A single center retrospective observational study on the accuracy of the MuLBSTA score in predicting mortality among COVID-19 confirmed moderate to critical pneumonia cases

Abstract

Background: The coronavirus disease (COVID-19) is a global pandemic that caused millions of deaths worldwide. There is no standard risk stratification score for COVID-19 pneumonia. This study aims to determine the accuracy of the MuLBSTA score in predicting the risk of mortality in COVID-19 confirmed moderate to critical pneumonia cases.

Methodology: A total of 168 COVID-19-confirmed moderate to critical pneumonia patients admitted at Cardinal Santos Medical Center from January 1, 2021 to April 30, 2021 were included by chart review. The MuLBSTA score was determined for each patient using the following information: age, smoking history, co-morbidities, complete blood count, sputum culture, blood culture, chest xray and chest CT scan. All clinical outcomes were based on patient status by the end of the hospital stay (survival versus death). Thereafter, logistic regression was done using the MuLBSTA score and mortality to determine any correlation. In addition, modified regression was used to find any correlation with the MuLBSTA score and patient co-morbidities as predictors of mortality. Chi-square tests of independence were conducted to assess the specific cut-off values of the MuLBSTA score in predicting mortality.

Results: The MuLBSTA score is a significant predictor of mortality (73.08%) and survivability (66.67%). It was determined that the MuLBSTA score's accuracy in predicting mortality increases with diabetics [b = .26, p < .05]. In addition, the intervention of hemoperfusion can skew the predictive accuracy of the scoring [b = -.45, p<.01]. The study showed that a MuLBSTA score of 8 as a cut-off value to delineate high risk patients was more accurate in COVID-19 pneumonia patients compared to the previously established score cut-off of 12 in viral pneumonia [1].

Conclusion: The MuLBSTA score may be used for risk stratification in predicting mortality in COVID-19 pneumonia, especially among diabetic patients. A MuLBSTA score of 8 proves to be the more accurate cut-off in assessing risk of mortality in COVID-19. However, hemoperfusion makes the MulBSTA score inapplicable.