Cognitive performance is lower among individuals with overlap syndrome than in individuals with COPD or obstructive sleep apnea alone: association with carotid artery stiffness

Abstract

Chronic obstructive pulmonary disease (COPD) and obstructive sleep apnea (OSA) are both independently associated with increased cardiovascular disease (CVD) risk and impaired cognitive function. It is unknown if individuals with both COPD and OSA (i.e., overlap syndrome) have greater common carotid artery (CCA) stiffness, an independent predictor of CVD risk, and lower cognitive performance than either COPD or OSA alone. Elevated CCA stiffness is associated with cognitive impairment in former smokers with and without COPD in past studies. We compared CCA stiffness and cognitive performance between former smokers with overlap syndrome, COPD only, OSA only and former smoker controls using analysis of covariance (ANCOVA) tests to adjust for age, sex, body mass index (BMI), pack years, and postbronchodilator FEV₁/FVC. We also examined the association between CCA stiffness and cognitive performance among each group separately. Individuals with overlap syndrome (n = 12) had greater CCA β-stiffness index (P = 0.015) and lower executive function-processing speed (P = 0.019) than individuals with COPD alone (n = 47), OSA alone (n = 9), and former smoker controls (n = 21), differences that remained significant after adjusting for age, BMI, sex, pack years, and FEV₁/FVC. Higher CCA β-stiffness index was associated with lower executive function-processing speed in individuals with overlap syndrome (r = −0.58, P = 0.047). These data suggest that CCA stiffness is greater and cognitive performance is lower among individuals with overlap syndrome compared with individuals with COPD or OSA alone and that CCA stiffening may be an underlying mechanism contributing to the lower cognitive performance observed in patients with overlap syndrome.

NEW & NOTEWORTHY Previous studies have demonstrated greater carotid artery stiffness and lower cognitive function among individuals with COPD alone and OSA alone. However, the present study is the first to demonstrate that individuals that have both COPD and OSA (i.e., overlap syndrome) have greater carotid artery stiffness and lower executive function-processing speed than individuals with either disorder alone. Furthermore, among individuals with overlap syndrome greater carotid artery stiffness is associated with lower executive function-processing speed.

INTRODUCTION

Chronic obstructive pulmonary disease (COPD) and obstructive sleep apnea (OSA) are two of the most common causes of morbidity in adults, affecting ∼10% and 30% of adults in the US, respectively (1–3). Although primarily respiratory disorders, both COPD and OSA have detrimental effects on organ systems outside of the respiratory tract and are independently associated with elevated cardiovascular disease (CVD) risk (4, 5) and lower cognitive function (6–9). Indeed, individuals with COPD alone have a threefold greater risk of ischemic heart disease, stroke, and sudden cardiac death (10). Furthermore, individuals with OSA alone have a two- to fourfold greater risk of coronary artery disease, heart failure, and stroke (11, 12). The prevalence of cognitive impairment among individuals with COPD is high, affecting 40%–70% of all individuals in this group (9, 13). The prevalence of cognitive impairment in individuals with OSA is also elevated and has been estimated to be present in up to 30% of all individuals between the ages of 30–75 yr with a clinical diagnosis of OSA (2, 14, 15). Importantly, up to 65% of all individuals with COPD also have a clinical diagnosis of OSA. This combination is termed “overlap syndrome” (16). Individuals with overlap syndrome have an increased prevalence of COPD exacerbations, poorer quality of life, and increased CVD morbidity compared with individuals with either disorder alone (3, 16). Common carotid artery (CCA) stiffness is a strong predictor of CVD events in adults, including incident stroke (17), and is significantly greater in individuals with COPD alone (18) and OSA alone compared with controls (19). However, CCA stiffness remains to be elucidated among individuals with overlap syndrome.

Greater CCA stiffness is itself associated with cognitive impairment, even in individuals without CVD (20–22). The common carotid arteries are large, elastic vessels that dampen the pulsatile blood flow from the left ventricle during systole, before it reaches the high-flow, low-resistance vessels of the brain. With aging and disease, however, the CCA stiffens, resulting in increased pulsatile pressure and flow into the cerebral microcirculation (22). Increased pulsatile energy in the cerebral microcirculation is associated with the development of regions of ischemic damage in the brain that likely contribute to the development of cognitive impairment (22–26), particularly reductions in executive functioning and processing speed given susceptibility of subcortical regions of the brain to ischemia. Consistent with this, we developed a study to examine the link between changes in vascular physiology and cognition among former smokers with and without COPD (27). We recently demonstrated that CCA stiffness was associated with lower executive function-processing speed in these former smokers independent of sex, smoking history, airflow limitation and estimated pre-morbid IQ (27). However, it is not known if cognitive function is lower in individuals with overlap syndrome than in COPD or OSA alone and if greater CCA stiffness contributes to the lower cognitive performance observed in this group.

In the present study, we aimed to determine if individuals with overlap syndrome have greater CCA stiffness and lower cognitive function than individuals with either COPD or OSA alone. Given the heightened CVD risk observed in individuals with overlap syndrome, we hypothesized that 1) individuals with overlap syndrome would have greater CCA stiffness and reduced cognitive performance compared with either COPD or OSA alone and 2) greater CCA stiffness would be associated with reduced cognitive performance among all groups but would be most pronounced among individuals with overlap syndrome.

