Excess intravenous water and sodium may be associated with postoperative complications and an adverse outcome. However, the effect of the magnitude of the surgery on such a relation has not been studied. This study assesses current practice in intravenous fluid and sodium administration after colonic and rectal resection and its relation to the postoperative outcome. A series of 100 consecutive patients undergoing elective colonic (n = 44) or rectal resection (n = 56) were included in a retrospective case-cohort study. The volumes of water and sodium from intravenous fluid and antibiotic administration on the day of surgery and the next 5 days were recorded together with the clinical outcome. The mean +/- SEM fluid and sodium administration on the day of operation was greater after rectal than colonic resection (4.6 +/- 0.2 vs. 3.6 +/- 0.2 liters and 507 +/- 34 vs. 389 +/- 22 mmol, respectively (p <0.05). The mean +/- SEM rate of daily fluid and sodium administration for the 5 subsequent days was greater following rectal than colonic resection (2.1 +/- 0.1 vs. 1.8 +/- 0.1 L/day and 155 +/- 8.7 vs. 128 +/- 8.0 mmol/day; p <0.05). For all resections, there were no differences in fluid and sodium administration on the day of surgery in patients with or without postoperative complications. During the subsequent 5 days, patients with complications after colonic resection had a higher postoperative mean rate of intravenous sodium administration than those who did not (149 +/- 12 vs. 115 +/- 10 mmol; p <0.05). A similar pattern was not observed following rectal resection. Current postoperative intravenous fluid prescription delivers approximately 2 liters of fluid and 140 mmol of sodium per day. Complications after colonic, but not rectal, resection are associated with more early postoperative daily intravenous sodium administration. Because colonic resection poses less of a physiologic insult than rectal resection, the overall outcome in the former group may be more sensitive to the interplay between fluid and sodium overload and patient co-morbidity.