Effect of cardiac rehabilitation with exercise training on fasting plasma glucose and HbA1c in patients with chronic heart failure

Abstract

Purpose: It has long been recognized that chronic heart failure (CHF) and glucose abnormalities are linked. However, the effect of physical exercise training on glucose metabolism in CHF has not been extensively investigated. The aim of the presented analysis is to describe a large cohort of CHF patients that engaged in rehabilitation in terms of exercise tolerance, fasting plasma glucose and HbA1c. Methods: Data concerning cardiopulmonary exercise testing, blood analysis and training were extracted from medical files. Maximal exercise testing with gas analysis was performed on a bicycle ergometer, venous blood samples were taken in a fasting state at the start and after 6 weeks of rehabilitation. Result: In total, 124 CHF patients were included in rehabilitation, of which 38 (31%) dropped out of the programme before the second exercise test after 6 weeks of training. The dropouts did not differ significantly in terms of age, left ventricular ejection fraction and percentage of predicted peak oxygen uptake. The prevalence of diagnosed diabetes was 10% in the dropout group (vs 23%). Nevertheless, baseline glucose (113 ± 39 vs 103 ± 18) and HbA1c (6.07 ± 1.07 vs 5.91 ± 0.66) tended to be higher in the dropout group, although the difference was not statistically significant. Eighty-six patients finished a second exercise test after 6 weeks, of which 20 patients had diagnosed diabetes (table 1). The diabetic group did not differ significantly in terms of age (66.4 ± 9.7 years vs 62 ± 12.3 years, p=.101), left ventricular ejection fraction (29 ± 13% vs 32 ± 13%, p=.422) and body mass index (28.1 ± 5.2 kg/m2 vs 27.1 ± 5.5 kg/m2, p=.48). Although both groups had a significant increase in peak oxygen uptake (+8.3 ± 8.1 % predicted in diabetics, p<.001 vs +9.45 ± 14.3 % predicted in non-diabetics, p<.0001), this was significantly lower at both tests in the diabetic group (p<.05). Fasting plasma glucose in the diabetic group was unaltered after training (-0.1 ± 41.5 mg/dL, p=.935), but tended to increase in the non-diabetic group (+4.1 ± 18.7, p=.222). HbA1c tended to decrease after training in both groups (-0.28 ± 0.47, p=.091 in the diabetic group vs -0.08 ± 0.29, p=.239 in the nondiabetic group). Conclusions: The prevalence of diabetes is considerable in a CHF population engaged in rehabilitation. Patients without diagnosed diabetes also showed high fasting plasma glucose, which did not decrease after exercise training. Further study will show what is the role of cardiac rehabilitation in improving glycemic control and what would be the best method to detect changes in glycemic control after intervention.