Prognostic Markers of Geriatric Patients in the Icu with Respiratory Failure Secondary to Covid-19

ABSTRACT

INTRODUCTION: SARS-CoV-2 (COVID-19) has continued to be a public health emergency, affecting almost 450 million people worldwide, with a disproportionate significant disease burden in the elderly community. Our objective is to provide population specific prognostic markers upon description of demographic factors, clinical characteristics, diagnostic variables, treatment characteristics and outcome variables in critically ill geriatric patients with acute hypoxic respiratory failure due to COVID-19 infection. METHOD(S) This is a retrospective chart review of 165 patients admitted to a single institution's medical and cardiovascular intensive care unit between the dates of March 01, 2020 and December 31, 2020. Inclusion criteria was patients age greater than or equal to 65 years, documented positive COVID-19 polymerase chain reaction test result and a diagnosis of acute hypoxic respiratory failure. Our primary end point evaluated the rate of mortality in relation to multiple variables during intensive care unit admission. RESULT(S) Of 165 patients, 45 patients were excluded. Of the remaining 120 patients, 41 were females and 79 were males. Four independent risk factors are significantly associated with higher odds of mortality for the concerned population presence of solid tumor (AOR 0.002, 95% CI < 0.001, 0.31), maximum value of PaCO2 (AOR 1.094, 95% CI 1.029, 1.163), Charlson comorbidity index (AOR 2.962, 95% CI 1.59, 5.52), and use of diuretics (AOR 0.015, 95% CI < 0.001, 0.49). CONCLUSION(S) It was to our surprise that the mortality rate among those intubated was not statistically significant. However, it has been shown in prior research, which is in alignment with our results, that mechanical ventilation does not necessarily result in increased mortality. Certain factors were found to be poor prognostic markers during intensive care unit admission, which may predict a higher rate of mortality in those patient populations.