ADVERTISEMENT

Issue Navigator

Volume 14 No. 10
Earn CME
Accepted Papers





Scientific Investigations

Physical Activity Following Positive Airway Pressure Treatment in Adults With and Without Obesity and With Moderate-Severe Obstructive Sleep Apnea

Yuan Feng, MD, PhD1,2; David Maislin, BS2; Brendan T. Keenan, MS2; Thorarinn Gislason, MD3,4; Erna S. Arnardottir, PhD3,4; Bryndis Benediktsdottir, MD3,4; Julio A. Chirinos, MD, PhD2,5; Raymond R. Townsend, MD2,5; Bethany Staley, RPSGT2; Francis M. Pack, RN, CCRC2,5; Andrea Sifferman, MPH2; Allan I. Pack, MBChB, PhD2,5; Samuel T. Kuna, MD2,5,6
1Sleep Medicine Center, Nanfang Hospital, Southern Medical University, Guangzhou, China; 2Center for Sleep and Circadian Neurobiology, University of Pennsylvania, Philadelphia, Pennsylvania; 3Department of Sleep, Landspitali - The National University Hospital of Iceland, Reykjavik, Iceland; 4Faculty of Medicine, University of Iceland, Reykjavik, Iceland; 5Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania; 6Department of Medicine, Corporal Michael J. Crescenz Veterans Affairs Medical Center, Philadelphia, Pennsylvania

ABSTRACT

Study Objectives:

To examine the level of physical activity (PA) before and following positive airway pressure (PAP) treatment in adults who have obstructive sleep apnea (OSA) with obesity versus without obesity.

Methods:

Simultaneous waist accelerometer and wrist actigraphy recordings were obtained in 129 adults with obesity and 69 adults without obesity and who had OSA prior to and following 4 months of PAP therapy and in 52 patients in a control group. Primary PA measurements were average steps per day on waist accelerometry and average counts per minute (CPM) per day on wrist actigraphy.

Results:

At baseline, participants with obesity and OSA exhibited fewer steps per day on waist accelerometer and fewer CPM per day on wrist actigraphy compared to participants without obesity and with OSA (despite similar apnea-hypopnea index between groups). Following PAP treatment, participants with OSA had modestly increased CPM per day on wrist actigraphy (17.69 [95% confidence interval (CI), 5.67–29.71], P = .005) and increased light PA time (0.26 [95% CI, 0.07–0.44] hours, P = .008) on waist accelerometer. Participants without obesity and with OSA had greater improvements in PA measures on average compared to participants with obesity and OSA, although the differences were not statistically significant. Weight increased following PAP treatment in the participants with obesity and OSA (1.71 [95% CI, 0.41–3.02] kg, P = .010) but was unchanged in the group without obesity (0.93 [95% CI, −0.89 to 2.76] kg, P = .311).

Conclusions:

Compared to study participants without obesity and with OSA, participants with obesity and OSA had reduced PA at baseline. PA increased significantly in participants without obesity, with OSA, and who were adherent to PAP treatment. Results indicate that treatment of OSA is unlikely to be associated with a change in PA in adults with obesity and OSA and help explain the absence of weight loss following PAP treatment in adults with OSA.

Clinical Trial Registration:

Registry: ClinicalTrials.gov, title: The Effects of Treating Obese and Lean Patients With Sleep Apnea (PISA), identifier: NCT01578031, URL: https://clinicaltrials.gov/ct2/show/NCT01578031

Citation:

Feng Y, Maislin D, Keenan BT, Gislason T, Arnardottir ES, Benediktsdottir B, Chirinos JA, Townsend RR, Staley B, Pack FM, Sifferman A, Pack AI, Kuna ST. Physical activity following positive airway pressure treatment in adults with and without obesity and with moderate-severe obstructive sleep apnea. J Clin Sleep Med. 2018;14(10):1705–1715.


BRIEF SUMMARY

Current Knowledge/Study Rationale: Previous studies report that body weight of individuals with obstructive sleep apnea (OSA) increases following positive airway pressure treatment. Physical activity, a factor influencing body weight, is reported to be decreased in adults with obesity and adults with untreated OSA.

Study Impact: We found that physical activity was generally lower in participants with OSA and with obesity, compared to those without obesity, prior to positive airway pressure treatment. Following 4 months of successful treatment, physical activity increased significantly in participants without obesity and with OSA. The results indicate that treatment of OSA is not associated with a change in PA in adults with obesity and with OSA. The findings help to explain the absence of weight loss in patients with OSA on PAP treatment.

INTRODUCTION

Obstructive sleep apnea (OSA) is a common sleep disorder, with an estimated 13% of middle-aged men and 6% of middle-aged women having moderate to severe disease (defined as an apnea-hypopnea index [AHI] ≥ 15 events/h).1 OSA is associated with an increased risk of hypertension, heart attack, stroke, depression, and motor vehicle accidents.25 The prevalence of OSA and its adverse consequences make it a major public health challenge.

Adiposity, a major risk factor for OSA, is inversely associated with physical activity (PA).69 Higher body mass index (BMI) in adults is associated with sedentary behavior.10,11 Obesity is reported to be associated with decreased moderate to vigorous PA.12,13 A stronger association between BMI and PA is observed in individuals with obesity than in individuals without obesity.14 Furthermore, longitudinal studies report that BMI and waist circumference are inversely associated with PA at follow-up.1519

The clinical features of OSA, including daytime sleepiness, decreased mood, and other functional impairments, may predispose patients to be physically inactive. In addition, OSA may reduce exercise tolerance20 and limit exercise capacity by reducing cardiac output and peak oxygen consumption21,22 and impairing muscle metabolism.23 Previous studies report that adults with OSA have decreased PA, as assessed by questionnaire2426 and objectively using armband accelerometry.27,28 In contrast, Mendelson et al.29 found that armband-based PA in adults with OSA was significantly related to obesity, but not to OSA severity or sleepiness.

