The Early Chronic Obstructive Pulmonary Disease (ECOPD) study (ChiCTR1900024643) is a multicenter cohort study that initially enrolled about 2000 subjects.11 The subjects in the current analysis were the first 1352 subjects who returned for a second visit approximately 1 year after their initial visit (up to December 2021). The key inclusion criteria were aged of 40 to 80 years, completion of the questionnaire interview, and complete quality-control spirometry and CT. The key exclusion criteria were acute exacerbation occurring within 4 weeks before study participation; chest, abdominal, or eye surgery within 3 months prior to screening; heart damage within 3 months before screening; and active pulmonary tuberculosis or malignant tumors. Finally, this study included 1352 subjects for baseline analysis (685 with spirometry-defined COPD and 667 with normal spirometry). Subjects included in the baseline analysis were followed up for 1 year until December 2021.

Written informed consent was obtained from all subjects prior to screening. The ECOPD study was approved by the Ethics Committee of The First Affiliated Hospital of Guangzhou Medical University (No.2018-53). This present study was in line with the principles of the Declaration of Helsinki.

High-resolution CT was performed following the previously published protocol,11 using a multidetector-row CT scanner (Siemens Definition AS Plus 128-slicer CT system; Siemens Healthineers, Erlangen, Germany and uCT 760 128-slicers CT system; United Imaging, Shanghai, China). The Parameters and acquisition details of CT images have been published previously.11 The CT images were quantitatively assessed using the Chest Imaging Platform (www.chestimagingplatform.org) on the 3D Slicer 4.11 software (www.slicer.org).12 The PMA was quantified in inspiratory CT at the level of the aortic arch by the software using predefined −50 and 90 Hounsfield unit (HU) attenuation ranges.10 The left and right pectoralis major muscles and pectoralis minor muscles were identified, and their areas (cm2) were measured. The PMA was presented as the aggregate area of the left and right pectoralis major and minor muscles assessed in the axial plane. Emphysema was defined by the percentage of low-attenuation areas below −950 HU in full-inspiratory CT (inspiratory LAA-950). Air trapping was defined by the percentage of low-attenuation areas below −856 HU in expiratory CT (expiratory LAA-856).13

A questionnaire interview was performed by well-trained staff. The standard respiratory epidemiological questionnaire in this study was modified from a COPD epidemiological study in China.14 The smoking status was classified as never smoker, ex-smoker, and current smoker. Never smoker was defined as having smoked <100 cigarettes over a lifetime. Ex-smoker was defined as having smoked >100 cigarettes but having not smoked within the last 6 months at baseline. Current smoker was defined as smoking at baseline. The smoking index was defined as smoking years multiplied by the daily number of packs of cigarettes. Biomass exposure was defined as cooking or heating using biomass such as crop residues and wood for >1 year.15 Occupational history of dust/gasses/fumes was defined as engagement in occupational exposure to dust/gasses/fumes for >1 year over a lifetime. Family history of respiratory diseases was defined as having parents, siblings, and/or children diagnosed with respiratory diseases by doctors (chronic bronchitis, emphysema, asthma, COPD, cor pulmonale, bronchiectasis, lung tumor, interstitial lung disease, or obstructive sleep apnea hypopnea syndrome). Respiratory symptoms were assessed using the modified British Medical Research Council (mMRC) dyspnea scale and COPD Assessment Test (CAT) score.16 Acute respiratory events or exacerbations were defined as new onset or aggravation of at least two of the following five symptoms: cough, sputum production, purulent sputum, wheezing, and dyspnea persisting for at least 48 h, excluding left and right heart insufficiency, pulmonary embolism, pneumothorax, pleural effusion, and arrhythmia.17,18 The severity of acute exacerbations of COPD was assessed and recorded by well-trained staff according to the following categories. Mild exacerbations were defined as those resulting in domiciliary management with COPD medications alone. Moderate exacerbations were defined as those resulting in outpatient or emergency department visits and the need for COPD medication. Severe exacerbations were defined as those resulting in hospitalization.

