Follow-up study of pulmonary function among COVID-19 survivors 1 year after recovery.

Dear Editor

Introduction

Coronavirus disease 2019 (COVID-19) is a novel systemic disease that affects multiple organs, with the lungs being most affected.1, 2, 3 Previous studies have demonstrated that carbon monoxide diffusing capacity (DLCO) is impaired in patients who had recovered from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection at the time of discharge. However, long-term pulmonary function in survivors is poorly understood. Here, we assessed pulmonary function in survivors who had recovered from SARS-CoV-2 infection 1 year previously.

Methods

In this cohort study conducted from March 16 to March 28, 2021, we followed up a total of 119 survivors with SARS-CoV-2 infection who had been hospitalized during January 24–March 18, 2020 in Huanggang, Hubei Province, China. Study inclusion criteria included a previous COVID-19 diagnosis (positive PCR result for SARS-CoV-2) and the willingness and ability to provide informed consent. Baseline demographics, smoking status, body mass index and comorbidities were extracted from the electronic medical record. The severity of the disease was defined according to the World Health Organization COVID-19 guidelines. Severe COVID-19 refers to fever or suspected respiratory infection, plus one of the following: respiratory rate > 30 breaths/min; severe respiratory distress; or SpO2 ≤ 93% on room air.

Lung function tests were performed by technicians in the lung function laboratory using the Master Screen Body (Jaeger, Germany). The procedure followed was in accordance with American Thoracic Society/European Respiratory Society guidelines.

This study was approved by the Hunan Provincial People's Hospital Ethics Commission. All participants provided their written or verbal consent to participate.

Results

A total of 119 survivors participated in this study (asymptomatic, n = 9; non-severe, n = 82; severe, n = 28) (Table 1 ). The median patient age was 52.97 (±12.17) years; 49 survivors (41%) were men and 70 (59%) were women. Twenty-four survivors (20%) had at least one chronic comorbidity, 10 (8%) with hypertension and 11 (9%) with diabetes; only 2 (2%) patients were reported as having chronic obstructive pulmonary disease. There were no statistically significant differences in age, sex, body mass index, and smoking status among the three groups.

Table 1

Demographics and pulmonary function characteristics of survivors with COVID-19.

Variable	Total (n = 119)	Asymptomatic cases (n = 9)	Non-severe cases (n = 82)	Severe cases (n = 28)	P-value*
Age, median(SD), y	52.97±12.17	46.44±10.48	52.66±12.39	56.00±11.40	0.111
Gender
Male, no, (%)	49 (41%)	3 (33%)	35 (43%)	11 (39%)	0.841
Female, no, (%)	70 (59%)	6 (67%)	47 (57%)	17 (61%)
Cigarette smoking
Never-smoker	86 (72%)	6 (67%)	58 (%)	22 (%)	0.737
Current smoker	15 (13%)	1 (11%)	10 (%)	4 (%)
Former smoker	18 (15%)	2 (22%)	14 (%)	2 (%)
BMI kg•m − 2	25.07±3.22	24.48±3.09	24.98±3.21	25.51±3.26	0.638
Comorbidities	24 (20%)	0	13 (16%)	11 (39%)	0.008
Hypertension	10 (8%)	0	6 (7%)	4 (14%)	0.331
Diabetes	11 (9%)	0	4 (5%)	7 (25%)	0.005
Cardiovascular diseases	2 (2%)	0	0	2 (7%)	0.037
Malignant tumor	1 (1%)	0	1 (1%)	0	0.797
COPD	2 (2%)	0	1 (1%)	1 (4%)	0.649
Liver disease	1 (1%)	0	1 (1%)	0	0.797
Chronic kidney disease	1 (1%)	0	1 (1%)	0	0.797
Spirometry
FVC% pred	97.7 ± 13.76	98.82±12.36	97.93±13.72	96.68±14.70	0.890
FVC <80% pred	11 (9%)	0	7 (9%)	4 (14%)	0.404
FEV1% pred	98.22±14.25	98.11±13.84	98.12±14.19	98.54±14.10	0.991
FEV1 <80% pred	11 (9%)	0	8 (10%)	3 (11%)	0.602
FEV1/FVC%	80.56±7.82	81.26±4.30	80.36±7.95	80.90±8.46	0.917
FEV1/FVC <70%	6 (5%)	0	5 (6%)	1 (4%)	0.672
MMEF75/25	77.60±26.06	80.24±16.59	76.70±26.89	79.38±26.64	0.855
MMEF75/25 <65%	41 (34%)	2 (22%)	30 (37%)	9 (32%)	0.661
Diffusion capacity
DLCO% pred	81.27±13.06	84.38±5.94	81.94±12.56	78.34±15.74	0.347
DLCO <80% pred	47 (39%)	1 (11%)	31 (38%)	15 (54%)	0.605
DLCO/VA% pred	103.74±16.86	106.21±10.84	103.66±16.94	103.18±18.56	0.895
DLCO/VA <80% pred	10 (8%)	0	7 (9%)	3 (11%)	0.600
Lung volume
TLC% pred	81.52±9.41	80.70±7.47	82.41±9.90	79.16±8.25	0.281
TLC <80% pred	50 (42%)	4 (44%)	34 (41%)	12 (43%)	0.980
RV% pred	70.67±17.61	61.49±11.93†	73.52±18.38	65.27±14.70†	0.026
RV <65% pred	50 (42%)	6 (67%)	28 (34%)	16 (57%)	0.031
RV/TLC% pred	85.36±20.11	75.30±11.48	87.61±19.59	82.02±22.73	0.132

Data are mean (SD), median (IQR), or n (%), unless otherwise specified.