Positive airway pressure (PAP), the first-line therapy for patients with OSA, improves daytime sleepiness,30,31 quality of life, mood state,32 blood pressure,3335 and, possibly, exercise capacity.36,37 Given these beneficial effects, it might be anticipated that patients with OSA would increase their PA following PAP treatment and that might lead to a decrease in body weight. However, previous studies report that PAP treatment is associated with an increase in body weight3840 and studies evaluating the effect of PAP treatment on PA in patients with OSA have yielded conflicting results.4145 The disagreement across the latter studies may in part be due to the disparate methods used to objectively assess PA.46,47

The primary purpose of the current study was to examine the level of PA in adults with untreated moderate to severe OSA with obesity versus those without obesity and investigate the change in PA following 4 months of PAP treatment. Simultaneous triaxial waist accelerometer and uniaxial wrist actigraphy recordings were obtained to objectively measure PA. To our knowledge, no previous studies have evaluated PA in adults with OSA using waist-worn triaxial accelerometry, which has a stronger relationship with energy expenditure than the uniaxial recorders used in previous studies.4648 We hypothesized that participants with OSA and without obesity, compared to participants with obesity, would have greater PA at baseline and a greater increase in PA following PAP treatment. Secondarily, we compared PA of adults with OSA prior to treatment to that of a control group.

METHODS

Participants

The analysis was conducted using data from the Penn Iceland Sleep Apnea research project (PISA), a two-site, longitudinal study of functional, cardiovascular, and metabolic measures following 4 months of PAP therapy in participants with moderate to severe OSA, with and without obesity. Consecutive, clinically stable adults aged 40 to 65 years were recruited from the sleep centers at the University of Pennsylvania and University of Iceland for evaluation of suspected OSA. Individuals with an apnea-hypopnea index (AHI) of at least 15 events/h but less than 75 events/h were recruited for the OSA group, and individuals with an AHI less than 10 events/h were recruited for the control group. Other inclusion criteria included no change in medications and no regular use (> 3 times/wk) of sedative or hypnotic medications in the past 2 months. Individuals were excluded from the study for the following reasons: BMI > 40 kg/m2 (the cutoff for performing the protocol's MRI of abdomen and thorax); diagnosis of another sleep disorder in addition to OSA; previous treatment of OSA; requiring oxygen or bilevel PAP for treatment of OSA; active infection, malignancy, or chronic inflammatory disorders; night-shift workers in situations or occupations where they regularly experience jet lag, or had irregular work schedules by history over the past 6 months; unable to perform tests due to inability to communicate verbally or inability to write and read in English; arm circumference ≤ 50 cm, the manufacturer's limit for performing ambulatory blood pressure (BP) recording; routine consumption of more than two alcoholic beverages per day; and positive urine toxicology. Women had to be postmenopausal and not be on hormone replacement treatment. The protocol was approved by the Institutional Review Board of the University of Pennsylvania and the University of Iceland, and written informed consent was obtained from all participants.

Study Protocol

Following enrollment, participants underwent an overnight polysomnography (PSG). They completed baseline questionnaires including the Epworth Sleepiness Scale (ESS).49 They wore a triaxial waist accelerometer (GT3X, ActiGraph, Pensacola, Florida, United States) and uniaxial wrist actigraph (Actiwatch 2, Philips Respironics, Murrysville, Pennsylvania, United States) simultaneously for 1 week during which they completed a sleep diary. Following the baseline assessment, the control group had completed the protocol and those in the OSA group were started on PAP treatment. Assessments were repeated in those in the OSA group who completed the 16-week PAP treatment and had an average daily PAP use of at least 4 h/d over at least 90 days.

Obesity Measurements

Participants were classified as obese or non-obese based on their waist circumference at baseline. Waist circumference was measured horizontally around the abdomen at the midpoint between the highest point of the iliac crest and lowest part of the costal margin in the midaxillary line with a nonmetallic constant tension tape while participants wore light clothing and stood with their feet together. Men and women were characterized as obese if they had a waist circumference > 107 cm and > 96 cm, respectively, and non-obese otherwise. Using these criteria, a total of 105 patients (69 with OSA, 36 in the control group) were classified as non-obese and 145 patients (129 with OSA, 16 in the control group) as obese. Waist circumference was chosen as an a priori definition of obesity in this study to better capture visceral obesity level. Sensitivity analyses were performed using cutoffs based on BMI; only 10 patients (19.2%) in the control group and 28 patients (14.1%) in the OSA group fell into different obesity groups using these two definitions.

Diagnostic Polysomnography

Overnight PSG tests were performed using standard techniques to ensure uniform data collection at the two clinical sites.50 The following signals were recorded: electroencephalograms (C3M2, C4M1, O2M1), bilateral electroculograms, electromyograms of the chin muscles and right and left anterior tibialis, movement of the rib cage and abdomen (piezoelectric crystal), oxygen saturation (SaO2) by pulse oximetry, electrocardiogram (lead 1), and body position. Airflow was assessed by nasal airway pressure and oronasal thermistry. All recordings were scored manually with the aid of computer software at the University of Pennsylvania sleep center using 2007 AASM standards for scoring sleep stages, arousals, and respiratory events.51 Apneas were defined as an absence of airflow on the oronasal thermistor and nasal pressure cannula for at least 10 seconds. If the oronasal thermistor signal was not available, the nasal pressure signal was used as the determinant of absent airflow. Hypopneas were defined as a greater than 30% reduction from baseline in airflow for at least 10 seconds associated with at least a 3% oxygen desaturation and or an arousal. AHI was calculated as the mean number of apneas and hypopneas per hour of sleep. Oxygen desaturation index (ODI) was calculated as the mean number of oxygen desaturation events ≥ 3% per hour of sleep.

Positive Airway Pressure Treatment

Following education about OSA and its treatment options, including PAP, participants were initiated on either continuous or auto-adjusting PAP treatment (S9, ResMed Inc.). Participants started on continuous positive airway pressure (CPAP) performed an overnight PSG according to current clinical practice guidelines to determine the appropriate pressure setting.52 The PAP device download was used to objectively monitor the participant's average daily adherence to PAP treatment. There were no restrictions regarding type of mask interface and participants were allowed to change the size and model of their mask interface throughout the study. To further assess treatment efficacy, participants underwent PSG on PAP treatment at the 4-month follow-up assessment.

