AMSTERDAM—The progression from infection and sepsis to shock is poorly understood, but a new study suggests that several clinical variables can help predict shock in febrile patients with infection.[1] Peak respiratory rate, nadir Glasgow Coma Scale score, and the presence of bacteremia were among the most useful predictors.

Ailko W. J. Bossink, MD, and colleagues from the Free University Hospital in Amsterdam conducted a study to evaluate the role of clinical and microbiologic information in predicting the development of shock in patients with fever and infection. Among the variables they assessed were the four criteria for the systemic inflammatory response syndrome (SIRS): abnormal body temperature, tachycardia, tachypnea, and abnormal white blood cell count.

All 212 patients in this study had what the researchers defined as a clinical infection—one that had been diagnosed through physical findings, imaging procedures, or other methods before microbiologic test results were available. Patients were considered to have a nosocomial infection if fever developed 72 hours or more after hospital admittance. Infection in the remaining patients was considered community-acquired.

The following clinical and laboratory variables were monitored at baseline and daily for two days: respiratory and heart rates, mean arterial blood pressure, body temperature, presence or absence of shaking chills, Glasgow Coma Scale score, hematocrit, hemoglobin concentration, platelet and white blood cell counts, erythrocyte sedimentation rate (ESR), and a variety of blood chemistry measurements. Additional blood samples were collected for culture at baseline and whenever considered necessary.

Two hundred of the 212 patients had SIRS. Fourteen of these patients developed shock within the first week after inclusion in the study; seven of these patients developed shock more than 48 hours after inclusion. None of the 12 patients who did not have SIRS developed shock.

Patients who had all four SIRS criteria within 48 hours of inclusion were more likely to develop shock than were patients with only two or three of these criteria. However, the peak white blood cell count and peak respiratory rate were better predictors of shock than were abnormal body temperature or peak heart rate.

Other variables that were strong predictors of shock were the nadir Glasgow Coma Scale score, the presence of bacteremia, and a low peak ESR. Although earlier researchers have questioned the value of bacteremia as a predictor of shock, Bossink et al believe that their results support this relationship. In fact, they suggested that bacteremia may play a causal role in the onset of shock “independent of the clinical indicators of the host response to infection.” They noted that recognizing these clinical risk factors can help physicians to assess disease severity and institute early treatment, thereby reducing the substantial morbidity and mortality that is associated with septic shock.

—Deborah L. O’Connor

Reference
1. Bossink AWJ, Groeneveld ABJ, Koffeman GI, Becker A. Prediction of shock in febrile medical patients with a clinical infection. Crit Care Med. 2001;29:25-31.

Return to table of contents