October 1996

Refeeding Hypophosphatemia in Critically Ill Patients in an Intensive Care UnitA Prospective Study

Author Affiliations

From the Departments of Critical Care Medicine (Dr Marik) and Nutrition and Food Services (Ms Bedigian), St Vincent Hospital, and University of Massachusetts, Worcester.

Arch Surg. 1996;131(10):1043-1047. doi:10.1001/archsurg.1996.01430220037007

Background:  Hypophosphatemia has been reported after refeeding of malnourished patients. Nutritional support is often delayed in patients in the intensive care unit (ICU) as a consequence of enteral intolerance and bowel hypomotility.

Objective:  To determine the incidence, risk factors, and clinical impact of refeeding hypophosphatemia in a heterogeneous group of patients in an ICU.

Design:  Prospective, noninterventional study.

Settings:  Surgical and medical ICUs of a university-affiliated community hospital.

Patients:  Sixty-two patients in the ICU who were refed after being starved for at least 48 hours were prospectively followed up.

Interventions:  None.

Main Outcome Measures:  Each patient had a nutritional assessment prior to the initiation of nutritional support. Serum phosphate, magnesium, and calcium levels were measured at baseline, and these measurements were repeated daily. Refeeding hypophosphatemia was considered to have developed in patients whose serum phosphorus level fell by more than 0.16 mmol/L to below 0.65 mmol/L.

Results:  Twenty-one patients (34%) experienced refeeding hypophosphatemia. In 6 patients, the serum phosphorus level fell below 0.32 mmol/L. The only risk factor studied that could predict the development of hypophosphatemia was the serum prealbumin concentration (mean ± SD, 127±34vs79±40 g/L;P<.001). Seventeen (81%) of these 21 patients in whom hypophosphatemia developed had a prealbumin concentration less than 110 g/L compared with that in 12 (30%) of the patients who did not experience this complication (P<.001). In those patients in whom refeeding hypophosphatemia developed, the serum phosphorus level reached a mean ± SD nadir of 1.9 ± 1.1 days after feeding was started. Although the Acute Physiology and Chronic Health Evaluation II score was similar (mean± SD, 19±6 vs 18±7), the length of mechanical ventilation (mean±SD, 10.5±5.2vs7.1±2.8 days; P=.04) and the length of hospital stay (mean±SD, 12.1±7.1 vs 8.2±4.6 days; P=.01) were significantly longer in those patients who experienced hypophosphatemia compared with those patients who did not experience this complication.

Conclusions:  Refeeding hypophosphatemia occurs commonly in critically ill patients in the ICU. Starvation for a period as short as 48 hours and poor nutritional status predispose to this syndrome. Patients at risk should be refed slowly, and the serum phosphorus level should be closely monitored and supplemented as required.

Arch Surg.  1996;131:1043-1047