Waist Accelerometer

Participants were instructed to wear the accelerometer (GT3X, ActiGraph) on an elastic belt over the right hip during all waking hours for 7 consecutive days, except during water-based activities. The accelerometers were initialized at a sampling rate of 30 Hz and the raw data were downloaded and extracted in 60-second epochs using Actilife v4.0.4 software (ActiGraph, Pensacola, Florida, United States).

A valid day was defined as at least 600 minutes of monitor wear time, and only participants with at least 5 valid days (including at least one weekend day) were included in the analysis. Daily wear time was defined by subtracting non-wear time from 24 hours. Non-wear time was defined as at least 60 consecutive minutes with no movement—defined as zero counts per minute (CPM) with allowance for a maximum of 2 minutes of movement with intensities up to 200 CPM. The criteria of Freedson et al.53 were used to classify the intensity of PA. The following average outcome measures per day were obtained: number of steps, CPM, energy expended by physical activity (Kcal), hours of sedentary time, hours of light physical activity (light PA time), and hours of moderate and vigorous physical activity (MVPA time).

Wrist-Worn Accelerometer

Participants were instructed to wear the wrist-worn accelerometer (Actiwatch 2, Philips Respironics) on their nondominant hand 24 hours per day for 7 days. The wrist accelerometer had a sampling rate of 32 Hz. The data were downloaded and analyzed using Actiware software (version 5.71.0; Philips Respironics) with a 30-second epoch. Rest intervals (period when study participant was trying to sleep or nap) and excluded intervals were manually edited based on the bedtime and wake time from sleep diary and combined with the recorded lux change when the diary was unavailable or unreliable. The strategy we used to score rest intervals was similar with that of Patel et al.,54 except that we required at least 10 consecutive epochs to define the change of lux or activity instead of 5. For each day, after removing rest intervals, the remaining time was included in the PA assessment. The criteria used to determine a valid day and valid data were the same as those for waist accelerometer. The average CPM per day was used to measure PA.

Statistical Analysis

The waist and wrist-worn accelerometer recordings were preprocessed prior to analysis. Quality assurance checks were performed by reviewing raw data and performing graphical screening. Descriptive statistics are presented using frequencies and percentages for categorical variables and means, standard deviations, and/or 95% confidence intervals for continuous variables. Variables are compared between groups using t tests (continuous) and chi-square or Fisher exact tests (categorical) at baseline.

Associations between baseline PA measures and OSA status or obesity were performed using linear regression models, with PA as an outcome variable. Results are presented as least squares mean estimates and 95% confidence intervals in each group, with associated P values comparing the two groups. Analyses were performed controlling for relevant covariates, including age, sex, baseline AHI, BMI, and site. As detailed previously, obesity status was determined using waist circumference. Separate analyses were also conducted using BMI to assess obesity, with a value of 30 kg/m2 or greater identifying participants with obesity. No clinically significant differences in results were observed when using this other metric of obesity (Table S3 and Table S4 in the supplemental material).

To examine the effect of PAP on PA measures, we calculated participant-specific change scores as follow-up minus baseline values among those in the OSA group who adhered to treatment. We assessed whether there were significant changes overall using linear mixed models evaluating the effect of time (0 = baseline, 1 = follow-up), restricted to individuals with observed values at both baseline and follow-up. Assessments of whether changes differed between groups defined by baseline obesity group or self-reported sleepiness status based on ESS score (sleepy [ESS ≥ 11] and nonsleepy [ESS < 11]) were performed using linear regression models with change scores as the outcomes. Results are presented as model estimated change scores and 95% confidence intervals, as well as P values testing whether the within-group changes are significantly different from zero and overall comparisons between groups where relevant. Analyses were performed controlling for similar covariates as baseline analyses, as well as baseline values of the change score of interest to control for potential regression to the mean. Our primary measures of PA were average number of steps per day on waist accelerometer and average CPM per day on wrist actigraph. We chose these primary measures because they are the most frequently used in studies of physical activity. Statistical significance for these primary metrics was based on a value of P < .05. Secondary activity measures from the waist accelerometer included CPM per day, Kcals per day, sedentary time, light PA time and MVPA time. For these measures, statistical significance was determined using the Hochberg step-up method to maintain overall family-wise error at α = .05.55 Briefly, for a given set of n null hypotheses (H0i) and associated P values (Pi), where i = 1, 2…n represents the rank of each P value, ordered from smallest to largest, statistical significance for each null hypothesis is determined through the following steps:

  • Step 1: If Pn < .05, reject H0i for all i = 1, 2,…n; else go to Step 2;

  • Step 2: If Pn− 1 < .05/2, reject H0i for all i = 1, 2,…n − 1; else go to Step 3;

  • […]

  • Step n: If Pi < .05/n, reject H0i for i = 1; else, no null hypotheses are rejected

A value of P < .05 was considered suggestive evidence of an association in secondary analyses. Analyses were performed using SAS, Version 9.4 (SAS Institute, Cary, North Carolina, United States).

RESULTS

Study Population

Of the 310 participants enrolled in the PISA study, a total of 250 participants had baseline waist accelerometer recordings and were included in the current analysis (Figure 1). Included participants were aged 53.3 ± 6.9 years, had an average BMI of 30.9 ± 4.4 kg/m2, 81.2% were male and 75.0% were Caucasian (Table 1).

Characteristics of participants in the control group versus OSA group.

jcsm.14.10.1705.t01.jpg

table icon
Table 1

Characteristics of participants in the control group versus OSA group.

(more ...)

Flow diagram illustrating the analysis populations.

AHI = apnea- hypopnea index, BMI = body mass index, OSA = obstructive sleep apnea, PAP = positive airway pressure.

jcsm.14.10.1705a.jpg

jcsm.14.10.1705a.jpg
Figure 1

Flow diagram illustrating the analysis populations.

(more ...)

Participants With OSA Versus Control Group

Table 1 also compares the characteristics of participants with OSA (n = 198) to those of participants in the control group (n = 52). The mean AHI was 35.8 ± 15.3 events/h in the OSA group and 4.1 ± 2.7 events/h in the control group. Compared to the control group, those with OSA had greater body weight, BMI, and waist and neck circumference (P < .001 for all measures), were more likely to be male (84.3% versus 69.2%, P = .017) and had higher levels of self-reported sleepiness based on ESS (9.6 versus 7.3, P = .002).

