Nocturnal Hypoxemia Is Associated with Sarcopenia in Patients with Chronic Obstructive Pulmonary Disease

Abstract

Rationale: Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide. Our previous studies have identified that nocturnal hypoxemia causes skeletal muscle loss (i.e., sarcopenia) in in vitro models of COPD. Objectives: We aimed to extend our preclinical mechanistic findings by analyzing a large sleep registry to determine whether nocturnal hypoxemia is associated with sarcopenia in patients with COPD. Methods: Sleep studies from patients with COPD (n = 479) and control subjects without COPD (n = 275) were analyzed. Patients with obstructive sleep apnea, as defined by apnea-hypopnea index ⩾ 5, were excluded. Pectoralis muscle cross-sectional area (PMcsa) was quantified using computed tomography scans performed within 1 year of the sleep study. We defined sarcopenia as less than the lowest 20% residuals for PMcsa of control subjects, which was adjusted for age and body mass index (BMI) and stratified by sex. Youden's optimal cut-point criteria were used to predict sarcopenia based on mean oxygen saturation during sleep. Additional measures of nocturnal hypoxemia were analyzed. The pectoralis muscle index (PMI) was defined as PMcsa normalized to BMI. Results: On average, males with COPD had a 16.6% lower PMI than control males (1.41 ± 0.44 vs. 1.69 ± 0.56 cm²/BMI; P < 0.001), whereas females with COPD had a 9.4% lower PMI than control females (0.96 ± 0.27 vs. 1.06 ± 0.33 cm²/BMI; P < 0.001). Males with COPD with nocturnal hypoxemia had a 9.5% decrease in PMI versus COPD with normal O₂ (1.33 ± 0.39 vs. 1.47 ± 0.46 cm²/BMI; P < 0.05) and a 23.6% decrease compared with control subjects (1.33 ± 0.39 vs. 1.74 ± 0.56 cm²/BMI; P < 0.001). Females with COPD with nocturnal hypoxemia had an 11.2% decrease versus COPD with normal O₂ (0.87 ± 0.26 vs. 0.98 ± 0.28 cm²/BMI; P < 0.05) and a 17.9% decrease compared with control subjects (0.87 ± 0.26 vs. 1.06 ± 0.33 cm²/BMI; P < 0.001). These findings were largely replicated using multiple measures of nocturnal hypoxemia. Conclusions: We defined sarcopenia in the pectoralis muscle using residuals that take into account age, BMI, and sex. We found that patients with COPD have a lower PMI than patients without COPD and that nocturnal hypoxemia was associated with an additional decrease in the PMI of patients with COPD. Additional prospective analyses are needed to determine a protective threshold of oxygen saturation to prevent or reverse sarcopenia due to nocturnal hypoxemia in COPD.

Keywords: COPD; cachexia; nocturnal hypoxemia; sarcopenia; skeletal muscle wasting.

Figure 1.

Flowsheet of patients with COPD from the Cleveland Clinic sleep registry. AHI = apnea–hypopnea index; COPD = chronic obstructive pulmonary disease; CT = computed tomography.

Figure 2.

Determining a threshold for hypoxemia to identify sarcopenia based on pectoralis muscle in patients with chronic obstructive pulmonary disease (COPD). 1) Our methods to define sarcopenia in a control population using pectoralis muscle. 2) Next, a hypoxemia threshold that predicts sarcopenia in patients with COPD was determined. 3) Our main findings are presented. Nocturnal hypoxemia was associated with lower pectoralis muscle index in both male and female patients with COPD compared with control subjects. Patients with COPD with nocturnal hypoxemia demonstrate further reductions in pectoralis muscle index. Collectively, these findings suggest an additive interaction between COPD and nocturnal hypoxemia in causing skeletal muscle loss. Image made with Biorender.com. CT = computed tomography.

Figure 3.

Pectoralis muscle cross-sectional area (PMcsa) increases with body mass index (BMI) in males with chronic obstructive pulmonary disease (COPD) and control/females with COPD. Univariate linear regression analysis was performed with PMcsa as the dependent variable and BMI as the independent variable. Model fit was determined by R². (A) Male data. (B) Female data. Regression lines that are not significant are not depicted. Blue/aqua indicates males; pink/purple indicates females.

Figure 4.

Pectoralis muscle index (PMI) is lower in both male and female patients with chronic obstructive pulmonary disease (COPD) compared with their control subjects. PMI was calculated from the PMcsa normalized to BMI. (A) A representative computed tomography scan used to determine pectoralis muscle area. (B) Lower PMI was noted in patients with COPD. Blue indicates male, pink indicates female. The box and whisker plot represents the median and 25–75% interquartile range. Asterisks indicate **P < 0.01 and ***P < 0.001, based on analysis of variance followed by Benjamini-Hochberg post hoc analysis. BMI = body mass index; Ctrl = control.

Figure 5.

Pectoralis muscle index (PMI) decreases based on age in control males with COPD and control females. Univariate linear regression analysis was performed with PMI as the dependent variable and age as the independent variable. Model fit was determined by R². Blue/aqua indicates males; pink/purple indicates females. (A) Male data. (B) Female data. Regression lines that are not significant are not depicted. BMI = body mass index; COPD = chronic obstructive pulmonary disease.

Figure 6.

Pectoralis muscle index (PMI) is lower in patients with chronic obstructive pulmonary disease (COPD) compared with control subjects and further reduced in the presence of nocturnal hypoxemia. Comparison of sex-stratified PMI in control subjects, patients with COPD with normoxemia, and patients with COPD with hypoxemia, as measured by mean Sa_O2 (male cutoff: 93%; female cutoff: 90%). Blue indicates male, pink indicates female. The box and whisker plot represents the median and 25–75% interquartile range. Asterisks indicate *P < 0.05 and ***P < 0.001 based on Benjamini-Hochberg post hoc analysis. Ctrl = control; nl = normal; PMI = pectoralis muscle cross-sectional area/body mass index; Sa_O2 = arterial oxygen saturation.