Background: The use of corticosteroids may help control the cytokine storm occurring in acute respiratory failure due to the severe form of COVID-19. We evaluated the postacute effect of corticosteroids used during the acute phase, such as impairment in pulmonary function parameters, on day 120 (D120)-follow-up, in participants who survived over 28 days.Methods: This is a parallel, double-blind, randomized, placebo-controlled phase IIb clinical trial carried out between April 18 and October 9, 2020, conducted in hospitalized patients with clinical–radiological suspicion of COVID-19, aged 18 years or older, with SpO2 ≤ 94% on room air or requiring supplementary oxygen, or under invasive mechanical ventilation (IMV) in a referral center in Manaus, Western Brazilian Amazon. Intravenous methylprednisolone (MP) (0.5 mg/kg) was given two times daily for 5 days to these patients. The primary outcome used for this study was pulmonary function testing at day 120 follow-up visit.Results: Out of the total of surviving patients at day 28 (n = 246) from the Metcovid study, a total of 118 underwent satisfactory pulmonary function testing (62 in the placebo arm and 56 in the MP arm). The supportive treatment was similar between the placebo and MP groups (seven [11%] vs. four [7%]; P = 0.45). At hospital admission, IL-6 levels were higher in the MP group (P < 0.01). Also, the need for ICU (P = 0.06), need for IMV (P = 0.07), and creatine kinase (P = 0.05) on admission also tended to be higher in this group. In the univariate analysis, forced expiratory volume on 1st second of exhalation (FEV1) and forced vital capacity (FVC) at D120 follow-up were significantly higher in patients in the MP arm, being this last parameter also significantly higher in the multivariate analysis independently of IMV and IL-6 levels on admission.Conclusion: The use of steroids for at least 5 days in severe COVID-19 was associated with a higher FVC, which suggests that hospitalized COVID-19 patients might benefit from the use of MP in its use in the long-term, with less pulmonary restrictive functions, attributed to fibrosis.Trial Registration:ClinicalTrials.gov, Identifier: NCT04343729.