Clinical Predictors of Lung Function in Patients Recovering From Mild COVID-19

Arturo Cortes-Telles; Esperanza Figueroa-Hurtado; Diana Lizbeth Ortiz-Farias; Gerald Stanley Zavorsky

BMC Pulm Med. 2022;22(294)

Conclusions

Nearly one-third of patients with mild COVID-19 have impaired DLCO 34 days post-diagnosis, and one-fifth of patients have lung restriction. The odds of having an impaired DLCO at follow-up increased by 10% for every one-year increase in age (from 25 to 83) and increased 12-fold if a restrictive spirometric pattern was evident. However, having excessive night sweats and a blocked/runny nose each reduced the probability of an impaired DLCO at follow-up by about 90%, demonstrating a protective effect against an impaired gas exchange. In a summary of 22 studies, having severe COVID-19 disease at diagnosis increased the percentage of those with impaired DLCO by 21%. And, if the study used < 80% of predicted to define DLCO impairment, then 13% more patients would be classified as having a poor gas exchange.

1 2 3 4 5 6

Next Section

Abbreviations
SARS-CoV2: Severe acute respiratory syndrome coronavirus 2; COVID-19: Coronavirus disease 2019; DLCO: Pulmonary diffusing capacity for carbon monoxide; LLN: Lower limit of normality; WHO: World Health Organization; BMI: Body mass index; FEV₁: Forced expiratory volume in 1 s; FVC: Forced vital capacity; FEF_25–75: Forced expiratory flow rate over the middle half of expiration; PEF: Peak expiratory flow; TLC: Total lung capacity; CVD: Cardiovascular disease.

Acknowledgements
Not applicable.

Funding
None.

Availability of data and materials
All data generated or analysed during this study are included in this article and/or its supplementary material files. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethics approval and consent to participate
Ethics approval was obtained from the Ethics Committee of the High Specialty Regional Hospital in Yucatan, Mexico (No. CONBIOETICA- 31-CEI-002-20170731). The Ethics Committee assigned this study as protocol number 2020-024. All patients signed a written informed consent to participate in the scheduled visit in compliance with the Helsinki declaration.

Consent for publication
Not applicable.

Table 1. Anthropometric and lung function data post mild COVID-19
	Males (n = 59)	Females (n = 87)
Age (yrs)	50 (13) [28–82]	50 (14) [25–83]
Weight (kg)	84.6 (15.0) [61.0–133.0]	70.0 (15.3) [42.0–118.0]
Height (cm)	168 (8) [150–182]	153 (8) [137–169]
Body mass index (kg/m²)	29.9 (4.0) [22.2–42.0]	30.0 (6.2) [16.6–44.1]
FVC (L)	3.97 (1.00) [1.20–6.73]	2.16 (0.66) [1.18–4.51]
FEV₁ (L)	3.26 (0.81) [1.06–5.18]	2.14 (0.59) [0.88–3.75]
FEV₁/FVC	0.82 (0.05) [0.64–0.91]	0.82 (0.06) [0.61–0.95]
PEF (L/s)	9.26 (2.15) [3.29–14.91]	5.75 (1.73) [2.07–9.70]
DLCO (mL min⁻¹ mmHg⁻¹)	26.6 (7.7) [7.6–53.6]	18.7 (4.8) [7.3–30.8]
VA (L)	6.03 (1.43) [1.88–9.53]	4.36 (0.89) [1.92–6.25
KCO mL min⁻¹ mmHg⁻¹ L⁻¹	4.44 (0.91) [2.46–6.84]	4.31 (0.84) [1.71–7.04]
Altitude of testing (m)	7	7

Data are showed as mean (S.D.). Brackets represent ranges

Table 2. Percent predicted for several lung function indices
	Males (n = 59)	Females (n = 87)
% predicted FVC	91 (16) [31–127]	93 (14) [61–131]
% predicted FEV₁	95 (17) [34–124]	90 (14) [56–120]
% predicted FEV₁/FVC	104 (6) [83–116]	101 (6) [78–116]
% predicted PEF	107 (24) [49–158]	105 (28) [53–180]
% predicted DLCO	81 (19) [25–136]	78 (17) [39–153]

Data are shown as mean (S.D.). Brackets represent ranges

Table 3. The number and percentage of males and females below the lower limit of normal (< 5th percentile)
	Males (n = 59)	Females (n = 87)	Total (n = 146)
FVC	15 (25%)	18 (21%)	33 (23%)
FEV₁	11 (19%)	14 (16%)	25 (17%)
FEV₁/FVC	1 (2%)	3 (3%)	4 (3%)
PEF	6 (10%)	9 (10%)	15 (10%)
DLCO	16 (27%)	28 (32%)	44 (30%)

A Fisher's Exact test was used to compare frequencies between males and females. No statistically significant differences were noted between males and females

Table 4. Variables that predict impaired DLCO in mild COVID-19 patients (n = 139 patients)
Variables	B	S.E	Wald	d_f	Sig	Odds ratio	95% CI for odds ratio lower–upper
Age	0.094 [0.053, 0.143]	0.02	17.0	1	0.000	1.10	1.05 to 1.15
Restriction (1 = yes, 0 = no)	2.49 [1.34, 3.79]	0.62	16.3	1	0.000	12.1	3.61 to 40.41
Blocked nose (yes = 1; no = 0)	− 2.29 [− 3.79, − 1.05]	0.69	11.0	1	0.001	0.10	0.03 to 0.39
Excessive sweating (yes = 1, no = 0)	− 2.51 [− 4.43, − 0.92]	0.89	8.0	1	0.005	0.08	0.01 to 0.46
Constant	− 5.91 [− 8.79, − 3.54]	1.33	19.8	1	0.000	0.003	0.0002 to 0.0365

Brackets represent the 95% CI. The model was accurate 81% of the time, i.e., correctly classifying 81% of the cases [85% (93/110) of the negative cases (true negative rate, specificity) and 69% (20/29) of the positive cases (true positive rate, sensitivity). The false-positive rate was 46%, and the false-negative rate was 9%. The positive predictive value was 54%, and the negative predictive value was 91%. The area under the curve was 0.89 (Standard error = 0.03) 95% CI = 0.83 to 0.94). Seven outliers (5%) were removed (Standardised residuals ≥ 3.0)

Table 5. Variables that predict the proportion of patients who had previous COVID-19 disease with an impaired DLCO at follow-up using the mean data from 21 previous studies
Model	Unstandardized coefficient B	Coefficient S.E	Standardized coefficient Beta	t	Sig	95% confidence interval for B
Constant	− 33.47	21.08		− 1.59	0.12	− 76.5 to 9.52
History of mild vs severe COVID-19	20.55	5.26	0.49	3.91	< 0.001	0.97 to 1.03
Mean age of study (years old)	1.21	0.38	0.40	3.14	0.004	0.95 to 1.05
Criteria to define impaired DLCO	13.13	5.23	0.32	2.51	0.017	0.99 to 1.01
Mean number of days between receiving a diagnosis of COVID-19 and testing	− 0.12	0.68	− 0.23	− 1.76	0.088	0.94 to 1.06