PROSPECTIVE OBSERVATIONAL STUDY COMPARING SEQUENTIAL ORGAN FAILURE ASSESSMENT SCORE WITH PH AND SEQUENTIAL ORGAN FAILURE ASSESSMENT SCORE FOR OUTCOME PREDICTION IN ICU

Abstract

Background : The SOFA (Sequential Organ Failure Assessment) score is widely used in ICUs (Intensive Care Units) to assess organ dysfunction and predict patient outcomes. The addition of blood pH as a parameter may enhance its predictive accuracy. This study aims to compare the SOFA score with and without pH for prognosis, mortality prediction, and risk estimation of hospital death in ICU patients.

Methods : This prospective, observational clinical study included ICU patients aged 18 years and older who were admitted for more than 48 hours after obtaining informed consent. A thorough general and systemic examination was conducted, and routine investigations were performed based on clinical findings. The SOFA score and SOFA score with pH were calculated using laboratory data and patient characteristics at admission (0 hours) and 48 hours. Outcomes, including ICU length of stay, hospital length of stay, and survival status, were recorded. The predictive ability of the scores was analyzed and correlated with actual patient outcomes.

Results : The overall mortality rate in the study was 27%. The mean age of survivors and non-survivors was 63.47 and 62.0 years, respectively (p = 0.725), showing no significant difference. No difference was observed between survivors and non-survivors in terms of ICU and hospital stay (p > 0.05). The mean SOFA score at baseline was significantly higher in non-survivors (8.33 vs. 5.9; p < 0.01), and this difference increased at 48 hours (11.67 vs. 3.66; p < 0.01). Similarly, the SOFA score with pH at baseline was higher in non-survivors (10.11 vs. 7.49; p < 0.01), and the difference was more pronounced at 48 hours (14.63 vs. 4.29; p < 0.01). Receiver operating characteristic (ROC) curve analysis demonstrated that the predictive accuracy of the SOFA score with and without pH was higher at 48 hours compared to baseline. At 48 hours, the diagnostic accuracy of SOFA with pH (AUC = 0.994, p < 0.01) was slightly superior to SOFA alone (AUC = 0.988, p < 0.01). The sensitivity and specificity of SOFA >6 at baseline were 66.7% and 71.2%, respectively, with an overall diagnostic accuracy of 70%. For SOFA with pH >6 at baseline, sensitivity and specificity were 70.4% and 69.9%, respectively, with a diagnostic accuracy of 70%. At 48 hours, SOFA >6 had a sensitivity and specificity of 92.6% and 95.9%, respectively, with an overall accuracy of 95%. The addition of pH to SOFA further improved predictive performance, with SOFA + pH >6 at 48 hours achieving 100% sensitivity, 97.3% specificity, and 98% diagnostic accuracy.

Conclusion : Sequential assessment of organ dysfunction using the SOFA score is a highly reliable predictor of ICU patient outcomes. Evaluating the score at 48 hours significantly enhances its prognostic value. The inclusion of pH further improves the predictive accuracy. Therefore, we recommend using the SOFA score at admission and at 48 hours, with the addition of pH preferred for more accurate mortality prediction.