Homeostasis of body weight in a diabetes clinic population.

Abstract

Body weights, blood glucose concentration, and therapeutic and clinical data were summarized by computer for each patient visit to our diabetes clinic between 1965 and 1970. Homeostasis of body weight was examined by plotting regression of body weight on time for patients making five or more visits to the clinic during the study period. A similar analysis was made of blood glucose values. Data from 131 patients treated with oral agents or by dietary regulation alone and forty-one patients treated with insulin were analyzed. Precision of regulation was assessed by calculating the coefficient of variation of body weight while the trend of body weight was assessed by calculating the slope of the linear regression equation relating body weight and time. In both treatment groups the individual slopes for body weight clustered closely around zero and the mean slopes did not differ significantly from zero. The coefficient of variation of body weight was also small, averaging 4.6 and 3.7 per cent of body weight in the oral agent and insulin treatment groups respectively. In contrast, blood glucose fluctuated widely in the two groups. Mean slopes for glucose versus time did not differ significantly from zero, however. The remarkable stability of body weight in this diabetic population despite widely varying modes of therapy, glucose concentrations, and degrees of body adiposity suggests that signals controlling long-term feeding behavior are either not related to these variables or, more likely, are multiple and sufficiently redundant to permit adaptation to perturbations of several metabolic variables.