METHODS

Participants

Former smokers were recruited from the COPDGene cohort at National Jewish Health in Denver, Colorado to participate in a separate study (“Cardiopulmonary Mechanisms Affecting Cognition in COPD”; NIH HL095658) with new informed consent provided before participating in the present study. Current smokers were excluded to avoid the acute effects of cigarette smoking on vascular function. Smoking status was confirmed by urine cotinine assays. Participants were required to have at least an 8th grade education, speak English at the level of a native speaker, and have normal or corrected hearing and vision. Participants were excluded if they had a previous history of the following: diagnosis with a cognitive disorder; neurological disorder (e.g., stroke, movement disorder); traumatic brain injury with loss of consciousness >10 min; life history of major psychiatric disorder (e.g., schizophrenia, bipolar disorder, substance use disorder other than tobacco use); change in treatment for depression or anxiety in the previous 3 mo; major medical condition other than COPD, asthma, or OSA (e.g., renal failure, active cancer, previous radiation treatment); arrhythmia; left sided heart failure; and COPD exacerbation within the past month that required a physician or ER visit and/or treatment with antibiotics or oral corticosteroids.

One hundred and four former smokers were enrolled. Three of the 104 subjects who attended the first study visit were withdrawn from further participation: two for clinically significant hypertension at the time of their initial study visit and one for a medication change after visit 1. One hundred and one participants completed two study visits. Twelve participants were excluded for the following reasons: three had current nicotine use identified via urine cotinine; three classified as GOLD stage 0, but had evidence of significant emphysema and air-trapping on their past COPDGene Phase 1 lung computed tomography (CT) scan; three with exclusion criteria identified during study visits (1 active illicit substance use, 2 traumatic brain injury), one outlier with low MMSE score and low estimated IQ at a young age, and two whose CCA ultrasound scans were not analyzable due to large plaques. The current analysis included 89 former smokers including 12 with overlap syndrome, 47 with COPD only, 9 with OSA only, and 21 former smoker controls without COPD or OSA.

Procedures and Measures

Participants completed study visits on two separate occasions that occurred within 3 wk of each other (median 5 days). On the first visit, participants completed the following measurements: self-reported medical history, symptom questionnaires, neuropsychological assessment, basic vitals, pre- and postalbuterol spirometry and a 6-min walk test (6MWT). Before the second visit, participants fasted for 8 h, held antihypertensive medications for 12 h or PDE inhibitors for 72 h. Participants then completed urine cotinine testing to confirm smoking abstinence and high-resolution ultrasonography measurements to determine CCA stiffness. Between the first and second visits participants also completed two consecutive nights of nocturnal pulse oximetry recordings. OSA status was determined by self-report of a previous clinical diagnosis with OSA via patient health history interview question.

Pre- and Postbronchodilator spirometry.

Spirometry was performed with the EasyOne spirometer (ndd, Medical Technologies Inc., Andover, MA), in accordance with American Thoracic Society guidelines (28), and was used to determine the presence or absence of COPD. After baseline FVC testing, two puffs of albuterol (180 µg) were administered from a metered dose inhaler with a spacer. After 20 min, the participants were prompted to perform three additional acceptable FVC maneuvers. Results of questionable quality (rating of C or below) were reviewed by a pulmonologist (BJM). For 3 of the 89 participants, previous spirometry results from the individuals’ Phase 1 COPDGene visit were entered due to suboptimal quality of current testing.

Common carotid artery stiffness.

Participants underwent CCA ultrasound imaging with a 7–12 MHz linear transducer (LOGIQ 7, GE Healthcare) used to calculate CCA β-stiffness index from 30-s ultrasound images as previously described (29). Briefly, maximal systolic and minimum diastolic diameters were analyzed ∼2 centimeters from the carotid bulb (Vascular Analysis Tools Analyzer 5.5, Medical Imaging Applications, LLC, Coralville, IA). Systolic and diastolic BP measurements were obtained from a brachial sphygmomanometer (Dinamap, Johnson & Johnson). To determine the stiffness of the vessel wall without the influence of BP, CCA β-stiffness index was calculated as follows:β = In (brachial systolic BP/ brachial distolic BP / Δd) × (d), Δd represents the change in diameter (end systolic diameter minus end diastolic diameter) and d represents the end diastolic diameter.

Neuropsychological assessment.

The cognitive measures were chosen because they are frequently used in clinical and experimental neuropsychology research and have normative data for healthy controls. The cognitive tests were administered and scored by a trained research assistant using standard administration and scoring criteria under the supervision of a clinical neuropsychologist (KFH). Consistent with common practice in neuropsychological assessment (30–32), single word reading (i.e., WRAT4 Reading Standard Score was included as an indicator of premorbid intellectual functioning (31). Other individual neuropsychological test scores for cognitive functioning were grouped into four domains as follows:

1. Executive function-processing speed: Trail Making Test Part B, COWA, Stroop Color Word Test Interference, Wechsler Adult Intelligence Scale-IV (WAIS-IV) Coding.

