Assessment of Metformin Failure Among Patients With Type 2 Diabetes Mellitus at a Tertiary Care Center in Central India

OBJECTIVES: Metformin is generally recommended as first-line therapy for T2DM due to high efficacy, low cost and additional benefits, but primary or secondary metformin failure is common in clinical practice. This study was carried out to characterize the failure of metformin monotherapy, and identify factors that predict likelihood of failure to optimise anti-hyperglycaemic therapy.

METHODS: A case-control design was used and adult T2DM patients who experienced primary or secondary metformin failure were recruited as cases and those who were controlled on metformin monotherapy, (HbA1c ≤ 7 %, or FBS ≤ 140 mg/dl) with adequate adherence were recruited as controls over a 15 month study period. Data was collected from the prescriptions / OPD diaries, analyzed for descriptive measures and odds ratio for factors predicting metformin failure.

RESULTS: A total of 124 participants were enrolled - 63 cases and 61 controls. Gender distribution showed predominance of males in both groups (59% vs 54%) and mean age was also comparable but odds of metformin failure were relatively higher for age < 40 or 50 years. Odds of metformin failure were also relatively higher (OR: 1.62) for a BMI ≥ 23 kg/m². HbA1c % at metformin initiation was significantly higher in the metformin failure group (8.87 ± 1.63) versus controls (7.89 ± 1.45); p = 0.02, OR of 4.33 for HbA1c ≥ 7% and similar difference was seen for FBS (p = 0.03). The mean blood pressures and lipid parameters were not significantly different in the two groups. In the logistic regression model, duration of diabetes, HbA1c at metformin initiation and metformin dose at initiation came out to be significant (p≤0.05).

CONCLUSIONS: Our results show that HbA1c / FBS at initiation, age, BMI, dose at initiation could be predictors to identify patients likely to have metformin failure. Further studies with higher sample size shall validate the results.