Total free living energy expenditure was compared in lean and obese women by the new doubly labelled water method and partitioned into basal metabolism and thermogenesis plus activity by whole body calorimetry. Average energy expenditure was significantly higher in the obese group (10-22 versus 7-99 MJ/day (2445 versus 1911 kcal/day); p<0-001) resulting from an increase in the energy cost of both basal metabolism and physical activity. Self recorded energy intakes were accurate in the lean subjects but underestimated expenditure by 3-5 MJ/day (837 kcal/day) in the obese group. Basal metabolic rate and energy expenditure on thermogenesis plus activity were identical in the two groups when corrected for differences in fat free mass and total body mass.In the obese women in this series there was no evidence that their obesity was caused by a metabolic or behavioural defect resulting in reduced energy expenditure.

Twelve women were studied before pregnancy and at 6-wk intervals from 6 to 36 wk gestation. Total energy expenditure (TEE) by the doubly labeled water method, basal metabolic rate (BMR), energy intake, and body composition were assessed on each occasion. There was substantial interindividual variation in the response to pregnancy. Mean total energy costs were as follows: delta BMR 112 +/- 104 MJ (range -53 to 273), delta TEE 243 +/- 279 MJ (range -61 to 869 MJ), and fat deposition 132 +/- 127 MJ (range -99 to 280 MJ). The mean total cost of pregnancy (cumulative TEE above baseline+energy deposited as fat and as products of conception) was 418 +/- 348 MJ (range 34-1192 MJ). This was much higher than current recommendations for incremental energy intakes. Self-recorded incremental intakes (208 +/- 272 MJ) seriously underestimated the additional costs. The variability in response emphasizes the problems in making prescriptive recommendations for individual women, because there is no way of predicting metabolic or behavioral responses to pregnancy.

The mechanisms by which well-nourished women meet the additional energy costs of lactation were studied by measuring energy intake (EI), basal metabolic rate (BMR), total energy expenditure by doubly labeled water (TEE), physical activity plus thermogenesis (TEE-BMR), changes in body fat stores, and milk energy transfer. Ten women were studied at 36 wk gestation; 4, 8, and 12 wk lactation (L4, L8, L12); and when nonpregnant and nonlactating (NPNL) after weaning. At L4, L8, and L12 the energy transferred in milk averaged 2245, 2225, and 2217 kJ/d with an additional 445 kj/d (106 kcal/d) estimated as being necessary for synthesis. EI was 1360, 1740, and 1275 kJ/d higher than the NPNL values, representing 56% of the costs of lactation. The remaining 44% was met by a reduction in TEE (-945, -688, and -826 kJ/d vs NPNL) caused largely by a reduction in physical activity because BMR was essentially unchanged (+29, -12, and -218 kJ/d). The energy-balancing strategies adopted by different women varied markedly.

The hypothesis that the energy cost of human pregnancy can be minimized by energy-sparing metabolic adaptations was tested using serial 24 h whole-body calorimetry. Eight healthy, well-nourished women were studied prepregnant and at 6, 12, 18, 24, 30 and 36 weeks gestation. Basal metabolic rate (BMR) showed highly characteristic changes within each subject and large inter-individual differences (F 3.5, P < 0.01). Some subjects showed a highly significant depression of metabolism up to 24 weeks gestation in support of the initial hypothesis. At 36 weeks BMR ranged from +8.6 to +35.4% relative to the prepregnant baseline. This wide variability was not explained by differences in the amount of lean tissue gained. Women displaying the energy-sparing suppression of BMR tended to be thin, suggesting that changes in metabolism may be responsive to initial energy status (ABMR Y. prepregnant body fat: r 0.84, P < 0.005). Changes in 24 h energy expenditure closely paralleled changes in BMR (v 0.98, P < 0-OOl), since the energy cost of minor voluntary activity and thermogenesis remained very constant within each individual. Pregnancy decreased the net cost of weight-dependent and weight-independent standard exercises when expressed per kg body-weight: stepping -10 (SD 2)%, P < 0.001 at 18-36 weeks, cycling -26 (SD 7) %, P < 0.01 at 12-36 weeks. The average integrated maintenance costs of pregnancy matched previous group estimates from well-nourished women, but individual estimates ranged from -16 to +276 MJ (coefficient of variation 93%). This high level of variability has important implications for the prescription of incremental energy intakes during pregnancy. It may also have had evolutionary significance.Energy-sparing adaptations : Pregnancy : Whole-body calorimetry.

1. Four men were each studied continuously over 12 d in a whole-body calorimeter. Dietary intake and daily activities were kept constant throughout the study. 2. Day-to-day coefficients of variation in energy expenditure within subjects were found to be 1.97% over 24 h, 5.93% during basal metabolic rate measurement, 2.40% overnight and 3.22% in exercise. 3. The contribution of measurement system noise to the observed variability was analysed and shown to be generally small. The source of this noise was considered. 4. The results reinforce and extend other comparable reports and show that within-subject variability forms a small part of reported observations of between-subject variability.