Long-term changes in spirometry and diffusing capacity in Mexican Hispanics with previous severe COVID-19
Description
Background: This longitudinal study (April 2020 - January 2023) investigated lung function (spirometry, pulmonary diffusing capacity) in Mexican Hispanics who experienced severe COVID-19. It focused on how recovery time affects lung function improvements, hypothesizing that patients with a longer recovery between diagnosis and pulmonary testing would show better lung function than those tested earlier. Methods: At a COVID-19 follow-up clinic in Yucatan, Mexico, lung function and symptoms were assessed in patients recovered from severe COVID-19. We used z-scores and Wilcoxon signed rank tests to analyze lung function changes over time. Lung function was measured twice in 82 patients: at a median of 94 and 362 days after COVID-19 diagnosis. High-resolution computed tomography (HRCT) was conducted in 44 of these 82 subjects, with a median time of 38 days between CT scanning and a pulmonary function test. Z-scores were determined using reference equations for spirometry (doi: 10.1164/rccm.202205-0963OC) and DLCO [Gochicoa-Rangel L. G. et al., (2024). Reference equations for DLNO & DLCO in Mexican Hispanics: Influence of altitude and race. BMJ Open Respir Res, in press as of September 2024]. Seven pulmonary ailments were assessed, identified based on the 2022 ERS/ATS interpretation strategies (doi: 10.1183/13993003.01499-2021): (A) Restrictive spirometry pattern (FEV1/FVC > LLN, FVC < LLN); (B) Airflow obstruction (FEV1/FVC < LLN, FVC > LLN); (C) Mixed disorder (FEV1/FVC < LLN, FVC < LLN); (D) Loss of alveolar-capillary structure with loss of lung volume (DLCO < LLN, VA < LLN, KCO < ULN); (E) Localized loss of lung volume or incomplete lung expansion (DLCO < LLN, VA < LLN, KCO > ULN); (F) Pulmonary vascular abnormality (DLCO < LLN, VA normal); and (G) Alveolar hemorrhage, polycythemia, increased blood flow (left to right shunt, or post-exercise). Results: Initially, 61% of patients exhibited at least one of seven pulmonary function abnormalities (LLN = –1.645), which decreased to 22% by 390 days post-recovery. Considering day-to-day variability, 68% of patients showed improvement by the final visit, while 30% had unchanged lung function from the initial assessment. Computed tomography (CT) scans revealed ground-glass opacities in 33% of patients. One year after infection, DLCO z-scores accounted for 29% of the variation in HRCT fibrosis scores. No significant correlation was found between recovery length and lung function improvement based on z-scores. Conclusion: Twenty-Two percent of patients who recovered from severe COVID-19 continued to show at least one lung function abnormality one year after recovery, indicating a prolonged impact of COVID-19 on lung health. The data is in SPSS (.sav) format. Once the file is open you will find the labels for each parameter under the "VARIABLE VIEW" tab.
Files
Steps to reproduce
Severe COVID-19 in adults, as defined by WHO criteria, includes any of the following: oxygen saturation below 90% on room air, severe pneumonia, or signs of severe respiratory distress (e.g., use of accessory muscles, inability to complete full sentences, respiratory rate >30 breaths per minute; WHO, 2023). Exclusion criteria were patients with pneumonia from non-SARS-CoV-2 causes, mild or moderate COVID-19, or only one follow-up evaluation. Pulmonary function testing was scheduled at approximately 1, 3, 6, and 12 months post-diagnosis. Height and weight were measured using a calibrated eye-level scale. This study followed the 2022 ATS/ERS guidelines using z-scores to evaluate the persistence and recovery of pulmonary abnormalities in Mexican Hispanics. Z-scores are preferred over percent predicted values, as the LLN of percent predicted changes with age. Seven pulmonary ailments were assessed based on 2022 ERS/ATS strategies, including restrictive spirometry pattern, airflow obstruction, mixed disorders, loss of alveolar-capillary structure, localized lung volume loss, pulmonary vascular abnormality, and alveolar hemorrhage. The furthest apart dates were selected for patients with fewer than two tests. Significant changes in pulmonary function were determined by comparing z-scores between visits using paired t-tests for normally distributed data and Wilcoxon signed rank tests otherwise. The Shapiro-Wilk Test confirmed normality. For multiple t-tests, the Benjamini-Hochberg procedure controlled the false discovery rate. Changes in lung function were examined by summing z-scores at each visit, with 95% CIs derived from 1000 bootstrapped samples. Linear regression assessed the relationship between lung function changes and time between tests, checking assumptions of linearity, homoscedasticity, and normality. Binary logistic regression analyzed associations between meaningful z-score changes and variables like sex, height, weight, age, or BMI. Patients reported symptoms such as fatigue, cough, or chest pain, which were compared using Wilcoxon Signed Rank Tests and correlated with z-score changes. High-resolution CT (HRCT) scans at 12 months assessed fibrosis, scored on a system adapted from Goh (Am J Respir Crit Care Med 177, 1248-1254) examining reticular patterns and fibrosis severity. The Goh fibrosis score was correlated with DLCO z-scores, all evaluated by the same radiologist. Statistical analyses used IBM SPSS (Version 29.0.1.0) with significance set at p < 0.05, and figures were created using GraphPad Prism (version 10.3.0.507). Z-scores for pulmonary function were calculated using GLI GLOBAL reference equations (doi: 10.1164/rccm.202205-0963OC) and DLCO [Gochicoa-Rangel L. G. et al., (2024). Reference equations for DLNO & DLCO in Mexican Hispanics: Influence of altitude and race. BMJ Open Respir Res, in press as of September 2024].Cases below the LLN were considered abnormal.