 diabetes. Of these, 2487 had accumulated some exposure to metformin by the end of the study. The anemia event rate was 41.8% over a median 8.3 years of follow-up, with these events occurring in 51.1% of patients currently taking metformin, 13.3% of those who had formerly taken metformin, and 35.6% of those who had never taken it.
With increasing cumulative exposure to metformin, there was a higher risk for moderate anemia, which was linear following an initial high rate in the first year.
“It is more likely that this initial greater risk of anemia with metformin can be attributed to an immediate effect of metformin on [hemoglobin], particularly in light of the ADOPT data where we see a significant change at 6 months,” says the team.
Pearson et al also believe it is unlikely that the mechanism underlying the early effects of metformin on hemoglobin is secondary to B12 deficiency “because individuals should have enough B12 stored to last for between 2 and 5 years,” they explain.
Added to this, the individuals in the metformin-exposed group in the GoDARTS study who developed anemia were more likely to have microcytic anemia and less likely to have macrocytic anemia, compared with those not exposed.
medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2020 Springer Healthcare Ltd, part of the Springer Nature Group
Diabetes Care 2020; doi:10.2337/dc20-1104