Comparisons between continuous variables were performed with one-way ANOVA. Chi-squared test and Fisher's exact test were applied to categorical variables as appropriate.

*Difference among all types.

†P<0.05 versus non-severe cases;.

Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; FVC, forced vital capacity; pred, predict; FEV1, forced expiratory volume in the first second; MMEF, maximum mid-expiratory flow; DLCO, diffusing capacity of the lung for carbon monoxide; DLCO/VA: DLCO corrected for alveolar volume; TLC, total lung volume; RV, residual volume.

Anomalies were found for the percent predicted DLCO (n = 47, 39%), DLCO/alveolar volume (n = 10, 8%), percent predicted total lung capacity (TLC; n = 50, 42%), percent predicted residual volume (n = 50, 42%), percent predicted forced expiratory volume in 1 second (FEV1; n = 11, 9%), maximal mid-expiratory flow (MMEF) 75/25 (n = 41, 34%), percent predicted forced vital capacity (FVC; n = 11, 9%), and FEV1/FVC (n = 6, 5%).

As shown in Table 1, there was no statistically significant difference in damaged diffusing capacity among groups with different disease severity, with 11% in the asymptomatic group, 38% in the non-severe group, and 54% in the severe group, respectively (P = 0.605). However, the gradual decline in lung diffusion capacity among survivors was consistent with varying degrees of severity. There was no significant difference in other measures (TLC, RV/TLC, FVC, FEV1, FEV1/FVC, and MMEF 75/25) among COVID-19 survivors with different disease severity.

Discussion

Previous studies have shown that survivors of COVID-19 may have lung damage.4, 5, 6, 7 In follow-up studies lasting 3–6 months among rehabilitating COVID-19 severe/critical patients, DLCO damage was the most common abnormality, accounting for 56%–82% of cases, followed by TLC deficiencies.6, 7, 8 They found significant differences in diffusing capacity damage among groups with different disease severity.

Alessia et al. found that 10 of 13 patients with COVID-19 pneumonia were damaged at the time of discharge. After 6 weeks, lung function was improved but a certain degree of restrictive changes remained. In this cohort study, lung functional impairment are highly prevalent in survivors with COVID-19 1 year after discharge. Forty-seven (39%) survivors had impaired diffusing capacity during the 1-year follow-up, with no significant difference between the severe and non-severe groups. This may indicate that pulmonary function damage from COVID-19 can improve over time.

DLCO abnormalities occurred in 39% of survivors, indicating damaged intra-alveolar diffusion pathways. Autopsy in patients who died from SARS-CoV-2 infection showed diffuse alveolar injury, accompanied by thrombosed small vessels with remarkable associated hemorrhage. Changes in lung pathology can explain the diffusing capacity damage to a certain extent. Moreover, a proportion of patients with COVID-19 developed acute respiratory distress syndrome (ARDS). Pulmonary fibrosis can develop as a result of chronic inflammation of the lungs owing to ARDS. Pulmonary fibrosis associated with ARDS in COVID-19 patients may damage alveolar-capillary units, causing loss of alveolar units and impaired gas exchange.

Patients with severe or critical COVID-19 may need to use ventilators in the intensive care unit for several weeks. The breathing muscles are affected, which weakens the ability to breathe. Pulmonary rehabilitation involves suggestions for physical exercise and management of symptoms and is important to help survivors fully recover.

Our study had several limitations. First, the lack of baseline pulmonary function data before the illness onset made it difficult to conduct comparisons with post-illness results. Moreover, we only carried out 1 year of follow-up; the long-term dynamic changes of pulmonary function after SARS-CoV-2 infection need further study.

In summary, in this cohort study, we found that lung functional impairment are highly prevalent in survivors with COVID-19 1 year after discharge, and persistent lung function impairment was found in about 40% of survivors. Lung damage might be related to pulmonary fibrosis. Further long-term research is needed to understand the mechanisms underlying long-term SARS-CoV-2-related pulmonary function damage.

Author contributions

YZhu, XH, YZeng, and XY generated the research question and analysis plan. XY, HH, CW, ZJ, ZZ, JH, SY, MF, JH, FC were involved in data collection and clinical appointments. XY, HH, CW, ZJ, ZZ, JH, and SY were involved in data analysis. All authors were involved in the final manuscript preparation.

Funding

This study was supported by the Huanggang Municipal Headquarters for COVID-19 Epidemic Prevention and Control, Key Project of Hunan Provincial Science and Technology Innovation (No. 2020SK1011, 2020SK1010), and Program of Hunan Science and Technology Department (No. 2020SK3011).

Declaration of Competing Interest

No conflicts of interests declared by an author.