2. Memory: California Verbal Learning Test-II Trials 1-5 immediate recall, Brief Visuospatial Memory Test- Revised Trials 1-3 total immediate recall; California Verbal Learning Test-II Delayed Free Recall, Brief Visuospatial Memory Test- Revised Trials 1-3 Delayed Free Recall

3. Language: Boston Naming Test, Category Fluency

4. Visual-spatial skills: WAIS-IV Block Design, Hooper Visual Organization Test.

A description of each neuropsychological task is available in Table 1. Raw test scores were transformed into t scores that have a mean of 50 and standard deviation of 10, using previously published age-referenced normative data for healthy adults (27). Cognitive domain scores were calculated as the mean of the normatively derived t scores for all of the tests in that domain. Thus, a domain score of 50 indicates exactly average performance, with a score of 40, for example, indicating performance that is 1 standard deviation below the mean relative to healthy age-matched peers. The executive function and processing speed cognitive domain scores were prorated for three individuals due to missing data and the memory domain score was prorated for one individual due to missing data.

Table 1.

Description of neuropsychological tasks by cognitive domain with references to age-adjusted normative data used for scoring

	Assesses
Executive Functioning-Processing Speed Domain
Trail Making Test Part B (33)	This test measures cognitive flexibility and sequencing. It involves connecting randomly arranged numbers and letters on a page.
Controlled Oral Word Association (34)	This test measures ability to initiate and maintain effort. It involves rapidly producing words beginning with specified letters.
Stroop Color Word Test-Interference (35)	This test measures ability to shift perceptual sets and inhibit overlearned responses. It involves stating the ink color in which an incongruent word is printed. For example, the word “red” is printed in green ink.
WAIS-IV Coding (36)	This test measures psychomotor speed and visual-motor coordination. It involves matching symbols with numbers according to a key at the top of the page and thus, includes an executive functioning component.
Memory Domain
California Verbal Learning Test-II (37) Trials 1-5 Immediate Recall	Immediate recall of a list of words.
Brief Visuospatial Memory Test-R (38) Trials 1-3 Total Immediate Recall	Immediate recall of a display of figures.
California Verbal Learning Test-II (37) Delayed Free Recall	Delayed recall for a list of words.
Brief Visuospatial Memory Test-R Delayed Free Recall	Delayed recall of a display of figures.
Language Domain
Boston Naming Test (39)	Ability to name pictures of objects.
Category Fluency (40)	Timed word generation task based on a category (i.e., animals).
Visuospatial Skill Domain
WAIS-IV Block Design (36)	Visuospatial construction and visual abstract problem solving using blocks.
Hooper Visual Organization Test (41)	Ability to visually integrate information into whole perceptions.

Nocturnal continuous pulse oximetry.

Participants wore a WristOx Pulse Oximeter (Respironics, Inc.) for two consecutive nights. Data collected from the WristOx Pulse Oximeter were analyzed using a semiautomated process. Aberrant data were identified by ProFox software and secondarily by manual review. Oxygen desaturation and sampling time were then automatically calculated by ProFox software. Oxygen desaturation index was the primary measure of nocturnal oxygen which reflects the average number of decreases in peripheral blood oxygen saturation ($\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$) of 4 or more during each hour of valid sampling time (averaged over 2 nights of recording). The percentage of nocturnal time spent with a $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ less than 88% was calculated for each participant. Participants were allowed to use supplemental oxygen and/or a continuous positive airway pressure (CPAP) as prescribed by their physician during the nocturnal continuous pulse oximetry testing.

Hospital anxiety and depression scale.

The Hospital Anxiety and Depression Scale (HADS) was used to determine severity of symptoms of depression and anxiety. The HADS has been used in many studies of COPD and OSA (42–44). The total score ranges from 0–21, with scores below 14 considered “normal.”

6MWT.

Participants completed a 6MWT according to ATS guidelines (45) while wearing a WristOx Pulse Oximeter to record heart rate (HR) and $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ at each minute of the walk. Briefly, baseline $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ and HR values were recorded by a trained research assistant administering the test. Participants were then instructed to walk up and down a hallway to determine the maximum distance that the participant could walk in 6 min. Participants were allowed to stop and rest during the test and the amount of time spent resting was recorded by the test administrator. $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ and HR were recorded at each minute of the walk. At the end of the 6 min, participants were instructed to stop walking and to sit in a chair to rest for at least 3 min. A trained research assistant later documented participant’s HR and $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ values at each minute of the walk and after each minute of recovery, using the recorded data and ProFox software. Participants were allowed to use supplemental oxygen during the walk, consistent with their typical treatment conditions. Therefore, 30 participants used supplemental oxygen during the test.

Statistical Analyses

All continuous data are reported as means ± standard deviation (SD). Demographic and clinical characteristics of the groups were compared using a one-way analysis of variance (ANOVA) for continuous variables or χ² for categorical variables. Statistical comparisons between CCA stiffness and the cognitive function domains with the four study groups were made using one-way ANOVA tests and analysis of covariance (ANCOVA) tests to adjust for age, sex, body mass index (BMI), pack years, and postbronchodilator FEV₁/FVC, a measurement of airflow limitation. Bonferroni post hoc analysis was used where significant main effects were observed to determine where differences occurred between groups. Bivariate correlations between CCA stiffness and the cognitive domains were performed. Data were analyzed with IBM SPSS software v. 25 and statistical significance was set at P < 0.05.

