American Journal of Respiratory and Critical Care Medicine

Neither the association between obstructive airways disease (OAD) and sleep apnea–hypopnea (SAH) nor the sleep consequences of each disorder alone and together have been characterized in an adult community setting. Our primary aims were (1) to determine if there is an association between OAD and SAH and (2) identify predictors of oxyhemoglobin desaturation during sleep in persons having OAD with and without SAH. Polysomnography and spirometry results from 5,954 participants in the Sleep Heart Health Study were analyzed. OAD was defined by a FEV1/FVC value less than 70%. Assessment of SAH prevalence in OAD was performed using thresholds of respiratory disturbance index (RDI) greater than 10 and greater than 15. A total of 1,132 participants had OAD that was predominantly mild (FEV1/FVC 63.81 ± 6.56%, mean ± SD). SAH was not more prevalent in participants with OAD than in those without OAD (22.32 versus 28.86%, with and without OAD, respectively, at RDI threshold values greater than 10; and 13.97 versus 18.63%, with and without OAD, respectively, at RDI threshold value greater than 15). In the absence of SAH, the adjusted odds ratio for sleep desaturation (> 5% total sleep time with saturation < 90%) was greater than 1.9 when FEV1/FVC was less than 65%. Participants with both OAD and SAH had greater sleep perturbation and desaturation than those with one disorder. Generally mild OAD alone was associated with minimally altered sleep quality. We conclude that (1) there is no association between generally mild OAD and SAH; (2) exclusive of SAH and after adjusting for demographic factors and awake oxyhemoglobin saturation, an FEV1/FVC value less than 65% is associated with increased risk of sleep desaturation; (3) desaturation is greater in persons with both OAD and SAH compared with each of these alone; and (4) individuals with generally mild OAD and without SAH in the community have minimally perturbed sleep.

Obstructive airway disease (OAD) has been estimated to affect 14 to 16 million individuals in the United States, causing substantial morbidity and mortality (14). Sleep apnea–hypopnea (SAH) is also prevalent in the community (510). By chance alone, some persons would be expected to have both conditions, previously termed an “overlap syndrome” (11). Chaouat and coworkers have suggested that the prevalence of OAD in patients with SAH exceeds the prevalence of OAD in the general population (12). Conversely, an unexpectedly high prevalence of SAH has also been reported in patients with OAD (13).

A putative association between OAD and SAH could be due to the role of tobacco smoking. Some, but not all, studies have suggested that tobacco use is a risk factor for both entities (1420). In addition, OAD has been associated with nocturnal hypoxemia, poor sleep quality, and insufficient or disrupted sleep (2125). The sleep-related physiologic disturbances in patients with OAD may be relevant to the pathogenesis of SAH. Some investigations have implied that these disturbances may be associated with abnormal ventilatory control and upper airway instability during sleep (2631). A number of studies have suggested that the presence of both OAD and SAH leads to greater blood gas and pulmonary hemodynamic perturbations than found in individuals with OAD or SAH alone (12, 3235), thereby increasing risk for cor pulmonale. On the basis of these studies as well as data suggesting a specific association between OAD and SAH, some authors have suggested that a diagnostic evaluation for OAD should be conducted in all SAH patients (12).

Prior studies of breathing during sleep and objectively assessed sleep quality in patients with OAD have evaluated relatively small samples, have focused on patients with severe disease, or have included individuals referred to sleep laboratories or pulmonary clinics, introducing the potential of a selection bias (2124). The baseline data of the Sleep Heart Health Study (SHHS), which includes polysomnography (PSG) and spirometry results on a large and diverse middle-aged to elderly community population, provide a unique opportunity to examine the interrelationships between OAD, sleep, and SAH. The two primary objectives of this analysis were (1) to test the hypothesis that OAD and SAH coexist more often in a community sample than would be expected by chance alone, and (2) to identify predictors of sleep-related oxyhemoglobin desaturation in community-dwelling OAD patients with and without objective evidence of SAH. The secondary aim of this analysis was to characterize sleep quality and architecture in OAD patients in the community who do not have SAH.

Study Participants

This investigation uses data from the SHHS, a prospective multicenter cohort study that was initiated to assess whether sleep-disordered breathing is a risk factor for hypertension and cardiovascular disease in adults. Details regarding the design and methodology have been reported (3638). In brief, participants were recruited from ongoing cohort studies, including the Cardiovascular Health Study (39), the Framingham Heart Study (40), the Tucson Epidemiologic Study of Obstructive Airways Disease (41), the Strong Heart Study (42), the Atherosclerosis Risk in Communities Study (43), the Health and Environment Cohort Study in Tucson, as well as three New York City cohorts undergoing evaluation for the impact of psychosocial risk factors on cardiovascular disease. Criteria for inclusion in SHHS included age of 40 years or older and having not received positive airway pressure treatment for sleep apnea, a tracheostomy, or supplemental oxygen. Oversampling of participants younger than 65 years who had a history of snoring was conducted to increase the prevalence of SAH in the study group and optimize the statistical power. Participants underwent unattended PSG at home between December 1995 and January 1998.


Unattended home PSGs were performed with the Compumedics PS-2 system (Compumedics Pty. Ltd, Abbotsville, Australia). The recorded variables included the electroencephalogram (montage: C3/A1 and C4/A2), right and left electrooculograms, submental electromyogram using bipolar electrodes, nasal/oral airflow recorded by thermocouple (Protec, Woodenville, WA), rib cage and abdominal movement recorded by inductive plethysmography, oxyhemoglobin saturation (SpO2) by pulse oximetry (Nonin, Minneapolis, MN), and electrocardiogram using a bipolar lead. In addition, a mercury gauge recorded body position, and a light sensor recorded ambient light. Leg movements were not recorded. Standardized techniques for sensor attachment and quality assurance were used and have been previously described (36, 44). PSG data were stored on PCMCIA cards and sent to a central scoring center.

PSG Scoring and Sleep Parameters

Scoring of sleep stages followed the guidelines of Rechtschaffen and Kales (45). Arousals were identified according to criteria published by Bonnet and coworkers (46). Apnea was defined as complete or almost complete cessation of airflow (to ⩽ 25% of baseline) and associated with 4% desaturation or more. Hypopnea was defined as a decrease to less than 70% of baseline on either inductance plethysmography channel or thermocouple channels, for 10 seconds or longer and associated with 4% desaturation or more. The Respiratory Disturbance Index (RDI) was calculated by computing the average number of apneas plus hypopneas per hour of sleep.

