The aim of the present study was to measure the changes in resting energy expenditure (REE) induced by malaria and to assess to what extent they are related to fever and nutritional status. The REE of 19 Gambian children (mean age +/- SEM, 9 +/- 1 y; weight, 24 +/- 2 kg; expected weight for height 86 +/- 1%) were measured with a hood system at repeated intervals at the onset of malaria crisis (test A), 3 to 4 d after therapy (test B), and 14 to 21 d later (test C). Axillary temperature averaged 39.2 +/- 0.1, 36.6 +/- 0.1, and 36.7 +/- 0.1 degrees C in the tests A, B, and C, respectively. REE in test A was significantly higher than REE in test B (223 +/- 10 versus 174 +/- 8 kJ/kg.d, p less than 0.0001), but in test C (169 +/- 8 kJ/kg.d), it did not differ from that observed in test B. The percentage of increase in REE was significantly correlated with the difference in axillary temperature (r = 0.46, p less than 0.05); the slope of the regression line indicated an increase of 6.9% in REE/degree C of fever. Furthermore, the individual increase in REE/degree C was correlated to the percentage of weight for height of the children (r = 0.54, p less than 0.05), indicating that the child's nutritional status influences the magnitude of the hypermetabolism due to fever. We concluded that Gambian children suffering from an acute episode of malaria have an increase in REE averaging 30%; however, REE promptly returns to baseline value a few days after the beginning of therapy.