RESULTS

Participant Characteristics

Participant characteristics and medication use are described in Table 2. Participants were on average 69 yr old with average premorbid intellectual functioning (WRAT4 Reading Standard Score: 101 ± 7; means ± SD) and had minimal current symptoms of depression and anxiety (HADS-D: 4 ± 3; HADS-A: 3 ± 3; all means ± SD). All participants quit smoking at least 6 mo before the current study visits. One-way ANOVA tests demonstrated that the four groups differed by BMI (P < 0.001), HR (P = 0.01), depressive symptom severity (P = 0.01), smoking history (pack years; P = 0.03), and inhaled bronchodilator use (P < 0.001). Post hoc tests demonstrated that individuals with COPD only and former smoker controls had lower BMI than individuals with overlap syndrome (COPD only vs. overlap: P < 0.001; former smoker controls versus overlap: P = 0.016). Individuals with COPD only also demonstrated lower BMI than individuals with OSA only (P = 0.006). Former smoker controls had lower HR than individuals with COPD only (P = 0.016) and lower depressive symptom severity than individuals with overlap syndrome (P = 0.015). Former smoker controls had a significantly lower pack year smoking history than individuals with COPD only (P = 0.03). There were no differences in age, sex, BP, anxiety severity, BP medication use, cholesterol medication use or self-reported history of HTN, high cholesterol or type II diabetes between groups (all P values >0.05).

Table 2.

Participant characteristics and medication use

	Overlap Syndrome n = 12	COPD Only n = 47	OSA Only n = 9	Former Smokers Controls n = 21	P Value
Demographic and clinical variables
Age, yr	72 ± 7	70 ± 7	67 ± 8	67 ± 6	0.09
Race, n, % black	2 (17)	5 (11)	0 (0)	1 (5)	0.50
Sex, n, % male	10 (83)	23 (51)	7 (58)	13 (62)	0.24
BMI, kg/m2	33 ± 5	27 ± 5*	32 ± 6†	28 ± 3*	<0.001
Smoking history, pack years	59.3 ± 40.7	58.9 ± 33.6	44.4 ± 27.3	35.0 ± 22.4†	0.03
Brachial systolic BP, mmHg	126 ± 9	126 ± 12	129 ± 7	120 ± 13	0.18
Brachial diastolic BP, mmHg	74 ± 9	75 ± 8	74 ± 8	72 ± 9	0.71
Heart rate, beats/min	64 ± 12	66 ± 9	62 ± 13	58 ± 8†	0.03
Depressive symptom severity (HADS)	4.1 ± 3.2	3.2 ± 2.8	3.1 ± 1.9	1.3 ± 1.1*	0.01
Anxiety symptom severity (HADS-A)	4.6 ± 4.1	3.9 ± 3.2	4.7 ± 2.4	3.1 ± 2.2	0.49
Education, yr	14.3 ± 2.1	13.9 ± 2.2	14.8 ± 2.3	14.7 ± 2.4	0.46
Self-reported medical history
HTN, n, % yes	7 (58%)	20 (43%)	2 (22%)	5 (24%)	0.15
High cholesterol, n, % yes	7 (58%)	27 (57%)	3 (33%)	15 (71%)	0.28
Type II diabetes, n, % yes	3 (25%	3 (6%)	1 (11%)	2 (10%)	0.30
Medication use
ACE inhibitors, n, % yes	3 (25)	9 (19)	1 (11)	4 (19)	0.89
Calcium channel blockers, n, % yes	2 (17)	5 (11)	0 (0)	2 (10)	0.66
β Blockers, n, % yes	3 (25)	9 (19)	2 (22)	5 (24)	0.96
Statins, n, % yes	8 (67)	26 (55)	4 (44)	10 (48)	0.69
Diuretics, n, % yes	6 (50)	9 (19)	2 (22)	5 (24)	0.18
Inhaled ICS/LABA, n, % yes	4 (33)	24 (51)	0 (8)	0 (0)	<0.001
Inhaled LAMA, n, % yes	3 (25)	24 (53)	1 (8)	0 (0)	<0.001
Inhaled SABA, n, % yes	3 (25)	18 (38)	1 (11)	1 (5)	0.13

Data are means ± SD. BMI, body mass index; BP, blood pressure; HADS, hospital anxiety and depression scale; 6MWT, 6-minute walk test; ACE, angiotensin-converting enzyme; ICS/LABA, inhaled corticosteroid/long-acting β agonist combination therapy; LAMA, long-acting muscarinic antagonist; SABA, short acting-β agonist. One-way analysis of variance (ANOVA) tests were used for all continuous variables and χ² tests were used for categorical variables. *P < 0.05 vs. overlap syndrome group; †P < 0.05 vs. COPD group. n = number of subjects.

Pulmonary Outcomes

Pulmonary characteristics are described in Table 3. As expected, groups differed by airflow limitation (FEV₁/FVC and FEV₁; P < 0.001), 6MWT total distance walked (P < 0.001), daytime oxygen use (P = 0.003), nocturnal oxygen use (P < 0.001), and nocturnal CPAP use (P < 0.001). Individuals with COPD only had lower FEV₁/FVC (P = 0.001) and individuals with OSA only and former smoker controls had higher FEV₁/FVC than individuals with overlap syndrome (OSA versus overlap P = 0.005; former smoker controls versus overlap P < 0.001). Former smoker controls had a greater 6MWT total distance walked compared with individuals with overlap syndrome (P = 0.005) and individuals with COPD only (P < 0.001). Nocturnal oxygen desaturation index, resting daytime $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$, nocturnal mean $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ and time spent at a nocturnal $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ less than 88% did not differ between groups under typical treatment conditions (all P value >0.05).