The arousal index was calculated as the average number of arousals per hour of sleep. Approximately 15% of studies were identified by the scorer to have potentially unreliable arousal data due to difficulties in discerning episodic changes in electroencephalogram from fluctuations in background electroencephalogram or less often because of difficulties in reliably distinguishing specific sleep stages due to excessive electroencephalogram artifact. Therefore, analyses of arousal data were restricted to 4,670 individuals in whom the electroencephalogram data were deemed to be of sufficient quality to permit reliable identification of arousals. Sleep latency was defined as the interval between “lights-out” and the first three consecutive 30-second intervals (epochs) scored as sleep. This measure was available for 3,612 participants in whom the light meter indicated a clear transition of ambient light before sleep onset. Sleep efficiency was defined as the percent of the Total Sleep Time (TST) divided by the time from lights-out until the final morning awakening. Analyses of sleep efficiency were restricted to those individuals in whom the total sleep period was considered to have been captured during the recording (i.e., the subject awoke before termination of the recording, and no intervening periods of lost data were identified between sleep onset and final awakening).


Spirometry was performed either as part of the parent study protocol or specifically for SHHS in accordance with published guidelines (39, 4751).

Questionnaire and Demographic Data

At the time of the PSG, participants completed a health questionnaire that included queries regarding underlying medical conditions (including OAD) and relevant exposures such as smoking. The questionnaire also included the Epworth Sleepiness Scale, which was used to assess sleep propensity during normal daily activities (36, 5254). Participants' weight was measured using a calibrated scale at the time of PSG. Race was determined by self-report. Participants' height was taken from parent study data obtained within 3 years of the sleep study and used to calculate the body mass index (BMI) (kg/m2).

Statistical Analyses

Home PSG was performed on 6,443 SHHS participants (including data from two individuals aged 37 and 39 years). One hundred and twenty four participants (2%) with self-reported congestive heart failure were excluded from analyses. Of the remaining participants, 5,954 had complete spirometry data. OAD was defined as the ratio of the FEV1 to FVC being less than 70% (5557). Predicted values for the FEV1 were calculated from the formulas of Hankinson and colleagues (58). The formula described for white individuals was applied to the “other” category of our study population (usually reflecting participants of mixed ancestry and those who were uncertain of their racial background).

Because of the variable time interval between spirometry and PSG (3.03 ± 2.4 years, mean ± SD, before PSG), analyses were repeated with the data obtained from the 3,496 (59%) participants who had spirometry within 3 years of the PSG. Results based on the restricted population were similar to those obtained from the entire sample, and therefore analyses based on the entire sample are presented.

Additional statistical methodology is described in the online supplement. All analyses were performed using SAS for Windows Version 6.12. Data are presented as mean ± SD.

The SHHS has been approved by the institutional review board at each participating site. Informed written consent has been obtained from all participants.

Characteristics of the Study Population

The overall FEV1/FVC for the sample population was 75.5 ± 7.9% (mean ± SD). Participants classified as having OAD had a mean FEV1/FVC of 63.8 ± 6.6%, and those without OAD had a mean FEV1/FVC of 78.3 ± 5.3%. The distribution of FEV1/FVC ratio values across the entire study population shows that only a small number of participants (n = 226, 3.8%) had an FEV1/FVC ratio less than 60% (Figure 1)


The characteristics of participants with and without spirometric evidence of OAD regardless of SAH status are shown in Table 1

TABLE 1. Characteristics of entire study population by spirometric evidence of obstructive airway disease

FEV1/FVC ⩾ 70%
 (n = 4,816)

FEV1/FVC < 70%
 (n = 1,138)
Age, yr (mean ± SD)*62.24 ± 10.966.46 ± 9.79
Men, %44.9557.21
Race, %
African American7.608.44
Native American9.269.49
BMI, (mean ± SD)*28.78 ± 5.4427.42 ± 4.96
Smoking status, %§

*p < 0.0001.

p < 0.001.

p = 0.0089.

§p = 0.005, overall p value difference for the χ-square test.

Definition of abbreviation: BMI = body mass index.

. The characteristics of those individuals with and without OAD who did not have SAH (RDI ⩽ 15) are shown in Table 2

TABLE 2. Characteristics of the study population by spirometric evidence of obstructive airways disease, excluding those with respiratory disturbance index greater than 15

FEV1/FVC ⩾ 70%
 (n = 3,917)

FEV1/FVC < 70%
 (n = 979)
Age, yr (mean ± SD)*61.75 ± 11.066.32 ± 9.96
Men, %*40.5953.12
Race, %
African American7.588.48
Native American8.509.40
BMI, (mean ± SD)*28.11 ± 5.0926.92 ± 4.64
Smoking status (%)

*p < 0.0001.

p = 0.0071.

p < 0.001.

For definition of abbreviations see Table 1.


Considering the entire study group (regardless of SAH status), participants with OAD (FEV1/FVC < 70%) were significantly older, more likely to be male, and had a lower mean BMI compared with those without OAD (Table 1). In the absence of SAH, participants with OAD were predominantly male and had a lower BMI than those without OAD (Table 2).

Association between OAD and SAH

Participants with OAD had a significantly lower mean and median RDI than those without OAD. In addition, the percentage of participants with an RDI greater than 10 and an RDI greater than 15 was significantly lower in the group of participants with OAD compared with those without OAD (Table 3)

TABLE 3. Respiratory disturbance index by spirometric evidence of obstructive airways disease

FEV1/FVC ⩾ 70%
 (n = 4,816)

FEV1/FVC < 70%
 (n = 1,138)
Mean ± SD9.13 ± 12.597.49 ± 11.87*
Median (interquartile range)4.51 (1.36, 11.59)3.51 (1.35, 8.81)
Participants with RDI > 10 events/hr, %28.8622.32*
Participants with RDI > 15 events/hr, %

Median (interquartile range) is presented due to the skewed RDI distribution.

*p < 0.0001.

p < 0.001.

§p < 0.0002.

