SEVERE SEPSIS MORTALITY IN A COMMUNITY TEACHING HOSPITAL - 1983-1985 AND 1989-1990

Despite better understanding of the pathophysiology and hemodynamic ab normalities of severe sepsis, mortality remains high. We examined the severity-adjusted outcome and use of pulmonary artery catheterization (PAC) in patients with severe sepsis during 1983 to 1985 and 1989 to 1 990 in an intensive care unit (ICU) staffed by full-time intensivists. A retrospective chart review was conducted on severe sepsis patients admitted to a multidisciplinary ICU of an 850-bed community teaching h ospital. Patients were identified at ICU admission and at 24 and 48 ho urs after ICU admission. Patients were stratified by severity of illne ss using the Mortality Probability Model (MPM) and presence or absence of a pulmonary artery catheter within the first 48 hours of ICU admis sion. In the years 1983 through 1985 and 1989 through 1990, respective ly, 245 and 132 patients with severe sepsis were identified. The sever ity-adjusted outcome and use of PAC was analyzed both within and betwe en these two periods. The severity adjusted mortality from sepsis was unchanged over time (63% in 1983-1985 and 51% in 1989-1990; P = 0.491) . Pulmonary artery catheter use increased from 52.2% in 1983 to 1985 t o 64.4% in 1989 to 1990 (p = 0.023). Invasive monitoring with PAC was used in a less severely ill population; during 1983 to 1985, mean seve rity estimated by MPM was 0.43 compared with an MPM of 0.32 in 1989 to 1990 (P = 0.003). A significant increase in the therapeutic use of va soactive drugs (65 vs 81%) to treat severe sepsis was observed over th e two periods (P = 0.0005). Outcome of severe sepsis was unchanged dur ing the two periods. Applying an ICU severity-of-illness adjustment mo del, early employment of pulmonary artery catheters and increased use of vasopressors did not effect outcome in an intensivist-managed ICU. Insertion of pulmonary artery catheters in patients with less severe i llness also did not impact outcome. Use of pulmonary artery catheters to monitor and guide treatment needs further rigorous prospective eval uation to define the specific conditions in severely septic patients f or which this intervention is most applicable.