Table 3.

Pulmonary characteristics

	Overlap Syndrome n = 12	COPD Only n = 47	OSA Only n = 9	Former Smoker Controls n = 21	P Value
Daytime pulmonary measures
COPD GOLD Stage
Stage 1, n, %	1 (8)	3 (7)	0 (0)	0 (0)
Stage 2, n, %	8 (67)	19 (42)	0 (0)	0 (0)
Stage 3, n, %	2 (17)	12 (27)	0 (0	0 (0)
Stage 4, n, %	1 (8)	11 (24)	0 (0)	0 (0)
FEV1/FVC, %	61.1 ± 10.5	48.7 ± 12.0*†	76.4 ± 4.5*†	79.7 ± 4.8*†	<0.001
FEV1, % predicted	57.1 ± 17.7	49.6 ± 21.0	80.3 ± 17.1*†	97.3 ± 17.3*†‡	<0.001
Daytime resting O2 saturation, %	92.2 ± 2.5	93.0 ± 3.3	93.7 ± 3.3	94.4 ± 2.6	0.18
O2 use while sitting, n, %	5 (42)	19 (40)	1 (11)	0(0)	0.003
6MWT, total distance walked, m	389 ± 83	395 ± 102	461 ± 120	514 ± 86*†	<0.001
Nocturnal continuous pulse oximetry
O2 use during nocturnal recording, n, %	9 (75)	27 (60)	3 (33)	1 (5)	<0.001
CPAP use during nocturnal recording, n, %	8 (67)	0 (0)	3 (33)	0 (0)	<0.001
Mean Nocturnal $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$, %	89.4 ± 5.4	91.0 ± 4.4	90.9 ± 1.7	90.2 ± 2.5	0.64
Oxygen desaturation index (ODI)	0.10 ± 0.1	0.11 ± 0.1	0.19 ± 0.2	0.18 ± 0.1	0.08
Time spent $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ <88, %	27.9 ± 33.5	25.2 ± 33.2	14.3 ± 15.1	16.2 ± 22.5	0.51

Data are means ± SD. COPD, chronic obstructive pulmonary disease; GOLD, global initiative for chronic obstructive lung disease; FEV₁, forced expiratory volume in one second; FVC, forced vital capacity; CPAP, continuous positive airway pressure; O₂, oxygen; $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$, peripheral blood oxygen saturation. One-way analysis of covariance (ANCOVA) tests were used for all continuous variables and χ² tests were used for categorical variables.*P < 0.05 vs. overlap syndrome group; †P < 0.05 vs. COPD group; ‡P < 0.05 vs. OSA group. n = number of subjects.

CCA Stiffness and Cognitive Performance

Groups significantly differed by CCA β-stiffness index (one-way ANOVA: P < 0.001) with individuals with overlap syndrome demonstrating greater CCA β-stiffness index compared with individuals with COPD alone (P = 0.001), individuals with OSA alone (P = 0.007), and former smoker controls (P < 0.001). Importantly, this difference remained significant after adjusting for age, BMI, sex, pack years, and postbronchodilator FEV₁/FVC (P = 0.02; Fig. 1). Individuals with overlap syndrome trended to have greater supine brachial pulse pressure (PP), greater CCA diameter, and lower CCA distention and compliance, however these differences did not reach statistical significance (all P values >0.05; Table 4).

Figure 1.

One-way analysis of covariance (ANCOVA) tests were used to compare carotid β-stiffness index between individuals with overlap syndrome (n = 12), individuals with COPD only (n = 47), individuals with OSA only (n = 9), and former smoker controls (n = 21). Data are means ± SE. COPD, chronic obstructive pulmonary disease; OSA, obstructive sleep apnea. n = number of subjects. *P < 0.05.

Table 4.

Vascular measurements

Vascular Measurement	Overlap Syndrome(n = 12)	COPD Only(n = 47)	OSA Only(n = 9)	Former Smoker Controls(n = 21)	One-WayANOVA P Value
Brachial PP, mmHg	52 ± 3	51 ± 2	55 ± 2	48 ± 9	0.30
Carotid diameter, mm	7.58 ± 0.91	6.95 ± 0.78	6.90 ± 0.77	7.0 ± 0.75	0.11
Carotid distension, %	0.36 ± 0.18	0.43 ± 0.12	0.52 ± 0.22	0.46 ± 0.11	0.06
Carotid compliance, mm2·mmHg−1	0.08 ± 0.01	0.10 ± 0.04	0.10 ± 0.04	0.01 ± 0.04	0.27
Carotid β-stiffness index, U	13.57 ± 1.68	8.99 ± 0.47	8.37 ± 0.96	7.99 ± 0.54	<0.001*

Data are means ± SE. COPD, chronic obstructive pulmonary disease; OSA, obstructive sleep apnea; ANOVA, one-way analysis of variance (ANOVA). n = number of subjects. *P < 0.05.