Definition of abbreviations: RDI = respiratory disturbance index.

. However, after stratification by BMI quartile, RDI values were similar in the participants with and without OAD (Table 4)

TABLE 4. Median respiratory disturbance index (interquartile range) determined by spirometric evidence of obstructive airway disease and body mass index quartiles

BMI Quartile (Interquartile Range)

Median RDI (Range)
 FEV1/FVC ⩾ 70%
 (n = 4,816)

Median RDI (Range)
 FEV1/FVC < 70%
 (n = 1,138)
1 (9.78–24.84)1.94 (0.52–6.26)1.79 (0.64–5.02)*
2 (24.85–27.76)3.37 (1.08–8.99)2.89 (1.04–6.64)*
3 (27.77–31.20)5.18 (1.95–11.64)4.7 (1.83–9.24)*
4 (31.21–58.88)
8.62 (3.75–18.17)
7.96 (3.17–17.12)*

*No significant difference.

Definition of abbreviations: BMI = body mass index; RDI = respiratory disturbance index.

. As expected, the RDI increased with higher BMI quartile in participants with and without OAD.

The relationship between RDI and FEV1/FVC was also evaluated by analysis of variance after adjusting for BMI, age, sex, and race as well as smoking status. The multivariable models revealed no significant interaction between BMI and FEV1/FVC (p = 0.40) in their relationship with RDI. However, there was a significant positive relationship between RDI and the FEV1/FVC ratio (β = 2.05, 95% confidence interval = 0.44 to 3.66), suggesting that a lower FEV1/FVC ratio is independently associated with a lower RDI.

We considered the possibility that some normal, elderly individuals might be misclassified as having OAD by defining this disorder as existing when the FEV1/FVC ratio is less than 70%. Therefore, we also examined and compared the mean and median RDI, as well as the percentage of participants with RDI greater than 10 and RDI greater than 15 employing a threshold value of FEV1/FVC at less than 65% versus greater than or equal to 65%. The results were similar to those described previously.

The Impact of OAD on Sleepiness and Sleep Variables in Participants with and without SAH

The Epworth Sleepiness Scores (ESS) and sleep variables in participants with OAD only, SAH only, neither disorder, and both disorders are shown in Table E1 in the online supplement. In the absence of SAH, after adjusting for age, sex, height, weight, race, and smoking status, there were statistically significant but small differences between participants with and without OAD with regard to TST, but no differences were observed with regard to ESS, sleep latency, sleep efficiency, and %TST spent in rapid eye movement or in Stages 1, 2, 3/4, or the arousal index (Table E1 in the online supplement, comparing values in Columns 1 and 2).

To examine the impact of coexistent OAD and SAH on sleepiness and sleep architecture, we compared data from participants having both disorders with data from participants with each disorder alone. After adjusting for age, sex, height, weight, race, and smoking status, significant but small differences were observed between participants who had SAH alone and those who had both disorders (SAH + OAD). The former group had higher sleep efficiency and lower %TST in Stage 1 (see Table E1 in the online supplement). In contrast, differences between participants with both SAH and OAD and those with OAD alone were expressed more broadly across sleep variables (see Table E1 in the online supplement). After adjusting for age, sex, height, weight, race, and smoking status, participants who had both SAH and OAD had significantly higher ESS, lower TST, lower sleep efficiency, lower %TST in Stages rapid eye movement and 3/4 sleep, greater %TST in Stage 2 sleep, and higher arousal index than those with OAD alone.

Comparison of participants with single disorders (e.g., OAD only, SAH only) (see Table E1 in the online supplement) indicated that those with SAH alone had greater perceived sleepiness by ESS, lower %TST in rapid eye movement as well as in Stage 3/4 sleep, and greater %TST in Stage 2 sleep than those with OAD alone. In addition, individuals with SAH alone had a markedly higher arousal index than those with only OAD. Thus, the most notable differences in sleep variables across the four participant groups were reflected in the comparisons between groups with and without SAH, regardless of whether or not there was coexistent OAD.

To determine if sleepiness and sleep architecture are influenced by the severity of OAD in the absence of SAH, data from the 976 participants with OAD (defined by an FEV1/FVC ratio < 70%) but without SAH were analyzed by quartile of percent-predicted FEV1. TST and sleep efficiency were slightly but statistically less in the lowest compared with highest FEV1 quartile. No other significant differences were noted from the highest to lowest percent-predicted FEV1 quartiles (see Table E2 in the online supplement). Similar findings were obtained from analyses of the 421 participants with OAD, defined as existing when the FEV1/FVC ratio is less than 65% (see Table E3 in the online supplement).

The Impact of OAD on Sleep-related Oxyhemoglobin Saturation in Participants with and without Coexistent SAH

After adjusting for age, sex, height, weight, smoking status, and awake SpO2, the odds ratios (OR) for experiencing more than 5% of TST with SpO2 less than 90% were calculated across a range of FEV1/FVC values. In the absence of SAH, the adjusted OR for nocturnal oxyhemoglobin desaturation increased at levels of FEV1/FVC below 65% (Table 5)

TABLE 5. Adjusted odds ratio* for greater than 5% total sleep time with oxyhemoglobin saturation less than 90% by fev1/fvc, excluding participants with respiratory disturbance index greater than 15

FEV1/FVC (%)


Adjusted Odds Ratio
 (95% Confidence Interval)
⩾ 801,2981.00 (Reference)
75.0–79.91,4580.92 (0.60, 1.44)
70.0–74.91,1551.01 (0.66, 1.56)
65.0–69.95521.32 (0.81, 2.14)
60–64.92241.92 (1.10, 3.34)
< 60
3.36 (1.98, 5.70)

*Adjusted for age, sex, height, weight, smoking, and awake oxyhemoglobin saturation.

. In participants with an FEV1/FVC of 60 to 65% the adjusted OR for desaturation to less than 90% for more than 5% TST was 1.92 (confidence interval: 1.1, 3.34). The adjusted OR conferred by an FEV1/FVC less than 60% was 3.36 (confidence interval: 1.98, 5.7).

