Prognostic factors of severe infectious purpura in children

Abstract

The French Club of Pediatric Intensive Care has prospectively studied 90 cases of infectious purpura which were hospitalized in 1981; the purpose of this study was to determine prognostic factors. The statistical study (X2 test) of all these cases is in agreement with data in the literature and shows that the mortality is significantly higher when there is: shock (p less than 0.001), coma (p less than 0.05), ecchymotic or necrotic purpura (p less than 0.01), temperature less than 36 degrees C (p less than 0.05), no clinical meningism (p less than 0.001), white cell count less than 10,000/mm3 (p less than 0.05), thrombocytopenia less than 100,000 (p less than 0.01), fibrinogen less than 1.5 g/l (p less than 0.001), kalemia greater than 5 mEq/l (p less than 0.01), spinal fluid cell count less than 20/mm3 (p less than 0.01). Because shock is one of the main prognostic factors (23 deaths in 55 shocked patients, versus 2 in 35 non-shocked) we have performed another statistical study (with the Benzecri method) to determine a prognostic index for patients in shock. For its determination, five initial parameters are used: age, kalemia, white cell count, clinical meningism, platelet count. The predictive value for survival is 91%. The predictive value for death is 87%. The score was applied on the patients hospitalized in shock in 1982: the predictive value for survival is 75%, the predictive value for death is 61%.