Groups demonstrated different cognitive performance in the executive function-processing speed domain (one-way ANOVA: P = 0.001) with individuals with overlap syndrome demonstrating lower executive function-processing speed compared with individuals with COPD alone (P = 0.04), individuals with OSA alone (P = 0.003), and former smoker controls (P = 0.004). This difference remained significant after adjusting for age, sex, BMI, pack years, and postbronchodilator FEV₁/FVC (P = 0.02; Fig. 2). There were no differences observed between groups in any of the other cognitive domains (i.e., memory, language and visuospatial skills; all P values >0.05; Table 5). Higher CCA β-stiffness index was associated with lower executive function-processing speed performance in the entire cohort (r = −0.30, P = 0.004; Fig. 3) as previously reported (27) and in individuals with overlap syndrome (r = −0.58, P = 0.047; Fig. 4). There was no relation between CCA β-stiffness index and executive function-processing speed in individuals with COPD alone (r = −0.06, P = 0.71), OSA alone (r = −0.17, P = 0.67), or former smoker controls (r = −0.01, P = 0.98; see Fig. 4).

Figure 2.

One-way analysis of covariance (ANCOVA) tests were used to compare cognitive performance in the executive function-processing speed domain between individuals with overlap syndrome (n = 12), individuals with COPD only (n = 47), individuals with OSA only (n = 9), and former smoker controls (n = 21). Data are means ± SE. COPD, chronic obstructive pulmonary disease; OSA, obstructive sleep apnea. n = number of subjects. *P < 0.05.

Table 5.

Cognitive Domains

Cognitive Domain	Overlap Syndrome(n = 12)	COPD Only(n = 47)	OSA Only(n = 9)	Former Smoker Controls(n = 21)	One-WayANOVA P Value	Correlation with Carotid β-Stiffness Index
Memory	48.6 ± 2.1	47.9 ± 1.2	50.2 ± 3.1	50.2 ± 2.0	0.72	r = −0.10, P = 0.35
Language	51.4 ± 1.9	51.0 ± 0.9	54.8 ± 2.6	53.9 ± 7.6	0.26	r = −0.11, P = 0.30
Visuospatial	53.3 ± 1.5	52.2 ± 0.8	53.7 ± 2.2	52.4 ± 0.6	0.79	r = 0.05, P = 0.64
Executive Function-Processing Speed	44.9 ± 1.4	49.2 ± 0.74	51.8 ± 2.0	50.2 ± 1.2	0.001*	r = 0.30, P = 0.004*

Data are means ± SE COPD, chronic obstructive pulmonary disease; OSA, obstructive sleep apnea; ANOVA, one-way analysis of variance (ANOVA). n = number of subjects. *P < 0.05.

Figure 3.

Bivariate correlation between carotid β-stiffness index and executive function-processing speed performance (P = 0.002) in the entire cohort (n = 89). n = number of subjects.

Figure 4.

Bivariate correlations between carotid β-stiffness index and executive function performance among individuals with overlap syndrome (n = 12), individuals with COPD only (n = 47), individuals with OSA only (n = 9), and former smoker controls (n = 21). COPD, chronic obstructive pulmonary disease; OSA, obstructive sleep apnea. n = number of subjects.

DISCUSSION

The major and novel finding of the present study was that CCA stiffness was greater and executive function-processing speed was lower among individuals with overlap syndrome (i.e., COPD and OSA) compared with individuals with COPD alone, OSA alone and former smoker controls. These differences remained significant after adjusting for age, BMI, sex, pack-years and post-bronchodilator FEV₁/FVC. We have previously reported that greater CCA stiffness is associated with lower executive function-processing speed in former smokers (27), and here we identify that the association is, in part, driven by individuals with overlap syndrome.

CCA stiffness is a predictor of stroke in adults independent of other CVD risk factors, including aortic stiffness (17), and is significantly greater in individuals with either COPD or OSA alone compared with controls (18, 19). A recent study by Shiina et al. (46) demonstrated greater brachial-ankle pulse wave velocity (ba-PWV) among those with overlap syndrome compared with individuals with OSA alone. However, ba-PWV reflects both central and peripheral artery stiffness. In contrast, the carotid arteries are the primary vessels supplying blood from the heart to the cerebral microcirculation. Given the proximity of the CCA to the cerebral circulation, it is plausible that CCA stiffness may play a more important role in contributing to cognitive dysfunction than aortic or peripheral stiffness. Furthermore, the relation between ba-PWV and cognitive function, independent of other CVD risk factors, remains unclear (47, 48). Although ba-PWV has been measured previously among individuals with overlap syndrome, CCA stiffness was the primary marker of stiffness used in the present study because it is a local measure of central elastic artery stiffness and has been associated with cognitive performance among other populations (20, 21). Importantly, we demonstrated that CCA stiffness, was greater in individuals with overlap syndrome than in individuals with either disorder alone, suggesting that CCA stiffness may potentially contribute in part to the heightened CVD risk demonstrated in overlap syndrome patients. Interestingly, CCA stiffness did not differ between individuals with COPD alone and OSA alone relative to former smoker controls. We believe that the former smoker status of the controls included in this study impacted the lack of a difference in CCA stiffness between those with COPD alone, OSA alone and controls. Although the former smokers did not meet the clinical criteria for COPD based on spirometry, it is likely that the presence of lung pathology from numerous years or smoking or the presence of additional comorbidities may have contributed to the lack of difference in CCA stiffness among these groups.

