Association Between Insomnia With or Without Comorbid Sleep Apnea and Endothelial Dysfunction in COPD Authors: Tetsuro Maeda, Chia-Ying Chiu, Gregory A. Payne, Vivian Y. Lin, Dongqi Xing, S. Justin Thomas, J. Michael Wells Published: 2026 To the Editor: COPD is characterized by systemic inflammation, which leads to vascular endothelial dysfunction, promoting development of cardiovascular disease (CVD). COPD often coexists with sleep-related disorders. Sleep apnea (SA) and insomnia are both associated with impaired endothelial function, with comorbid insomnia and SA (COMISA) shown to have a greater impact on CVD risk than either condition alone.1,2 In patients with COPD, OSA increases CVD risk,3 but whether insomnia or COMISA further heightens CVD risk is unknown. We hypothesized that endothelial function would be impaired in individuals with COPD and COMISA compared with COPD with comorbid OSA, insomnia, or neither. Methods We conducted a single-center, cross-sectional pilot study in a subpopulation within our parent research registry which included people with COPD, people who currently smoke and formerly smoke without COPD, and people who do not smoke without COPD. The participants with COPD who agreed to participate in future studies were contacted to participate in the current substudy. The institutional review board of the University of Alabama at Birmingham approved this study (No. IRB-300009849-005) and its parent research registry (No. IRB-170301002), and all participants provided informed consent. The research registry visits included collection of demographics, anthropometric data, medical history, and prebronchodilator and postbronchodilator spirometry according to American Thoracic Society/European Respiratory Society standards.4 For this substudy, we collected SA-related history based on participant self-report and review of electronic medical records. Participants completed the STOP-BANG questionnaire (3-4 points indicated moderate risk of OSA and ≥ 5 points indicated high risk of OSA),5 Insomnia Severity Index (ISI) (8-14 points indicated subthreshold insomnia and ≥ 15 points indicated clinically significant insomnia),6 and Pittsburgh Sleep Quality Index (PSQI) (> 5 points indicated poor sleep quality)7 and underwent digital thermal monitoring using the VENDYS II (Endothelix Inc). The device measures fingertip temperature change (temperature rebound) with arm arterial occlusion and calculates the vascular reactivity index (VRI) (maximum distance from the observed temperature rebound in the occluded arm and the zero reactivity curve), where VRI score > 2 is considered normal, 1 to 2 is intermediately reduced, and < 1 is low reactivity or endothelial function.8,9 We performed statistical analyses by the 3 VRI groups, using the χ2 test for categorical variables and Kruskal-Wallis test for continuous variables. Pearson correlation analyses were also performed to assess the correlation between sleep-related parameters (STOP-BANG, ISI, and PSQI) and endothelial function (VRI). Multiple linear regression analyses were performed to assess the impact of OSA and insomnia on the VRI, with STOP-BANG scores ≥ 5 and ISI scores ≥ 15 as surrogate measures, respectively. Two-tailed P < .05 was considered statistically significant. SAS software version 9.4 (SAS Institute Inc., Cary, North Carolina) was used. Results Twenty-four participants were included. The median age was 65 years, 12 (50%) were male, the median postbronchodilator FEV1 was 56% predicted (interquartile range [IQR], 49-63), and the FEV1/FVC was 0.56 (IQR, 0.44-0.66). Eight patients (33%) reported having had sleep study with diagnosis of SA, but further details were unavailable. The median STOP-BANG, ISI, and PSQI scores were 3.0 (IQR, 2.5-4.5), 10.5 (IQR, 4.5-19.5), and 10.5 (IQR, 7.5-12.5), respectively. Between the 3 groups by the VRI (4 participants with scores > 2, 8 participants with scores 1-2, and 12 participants with scores < 1), the prevalence of hypertension (P = .032) and PSQI score (P = .014) was significantly different, but lung function, smoking status, SA-related parameters (STOP-BANG