To examine the degree to which OAD and SAH independently and conjointly contribute to desaturation during sleep we assessed the risk for spending more than 5% of TST with SpO2 less than 90% and less than 85%, respectively, in the presence of single and combined disorders. After adjusting for age, sex, height, weight, race, smoking status, and awake SpO2, the OR for oxyhemoglobin desaturation below threshold levels of less than 90% and less than 85% for more than 5% of TST was considerably increased in the presence of SAH, with a relatively lower OR conferred by OAD in the absence of SAH (Table 6)

TABLE 6. Adjusted odds ratio of desaturation based on obstructive airway disease and sleep apnea–hypopnea status

SAH (+)

SAH ()

OAD (+)
 (n = 254)
OAD ()
 (n = 897)
OAD (+)
 (n = 884)
OAD ()
 (n = 3,919)
> 5% TST spent with SpO2 < 90%
People, %*42.9147.9411.436.30
Odds ratio (CI),8.06 (5.55, 11.69)8.98 (6.86, 11.74)1.80 (1.33, 2.45)1.0 (Reference)
Odds ratio (CI)§8.28 (5.78, 11.86)9.26 (7.14, 12.02)1.89 (1.40, 2.54)
> 5% TST spent with SpO2 < 85%
People, %*11.0210.590.790.41
Odds ratio (CI),30.08 (13.21, 73.18)15.83 (7.23, 34.67)3.15 (1.07, 9.26)1.0 (Reference)
Odds ratio (CI)§
28.73 (12.56, 65.70)
15.18 (7.17, 32.14)
2.85 (0.99, 8.18)

*Overall chi-square comparison significant at < 0.0001 level.

OR (95% CI) adjusted for age, sex, height, weight, race, smoking status (former and current); comparison group for each is −OAD/−SAH.

Adjusted for awake oxyhemoglobin saturation.

§Unadjusted for awake oxyhemoglobin saturation.

Definition of abbreviations: CI = confidence interval; OAD = obstructive airways disease; SAH = sleep apnea-hypopnea; TST = total sleep time.

. The OR for desaturation below 85% for greater than 5% TST was approximately 20-fold greater in participants with SAH alone compared with those who had neither disorder, and increased to approximately 30-fold in participants with both disorders (Table 6). A separate model examining the individual effects of OAD and SAH revealed no interaction between the two disorders. Thus, the observed effect of coexistent disorders on sleep desaturation was not greater than that which is expected based on the individual risks conferred by each alone.

The principal results of our study are (1) the prevalence of SAH, defined either as an RDI greater than 10 or greater than 15 is not greater in community-dwelling adults with objective evidence of predominantly mild OAD than in those without OAD; (2) in the absence of SAH, individuals with objective evidence of generally mild OAD do not perceive themselves to have greater sleep propensity during usual daily conditions than those without OAD; (3) in the absence of SAH, there are only minor differences in sleep quality and architecture between community-dwelling adults with generally mild OAD and those without OAD, and these may not be clinically significant; (4) independent of SAH and awake SpO2, the presence of OAD characterized by an FEV1/FVC ratio less than 65% is associated with an increased risk of oxyhemoglobin desaturation during sleep; and (5) the proportion of participants with notable sleep desaturation as well as the degree to which sleep is perturbed is greater in the presence of both disorders but is largely related to the contribution of SAH.

Our analyses support the hypothesis that when generally mild OAD and SAH coexist, it is on the basis of aggregation by chance rather than through a pathophysiologic linkage. In fact, the prevalence of individuals with an RDI greater than 10 or greater than 15 was lower in adults with OAD, compared with those without OAD. Although the participants with OAD had a significantly lower BMI, this did not completely explain the difference in RDI between the groups. Across all BMI strata, the median RDI was consistently, although not significantly, lower in participants with OAD. Furthermore, after accounting for age, race, sex, resting SpO2, and BMI, RDI tended to be lower as the FEV1/FVC ratio decreased. This surprising observation requires confirmation by further studies.

OAD and SAH are both prevalent health problems with well-defined effects on health-related quality of life and cognitive function (5965). A clinical view that OAD is inherently associated with poor sleep quality, daytime sleepiness, insomnia, and nocturnal desaturation has evolved from some investigations addressing these issues (24, 25, 6675). The SHHS study participants represent a community population that was not identified by virtue of health care–seeking efforts and that has generally mild OAD, with a relatively small proportion of individuals having more severe disease. This large study population provides a unique opportunity to resolve existing controversies regarding the impact of mild OAD, with and without SAH, on sleep quality.

Our data suggest that after excluding individuals with SAH, community-dwelling adults with objective evidence of OAD do not perceive themselves to be sleepier than those without OAD, at least as measured by ESS. Moreover, although individuals with OAD alone had a shorter TST, there were no other significant differences in sleep architecture compared with participants with neither OAD nor SAH.

Although in general, the study population of OAD patients had predominantly mild disease severity, some individuals had more severe degrees of obstruction, with 3.8% of participants having FEV1/FVC less than 60%. Aside from a shorter TST, there were no apparent trends regarding ESS or parameters of sleep architecture from the highest to the lowest percent-predicted FEV1 quartiles either in participants with an FEV1/FVC less than 70% or in those with an FEV1/FVC less than 65%. This suggests no difference in these regards with increasing OAD severity (see Tables E2 and E3 in the online supplement). To the extent that comparisons of certain quartile pairs yielded significant differences at p values less than 0.05, it should be recalled that no correction was made for multiple comparisons, reinforcing the absence of meaningful differences. In this light, although not normal, sleep architecture was remarkably well preserved in those SHHS participants with generally mild OAD, suggesting that clinicians should not conclude that sleep-related complaints in patients with similar degrees of OAD are attributable to the underlying OAD per se without exploring other diagnostic possibilities.