The past literature has found that individuals with COPD alone and OSA alone both demonstrate particularly reduced performance on cognitive tasks pertaining to executive function, processing speed and memory retrieval compared with controls (6, 8). The results from the present study suggest that having both COPD and OSA may have additive detrimental effects on executive function and processing speed and that CCA stiffness may not only contribute to the heightened CVD risk in overlap syndrome patients, but also to lower executive function-processing speed performance. Executive functioning and processing speed are cognitive domains that are commonly reduced among individuals with cerebrovascular disease and have been associated with elastic arterial stiffening (22–24, 26, 49–52). Subcortical regions of the brain, which modulate executive function and processing speed via frontal-subcortical networks, are particularly susceptible to vascular insults (26, 53) such as variability in pulsatile pressure that accompanies carotid artery stiffening. Subcortical structures receive blood supply from small cerebral vessels that are not structured to adequately handle variability of pulse pressure and have few interconnections that would aid in maintaining perfusion if ischemic injury were to occur (54). Interestingly, while we hypothesized that there would be an association between carotid artery stiffness and cognitive performance among all groups, the association between carotid β-stiffness index and executive function-processing speed was only statistically significant among individuals with overlap syndrome. The reasons why we did not observe an association in individuals with COPD alone, OSA alone and former smoker controls are not clear. It is possible that individuals with overlap syndrome are more susceptible to the deleterious effects of greater carotid artery stiffness compared with individuals with either disorder alone. Among individuals with COPD alone, it is possible that excluding individuals with a history of stroke impacted the lack an association between carotid artery stiffness and executive function-procession speed as individuals with a history of a stroke may have had the highest carotid stiffness and were thus excluded from the study. It is also possible that looking at COPD as a unitary diagnosis when there is heterogeneity in COPD may obscure a strong association among as subset of COPD patients who have a disease phenotype with predominant systemic inflammation and vascular changes. Thus, more nuanced future work that examines COPD phenotype is needed. Finally, among individuals with OSA alone it is possible that other mechanisms may contribute to cognitive performance in this group such as intermittent hypoxia resulting from recurring nocturnal apneas. Furthermore, the small sample size among this group may have contributed to the lack of a statically significant association between carotid artery stiffness and executive function. Thus, we would expect that future studies would observe associations between stiffness and executive function-speed in OSA alone and COPD alone and that the association may be most pronounced in those with overlap syndrome.

Most past research has suggested that large elastic arterial stiffness preferentially impacts executive function and processing speed; however, some studies have found that aortic stiffness is associated with lower performance on memory tasks (26, 49–52). The present study found an association between carotid β-stiffness index and executive function-processing speed but not memory (20, 27). One potential factor that may influence discrepant results regarding stiffness and memory is the type of memory function being tested (e.g., memory recall versus discriminability). In addition, our sample was designed so that individuals with a previously known history of cognitive impairment (e.g., mild cognitive impairment) were excluded and by design this likely lead to a sample of older adults who had a low rate of memory disorders such as early neurodegenerative disease. Future studies should include cognitive measures across domains and further explore the relation between carotid β-stiffness index and memory. Altogether, the greater carotid artery stiffness demonstrated among individuals with overlap syndrome may explain in part why individuals in this group demonstrated lower cognitive performance specifically in the executive function-processing speed domain and not other cognitive domains that are not as susceptible to vascular insults such as language and visuospatial skills.

The mechanisms contributing to the greater CCA stiffness demonstrated in individuals with overlap syndrome in the present study remain unclear. Chronic and intermittent hypoxemia are both associated with greater large elastic artery stiffness in individuals with COPD alone and OSA alone (55–57). Therefore, it is plausible that chronic hypoxemia, in addition to recurrent nocturnal apneas, may exacerbate CCA stiffening in individuals with overlap syndrome. In the present study, there were no differences in daytime mean $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$, nocturnal mean $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ or nocturnal oxygen desaturation index in individuals with overlap syndrome versus the other groups. This is likely because recordings were made under normal treatment conditions and ∼40% of the entire cohort reported using supplemental oxygen and/or CPAP during the nocturnal oximetry monitoring. Individuals with overlap syndrome did spend significantly more time with a nocturnal $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ below 80%. However, this difference was largely driven by one outlier who reported using both supplemental oxygen and a CPAP at night but spent most of the night with an $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ below 80%. When this outlier is excluded there are no differences between groups. A more thorough analysis of nocturnal and daytime $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$without the confounding influence of nocturnal oxygen or CPAP use is needed to determine the effects of hypoxemia on CCA stiffness and cognitive function in overlap syndrome patients. However, it is likely that several different mechanisms contribute to the greater CCA stiffness observed in individuals with overlap syndrome possibly including elevated sympathetic nerve activity (SNA) and/or heightened inflammation.

