Identifying CT predictors of mortality in nonelderly healthy patients with COVID-19 pneumonia will aid to distinguish the most vulnerable patients in this age group and thus alter the management. We aimed to evaluate the prognostic value of multiple CT features of COVID-19 pneumonia on initial presentation in nonelderly patients without underlying medical conditions. In this retrospective case-control study, thirty laboratory-confirmed COVID-19 patients with no known major underlying disease who underwent a chest CT scan and expired of pneumonia within the following 30 days after admission, were included as case group. Sixty control subjects individually matched on their age, gender, without underlying medical conditions, who received same-criteria standard care and were discharged from the hospital in 30-day follow-up were included in the control group. A conditional logistic regression model was applied. Applying a univariate conditional logistic regression model, it was revealed that bilateral lung disease, anterior involvement, central extension, GGO, consolidation, air bronchograms, pleural effusion, BMI ≥ 25 kg/m² and CT severity score were the significant preliminary predictors (all p-values < 0.05). Next, by applying a multivariate conditional logistic regression model, it was determined that the CT severity score is the only statistically significant CT predictor of mortality (Odds Ratio = 1.99, Confidence Interval: 1.01-4.06, p-value < 0.05). The ROC curve analysis revealed a score of 7.5 as the cut-off point of CT severity score with the highest sensitivity (0.83) and specificity (0.87). Our study demonstrates that CT severity score is a reliable predictor factor of mortality in nonelderly previously healthy individuals with COVID-19 pneumonia. Assessment of disease extension in addition to the morphological pattern is necessary for CT reports of COVID-19 patients. This may alert the clinicians to alter the management for this specific group of patients, even when they are clinically silent or have a mild presentation. Copyright © 2020 Elsevier B.V. All rights reserved.