In fully adjusted analyses (Table 2), no statistically significant differences were found at baseline between the OSA and control groups for primary or secondary waist accelerometer measurements; differences in steps per day were trending (P = .062, Table S1 in the supplemental material) prior to BMI adjustment. In contrast, participants in the control group had greater average daily CPM on the wrist actigraphy than participants with OSA (P = .005 without BMI adjustment, P = .046 after BMI adjustment). Neither waist accelerometer nor wrist actigraph wear times during wakefulness at baseline were significantly different between the two groups.

Adjusted daily activity measurements in participants in the control group versus OSA group.

jcsm.14.10.1705.t02.jpg

table icon
Table 2

Adjusted daily activity measurements in participants in the control group versus OSA group.

(more ...)

Participants With OSA, Obese Versus Non-Obese

Of the 198 participants with OSA, 129 were obese and 69 were non-obese based on waist circumference (Table 3). Although obesity was defined based on waist circumference, BMI was also significantly different between the obese (33.7 ± 3.2 kg/m2) and non-obese (27.9 ± 2.5 kg/m2) groups (P < .001). The mean AHI of all participants with OSA was 35.8 ± 15.3 events/h, and AHI was similar in the obese and non-obese groups (P = .392).

Characteristics of all participants with moderate to severe OSA and comparison of those with and without obesity.

jcsm.14.10.1705.t03.jpg

table icon
Table 3

Characteristics of all participants with moderate to severe OSA and comparison of those with and without obesity.

(more ...)

Table 4 compares PA measurements at baseline. In general, participants who were obese were less physically active than those who were non-obese (Figure 2). The waist accelerometer results showed that participants who were obese had significantly fewer average steps per day (P = .0003), fewer average CPM per day (P = .009), less average light PA time per day (P = .003), and less MVPA time per day (P = .011). The wrist actigraphy results showed similar results, with participants who were obese having fewer average CPM per day than participants who were non-obese (P = .011). There was no difference in device wear time for either the waist accelerometer (P = .106) or wrist actigraphy (P = .892). Results were generally similar when defining obesity based on BMI instead of waist circumference (Table S2 in the supplemental material).

Adjusted average daily activity measures of participants with moderate to severe OSA, with and without obesity, at baseline.

jcsm.14.10.1705.t04.jpg

table icon
Table 4

Adjusted average daily activity measures of participants with moderate to severe OSA, with and without obesity, at baseline.

(more ...)

Primary PA measures comparison between obese and non-obese groups at baseline.

Whiskers represent 95% confidence intervals. P values are for between-group difference. CPM = counts per minute, OSA = obstructive sleep apnea, PA = physical activity.

jcsm.14.10.1705b.jpg

jcsm.14.10.1705b.jpg
Figure 2

Primary PA measures comparison between obese and non-obese groups at baseline.

(more ...)

Participants With OSA and Adherent to PAP, Obese Versus Non-Obese

Of the 118 patients with OSA (100 males, 18 females) who were adherent to PAP treatment over the 4-month intervention (ie, average PAP adherence > 4 h/d for at least 90 days), 40 were non-obese and 78 were obese (Table S3). At baseline, no significant differences were present with respect to OSA severity (AHI P = .753, ODI P = .435, % time SaO2 < 90% P = .267, SaO2 nadir P = .777) or average ESS scores (P = .178). Specifically, the AHI after treatment was 5.68 (4.41, 6.95) events/h in participants who were obese and 5.02 (3.24, 6.79) events/h in participants who were non-obese. Average PAP use was 5.7 ± 1.0 h/d overall, with no difference between obese and non-obese groups (P = .642).

Following PAP treatment among all participants with OSA, ESS score was significantly reduced (−4.18 points P < .001). On the contrary, there were slight but statistically significant increases in weight (1.44 kg P = .006) and BMI (0.39 kg/m2 P = .019) following PAP treatment in all participants with OSA (Table S4) that was primarily due to an increase in weight of the participants who were obese. Participants who were obese had a weight increase of 1.72 (95% CI 0.46–2.99) kg (P = .008), or 1.87 (95% CI 0.26–3.48) % of the baseline (P = .023), whereas the change in weight in the participants who were non-obese was neither significant in absolute value (0.88, 95% CI −0.89, 2.65 kg, P = .328) nor in the percentage (1.11, 95% CI −1.14, 3.36 %, P = .328). The change in weight between groups was not significant (P = .445 and .588 for absolute and percentage of weight change respectively).

Table 5 shows the adjusted changes in PA measures following 4 months of PAP treatment for the participants who were adherent to PAP. There were no significant changes in the hours (P = .727) or days (P = .445) that the waist accelerometer was worn or in the number of days the wrist actigraph was worn (P = .540) from baseline and follow-up. Similarly, there were no significant differences in wear time changes between obese or non-obese groups.

Adjusted changes in characteristics and PA measures of all participants with OSA who were adherent to PAP and those with and without obesity following 4 months of PAP treatment.

jcsm.14.10.1705.t05.jpg

table icon
Table 5

Adjusted changes in characteristics and PA measures of all participants with OSA who were adherent to PAP and those with and without obesity following 4 months of PAP treatment.

(more ...)

Among all patients who were adherent to PAP, after controlling for relevant covariates, we found significant increases in CPM per day on wrist actigraphy (P = .005), but no significant change in steps per day on the waist accelerometer (P = .331, Figure 3). For secondary measures on the waist accelerometer, there was suggestively decreased sedentary time (P = .023) and significantly increased light PA time (P = .008) after PAP treatment. Thus, PAP does have some influence on improving PA measures among patients with OSA who are adherent to the treatment.

Changes from baseline in primary PA measures in all participants, and in those with OSA with and without obesity following PAP therapy.

Upper panel shows changes in steps per day on waist accelerometer. Lower panel shows changes in CPM per day on wrist actigraphy. Whiskers represent 95% confidence intervals. P values inside brackets are for the change from baseline in each group. P values without brackets are for between-group difference. CPM = counts per minute, OSA = obstructive sleep apnea, PA = physical activity, PAP = positive airway pressure.

jcsm.14.10.1705c.jpg

jcsm.14.10.1705c.jpg
Figure 3

Changes from baseline in primary PA measures in all participants, and in those with OSA with and without obesity following PAP therapy.

