OBJECTIVE: To evaluate serum soluble L-selectin as a prognostic factor for survival in patients with sepsis.

DESIGN: A prospective study of mortality in patients with sepsis whose serum levels of sL-selectin were measured on admission to an intensive care unit (ICU) and 4 days later. Follow-up data on mortality were obtained from the Danish Central Office of Civil Registration.

SETTING: A tertiary referral university hospital ICU in Copenhagen.

PATIENTS: Sixty-three patients meeting the criteria for systemic inflammatory response syndrome (SIRS) with a suspected or verified infection in one or more major organs, and 14 control subjects.

MEASUREMENTS AND RESULTS: On admission to the ICU the Simplified Acute Physiology Score (SAPS) II was calculated, and relevant microbial cultures were performed. Mortality was registered at various follow-up points: 7 days after admission, at discharge from hospital, and 3 and 12 months after admission. Serum sL-selectin levels were significantly lower in the patients than in the controls. Sepsis nonsurvivors had significantly lower levels than survivors. Efficiency analysis and receiver operation characteristics showed that the ideal cutoff point for sL-selectin as a test for sepsis survival was 470 ng/ml. The accumulated mortality in patients with subnormal sL-selectin levels on admission was significantly increased. No correlation was found between clinical or paraclinical markers, including SAPS II and sL-selectin, and no relationship to the microbial diagnosis was found.

CONCLUSIONS: Serum sL-selectin is a predictor of survival in patients with sepsis. Those admitted with low sL-selectin (<470 ng/ml) are characterized by a high mortality within the subsequent 12-month period.