Validité des scores de gravité chez les patientes obstétricales en soins intensifs

Assessing the prognostic value of the scores Simplified Acute Physiology Score (SAPS II), Acute Physiology And Chronic Health Evaluation (APACHE II), Mortality Predicting Model (MPM II0 et MPM II24) and Organ Dysfunctions and/or Infection (ODIN) in obstetric patients admitted in intensive care unit (ICU). A retrospective study conducted in the ICU of Sfax. All obstetrical patients between 01/01/1993 and 31/12/2012 were included. The 5 scores were calculated on admission. The discriminatory power was studied by the method of ROC curves. The calibration was based on the comparison of the observed mortality (OM) and the predicted mortality (PMR) and the calculation of the standardized mortality ratio (SMR) (SMR = OM/PMR). Calibration was evaluated by the test of Lemeshow Goodness of fit-Hosmer (H). We included 725 patients. ICU mortality was 6.7%. The areas under the ROC curve for scores SAPS II, APACHE II, MPM II0, MPM II24 and ODIN were respectively 0.88, 95% CI [0.82,0.93], 0.84 95% CI [0.77,0.89], 95% CI 0.86 [0.81,0.91], 95% CI 0.94 [0.90,0.91] and 95% CI 0.93 [0.90,0.95]. SMR was close to 1 for SAPS II and MPM II24 scores (respectively: 0.90 and 0.97), whereas it was 0.53, 0.54 and 0.6 respectively for APACHE II, MPM II0 and ODIN scores. The value of H was 13.65 (P > 0.05) and 2.14 (P > 0.05) respectively for scores SAPS II and MPM II24 while it was 32.48 (P < 0.001), 32.47 (P < 0.001) and 37.56 (P < 0.001) respectively for scores APACHE II, MPM II0 and ODIN. For each of the 5 studied scores, there was an excellent discriminating power in terms of mortality. SAPS II and MPM II24 had perfect calibration. APACHE II, MPM II0 and ODIN scores gave an overestimation of mortality.