(more ...)

When comparing changes in PA measures between the obese and non-obese groups who were adherent to PAP (Table 5, Figure 3), we observed suggestive, but not statistically significant, differences in changes in steps per day (P = .078), CPM per day (P = .052), and sedentary time (P = .033). For each measure, the participants who were non-obese had improvements at the 4-month follow-up (including steps per day P = .043, CPM per day P = .032 and sedentary time P = .002), compared to no significant change among those who were obese.

Participants With OSA and Adherent to PAP, Sleepy Versus Non-Sleepy

Changes in PA measures were compared in participants who were adherent to PAP stratified by level of sleepiness at baseline (Table S5 in the supplemental material). We found no significant differences in the change in PA measurements based on waist accelerometer or wrist actigraphy, suggesting baseline sleepiness status does not significantly affect PAP response. Correlations between ESS change and change in PA showed that neither change of steps on waist accelerometer (rho = −.18) nor change of CPM on wrist actigraph (rho = −.07) had a significant correlation with ESS change. A modest negative correlation with ESS was observed for MVPA time (rho = −.22, P = .028), Kcals per day (rho = −.19, P = .048), and CPM per day (rho = −.23, P = .020) on waist accelerometer.

DISCUSSION

The primary purpose of this study was to compare PA before and following PAP treatment in adults with and without obesity and with OSA. We found that adults with obesity and with OSA had decreased PA levels at baseline compared to adults without obesity and with OSA, despite similar AHI between the two groups. Following 4 months of PAP treatment, participants without obesity and with OSA showed significant improvements in steps per day on waist accelerometer and in CPM per day on wrist actigraph; however, differences between participants with and without obesity did not reach statistical significance. The differences in PA between participants with and without obesity at baseline and improvement in PA outcomes in only those without obesity following PAP treatment suggest that PA is associated with both obesity and OSA status. But PA measures were not associated with self-reported daytime sleepiness at baseline or change in sleepiness following PAP treatment.

Body weight and BMI increased in our participants with OSA following PAP treatment, in agreement with previous studies.3840,56 Most of the participants in those previous studies were adults who were obese with moderate to severe OSA. In our study, a significant weight increase occurred in our participants who were obese but not in the participants who were non-obese. Our finding that treatment of OSA is not associated with a change in PA in participants with obesity and OSA helps to explain the absence of weight loss in participants with obesity and OSA on PAP treatment. Although weight was unchanged following PAP treatment in participants without obesity, we did find modest but significant changes of PA in those participants following PAP treatment, including an average increase of 763.5 steps per day on waist accelerometer (12% change from baseline). However, this increase in PA without change in caloric intake would not be expected to result in weight loss. A previous study reported that an increase of more than 2100 steps walked per day only decreased BMI by a modest 0.38 kg/m2.57 In addition, MVPA time did not increase in our participants with and without obesity with OSA following PAP treatment, and this measure of PA is most closely associated with weight loss.8

We found no significant difference in PA measured by waist accelerometer between untreated adults with moderate to severe OSA and in participants in the control group. However, participants with OSA had fewer CPM per day on wrist actigraphy than those in the control group. Our findings support previous studies that compared the relationship between PA and OSA severity. Mendelson and colleagues29 found PA measured by armband activity was inversely correlated with BMI, but was not correlated with level of sleepiness (ESS) or other indices of OSA severity. Verwimp et al.28 also used armband actigraphy to objectively measure PA in 75 adults with moderate to severe OSA. Controlling for age and BMI, the association between PA and OSA severity was only significant for AHI in rapid eye movement sleep. Similarly, a study of patients with OSA that measured cardiorespiratory fitness, a proxy for PA especially when PA is of relatively low intensity, found that greater METs were associated with lower AHI at baseline, but the relationship was no longer significant after adjustment for weight.58

Previous studies evaluating the change in PA activity following PAP treatment have yielded conflicting results, perhaps because of the disparate methods to assess PA.41,42,44,5961 Batool-Anwar et al.42 reported that treatment with CPAP did not substantially change self-reported PA (via questionnaire) of patients with OSA. West et al.41 found no improvement in PA among 36 men with OSA who completed 1 week of actigraphy before and following either CPAP or sham CPAP. Mendelson and colleagues60 did not find a change in PA measured by armband activity following PAP treatment in 107 adults with OSA. Using armband activity, Bamberga et al.44 found no change in steps per day and energy expenditure following 6 months of CPAP therapy in adults with obesity and with OSA. On the contrary, a study using both armband actigraphy and questionnaire to measure the PA in 23 patients with OSA and type 2 diabetes found that 1 month of PAP therapy improved PA compared to sham PAP.43 Most recently, Jean et al.45 reported a progressive improvement in PA measured with a pedometer in 62 adults with OSA following 3 and 6 months of PAP treatment.

To our knowledge, this is the first study to evaluate PA in adults with OSA using a waist-worn triaxial accelerometer. Previous studies have used either a wrist or armband actigraph. We obtained simultaneous recordings of waist-worn accelerometer and wrist-worn actigraph to objectively measure PA. The results of the two measurement techniques differed when comparing results at baseline in participants with OSA and control participants; however, comparisons of participants with OSA with and without obesity at baseline and the trend of changes following PAP treatment were similar between the two techniques. The waist accelerometer had a triaxial capacitive microelectromechanical system (MEMS) sensor, whereas the wrist actigraph had a piezoelectric element.62 Differences in the sensors, processing algorithms, and placement locations of the two devices are likely to explain the differences in their results. Ideally, a method for assessing PA would provide information regarding the frequency, duration, intensity, and mode of PA over several days, which in combination would provide an index approximating energy expenditure.63 For these reasons, a triaxial accelerometer is preferred for PA assessment. The wrist-worn uniaxial accelerometer is less accurate when estimating energy expenditure than a triaxial waist-worn device.64 However, wrist-worn devices are more convenient to wear and can also be used to assess sleep time. Some studies indicate that a wrist-worn triaxial accelerometer might be as acceptable as waist-worn devices to assess PA induced energy expenditure.65

This study has several strengths. To our knowledge, this is the first study comparing PA between adults with OSA, with and without obesity. Stratification for obesity allowed us to evaluate the relationship between PA and OSA and assess the effect of obesity on PA in patients with OSA. We used objective measurements of PA and the data underwent stringent and objective data reduction procedures, which ensure valid and reliable results. Moreover, to our knowledge, this is the first study to compare wrist actigraphy and waist accelerometry to measure PA in patients with OSA. In addition to the PA measures usually reported, including steps, energy expenditure, and MVPA time, we also report light PA and sedentary time. This information could help better link PA exposure to health outcomes. Finally, we obtained follow-up assessments in participants with OSA who were adherent to PAP treatment. Therefore, the potential effect of OSA on PA could be accurately assessed.

