Residual lung abnormalities in recovered COVID-19 cases at 1-year follow-up: A single-center final radiological outcome phenotype study in India

Abstract

Abstract INTRODUCTION: The radiological phenotype represents the radiological characteristics of coronavirus disease 2019 (COVID-19) pneumonia. Radiological outcomes initially interpreted as post-COVID-19 lung fibrosis have been detected in short-term follow-up of recovered COVID-19 cases, but long-term radiological evaluation suggests that these abnormalities should be defined as post-COVID-19 sequelae. The present study aimed to evaluate the radiological phenotypes of post-COVID-19 residual radiological lung abnormalities (RLAs) in recovered COVID-19 cases. METHODS: This prospective, observational study included 3000 patients with polymerase chain reaction-confirmed COVID-19, with documented lung involvement and radiological severity categorized as mild, moderate, or severe by high-resolution computed tomography (HRCT) at admission. Final radiological outcomes were based on radiological assessments at 6 months and 1 year after hospital discharge. Final radiological phenotypes were defined as RLAs and categorized as resolving, persistent, or progressive, based on HRCT imaging at 1 year of follow-up. A total of 622 cases with RLAs were finally included. The final outcome phenotypes were evaluated in relation to age, sex, comorbidities, laboratory parameters, and interventions, including oxygen and ventilator-support requirements during hospitalization. The data were analyzed using Chi-square tests. RESULTS: RLAs were documented in 622 of the 3000 COVID-19 pneumonia cases (20.73%). The final radiological outcomes were defined as resolving in 77.17% (480/622), persistent in 15.43% (96/622), and progressive in 7.39% (46/622). Radiological outcome phenotypes and RLAs were significantly associated with the duration of illness at admission (P < 0.00001), laboratory parameters (D-dimer, C-reactive protein, and interleukin-6) at admission (P < 0.00001), radiological severity on computed tomography (P < 0.00001), and inpatient interventions (P < 0.00001). Radiological outcome phenotypes and RLAs were also significantly associated with age, sex, diabetes mellitus, ischemic heart disease, hypertension, chronic obstructive pulmonary disease, obesity, and lung functions assessed by spirometry at 1-year follow-up (P < 0.00001). CONCLUSIONS: Radiological outcome phenotypes have a crucial role in assigning final radiological outcomes in patients with COVID-19 pneumonia and differentiating between post-COVID-19 lung fibrosis and post-COVID-19 lung sequelae. RLAs, categorized as resolving, persistent, and progressive, provide a novel radiological classification of post-COVID-19 lung sequelae. Final phenotypic classification may play an important role in further workups and treatment planning for RLAs; some patients with progressive phenotypes may have underlying rheumatological predispositions and many cases may have resolving phenotypes requiring no further treatment, whereas cases with persistent RLAs require further follow-up.