RESUMEN
INTRODUCTION: The coronavirus disease 2019 (COVID-19) pandemic continues to have a global impact. The behavior and viral course of severe acute respiratory syndrome coronavirus (SARS-CoV-2) remains unpredictable. We aimed to investigate the prediction factors associated with prolonged viral shedding in COVID-19 patients. METHODOLOGY: This is a retrospective, nested, case-control study with 155 confirmed COVID-19 infected patients divided into two groups based on nucleic acid conversion time (NCT), a prolonged group (viral RNA shedding >14 days, n = 31) and a non-prolonged group (n = 124). RESULTS: The mean age of participants was 57.16 years, and 54.8% were male. Inpatient numbers were 67.7% across both groups. No statistically significant differences between the two groups were observed in terms of clinical manifestation, comorbidities, computer tomography, severity index, antiviral treatment, and vaccination. However, C-reactive protein and D-dimer levels were significantly higher in the prolonged group (p = 0.01; p = 0.01). Using conditional logistic regression analysis, D-dimer and bacterial co-infection were found to be independent factors associated with the prolonged NCT (OR: 1.001, 95% CI: 1.000-1.001, p = 0.043; OR: 12.479, 95% CI: 2.701-57.654, p = 0.001 respectively). We evaluated the diagnostic value of the conditional logistic regression model by using receiver operating characteristic curve analysis. The area under the curve was 0.7 (95% Cl: 0.574-0.802; p < 0.001). CONCLUSIONS: Our study design included controlling confounders. We showed a clear result associating predicting factors with prolonged NCT of SARS-CoV-2. D-dimer level and bacterial co-infection were considered as independent predictors of prolonged NCT.