There are also several limitations in the current study. This was an ancillary study and the parent study's design did not include a control group, that is, patients with OSA who were not treated with PAP. As a result, we do not know whether the modest changes that we observed following PAP treatment among all participants with OSA were indeed associated with PAP treatment. The parent study only recruited volunteers with moderate and severe OSA (AHI > 15 events/h) and individuals with BMI ≤ 40 kg/m2. Therefore, our results may not generalize to all patients with OSA. Furthermore, we assigned individuals with an AHI < 10 events/h to our control group, which technically would include individuals with mild OSA (AHI 5 to < 10 events/h). In addition, participants who were not adherent to PAP therapy were not included in the follow-up assessment. Therefore, we were unable to compare PA changes from baseline between individuals who were adherent versus nonadherent to PAP. Furthermore, we did not monitor the time of year that the data were collected; seasonal changes of PA could have influenced the results. Finally, the number of participants in our control group, particularly those who were obese, was relatively low. In total, 52 participants were included in our control group, and 16 of these were obese. These low numbers of participants limited our ability to robustly evaluate PA in adults with obesity in our control group.

In summary, the findings indicate that both obesity and OSA affect PA. Compared to controls, participants with OSA who were obese had reduced PA at baseline. PA increased significantly following PAP treatment among participants with OSA who were non-obese but the change in PA in participants with OSA following PAP treatment who were obese was not significant. The results help explain the absence of weight loss in patients with OSA following PAP treatment in this and previous studies. The minimal effect of PAP treatment on PA in participants with obesity and OSA emphasizes the importance of enrolling these patients in programs that target weight loss and increased PA, including MVPA, in order to achieve a healthier lifestyle.

DISCLOSURE STATEMENT

All listed authors have seen and approved the manuscript. Source of support: National Institutes of Health (NIH HL094307); China Scholarship Council (CSC No.201508440455). Dr. Pack is The John L. Miclot Professor of Medicine at the University of Pennsylvania. Funds for this endowment were provided by the Philips Respironics Foundation. Dr. Arnardottir is a consultant for Nox Medical, Reykjavik, Iceland (unrelated to manuscript). The remaining authors report no conflicts of interest.

ABBREVIATIONS

AASM

American Academy of Sleep Medicine

AHI

apnea-hypopnea index

BMI

body mass index

CPAP

continuous positive airway pressure

CPM

counts per minute

ESS

Epworth Sleepiness Scale

Kcal

kilocalorie

METs

metabolic equivalent

MEMS

microelectromechanical system

MVPA

moderate and vigorous physical activity

ODI

oxygen desaturation index

OSA

obstructive sleep apnea

PA

physical activity

PAP

positive airway pressure

PISA

Penn Iceland Sleep Apnea

PSG

polysomnography

SaO2

oxygen saturation

REFERENCES

1 

Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, Hla KM. Increased prevalence of sleep-disordered breathing in adults. Am J Epidemiol. 2013;177(9):1006–1014. [PubMed Central][PubMed]

2 

Senaratna CV, English DR, Currier D, et al. Sleep apnoea in Australian men: disease burden, co-morbidities, and correlates from the Australian longitudinal study on male health. BMC Public Health. 2016;16 Suppl 3:1029[PubMed]

3 

Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med. 2000;342(19):1378–1384. [PubMed]

4 

Hla KM, Young T, Hagen EW, et al. Coronary heart disease incidence in sleep disordered breathing: the Wisconsin Sleep Cohort Study. Sleep. 2015;38(5):677–684. [PubMed Central][PubMed]

5 

Bassetti C, Aldrich MS, Chervin RD, Quint D. Sleep apnea in patients with transient ischemic attack and stroke: a prospective study of 59 patients. Neurology. 1996;47(5):1167–1173. [PubMed]

6 

Besson H, Ekelund U, Luan J, et al. A cross-sectional analysis of physical activity and obesity indicators in European participants of the EPIC-PANACEA study. Int J Obes (Lond). 2009;33(4):497–506

7 

Pate RR, Taverno Ross SE, Liese AD, Dowda M. Associations among physical activity, diet quality, and weight status in US adults. Med Sci Sports Exerc. 2015;47(4):743–750. [PubMed Central][PubMed]

8 

Van Dyck D, Cerin E, De Bourdeaudhuij I, et al. International study of objectively measured physical activity and sedentary time with body mass index and obesity: IPEN adult study. Int J Obes (Lond). 2015;39(2):199–207

9 

Bann D, Hire D, Manini T, et al. Light intensity physical activity and sedentary behavior in relation to body mass index and grip strength in older adults: cross-sectional findings from the Lifestyle Interventions and Independence for Elders (LIFE) study. PLoS One. 2015;10(2):e0116058[PubMed Central][PubMed]

10 

Healy GN, Wijndaele K, Dunstan DW, et al. Objectively measured sedentary time, physical activity, and metabolic risk: the Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Diabetes Care. 2008;31(2):369–371. [PubMed]

11 

Dunton GF, Berrigan D, Ballard-Barbash R, Graubard B, Atienza AA. Joint associations of physical activity and sedentary behaviors with body mass index: results from a time use survey of US adults. Int J Obes (Lond). 2009;33(12):1427–1436

12 

Kwon S, Wang M, Hawkins M. Association between self-reported physical activity and obesity among White, Black, Hispanic, and Asian Americans: 2007 and 2009 brfss. Ethn Dis. 2013;23(2):129–135. [PubMed]