SNA and inflammation are elevated in individuals with COPD alone and OSA alone (58–60) and are both independently associated with greater large elastic artery stiffness in adults (61–64). Interestingly, individuals with overlap syndrome have greater SNA than individuals with COPD alone and OSA alone (65). However, SNA and inflammation were not assessed in the present study and, therefore, the potential roles of elevated SNA and inflammation in contributing to greater CCA stiffness in individuals with overlap syndrome cannot be determined. In addition to greater SNA and inflammation, the influence of psychosocial factors, cardiorespiratory fitness, and medication usage on CCA stiffness must also be considered. Depression and worse cardiorespiratory fitness are both associated with greater large elastic artery stiffness (66, 67). Individuals in the present study had minimal depressive symptoms, which is expected given inclusion/exclusion criteria requiring stable treatment for depression. Individuals with overlap syndrome did have statistically significantly higher scores on the HADS depression subscale than former smoker controls, but not with individuals with either COPD or OSA alone. However, severity of depressive symptoms was not associated with CCA stiffness in the entire cohort or in individuals with overlap syndrome, suggesting that depression severity did not account for the exacerbated CCA stiffness demonstrated in individuals with overlap syndrome in our sample. Individuals with overlap syndrome also had a lower total distance walked on the 6MWT compared with individuals with OSA alone and former smoker controls. However, there were no differences in 6MWT total distance walked between individuals with overlap syndrome and individuals with COPD. Furthermore, CCA stiffness was not associated with 6MWT total distance walked in the entire cohort or among individuals with overlap syndrome only. Finally, among individuals with COPD, bronchodilator therapy improves respiratory symptoms, lowers the risk of exacerbations and reduces aortic PWV in patients with a very high PWV (≥11 m/s) (68). However, there have been no studies to date demonstrating an association between bronchodilator therapy and lower CCA stiffness among individuals with COPD. In the present study, there were no differences in carotid β-stiffness index among individuals that reported either LAMA or ICS/LABA bronchodilator use compared with individuals who did not report using these bronchodilators (all P values >0.05) suggesting that bronchodilator use does not explain group differences in CCA stiffness.

The results of the present study should be interpreted in the context of several limitations. First, to avoid the effects of other major diseases and recent COPD exacerbations on vascular and cognitive function, we excluded individuals with a major medical condition other than COPD/asthma and individuals who had a recent COPD exacerbation. This exclusion criteria likely contributed to the relatively small sample size of individuals with overlap syndrome (n = 12) because individuals in this group are more likely to suffer from COPD exacerbations and cardiovascular comorbidities such as stroke. Second, although a power analysis was conducted for the overall sample size, we did not select the sample sizes for individuals with overlap syndrome, COPD alone, OSA alone and former smoker controls. Despite having unbalanced groups we feel the analysis contributes a novel approach to the current literature. Many previous studies examining large elastic artery stiffness and cognitive performance among COPD patients have either excluded patients who also have OSA or have not differentiated between individuals that have COPD alone or both COPD and OSA (i.e., overlap syndrome). Therefore, there has been a lack of studies examining the effects of overlap syndrome on large elastic artery stiffness and cognitive performance and the association between large elastic artery stiffness and cognitive performance among this group. Third, the cross-sectional nature of the present study prevents the determination of a causal relation between CCA stiffness and cognitive performance in individuals with overlap syndrome. Fourth, identification of OSA was based on self-report of a clinical diagnosis of OSA and could not be confirmed with nocturnal polysomnography given that nocturnal measures were obtained under usual treatment conditions (e.g., supplemental oxygen and/or CPAP). Thus, it is possible that relying only on self-report for the diagnosis of OSA did not fully identify everyone with OSA in the present study. However, because of the strong group differences observed in the present study between individuals with overlap syndrome versus the other groups, it is likely that future studies that identify individuals with OSA based on nocturnal polysomnography will show even more pronounced findings than were observed in the present study. Finally, the participants in the present study were recruited from the COPDGene cohort which only includes non-Hispanic white and black individuals. Although participation of non-Hispanic black individuals was encouraged, the majority of participants in the present study were non-Hispanic white. Therefore, further studies among persons of non-Caucasian and Hispanic race/ethnicity are needed.

In summary, findings from the present study suggest that CCA stiffness is greater and cognitive performance is lower among individuals with overlap syndrome compared with individuals with either COPD or OSA alone and that CCA stiffness may be at least one underlying mechanism that contributes in part to lower cognitive performance observed in this group. These findings demonstrate the clinical importance of early identification and intervention in individuals with overlap syndrome because individuals in this group may be at a greater risk of CVD events associated with CCA stiffness, such as stroke, and also a greater risk of cognitive decline.

DISCLAIMERS

Data were managed using REDCap hosted at the University of Colorado. Contents are the authors’ sole responsibility and do not necessarily represent official NIH views.

GRANTS

The current study was supported by National Institutes of Health grants K23 HL095658, R01 HL089897, NCATS Colorado CTSI Grant UL1 TR001082, and NCATS Iowa ICTS Grant U54 TR001356. The COPDGene study (NCT00608764) is also supported by the COPD Foundation through contributions made to an Industry Advisory Board comprised of AstraZeneca, Boehringer-Ingelheim, Novartis, Pfizer, GlaxoSmithKline, Siemens and Sunovion.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

R.E.L., J.D.C., and K.F.H. conceived and designed research; K.F.H. and K.L.M. performed experiments; R.E.L. and K.F.H. analyzed data; R.E.L., K.L.M., G.L.P., E.K., and K.F.H. interpreted results of experiments; R.E.L. prepared figures; R.E.L. drafted manuscript; R.E.L., K.L.M., G.L.P., F.W., M.A., H.D.W., B.M., R.B., J.D.C., K.M., E.K., D.J.M., and K.F.H. edited and revised manuscript; R.E.L., K.L.M., G.L.P., F.W., M.A., H.D.W., B.M., R.B., J.D.C., K.M., E.K., D.J.M., and K.F.H. approved final version of manuscript.