PREDICTIVE POWER OF THE MODIFIED BRIXIA CHEST X-RAY SCORING SYSTEM IN PROGNOSTICATING IN-HOSPITAL MORTALITY IN ICU-ADMITTED COVID-19 PATIENTS

Purpose:
Given the emergence of new strains (some of concern) of SARS-COV-2 every now and then, and considering the fast viral transmittion (especially in newer strains), implementing a dependable diagnostic/staging method is of great importance in risk-stratifying and allocating resources accordingly. The aim of this study was to determine the prognostic value of the modified Brixia CXR scoring system for predicting final in-hospital outcome in ICU-Admitted COVID-19 patients.
Methods:
In this cross-sectional, observational, single-center study, 108 ICU-admitted, RT-PCR confirmed cases of COVID-19 who underwent on-admission plain chest radiograph were enrolled. Relevant available clinical, laboratory and historical data were extracted from patients’ paper and electronic records, and via phone contact in cases of missing data. Plain chest radiographs were inspected by two radiologists, and cases of mismatch were resolved by consensus. Finally, the predictive power of modified Brixia chest X-Ray scoring system to prognosticate the ICU length of stay and mortality in studied participants was investigated.
Results:
The CXR score was meaningfully predictive of mortality, with a cut-off point of 5.5 and an area under the curve of 0.925. Diabetes (P = 0.001), cardiovascular disease (P = 0.006), central pulmonary involvement (P < 0.001), bilateral pulmonary involvement (P < 0.001) and pleural effusion (P < 0.001) were significantly associated with higher CXR scores. Additionally, CXR score was significantly associated with a higher rate of subsequent tracheal intubation (P < 0.001); However, it was not significantly correlated with ICU length of stay (P = 0.05).
Conclusion:
The present study showed that the modified Brixia CXR scoring system is a powerful predictor of mortality in ICU-Admitted COVID-19 patients, and might be of help to triage patients and determine high-risk cases for the purpose of resource allocation.