Identifying patients who experience sleep-related oxyhemoglobin desaturation is clinically important due to the known adverse physiologic and clinical health consequences of hypoxemia as well as the therapeutic benefits of supplemental oxygen. It is noteworthy in this regard that the threshold degree and duration of sleep-related desaturation that results in adverse health outcomes has not been established. As recently described by Gries and Brooks (76), there is considerable variability in reported normative values of sleep-related SpO2 during sleep. Based on evidence of reduced survival in OAD patients who experienced overnight desaturation to less than 90% for 5 minutes, reaching a nadir of less than or equal to 85% at least once (77, 78), we elected to examine the prevalence of desaturation across our study groups employing the criteria of greater than 5% of TST spent with SpO2 less than 90% and less than 85%. Uncertainties regarding the threshold defining unacceptable sleep-related oxyhemoglobin desaturation notwithstanding, considerable efforts have been made to predict which OAD patients experience hypoxemia during sleep (23, 7983). Many, but not all of these, studies found that awake SpO2 was the best predictor of sleep-related desaturation. We observed that an FEV1/FVC ratio less than 65% was associated with a considerably increased OR for spending more than 5% TST with SpO2 less than 90% even after adjusting for awake oxyhemoglobin saturation and other factors (Table 6). Thus, our observations suggest that even when SAH is not clinically suspected, overnight oximetry should be considered in OAD patients with this degree of disease.

Our data reinforce the results of previous studies indicating greater perturbations of sleep quality and architecture and increased risk of notable sleep desaturation in individuals with both SAH and even generally mild OAD (overlap syndrome) relative to each disorder alone. The abnormalities are mostly, although not exclusively, attributable to the SAH. Thus, although OAD is not a risk factor for SAH or the converse, when one of these disorders is diagnosed, clinicians should consider the possible independent presence of the other due to the risk for additional adverse physiologic impact.

Considerations and Potential Limitations of the Study

Several issues and potential limitations were considered in interpreting our data: (1) the SHHS population in this study consisted of community-dwelling individuals, none of whom were under treatment with continuous positive airway pressure or supplemental oxygen. The relatively low percentage of subjects with severe OAD may limit the generalizability of our findings to more severely impaired individuals. Nonetheless, our data are representative of a heterogeneous middle-aged and older community population, unencumbered by potential selection bias associated with recruitment of individuals who were seeking health care in pulmonary, general medical, or sleep clinics or laboratories. Published studies have addressed sleep and breathing in patients with severe OAD, but none to our knowledge provide insight with respect to a patient population with predominantly mild OAD. (2) As previous investigators have done, we defined OAD as occurring when the FEV1/FVC ratio is less than 70%, (48, 5557, 84). Because a decline in this ratio may accompany normal aging, it is possible that some normal participants were misclassified as having OAD. However, our data indicate that at least in individuals with generally mild OAD but without SAH, sleep quality, architecture, and continuity are not notably impaired (see Table E2 in the online supplement). Similarly, irrespective of the value of the FEV1/FVC ratio that was chosen as the threshold for defining OAD, in the absence of SAH, the risk of sleep-related desaturation increased as the ratio declined (Table 5). Due to the relatively small number of participants with FEV1/FVC less than 60% without SAH (n = 197) we were unable to analyze subquartiles of the ratio below this value. Similarly, the number of individuals with more than 5% TST spent with SpO2 less than 85% was too small to stratify further by severity of OAD. (3) We considered the possibility that our conclusions might be influenced by survival bias. Such a bias could result in failure to detect an association between OAD and SAH due to underrepresentation of participants with both disorders in the study population on the basis of mortality. We believe, however, that the large size and heterogeneity of the study population as well as the wide age distribution make selection bias unlikely. We are not certain why there seems to be a higher than expected percentage of nonsmokers among those participants with spirometrically evident airway obstruction. Perhaps this is a consequence of misclassification due to the nature of self-reported smoking data. It is also possible that the prevalence of non-smokers with FEV1/FVC less than 70% reflects an effect of survival bias. Individuals participating in the SHHS represent a group of individuals (survivors, if you will) from other epidemiologic studies. Thus, those participants who smoked (current and past) and had a lower FEV1/FVC may have experienced a higher mortality, leaving those individuals with a FEV1/FVC ratio less than 70% and never-smokers behind. (4) Sleep monitoring was performed in an unattended environment with consequent potential limitations regarding completeness of data collection and quality. However, rigorous quality control measures maximized conformity in the performance and interpretation of PSGs (37, 38). Furthermore, studies were coded regarding possible problems with signals or interpretability to identify potentially unreliable studies. Incomplete information on sleep latency and sleep efficiency occurred predominantly due to problems in interpreting data from the light meter (requiring proper calibration, positioning on the recording garment, and ambient light conditions in the participants' bedrooms that parallel “time in bed”). However, the group of subjects excluded from these subanalyses appears similar to those included. Likewise, the overall parallel findings for sleep stage differences (available for more subjects) with differences in sleep efficiency did not suggest a bias in representation within this subsample. (5) We did not analyze our data with respect to medications. Indeed, OAD patients may receive medications that could alter sleep quality and architecture. In this light, however, our observation of minimal differences in sleep quality and architecture between individuals with and without chronic obstructive pulmonary disease (in the absence of SAH) as well as across quartiles of %predicted FEV1 is even more interesting. (6) Although there is no consensus regarding normative data, on initial inspection we noted that the arousal index in our study population, even in the absence of SAH and OAD, seemed relatively high. Whether this reflects our polysomnographic monitoring or scoring technique, inclusion of an elderly population, or other unaccounted factors is not clear. Considerable care was exercised to analyze only the arousal data that were based on reliable scoring (3638). It should also be noted that our analyses did not identify and exclude participants on the basis of snoring, periodic limb movement disorder, psychiatric/neurologic diagnoses, medication, or caffeine consumption before PSG. It is reassuring, however, that the arousal frequency in our data set is similar to that observed by others in an unselected community population undergoing a first night of in-laboratory PSG (85) and in normal individuals after an acclimatization night (86). In addition, the arousal index in SHHS is similar to that which would be observed by combining the data reported by Boselli and collegues in healthy, middle-aged (arousal index: 17.8 ± 2) and elderly (27.1 ± 3.3) individuals without SAH, medical or psychologic disorders, or periodic limb movement disorder (87).

In conclusion, our data indicate that the associations between generally mild OAD and SAH occurs by chance and not by pathophysiologic linkage. The risk of oxyhemoglobin desaturation for more than 5% of sleep time in patients with coexistent OAD and SAH is equal to the combined risk from each disorder alone. We also observed that sleep is only marginally perturbed in patients with milder OAD without SAH. Consequently, attribution of sleep-related symptoms to underlying generally mild OAD without considering the possibility of other etiologies is not warranted.