Spirometry was performed following the normative operation methods and standard quality-control principles recommended by the American Thoracic Society and Europe Respiratory Society.19,20 We used a MasterScreen Pneumo PC spirometer (CareFusion, Yorba Linda, CA, USA) for spirometry measurements. Each subject needed to fit at least three acceptable measurement values and two repeatable measurement values (i.e., maximum and sub-maximum of forced vital capacity [FVC] and forced expiratory volume in 1 s [FEV1] within 150 mL or 5%). The post-bronchodilator test was performed 20 min after inhalation of a bronchodilator (400 μg salbutamol [Ventolin; GlaxoSmithKline, UK]). We defined spirometry-defined COPD and airflow limitation severity according to the GOLD Report.1 Subjects with a post-bronchodilator FEV1/FVC of ≥0.70 and FEV1% predicted of ≥80% were defined as having normal spirometry, and subjects with a post-bronchodilator FEV1/FVC of <0.70 were defined as having spirometry-defined COPD. We used the reference values of the European Coal and Steel Community 1993 for the FEV1% predicted, adjusted using conversion factors for the Chinese population (men, 0.95 and women, 0.93).21,22 The airflow limitation severity was classified as GOLD stages 1 to 4 using the post-bronchodilator FEV1% predicted (≥ 80%, ≥ 50% to < 80%, ≥ 30% to < 50%, and < 30%, respectively).1

Continuous variable with a normal distribution was expressed as mean ± standard deviation, and continuous variable with a non-normal distribution was expressed as median (interquartile range). Discontinuous variable was expressed as n (%). We evaluated the differences between the two groups using a two-sample t-test, chi-square test, or Fisher's exact test as required. Multivariate linear regression analyses were used to assess the PMA and airflow limitation severity, respiratory symptoms, lung function, emphysema, and air trapping after adjusting for confounding factors (i.e., age, sex, BMI, smoking status, smoking index, biomass exposure, family history of respiratory disease, and occupational history of dusts/gasses/fumes). Meantime, the GOLD ABE classification was used to assess the association of PMA and disease severity as a sensitivity analysis.1 Multivariate linear regression analysis was used to evaluate the PMA and the annual decline in lung function after adjustment. Cox proportional hazards analysis and Poisson regression analysis were used to evaluate the PMA and acute respiratory events/exacerbations adjusted by confounding factors. We repeated the above analyses for the pectoralis major muscle area and pectoralis minor muscle area. Statistical analyses were performed using SPSS 25.0 software (IBM Corp., Armonk, NY, USA). Two-sided P values of <0.05 indicated statistical significance.

Table 2 shows the association of the PMA with airflow limitation severity. The PMA was monotonically lower with progressive airflow limitation severity after adjusting for confounding factors (vs. normal spirometry; GOLD 1: β=−1.27, 95% confidence interval [CI], −2.41 to −0.14, P=0.028; GOLD 2: β=−2.29, 95%CI, −3.48 to −1.11, P<0.001; GOLD 3: β=−4.88, 95%CI, −6.94 to −2.83, P<0.001; GOLD 4: β=−6.47, 95%CI, −11.61 to −1.33, P=0.014). Age, sex, BMI, and smoking index were respectively associated with the PMA (all P<0.05). A similarly negative association was found in the pectoralis major muscle area and pectoralis minor muscle area, and in the GOLD ABE classification (Supplementary Table 1).

Table 3 shows the associations of the PMA with respiratory symptoms and lung function. Among all subjects, the PMA was negatively associated with the mMRC score (β=−0.005, 95%CI, −0.009 to −0.001, P=0.026) and CAT score (β=−0.06, 95%CI, −0.09 to −0.02, P=0.001) after adjusting for confounding factors. The PMA was positively associated with pre-bronchodilator or post-bronchodilator lung function including the FEV1, FEV1% predicted, FVC, and FEV1/FVC (all P<0.05). A similarly positive association was found for the pectoralis major muscle area and pectoralis minor muscle area.

Table 3 shows the association of the PMA with CT imaging. The PMA was negatively associated with emphysema (inspiratory LAA-950: β=−0.07, 95%CI, −0.10 to −0.04, P<0.001) and airflow trapping (expiratory LAA-856: β=−0.24, 95%CI, −0.35 to −0.12, P<0.001) in CT-quantitated imaging adjustment for confounding variables. The addition of the adjustment for confounding factors did not substantively alter the association between emphysema or air trapping and the pectoralis major or minor muscle area.

Of 685 subjects with spirometry-defined COPD, 593 had 1-year of spirometry measurements (follow-up rate, 87%) and 618 had 1-year of complete exacerbation assessment data (follow-up rate, 90%). Of 667 subjects with normal spirometry, 563 had 1-year of follow-up spirometry measurements (follow-up rate, 84%) and 622 had 1-year follow-up acute respiratory event assessment data (follow-up rate, 93%). The results are displayed in Table 4. A multivariable linear regression analysis model was used to explore the relationship between the PMA and the annual decline in lung function. The PMA was associated with a reduced annual decline of post-bronchodilator FEV1% predicted (β=0.022, 95%CI, 0.008 to 0.035, P=0.002) after adjusting for confounding factors. After the 1-year follow-up, the PMA was not associated with the annual rate of exacerbations (Table 5).