BACKGROUND: Differences in regulating factors and the clinical implications of body temperature variability (BTV) between subjects with and without diabetes have not been clarified to date. METHODS AND RESULTS: In 66 subjects with ischemic heart disease (33 with diabetes and 33 without diabetes), BTV, the difference between the highest and lowest temperature measurements, and body temperature standard deviation (BT SD) were measured from axillary body temperature (ABT) records of 3 consecutive days and followed for 16.4±8.4 months. In subjects without diabetes BTV and BT SD were closely associated with endothelial function as evaluated on flow-mediated dilation (BTV, R=0.33, P=0.026; BT SD, R=0.41, P=0.029), whereas there was a poor association in subjects with diabetes. In the absence of an interrelationship between vascular function and thermoregulation, the contribution of inflammation to BTV was increased in subjects with diabetes (BTV, 0.59±0.21°C for C-reactive protein [CRP] <0.08 mg/dl vs. 0.79±0.28°C for CRP >0.08 mg/dl, P=0.014). Event-free survival analysis showed that in subjects with diabetes higher BT SD was associated with shorter event-free survival (log-rank P=0.012), but this relationship was not found in subjects without diabetes. CONCLUSIONS: In subjects with diabetes, the interrelationship between thermoregulation and vascular function was disrupted and the effect of inflammation on thermoregulation was enhanced, so that BTV had a sufficient predictive value for cardiovascular events in diabetic subjects.
BACKGROUND: Differences in regulating factors and the clinical implications of body temperature variability (BTV) between subjects with and without diabetes have not been clarified to date. METHODS AND RESULTS: In 66 subjects with ischemic heart disease (33 with diabetes and 33 without diabetes), BTV, the difference between the highest and lowest temperature measurements, and body temperature standard deviation (BT SD) were measured from axillary body temperature (ABT) records of 3 consecutive days and followed for 16.4±8.4 months. In subjects without diabetes BTV and BT SD were closely associated with endothelial function as evaluated on flow-mediated dilation (BTV, R=0.33, P=0.026; BT SD, R=0.41, P=0.029), whereas there was a poor association in subjects with diabetes. In the absence of an interrelationship between vascular function and thermoregulation, the contribution of inflammation to BTV was increased in subjects with diabetes (BTV, 0.59±0.21°C for C-reactive protein [CRP] <0.08 mg/dl vs. 0.79±0.28°C for CRP >0.08 mg/dl, P=0.014). Event-free survival analysis showed that in subjects with diabetes higher BT SD was associated with shorter event-free survival (log-rank P=0.012), but this relationship was not found in subjects without diabetes. CONCLUSIONS: In subjects with diabetes, the interrelationship between thermoregulation and vascular function was disrupted and the effect of inflammation on thermoregulation was enhanced, so that BTV had a sufficient predictive value for cardiovascular events in diabetic subjects.