SHHS acknowledges the Atherosclerosis Risk in Communities Study, the Cardiovascular Health Study, the Framingham Heart Study, the Cornell/Mt. Sinai Worksite and Hypertension Studies, the Strong Heart Study, the Tucson Epidemiologic Study of Airways Obstructive Diseases, and the Tucson Health and Environment Study for allowing their cohort members to be part of the SHHS and for permitting data acquired by them to be used in the study. SHHS is particularly grateful to the members of these cohorts who agreed to participate in SHHS as well. SHHS further recognizes all the investigators and staff who have contributed to its success. A list of SHHS investigators, staff, and their participating institutions is available on the SHHS website, The authors also wish to express their appreciation of the support provided of James P. Kiley and Michael Twery and thanks to the SHHS participants who contributed their time to this study. They are appreciative of the efforts of the technical staff at the clinical sites for collecting the data and the Reading Center for their dedication to development and implementation of protocols for analyzing and coding the sleep studies.

1. Bang KM, George PJ, Kramer R, Cohen B. The effect of pulmonary impairment on all-cause mortality in a national cohort. Chest 1993;103:536–540.
2. American Thoracic Society. Definitions, epidemiology, pathophysiology, diagnosis, and staging. Am J Respir Crit Care Med 1995;152(Suppl):S77–S83.
3. Celli BR, Snider GL, Heffner J, Tiep B, Ziment I, Make B, Bramen S, Olsen G, Philips Y. ATS guidelines. Diagnosis and care of patients with COPD: I. definitions epidemiology; pathophysiology; diagnosis and prognosis. Am J Respir Crit Care Med 1995;152:S77.
4. National Heart Lung & Blood Institute. Morbidity and mortality: 2002 chart book on cardiovascular, lung and blood disorders. Available at
5. Redline S, Kump K, Tishler PV, Browner I. Ferrette: gender differences in sleep disordered breathing in a community-based sample. Am J Respir Crit Care Med 1994;149:722–726.
6. Ronald J, Delaive K, Manfreda J, Bahammam A, Kryger MH. Health care utilization in the 10 years prior to diagnosis in obstructive sleep apnea syndrome patients. Sleep 1999;22:225–229.
7. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230–1235.
8. Young T, Evans L, Finn L, Palta M. Estimation of the clinically diagnosed proportion of sleep apnea syndrome in middle-aged men and women. Sleep 1997;20:705–706.
9. Olson LG, Hensley MJ, King MT, Saunders NA. A community study of snoring and sleep disordered breathing: health outcomes. Am J Respir Crit Care Med 152:717–720.
10. Bearpark H, Elliott L, Grunstein R, Cullen S, Schneider H, Althaus W, Sullivan C. Snoring and sleep apnea: a population study in Australian men. Am J Respir Crit Care Med 1995;151:1459–1465.
11. Flenley DC. Sleep in chronic obstructive lung disease. Clin Chest Med 1985;6:651–661.
12. Chaouat A, Weitzenblum E, Krieger J, Ifoundza T, Oswald M, Kessler R. Association of chronic obstructive pulmonary disease and sleep apnea syndrome. Am Rev Respir Dis 1995;151:82–86.
13. Guilleminault C, Cummiskey J, Motta J. Chronic obstructive airflow disease and sleep studies. Am Rev Respir Dis 1980;122:397–406.
14. Wetter DW, Young TB, Bidwell TR, Badr MS, Palta M. Smoking as a risk factor for sleep-disordered breathing. Arch Intern Med 1994;154:2219–2224.
15. Phillips B, Danner F. Cigarette smoking and sleep disturbance. Arch Intern Med 1995;155:734–737.
16. Surgeon General. Chronic obstructive lung disease. The health consequences of smoking: US Department of Health and Human Services; 1984. Report No.: 84-50205.
17. Davis RM, Novotny TE. The epidemiology of cigarette smoking and its impact on chronic obstructive pulmonary disease. Am Rev Respir Dis 1989;140:S82–S84.
18. Newman AB, Nieto FJ, Guidry U, Lind BK, Redline S, Shahar E, Pickering TG, Stuart F, Quan SF. The relationship of sleep disordered breathing to cardiovascular disease risk factors: the Sleep Heart Health Study. Am J Epidemiol 2001;154:50–59.
19. Shahar E, Whitney CW, Redline S, Lee ET, Newman AB, O'Connor GT, Boland LL, Schwartz JE, Samet JM. Sleep-disordered breathing and cardiovascular disease: cross-sectional results of the Sleep Heart Health Study. Am J Respir Crit Care Med 2001;163:19–25.
20. Hoffstein V. Relationship between smoking and sleep apnea in clinic population. Sleep 2002;25:519–526.
21. Cormick W, Olson LG, Hensley MJ, Saunders NA. Nocturnal hypoxaemia and quality of sleep in patients with chronic obstructive lung disease. Thorax 1986;41:846–854.
22. Calverley PMA, Brezinova V, Douglas NJ, Catterall JR, Flenley DC. The effect of oxygenation on sleep quality in chronic bronchitis and emphysema. Am Rev Respir Dis 1982;126:206–210.
23. McKeon JL, Murree-Allen K, Saunders NA. Prediction of oxygenation during sleep in patients with chronic obstructive lung disease. Thorax 1988;43:312–317.
24. Fleetham J, West P, Mezon B, Conway W, Roth T, Kryger M. Sleep, arousals, and oxygen desaturation in chronic obstructive pulmonary disease: the effect of oxygen therapy. Am Rev Respir Dis 1982;126:429–433.
25. Klink M, Quan SF. Prevalence of reported sleep disturbances in a general adult population and their relationship to obstructive airways diseases. Chest 1987;91:540–546.
26. Schiffman PL, Trontell MC, Mazar MF, Edelman NH. Sleep deprivation decreases ventilatory response to CO2 but not load compensation. Chest 1983;84:695–698.
