Neutrophil-to-lymphocyte ratio as a prognostic marker of disease severity and in-hospital mortality in hospitalized patients with COVID-19: a retrospective Saudi cohort study

Abstract Background Coronavirus disease 2019 (COVID-19) can cause a wide range of health issues, from relatively mild symptoms to severe life-threatening conditions. It’s important to identify patients predisposed to adverse clinical outcomes so that their care can be managed better. The neutrophil-to-lymphocyte ratio (NLR), which is an inexpensive and highly accessible biomarker of systemic inflammation, has been suggested as a way to predict how someone with severity of COVID-19. This present study investigated the utility of NLR in prediction of COVID-19 disease severity and in hospital death in Saudi Arabia. Methods This study included COVID-19 patients hospitalized at King Fahd Specialist Hospital in Buraydah, Qassim, Saudi Arabia, from April to December 2021. Patients’ data including age, gender, comorbidities, lab test results, hospital stay length, and outcome (death or recovery) were gathered from patients’ electronic records. Data were analyzed using regression and survival analysis. Results A total of 204 patients who were hospitalized were included in the study, and out of these, 56 patients (which is 27.5%) had critical illness. Critical patients were much older than non-critical cases (72.7 ± 13.4 years versus 61.9 ± 18.8 years, P < 0.001), with higher levels D-dimer, aspartate aminotransferase, total bilirubin, lactate dehydrogenase, and amylase. The NLR was much higher in dead cases compared to survivors (the middle value was 7.2 versus 4.3, with a P-value of 0.002). Furthermore, higher NLR values were recognized in critical patients and those in the intensive care unit, with no statistical significance. The ROC analysis showed that the NLR had only a moderate ability to predict in-hospital mortality, with an AUC of 0.638 and the best cutoff value at 5.01.It also showed a similar but slightly lower ability to predict disease severity, with an AUC of 0.581 and the best cutoff value at 6.2. Conclusion NLR was independently associated with in-hospital mortality in multivariable logistic regression, but not with critical disease severity. However, NLR did not retain an independent association with time-to-death in the fully adjusted Cox proportional-hazards model. NLR is a low-cost, and easily available marker of inflammation that could help to assess a patient’s risk early on, especially when used together with other clinical information and lab tests. More research is needed, especially large studies involving multiple centers, that use detailed analysis of several factors to better confirm how useful this is in predicting outcomes.