BACKGROUND: Hypophosphatemia may cause organ derangements in the surgical intensive care unit. The purpose of this study was to determine the impact of a repletion protocol for hypophosphatemia based on admission weight and phosphorus level. STUDY DESIGN: All patients who presented to an 18-bed surgical intensive care unit with a serum phosphorus level of 2.2 mg/dL or less or who received phosphorus supplementation despite having normal levels were identified. In the preintervention phase between January and June 2001, 137 patients were retrospectively identified who met these criteria. A protocol was then designed giving a single intravenous dose of phosphorus based on weight and serum phosphorus. Repletion was given with sodium or potassium phosphorus based on presupplementation levels. After protocol implementation 141 patients met these criteria between September 2001 and February 2002, and treatment and postrepletion levels were followed prospectively. RESULTS: A total of 47 patients were repleted before the intervention with adequate followup and 22 (47%) attained a normal level. Supplementation success was 53% in moderate hypophosphatemia (2.2 mg/dL or less) and 27% in severe hypophosphatemia (less than 1.5 mg/dL). After protocol implementation, 111 patients were repleted with 84 (76%) correcting to a normal level (p = 0.002 compared with retrospective patients). Success was 78% in moderate hypophosphatemia and 62% in severe hypophosphatemia. Inappropriate supplementation of normal phosphorus levels decreased from 51 to 16 patients after protocol implementation. CONCLUSIONS: A protocol based on weight and serum levels successfully treated both moderate and severe hypophosphatemia in the majority of critically ill patients. Protocol implementation also decreased unnecessary supplementation of normal phosphorus levels.