Steroid pulse therapy and vital prognosis among pediatric cancer patients with sepsis

Abstract

The usefulness of glucocorticosteroid therapy in patients with sepsis has been controversial. We investigated the effect of steroid pulse therapy on the vital prognosis of pediatric patients with sepsis and followed the vital status up to one month after the use of pulse therapy. We reviewed the medical records of 89 pediatric cancer patients with sepsis treated at our hospital between 1988 and 1996. The risks of potential predictors were estimated by calculating crude and adjusted relative risk (RR). The total cumulative death was 33/89 (25%). All patients treated with steroid pulse therapy died (12/12). Patients with either interstitial pneumonia (IP), infection-associated hemophagocytic syndrome (IAHS) and graft-versus-host disease (GVHD) showed a significantly higher cumulative death (57% (12/21) vs 31% (21/68), p = 0.03). Increased risk of cumulative death was suggested for the older age group, inappropriate antimicrobial therapy, and the conditions requiring steroid pulse therapy (IP, IAHS, and GVHD), (crude RR were 1.6, 1.6, and 1.9, respectively). However, when adjusted for pulse therapy, these three factors no longer indicated risk elevation, (adjusted RR = 1.2, 1.2, and 0.3, respectively). On the other hand, steroid pulse therapy per se was independently associated with increased risk of cumulative death (crude RR = 3.6, adjusted RR = 10). Thus, the risk of the conditions requiring steroid pulse therapy (IP, IAHS, and GVHD) firstly observed could be regarded as an apparent association due to steroid pulse therapy.