score, reported diagnosis of SA and CPAP use), and ISI score were similar (Table 1). Although the STOP-BANG score was not correlated with the VRI (r = −0.23; P = .29), the ISI and PSQI scores were inversely correlated (r = −0.62, P = .001 and r = −0.56; P = .005, respectively). Multiple linear regression analysis revealed a significant decrease of 0.94 ± 0.42 units in the VRI for participants with ISI scores ≥ 15 vs < 15 (P = .036). In contrast, STOP-BANG scores ≥ 5 were not significantly associated with the VRI (β ± SE, −0.070 ± 0.46; P = .88) (Table 2).Table 1Demographics and Sleep-Related Data in Participants With COPD by VRI GroupsVariableVRI Score > 2 (n = 4)VRI Score 1-2 (n = 8)VRI Score < 1 (n = 12)Total (N = 24)P ValueDemographics Male sex3 (75.0)2 (25.0)7 (58.3)12 (50.0).19 Age, y73.0 (64.5-76.0)64.0 (60.5-70.5)64.0 (61.0-70.5)65.0 (60.5-71.0).44 Height, m1.66 (1.59-1.75)1.65 (1.60-1.73)1.73 (1.65-1.78)1.71 (1.63-1.75).34 Weight, kg81.6 (75.1-95.9)80.7 (75.7-83.9)78.2 (65.1-89.6)79.6 (70.8-86.2).82 Neck size, cm43.2 (36.3-49.5)36.6 (35.6-41.4)36.8 (35.6-40.9)36.8 (35.6-41.9).50 Postbronchodilator FEV1 % predicted49.0 (42.5-58.0)57.5 (48.5-77.0)57.5 (53.5-63.0)56.0 (48.5-63.0).40 Postbronchodilator FEV1/FVC0.5 (0.5-0.6)0.5 (0.4-0.7)0.6 (0.4-0.7)0.6 (0.4-0.7).94 Smoking status.52 Never0 (0.0)1 (12.5)0 (0.0)1 (4.2) Former1 (25.0)4 (50.0)6 (50.0)11 (45.8) Current3 (75.0)3 (37.5)6 (50.0)12 (50.0) Smoking, pack-years14.8 (6.8-68.0)27.5 (17.2-34.5)37.5 (22.3-52.5)30.0 (18.8-44.0).27 Exacerbation of COPD within 1 y0.0 (0.0-0.5)0.5 (0.0-1.0)0.0 (0.0-0.5)0.0 (0.0-1.0).58 Hypertension0 (0.0)6 (75.0)4 (33.3)10 (41.7).032 Any heart disease1 (25.0)2 (25.0)3 (25.0)6 (25.0)> .99Sleep-related data Reported diagnosis of SA2 (50.0)3 (37.5)3 (25.0)8 (33.3).63 CPAP use2 (50.0)1 (12.5)1 (8.3)4 (16.7).14 STOP-BANG3.5 (3.0-4.0)3.0 (2.5-4.5)3.5 (2.0-5.5)3.0 (2.5-4.5).97 ISI3.0 (1.5-8.0)11.5 (4.5-13.5)14.5 (8.5-22.5)10.5 (4.5-19.5).070 PSQI7.5 (4.0-10.5)7.5 (5.5-10.0)12.0 (10.5-16.0)10.5 (7.5-12.5).014For categorical variables, No. (%) and P values calculated by the χ2 test are reported. For continuous variables, median (interquartile range) and P values calculated by the Kruskal-Wallis test are reported. The bold font indicates statistically significant results. ISI = insomnia severity index; PSQI = Pittsburgh Sleep Quality Index; SA = sleep apnea; VRI = vascular reactivity index.Table 2Associations Between OSA (Represented by STOP-BANG Score ≥ 5) and Insomnia (ISI Score ≥ 15) and the Vascular Reactivity Index in COPDVariableCoefficientSEt ValueP ValueISI score ≥ 15−0.940.42−2.24.036STOP-BANG score ≥ 5−0.0700.46−0.15.88Multiple linear regression models were used. The bold font indicates statistically significant results. ISI = insomnia severity index. Table 1Demographics and Sleep-Related Data in Participants With COPD by VRI GroupsVariableVRI Score > 2 (n = 4)VRI Score 1-2 (n = 8)VRI Score < 1 (n = 12)Total (N = 24)P ValueDemographics Male sex3 (75.0)2 (25.0)7 (58.3)12 (50.0).19 Age, y73.0 (64.5-76.0)64.0 (60.5-70.5)64.0 (61.0-70.5)65.0 (60.5-71.0).44 Height, m1.66 (1.59-1.75)1.65 (1.60-1.73)1.73 (1.65-1.78)1.71 (1.63-1.75).34 Weight, kg81.6 (75.1-95.9)80.7 (75.7-83.9)78.2 (65.1-89.6)79.6 (70.8-86.2).82 Neck size, cm43.2 (36.3-49.5)36.6 (35.6-41.4)36.8 (35.6-40.9)36.8 (35.6-41.9).50 Postbronchodilator FEV1 % predicted49.0 (42.5-58.0)57.5 (48.5-77.0)57.5 (53.5-63.0)56.0 (48.5-63.0).40 Postbronchodilator FEV1/FVC0.5 (0.5-0.6)0.5 (0.4-0.7)0.6 (0.4-0.7)0.6 (0.4-0.7).94 Smoking status.52 Never0 (0.0)1 (12.5)0 (0.0)1 (4.2) Former1 (25.0)4 (50.0)6 (50.0)11 (45.8) Current3 (75.0)3 (37.5)6 (50.0)12 (50.0) Smoking, pack-years14.8 (6.8-68.0)27.5 (17.2-34.5)37.5 (22.3-52.5)30.0 (18.8-44.0).27 Exacerbation of COPD within 1 y0.0 (0.0-0.5)0.5 (0.0-1.0)0.0 (0.0-0.5)0.0 (0.0-1.0).58 Hypertension0 (0.0)6 (75.0)4 (33.3)10 (41.7).032 Any heart disease1 (25.0)2 (25.0)3 (25.0)6 (25.0)> .99Sleep-related data Reported diagnosis of SA2 (50.0)3 (37.5)3 (25.0)8 (33.3).63 CPAP use2 (50.0)1 (12.5)1 (8.3)4 (16.7).14 STOP-BANG3.5 (3.0-4.0)3.0 (2.5-4.5)3.5 (2.0-5.5)3.0 (2.5-4.5).97 ISI3.0 (1.5-8.0)11.5 (4.5-13.5)14.5 (8.5-22.5)10.5 (4.5-19.5).070 PSQI7.5 (4.0-10.5)7.5 (5.5-10.0)12.0 (10.5-16.0)10.5 (7.5-12.5).014For categorical variables, No. (%) and P values calculated by the χ2 test are reported. For continuous variables, median (interquartile range) and P values calculated by the Kruskal-Wallis test are reported. The bold font indicates statistically significant results. ISI = insomnia severity index; PSQI = Pittsburgh Sleep Quality Index; SA = sleep apnea; VRI = vascular reactivity index. Demographics and Sleep-Related Data in Participants With COPD by VRI Groups For categorical variables, No. (%) and P values calculated by the χ2 test are reported. For continuous variables, median (interquartile range) and P values calculated by the Kruskal-Wallis test are reported. The bold font indicates statistically significant results. ISI = insomnia severity index; PSQI = Pittsburgh Sleep Quality Index; SA = sleep apnea; VRI = vascular reactivity index. Table 2Associations Between OSA (Represented by STOP-BANG Score ≥ 5) and Insomnia (ISI Score ≥ 15) and the Vascular Reactivity Index in COPDVariableCoefficientSEt ValueP ValueISI score ≥ 15−0.940.42−2.24.036STOP-BANG score ≥ 5−0.0700.46−0.15.88Multiple linear regression models were used. The bold font indicates statistically significant results. ISI = insomnia severity index. Associations Between OSA (Represented by STOP-BANG Score ≥ 5) and Insomnia (ISI Score ≥ 15) and the Vascular Reactivity Index in COPD Multiple linear regression models were used. The bold font indicates statistically significant results. ISI = insomnia severity index. Discussion This pilot study provides clinically important data regarding sleep-related disorders in patients with COPD. In this cohort, overall sleep quality was poor as evidenced by high median PSQI values of 10.5, much higher than general cutoff values of 5. Clinically significant insomnia (ISI score ≥ 15 compared with < 15) and poor sleep quality (higher PSQI scores) were associated with lower endothelial function (VRI). In previous studies, a lower VRI score was associated with the Framingham Risk Score and coronary calcium score8 and significant coronary artery disease in multidetector coronary CT angiography.9 This suggests that insomnia and poor sleep quality may be an important risk factor for CVD in patients with COPD. The association between SA and endothelial function appears more complex. STOP-BANG questionnaire scores did not show significant association with the VRI, which conflicts with previous studies showing association of OSA and lower endothelial function.10 This may be explained by using patient-reported measures as the variables rather than sleep study-confirmed diagnosis of SA (eg, STOP-BANG questionnaire scores include multiple demographic factors which complicate analyses6). Also, although only a small proportion of this cohort reported its use, CPAP therapy has been previously demonstrated to improve endothelial function in patients with OSA, potentially influencing our results.10 The limitations of this study include its small sample size precluding adjustment for relevant covariables such as sex, age, medication use (eg, statins), cross-sectional design precluding adjudication of actual CVD events, lack of objective sleep study data or possible confounding factors for insomnia (eg, systemic steroid use), and lack of data to corroborate with other CVD risk scores (eg, cholesterol levels). Future studies would better inform development of clinical risk assessment tools for patients with COPD and sleep-related comorbidities. Insomnia (with or without comorbid OSA) may increase risk of endothelial dysfunction in patients with COPD. Future studies with more in-depth sleep-related data should be beneficial to inform CVD risk prediction in this complex patient population. Funding/Support This study was supported by the 10.13039/100000002National Institutes of Health [Grant R01HL148215 to J. M. W.] and salary support to C. C. from the UAB Comprehensive Cardiovascular Center 2024 Pilot Grant awarded to T. M. Financial/Nonfinancial Disclosures None declared.