Factors Predictive of Outcome in COVID-19 Pneumonia with Acute Respiratory Failure

Abstract

Rationale: Coronaviruses usually cause mild upper respiratory tract infections in humans, however, severe acute respiratory syndrome coronavirus 2 (SARSCoV-2) can replicate in the lower respiratory tract. The pathophysiology of SARS-CoV-2 is similar to its predecessor SARS-CoV inciting an aggressive and disproportionate host immune response leading to multi organ failure and death. Multiple factors have been studied for their association with outcome in COVID-19. Variable data exists in current literature regarding the impact of demographic factors, admission hemoglobin, creatinine, d-dimer, ferritin and BNP on patient's survival. The objective of this study was to identify whether these parameters affect mortality in COVID-19 with acute respiratory failure. Methods: Seventy-one patients admitted to our hospital with COVID-19 and acute respiratory failure were studied. Medical records were reviewed to obtain age, gender, body mass index (BMI), admission hemoglobin (Hb), white blood cell (WBC) count, d-dimer, C-reactive protein (CRP), ferritin, creatinine, brain natriuretic peptide (BNP), and outcome (survived or expired). Correlation analysis and t-test was performed to determine the impact of above parameters on outcome. Results: Of the 71 patients, 73% were male and 27% were females. Mean age was 47.7 + 16.7 years. Mean BMI was 32.27 + 2.73 kg/m2. Nineteen percent (19%) patients required invasive mechanical ventilation (MV);twenty-two percent (22%) of patients expired. Studied admission parameters in survivors and non-survivors are shown in the table below. Pearson's correlation analysis showed a significant correlation between mortality and variables such as age, BMI and WBC count. No correlation was observed with gender, admission Hb, creatinine, CRP, ferritin, d-dimer or BNP levels. Conclusions: Information on COVID-19 continues to evolve. Future studies aimed at determining clinical parameters at the time of hospital admission that can predict mortality can be helpful in optimizing treatment and monitoring of these patients. Our study did not show any correlation between mortality and various inflammatory markers including CRP, ferritin, d-dimer which could have been due to the limited number of patients in this study.