1. This study aimed to investigate the possible role of leptin in post-traumatic anorexia by making pre- and post-operative (0-8 days) measurements of circulating leptin concentrations in six patients undergoing elective total hip replacement for osteoarthritis. 2. Mean daily hunger rating (four categories) and food intake (assessed by food record charts) were measured pre-operatively, as well as post-operatively for the first 5 days (days 0-5). Leptin concentrations, circulating metabolites [glucose, non-esterified fatty acids, glutamine and 3-hydroxybutyrate] and insulin and cortisol concentrations were measured pre-operatively (day 0) and post-operatively (days 1, 2, 3, 5 and 8). 3. Mean leptin concentrations were significantly increased only on day 1 (56% increase compared with pre-operative values, P < 0.009), whereas food intake (only 0.6 MJ on day 0) and hunger (5/6 patients 'not hungry' on day 0) only gradually improved over the next few days. (The energy intake over the first 5 days was 56% of the pre-operative value.) 4. Circulating insulin and cortisol concentrations were elevated on day 1 compared with pre-operative values on day 0 (P < 0.05). Of the measured metabolites implicated in the control of food intake, circulating non-esterified fatty acids and 3-hydroxybutyrate were not significantly altered in the post-operative period, but significant hyperglycaemia was noted on day 1 compared with day 0 pre-operatively (8.8 compared with 6.4 mmol/l glucose; P < 0.01). 5. It is concluded that circulating leptin is involved in the early (< 24 h) acute-phase response after moderately severe surgical trauma (characterized biochemically by a substantial acute-phase protein response, hypoalbuminaemia, hyperglycaemia and hypoglutaminaemia). Therefore, leptin may be implicated in post-traumatic anorexia, although other factors are likely to be involved, especially after the first 24 h when circulating leptin concentrations are no longer elevated.