13 

Murabito JM, Pedley A, Massaro JM, et al. Moderate-to-vigorous physical activity with accelerometry is associated with visceral adipose tissue in adults. J Am Heart Assoc. 2015;4(3):e001379[PubMed Central][PubMed]

14 

Hemmingsson E, Ekelund U. Is the association between physical activity and body mass index obesity dependent? Int J Obes (Lond). 2007;31(4):663–668

15 

Kim Y, White T, Wijndaele K, Sharp SJ, Wareham NJ, Brage S. Adiposity and grip strength as long-term predictors of objectively measured physical activity in 93,015 adults: the UK Biobank study. Int J Obes (Lond). 2017;41(9):1361–1368

16 

Golubic R, Ekelund U, Wijndaele K, et al. Rate of weight gain predicts change in physical activity levels: a longitudinal analysis of the EPIC-Norfolk cohort. Int J Obes (Lond). 2013;37(3):404–409

17 

Golubic R, Wijndaele K, Sharp SJ, et al. Physical activity, sedentary time and gain in overall and central body fat: 7-year follow-up of the ProActive trial cohort. Int J Obes (Lond). 2015;39(1):142–148

18 

Petersen L, Schnohr P, Sorensen TI. Longitudinal study of the long-term relation between physical activity and obesity in adults. Int J Obes Relat Metab Disord. 2004;28(1):105–112. [PubMed]

19 

Bak H, Petersen L, Sorensen TI. Physical activity in relation to development and maintenance of obesity in men with and without juvenile onset obesity. Int J Obes Relat Metab Disord. 2004;28(1):99–104. [PubMed]

20 

Cheung YY, Tai BC, Loo G, et al. Screening for obstructive sleep apnea in the assessment of coronary risk. Am J Cardiol. 2017;119(7):996–1002. [PubMed]

21 

Vanhecke TE, Franklin BA, Zalesin KC, et al. Cardiorespiratory fitness and obstructive sleep apnea syndrome in morbidly obese patients. Chest. 2008;134(3):539–545. [PubMed]

22 

Przybylowski T, Bielicki P, Kumor M, et al. Exercise capacity in patients with obstructive sleep apnea syndrome. J Physiol Pharmacol. 2007;58 Suppl 5:(Pt 2):563–574. [PubMed]

23 

Vanuxem D, Badier M, Guillot C, Delpierre S, Jahjah F, Vanuxem P. Impairment of muscle energy metabolism in patients with sleep apnoea syndrome. Respir Med. 1997;91(9):551–557. [PubMed]

24 

Chasens ER, Umlauf MG, Weaver TE. Sleepiness, physical activity, and functional outcomes in veterans with type 2 diabetes. Appl Nurs Res. 2009;22(3):176–182. [PubMed]

25 

Quan SF, O'Connor GT, Quan JS, et al. Association of physical activity with sleep-disordered breathing. Sleep Breath. 2007;11(3):149–157. [PubMed]

26 

Simpson L, McArdle N, Eastwood PR, et al. Physical inactivity is associated with moderate-severe obstructive sleep apnea. J Clin Sleep Med. 2015;11(10):1091–1099. [PubMed Central][PubMed]

27 

Chasens ER, Sereika SM, Houze MP, Strollo PJ. Subjective and objective appraisal of activity in adults with obstructive sleep apnea. J Aging Res. 2011;2011:751819[PubMed Central][PubMed]

28 

Verwimp J, Ameye L, Bruyneel M. Correlation between sleep parameters, physical activity and quality of life in somnolent moderate to severe obstructive sleep apnea adult patients. Sleep Breath. 2013;17(3):1039–1046. [PubMed]

29 

Mendelson M, Tamisier R, Laplaud D, et al. Low physical activity is a determinant for elevated blood pressure in high cardiovascular risk obstructive sleep apnea. Respir Care. 2014;59(8):1218–1227. [PubMed]

30 

Kushida CA, Nichols DA, Holmes TH, et al. Effects of continuous positive airway pressure on neurocognitive function in obstructive sleep apnea patients: The Apnea Positive Pressure Long-term Efficacy Study (APPLES). Sleep. 2012;35(12):1593–1602. [PubMed Central][PubMed]

31 

Weaver TE, Maislin G, Dinges DF, et al. Relationship between hours of CPAP use and achieving normal levels of sleepiness and daily functioning. Sleep. 2007;30(6):711–719. [PubMed Central][PubMed]

32 

Antic NA, Catcheside P, Buchan C, et al. The effect of CPAP in normalizing daytime sleepiness, quality of life, and neurocognitive function in patients with moderate to severe OSA. Sleep. 2011;34(1):111–119. [PubMed Central][PubMed]

33 

Hla KM, Skatrud JB, Finn L, Palta M, Young T. The effect of correction of sleep-disordered breathing on BP in untreated hypertension. Chest. 2002;122(4):1125–1132. [PubMed]

34 

Martinez-Garcia MA, Capote F, Campos-Rodriguez F, et al. Effect of CPAP on blood pressure in patients with obstructive sleep apnea and resistant hypertension: the HIPARCO randomized clinical trial. JAMA. 2013;310(22):2407–2415. [PubMed]

35 

Duran-Cantolla J, Aizpuru F, Montserrat JM, et al. Continuous positive airway pressure as treatment for systemic hypertension in people with obstructive sleep apnoea: randomised controlled trial. BMJ. 2010;341:c5991[PubMed]

36 

Steiner S, Schueller PO, Schannwell CM, Hennersdorf M, Strauer BE. Effects of continuous positive airway pressure on exercise capacity in chronic heart failure patients without sleep apnea. J Physiol Pharmacol. 2007;58 Suppl 5:(Pt 2):665–672. [PubMed]

37 

De Sant'anna M Jr, Mafra Moreno A, Da Silva Soares PP. The effects of positive airway pressure on cardiovascular responses, perceived effort and time to exhaustion during high-intensity exercise in healthy subjects. J Sports Med Phys Fitness. 2013;53(6):604–611. [PubMed]

38 

Drager LF, Brunoni AR, Jenner R, Lorenzi-Filho G, Bensenor IM, Lotufo PA. Effects of CPAP on body weight in patients with obstructive sleep apnoea: a meta-analysis of randomised trials. Thorax. 2015;70(3):258–264. [PubMed]

