Metformin is linked to anaemia in type 2 diabetes patients

Metformin is linked to anaemia in type 2 diabetes patients

Metformin may increase the risk for moderate anaemia among adults with type 2 diabetes, according to the findings of a study published in the journal Diabetes Care.

Each gram per day of metformin use was found associated with a 2% higher risk of moderate anaemia per year, noted the researchers based on findings from a large, observational, population-based study with a maximum follow-up period of almost 20 years.

Researchers analysed data from the randomized controlled trials — A Diabetes Outcome Progression Trial (ADOPT) and the U.K. Prospective Diabetes Study (UKPDS) — to detect the effects of metformin at specific follow-up points. In ADOPT, adults recently diagnosed with type 2 diabetes (n = 3,967, 58% men) were randomly assigned to thiazolidinedione, metformin or sulfonylurea monotherapy. Blood samples were collected at baseline, 6 months and 1 year and then annually up to 5 years. In UKPDS, adults with newly diagnosed type 2 diabetes (n = 1,473, 47.1% men) were randomly assigned to a conventional diet with no medication or to insulin, sulfonylurea or metformin. Blood samples were collected at baseline, 3, 6 and 9 years.

Researchers also analyzed routine clinical data from the Genetics of Diabetes Audit and Research in Tayside Scotland (GoDARTS) study. The data included information on prescriptions and routine laboratory tests. Adults with type 2 diabetes (n = 3,485) diagnosed in 1996 or later were included if they had a baseline haemoglobin measurement with no anaemia. Researchers used these data to evaluate the effects of metformin exposure over time.

In ADOPT, the use of metformin (adjusted OR = 1.93) and TZDs (aOR = 4.18) increased the odds of developing moderate anaemia when compared with sulfonylureas. Older participants, those with lower baseline haemoglobin and men were at greater risk for moderate anaemia. The ADOPT group saw a similar drop in haemoglobin at 6 months after treatment started with both metformin and TZDs. No further decreases were reported after 3 years.

In the UKPDS group, metformin also increased the odds for moderate anaemia when compared with diet (aOR = 4.42). The ORs were lower for insulin (aOR = 1.79) and sulfonylureas (aOR = 0.53). Lower baseline haemoglobin was a predictor for moderate anaemia.

In UKPDS, the metformin group had a greater reduction in haemoglobin in the first measurement 3 years after treatment started compared with all other groups. Haemoglobin decreased from baseline for all groups at years 6 and 9, but the metformin group had the largest drop when compared with the diet group.

In GoDARTS, 2,487 of 3,485 individuals had been exposed to metformin by the end of follow-up. Increased cumulative metformin exposure of 1 g per day for 1 year slightly increased the odds of moderate anaemia (OR = 1.02).

The team thus observed that the findings were consistent across two randomised controlled trials and was also replicated in the GoDARTS, real-world study of routinely collected data.

Researchers noted that because the mechanisms for metformin-related anaemia are still unknown and the effects modest, they would not advocate discontinuation of metformin, even for users with anaemia, due to its benefits. However, patients using metformin to treat type 2 diabetes should anticipate a reduction in haemoglobin, stressed the researchers.