Clinical characteristics and pulmonary function tests of patients hospitalised with coronavirus disease 2019 (COVID-19), 6 months after symptom onset
| Features | Overall cohort | Patients with dyspnoea at 6 months | p-value | |
| Unexplained dyspnoea | Dyspnoea | |||
| Subjects n | 73 | 13 | 15 | |
| Demographics | ||||
| Age years | 65 (53–72) | 49 (34–67) | 66 (59–76) | 0.02 |
| Male sex n (%) | 44 (60) | 4 (31) | 9 (60) | 0.12 |
| Ever-smoker n (%) | 23 (32) | 2 (15) | 8 (53) | 0.06 |
| Comorbidities n (%) | ||||
| Hypertension | 27 (37) | 5 (39) | 8 (53) | 0.43 |
| Diabetes | 19 (26) | 3 (23) | 5 (33) | 0.69 |
| Chronic pulmonary disease# | 10 (14) | 0 | 4 (27) | 0.10 |
| Coronary heart disease | 7 (10) | 0 | 3 (20) | 0.23 |
| Malignancy | 8 (11) | 1 (8) | 1 (7) | 1.00 |
| Chronic kidney disease | 6 (8) | 1 (8) | 2 (13) | 1.00 |
| Respiratory symptoms | ||||
| UCSD dyspnoea score | 9 (3–31) | 31 (17–40) | 35 (23–46) | 0.27 |
| Cough VAS mm | 20 (10–37) | 10 (9–10) | 30 (16–44) | 0.07 |
| Patient-reported outcome measures | ||||
| EQ-5D health utility | 0.9 (0.8–0.9) | 0.83 (0.77–0.87) | 0.83 (0.76–0.87) | 0.79 |
| EQ-5D VAS | 80 (75–90) | 75 (70–90) | 75 (65–85) | 0.50 |
| PSQI | 5 (2–9) (n=72) | 9 (6–12) (n=12) | 7 (5–9) | 0.28 |
| PHQ-9 | 1 (0–6) | 6 (2–10) | 5 (1–7) | 0.39 |
| Pulmonary function tests | ||||
| FEV1 % predicted | 91±15 (n=72) | 88±14 | 83±14 | 0.31 |
| FVC % predicted | 93±16 (n=72) | 93±11 | 81±15 | 0.03 |
| FEV1/FVC % | 84±12 (n=72) | 84±13 | 84±11 | 0.88 |
| TLC % predicted | 87±13 (n=64) | 86±11 (n=12) | 77±13 (n=14) | 0.09 |
| DLCO % predicted | 79±18 (n=70) | 88±9 | 63±14 | <0.001 |
| Transthoracic echocardiogram (n=72) | ||||
| LVEF % | 60 (60–65) | 60 (60–64) | 65 (60–65) | 0.29 |
| PASP mmHg | 27 (23–30) | 19 (23–34) | 27 (23–30) | 0.77 |
Data are shown as mean±sd or median (IQR) unless otherwise stated. Data denoted in bold indicate p-value <0.05. Patients with dyspnoea (n=31) have been categorised into those with unexplained dyspnoea (i.e. UCSD >10 with DLCO ≥80% predicted) and those with dyspnoea (i.e. UCSD >10 with DLCO <80% predicted). There were three patients who did not have DLCO measurements and could not be categorised. Echocardiogram data are from 3 months after symptom onset. UCSD: University of California, San Diego shortness of breath questionnaire; VAS: visual analogue scale; EQ-5D: EuroQol- 5 Dimension; PSQI: Pittsburgh Sleep Quality Index; PHQ-9: Patient Health Questionnaire-9; FEV1: forced expiratory volume in 1 s; FVC: forced vital capacity; TLC: total lung capacity; DLCO: diffusing capacity of the lung for carbon monoxide; LVEF: left ventricular ejection fraction; PASP: pulmonary artery systolic pressure. #: asthma (n=3), COPD (n=4), interstitial lung disease (n=2) or previous pulmonary embolism (n=1).