27. White DP, Douglas NJ, Pickett CK, Zwillich CW, Weil JV. Sleep deprivation and the control of ventilation. Am Rev Respir Dis 1983;128:984–986.
28. Espinoza H, Thornton AT, Sharp D, Antic R, McEvoy RD. Sleep fragmentation and ventilatory responsiveness to hypercapnia. Am Rev Respir Dis 1991;144:1121–1124.
29. Spengler CM, Shea SA. Sleep deprivation per se does not decrease the hypercapnic ventilatory response in humans. Am J Respir Crit Care Med 2000;161:1124–1128.
30. Guilleminault C, Rosekind M. The arousal threshold: sleep deprivation, sleep fragmentation, and obstructive sleep apnea syndrome. Bull Eur Pathophysiol Respir 1981;17:341–349.
31. Stoohs RA, Dement WC. Snoring and sleep-related breathing abnormality during partial sleep deprivation. N Engl J Med 1993;328:1279.
32. Fletcher EC, Schaaf JW, Miller J, Fletcher JG. Long-term cardiopulmonary sequelae in patients with sleep apnea and chronic lung disease. Am Rev Respir Dis 1987;135:525–533.
33. Bradley TD, Rutherford R, Grossman RF, Lue F, Zamel N, Moldofsky H, Phillipson EA. Role of daytime hypoxemia in the pathogenesis of right heart failure in the obstructive sleep apnea syndrome. Am Rev Respir Dis 1985;131:835–839.
34. Bradley TD, Rutherford R, Lue F, Moldofsky H, Grossman RF, Zamel N, Phillipson EA. Role of diffuse airway obstruction in the hypercapnia of obstructive sleep apnea. Am Rev Respir Dis 1986;134:920–924.
35. Weitzenblum E, Krieger J, Apprill M, Vallee E, Ehrhart M, Ratomaharo J, Oswald M, Kurtz D. Daytime pulmonary hypertension in patients with obstructive sleep apnea syndrome. Am Rev Respir Dis 1988;138:345–349.
36. Quan SF, Howard BV, Iber C, Kiley JP, Nieto FJ, O'Connor GT, Rapoport DM, Redline S, Robbins J, Samet JM, et al. The Sleep Heart Health Study: design, rationale, and methods. Sleep 1997;20:1077–1085.
37. Redline S, Sanders M, Quan SF, Iber C, Gottlieb DJ, Bonekat WH, Rapoport DM, Smith PL, Kiley JP. Methods for obtaining and analyzing polysomnography data for a multicenter study. Sleep 1998;21:759–767.
38. Whitney CW, Gottleib DJ, Redline S, Norman RG, Dodge R, Shahar E, Surovec S, Nieto FJ. Reliability of scoring respiratory indices and sleep staging. Sleep 1998;21:749–757.
39. Fried LP, Borhani NO, Enright P, Furberg CD, Gardin JM, Kronmal RA, Kuller LH, Manolio TA, Mittelmark MB, Newman A. The Cardiovascular Health Study: design and rationale. Ann Epidemiol 1991;1:263–266.
40. Dawber TR, Kannel WB, Lyell LP. An approach to longitudinal studies in a community: the Framingham study. Ann NY Acad Sci 1963;107:539–556.
41. Lebowitz MD, Knudson RJ, Burrows B. Tucson epidemiologic study of obstructive lung disease. I: methodology and prevalence of disease. Am J Epidemiol 1975;102:137–152.
42. Lee ET, Welty TK, Fabsitz R, Cowan LD, Le NA, Oopik AJ, Cucchiara AJ, Savage PJ, Howard BV. The Strong Heart study—a study of cardiovascular disease in American Indians: design and methods. Am J Epidemiol 1990;132:1141–1155.
43. The ARIC Investigators. The atherosclerosis risk in communities (ARIC) study: design and objectives. Am J Epidemiol 1989;129:687–702.
44. Sleep Heart Health Study Research Group. Sleep Heart Health Study Reading Center manual of operations. Available at the following URL: SHHS Coordinating Center, Seattle, WA: 1996.
45. Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects, NIH publication no. 204. Bethesda, US Government Printing Office; 1968.
46. Bonnet M, Carley D, Carskadon M, Easton P, Guilleminault C, Harper R, Hayes B, Hirshkowitz M, Ktonas P, Keenan S, et al. ASDA Report: EEG arousals: scoring rules and examples. Sleep 1992;15:173–184.
47. Knudson RJ, Lebowitz MD, Holberg CJ, Burrows B. Changes in the normal maximal expiratory flow-volume curve with growth and aging. Am Rev Respir Dis 1983;127:725–734.
48. Shahar E, Folsom A, Melnick S, Tockman M, Comstock G, Gennaro V, Higgins MW, Sorlie PD, Ko W-J, Szklo M, and The Atherosclerosis Risk in Communities Study Investigators. Dietary n-3 polyunsaturated fatty acids and smoking-related chronic obstructive pulmonary disease. N Engl J Med 1994;331:228–233.
49. Enright PL, Kronmal RA, Higgins M, Schenker M, Haponik EF. Spirometry reference values for men and women 65 to 85 years of age: the Cardiovascular Health Study. Am Rev Respir Dis 1993;147:125–133.
50. American Thoracic Society. Standardization of spirometry: 1994 update. Am J Respir Crit Care Med 1995;152:1107–1136.
51. American Thoracic Society. Standardization of spirometry. 1987 update. Am Rev Respir Dis 1987;136:1285–1298.
52. Gottleib DJ, Whitney CW, Bonekat WH, Iber C, James GD, Lebowitz M, Nieto FJ, Rosenberg CE. Relation of sleepiness to respiratory disturbance index: the Sleep Heart Health Study. Am J Respir Crit Care Med 1999;159:502–507.
53. Johns MW. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep 1991;14:540–545.
54. Johns M. Daytime sleepiness, snoring, and obstructive sleep apnea: the Epworth Sleepiness Scale. Chest 1993;103:30–36.
55. Enright PL, Johnson LR, Connett JE, Voelker H, Buist AS. Spirometry and the lung health study. Am Rev Respir Dis 1991;143:1215–1223.
56. Enright PL, Connett JE, Kanner RE, Johnson LR, Lee WW, for the Lung Health Study. Spirometry in the lung health study: II. Determinants of short-term intraindividual variability. Am J Respir Crit Care Med 1995;151:406–411.
57. Enright PL. Office spirometry. In: UpToDate, Rose BD, editor. Wellesley, MA: UpToDate; 2002.
58. Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general US population. Am J Respir Crit Care Med 1999;159:179–187.
59. McSweeny AJ, Grant I, Heaton RK, Adams KM, Timms RM. Life quality of patients with chronic obstructive pulmonary disease. Arch Intern Med 1982;142:1470–1476.
60. Grant I, Heaton RK, McSweny AJ, Adams KM, Timms RM. Neuropsychological findings in chronic obstructive pulmonary disease. Arch Intern Med 1982;142:1470–1476.
61. Okubadejo AA, Jones PW, Wedzicha JA. Quality of life in patients with chronic obstructive pulmonary disease and severe hypoxaemia. Thorax 1996;51:44–47.
62. Roehrs T, Merrion M, Pedrosi B, Stepanski E, Zorick F, Roth T. Neurophysiological function in obstructive sleep apnea syndrome (OSAS) compared to chronic obstructive pulmonary disease (COPD). Sleep 1995;18:382–388.
63. Engleman HM, Cheshire KE, Deary IJ, Douglas NJ. Daytime sleepiness, cognitive performance and mood after continuous positive airway pressure for the sleep apnoea/hypopnoea syndrome. Thorax 1993;48:911–914.
64. Ferrer M, Alonso J, Morera J, Marrades RM, Khalaf A, Aguar MC, Carmen M, Plaza V, Prieto L, Anto JM. Chronic obstructive pulmonary disease stage and health-related quality of life. Ann Intern Med 1997;127:1072–1079.
65. D'Ambrosio C, Bowman T, Mohsenin V. Quality of life with obstructive sleep apnea. Chest 1999;115:123–129.
66. Wynne JW, Block AJ, Hemenway J, Hunt LA, Flick MR. Disordered breathing and oxygen desaturation during sleep in patients with chronic obstructive lung disease (COLD). Am J Med 1979;66:573–579.
67. Goldstein RS, Ramcharan V, Bowes G, McNicholas WT, Bradley D, Phillipson EA. Effect of supplemental nocturnal oxygen on gas exchange in patients with severe obstructive lung disease. N Engl J Med 1984;310:425–429.
68. D'Ambrosio CM, Mohsenin V. Sleep in asthma. Clin Chest Med 1998;19:127–137.
69. Sadeh A, Horowitz I, Wolach-Benodis L, Wolach B. Sleep and pulmonary function in children with well-controlled asthma. Sleep 1998;21:379–384.
70. Janson C, DeBacker W, Gislason T, Plaschke P, Bjornsson E, Hetta J, Kristbjarnarson H, Vermeire P, Boman G. Increased prevalence of sleep disturbances and daytime sleepiness in subjects with bronchial asthma: a population study of young adults in three European countries. Eur Respir J 1996;9:2132–2138.
71. Janson C, Gislason T, Boman G, Hetta J, Roos BE. Sleep disturbances in patients with asthma. Respir Med 1990;84:37–42.
72. van Keimpema AR, Ariaansz M, Nauta JJ, Postmus PE. Subjective sleep quality and mental fitness in asthmatic patients. J Asthma 1995;32:69–74.
73. Fitzpatrick MF, Martin K, Fossey E, Shapiro CM, Elton RA, Douglas NJ. Snoring, asthma and sleep disturbance in Britain: a community-based survey. Eur Respir J 1993;6:531–535.
74. Klink ME, Dodge R, Quan SF. The relationship of sleep complaints to respiratory symptoms in a general population. Chest 1994;105:151–154.
75. Dodge R, Cline MG, Quan SF. The natural history of insomnia and its relationship to respiratory symptoms. Arch Intern Med 1995;155:1797–1800.
76. Gries RE, Brooks LJ. Normal oxyhemoglobin saturation during sleep: how low does it go? Chest 1996;110:1489–1492.
77. Fletcher EC, Donner CF, Midgren B, Zielinski J, Levi-Valensi P, Btaghirolo A, Ida Z, Miller CC. Survival in COPD patients with a daytime PaO2 > 60 mm Hg with and without nocturnal oxyhemoglobin desaturation. Chest 1992;101:649–655.
78. Crockett AJ, Moss JR, Cranston JM, Alpers JH. Domiciliary oxygen for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2000;CD001744.
79. Little SA, Elkholy MM, Chalmers GW, Farouk A, Patel KR, Thomson NC. Predictors of nocturnal oxygen desaturation in patients with COPD. Respir Med 1999;93:202–207.
80. Mohsenin V, Guffanti EE, Hilbert J, Ferranti R. Daytime oxygen saturation does not predict nocturnal oxygen desaturation in patients with chronic obstructive pulmonary disease. Arch Phys Med Rehabil 1994;75:285–289.
81. Connaughton JJ, Catterall JR, Elton RA, Stradling JR, Douglas NJ. Do sleep studies contribute to the management of patients with severe chronic obstructive pulmonary disease? Am Rev Respir Dis 1988;138:341–344.
82. Stradling JR, Lane DJ. Nocturnal hypoxaemia in chronic obstructive pulmonary disease. Clin Sci 1983;64:213–222.
83. Mulloy E, Fitzpatrick M, Bourke S, O'Regan A, McNicholas WT. Oxygen desaturation during sleep and exercise in patients with severe chronic obstructive pulmonary disease. Respir Med 1995;89:193–198.
84. Shahar E, Folsom AR, Melnick SL, Tockman MS, Comstock GW, Shimakawa T, Higgins MW, Sorlie PD, Szklo M. Does dietary vitamin A protect against airway obstruction? Am J Respir Crit Care Med 1994;150:978–982.
85. Mathur R, Douglas NJ. Frequency of EEG arousals from nocturnal sleep in normal subjects. Sleep 1995;18:330–333.
86. Martin SE, Wraith PK, Deary IJ, Douglas NJ. The effect of nonvisible sleep fragmentation on daytime function. Am J Respir Crit Care Med 1997;155:1596–1601.
87. Boselli M, Parrino L, Smerieri A, Terzano MG. Effect of age on EEG arousals in normal sleep. Sleep 1998;21:351–357.
Correspondence and requests for reprints should be addressed to Mark H. Sanders, M.D., Division of Pulmonary, Allergy, and Critical Care Medicine, Montefiore University Hospital, University of Pittsburgh School of Medicine, North-1292, Pittsburgh, PA 15213. E-mail:


No related items
American Journal of Respiratory and Critical Care Medicine

Click to see any corrections or updates and to confirm this is the authentic version of record