39 

Quan SF, Budhiraja R, Clarke DP, et al. Impact of treatment with continuous positive airway pressure (CPAP) on weight in obstructive sleep apnea. J Clin Sleep Med. 2013;9(10):989–993. [PubMed Central][PubMed]

40 

Keenan BT, Maislin G, Sunwoo BY, et al. Obstructive sleep apnoea treatment and fasting lipids: a comparative effectiveness study. Eur Respir J. 2014;44(2):405–414. [PubMed]

41 

West SD, Kohler M, Nicoll DJ, Stradling JR. The effect of continuous positive airway pressure treatment on physical activity in patients with obstructive sleep apnoea: A randomised controlled trial. Sleep Med. 2009;10(9):1056–1058. [PubMed]

42 

Batool-Anwar S, Goodwin JL, Drescher AA, et al. Impact of CPAP on activity patterns and diet in patients with obstructive sleep apnea (OSA). J Clin Sleep Med. 2014;10(5):465–472. [PubMed Central][PubMed]

43 

Chasens ER, Korytkowski M, Sereika SM, Burke LE, Drumheller OJ, Strollo PJ Jr. Improving activity in adults with diabetes and coexisting obstructive sleep apnea. West J Nurs Res. 2014;36(3):294–311. [PubMed]

44 

Bamberga M, Rizzi M, Gadaleta F, Grechi A, Baiardini R, Fanfulla F. Relationship between energy expenditure, physical activity and weight loss during CPAP treatment in obese OSA subjects. Respir Med. 2015;109(4):540–545. [PubMed]

45 

Jean RE, Duttuluri M, Gibson CD, et al. Improvement in physical activity in persons with obstructive sleep apnea treated with continuous positive airway pressure. J Phys Act Health. 2017;14(3):176–182. [PubMed]

46 

Murphy SL. Review of physical activity measurement using accelerometers in older adults: considerations for research design and conduct. Prev Med. 2009;48(2):108–114. [PubMed]

47 

Trost SG, McIver KL, Pate RR. Conducting accelerometer-based activity assessments in field-based research. Med Sci Sports Exerc. 2005;37 11 Suppl:S531–S543. [PubMed]

48 

Plasqui G, Westerterp KR. Physical activity assessment with accelerometers: an evaluation against doubly labeled water. Obesity (Silver Spring). 2007;15(10):2371–2379

49 

Johns MW. Reliability and factor analysis of the Epworth Sleepiness Scale. Sleep. 1992;15(4):376–381. [PubMed]

50 

Kushida CA, Littner MR, Morgenthaler T, et al. Practice parameters for indications for polysomnography and related procedures: an update for 2005. Sleep. 2005;28(4):499–521. [PubMed]

51 

Iber C, Ancoli-Israel S, Chesson AL Jr, Quan SF; for the American Academy of Sleep Medicine. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. 1st ed. Westchester, IL: American Academy of Sleep Medicine; 2007.

52 

Kushida CA, Chediak A, Berry RB, et al. Clinical guidelines for the manual titration of positive airway pressure in patients with obstructive sleep apnea. J Clin Sleep Med. 2008;4(2):157–171. [PubMed Central][PubMed]

53 

Freedson PS, Melanson E, Sirard J. Calibration of the Computer Science and Applications, Inc. accelerometer. Med Sci Sports Exerc. 1998;30(5):777–781. [PubMed]

54 

Patel SR, Weng J, Rueschman M, et al. Reproducibility of a standardized actigraphy scoring algorithm for sleep in a US Hispanic/Latino population. Sleep. 2015;38(9):1497–1503. [PubMed Central][PubMed]

55 

Huang Y, Hsu JC. Hochberg's Step-Up Method: cutting corners off Holm's Step-Down Method. Biometrika. 2007;94(4):965–975

56 

Redenius R, Murphy C, O'Neill E, Al-Hamwi M, Zallek SN. Does CPAP lead to change in BMI? J Clin Sleep Med. 2008;4(3):205–209. [PubMed Central][PubMed]

57 

Bravata DM, Smith-Spangler C, Sundaram V, et al. Using pedometers to increase physical activity and improve health: a systematic review. JAMA. 2007;298(19):2296–2304. [PubMed]

58 

Kline CE, Reboussin DM, Foster GD, et al. The effect of changes in cardiorespiratory fitness and weight on obstructive sleep apnea severity in overweight adults with type 2 diabetes. Sleep. 2016;39(2):317–325. [PubMed Central][PubMed]

59 

Tachikawa R, Ikeda K, Minami T, et al. Changes in energy metabolism after continuous positive airway pressure for obstructive sleep apnea. Am J Respir Crit Care Med. 2016;194(6):729–738. [PubMed]

60 

Mendelson M, Vivodtzev I, Tamisier R, et al. CPAP treatment supported by telemedicine does not improve blood pressure in high cardiovascular risk OSA patients: a randomized, controlled trial. Sleep. 2014;37(11):1863–1870. [PubMed Central][PubMed]

61 

Diamanti C, Manali E, Ginieri-Coccossis M, et al. Depression, physical activity, energy consumption, and quality of life in OSA patients before and after CPAP treatment. Sleep Breath. 2013;17(4):1159–1168. [PubMed]

62 

John D, Freedson P. ActiGraph and Actical physical activity monitors: a peek under the hood. Med Sci Sports Exerc. 2012;44 1 Suppl 1:S86–S89. [PubMed Central][PubMed]

63 

Going SB. Physical activity measurements: lessons learned from the pathways study. J Public Health Manag Pract. 2010;16(5):420–425. [PubMed Central][PubMed]

64 

Swartz AM, Strath SJ, Bassett DR Jr, O'Brien WL, King GA, Ainsworth BE. Estimation of energy expenditure using CSA accelerometers at hip and wrist sites. Med Sci Sports Exerc. 2000;32 9 Suppl:S450–S456. [PubMed]

65 

Welch WA, Bassett DR, Thompson DL, et al. Classification accuracy of the wrist-worn gravity estimator of normal everyday activity accelerometer. Med Sci Sports Exerc. 2013;45(10):2012–2019. [PubMed Central][PubMed]


Supplemental Material

Supplemental Material

(.pdf | 114 KB)