Inverse relation of body weight and weight change with mortality and morbidity in patients with type 2 diabetes and cardiovascular co-morbidity: An analysis of the PROactive study population

Context: Although weight reduction is a recommended goal in type 2 diabetes mellitus (T2DM), weight loss is linked to impaired survival in patients with some chronic cardiovascular diseases. Objective: To assess the association of weight and weight change with mortality and non-fatal cardiovascular outcomes (hospitalisation, myocardial infarction and stroke) in T2DM patients with cardiovascular co-morbidity and the effect of pioglitazone-induced weight change on mortality. Setting and participants: We assessed in a post hoc analysis body weight and weight change in relation to outcome in 5202 patients from the PROactive trial population who had T2DM and evidence of pre-existing cardiovascular disease. Patients were randomized to treatment with pioglitazone or placebo in addition to their concomitant glucose-lowering and cardiovascular medication. Mean follow up was 34.5 months. Main outcome measure: The impact of body weight and bodyweight change on all-cause mortality, cardiovascular mortality, on non-fatal cardiovascular events and on hospitalisation. Results: The lowest mortality was seen in patients with BMI 30-35 kg/m(2) at baseline. In comparison to this (reference group), patients in the placebo group with BMI <22 kg/m(2) (Hazard Ratio (95% confidence intervals) 2.96 [1.27 to 6.86]; P = 0.012) and BMI 22 to 25 kg/m(2) (HR 1.88 [1.11 to 3.21]; P= 0.019) had a higher all-cause mortality. Weight loss was associated with increased total mortality (HR per 1% body weight: 1.13 [1.11 to 1.16]; P<0.0001), with increased cardiovascular mortality, all-cause hospitalisation and the composite of death, myocardial infarction and stroke. Weight loss of >= 7.5% body weight (seen in 18.3% of patients) was the strongest cut-point to predict impaired survival (multivariable adjusted HR 4.42 [3.30 to 5.94]. Weight gain was not associated with increased mortality. Weight gain in patients treated with pioglitazone (mean + 4.0 +/- 6.1 kg) predicted a better prognosis (HR per 1% weight gain: 0.96 [0.92 to 1.00] P = 0.037) compared to patients without weight gain. Conclusion: Among patients with T2DM and cardiovascular co-morbidity, overweight and obese patients had a lower mortality compared to patients with normal weight. Weight loss but not weight gain was associated with increased mortality and morbidity. There may be an "obesity paradox" in patients with type 2 diabetes and cardiovascular risk. The original PROactive trial is registered as an International Standard Randomized Controlled Trial (Number ISRCTN NCT00174993). (C) 2011 Elsevier Ireland Ltd. All rights reserved.