Comparative analysis of the prognostic significance of the CURB-65 and CRB-65 scores and their modifications in hospital mortality assessment in patients with community-acquired pneumonia

Abstract

Background. Mortality associated with community-acquired pneumonia (CAP) continues to be a crucial health problem worldwide. Correct assessment of CAP severity and the level of care is pivotal in the disease outcome.Aim. To evaluate the prognostic value of the CURB-65 and CRB-65 scores and their modifications in determining the risk of in-hospital mortality in patients with CAP.Materials and methods. The retrospective study included 1,412 patients with CAP aged over 18 years. In a population of 1,020 patients, which was subsequently split into test (n = 676) and training (n = 344) samples in the ratio 2 : 1, we compared the predictive value of the CURB-65 (confusion, urea > 7 mmol / l, respiratory rate ≥ 30 / min, low blood pressure (BP), and age ≥ 65 years) and CRB-65 (confusion, respiratory rate ≥ 30 / min, low blood pressure (BP), and age ≥ 65 years) scores in identifying patients at high risk of in-hospital death. The specified scoring systems were modified by changing the cut-offs for each criterion to increase their accuracy. For comparison, we used the ROC analysis with the calculation of the area under the curve (AUC).Results. The modified CURB-65 score with new cut-off values (age > 72 years, respiratory rate > 21 / min, urea level > 9.5 mmol / l, systolic blood pressure ≤ 105 mm Hg, and diastolic blood pressure ≤ 65 mm Hg) was more accurate than the original one in predicting death and was named CURB-72. The AUC for CURB-72 and CURB-65 was 0.946 (95% confidence interval (CI): 0.916–0.967) and 0.905 (95% CI: 0.869–0.934), respectively (p = 0.0034). The modified CRB-65 (CRB-72) score also outperformed the original model, but showed no statistically significant difference. While comparing the modified scoring systems, the new CURB-72 score surpassed the CRB-72 score and demonstrated maximum accuracy in identifying CAP patients at risk of in-hospital mortality (p = 0.0347).Conclusion. The modified CURB-65 (CURB-72) and CRB-65 (CRB-72) scores demonstrated potential for assessing the prognosis of CAP and are superior to classical scoring systems. CURB-72 showed the highest sensitivity and specificity.