Bilateral leg edema, pulmonary hypertension, and obstructive sleep apnea
A cross-sectional study
Results
Twenty-eight subjects enrolled in the study, 16 with pulmonary hypertension and 12 without. Findings regarding 15 of the 16 subjects with pulmonary hypertension were reported previously.3 The edema was mild (1+ or 2+ pitting) for most subjects, typically presenting as an incidental examination finding. Of the edematous patients recruited for enrollment, many more than the number who actually participated were ineligible because their echocardiograms did not allow an estimation of the pulmonary artery pressure.
Demographic information on the subjects with and without pulmonary hypertension is shown in Table 1. Subjects with pulmonary hypertension were older (mean age 63.4 ± 13.6 years versus 52.2 ± 9.9 years, P = .02). Most subjects in both groups were obese. There were no differences between the 2 groups in sex, race, education, marital status, body mass indices, or duration of edema.
Ten of 16 (63%) subjects with pulmonary hypertension and 9 of 12 (75%) subjects without pulmonary hypertension had obstructive sleep apnea (P = .48). There were no differences between the 2 groups in apnea-hypopnea indices, spirometry measurements, oxygen saturation, asthma, systemic hypertension, previous use of appetite suppressants, use of prescription medications, or Epworth sleepiness scale scores. Because Epworth sleepiness scale scores of 9 to 10 or less are considered mild,6 the low Epworth sleepiness scale scores in both groups indicate that many individuals with obstructive sleep apnea and edema lack symptoms of excessive daytime sleepiness. In the hierarchical logistic regression analysis, the probability associated with the adjusted regression coefficient for pulmonary hypertension status was .71, indicating that even with adjustment for potential confounding variables (age, duration of edema), there was no association between pulmonary hypertension and obstructive sleep apnea.
TABLE 1
Demographic characteristics, pathologic conditions, and laboratory data of subjects with bilateral leg edema
| Variable | Pulmonary hypertension (n = 16) | No pulmonary hypertension (n = 12) | P |
|---|---|---|---|
| Age (y) | 63.4 ± 13.6 | 52.2 ± 9.9 | .02 |
| Female sex | 69% | 75% | .72 |
| White race | 88% | 100% | .49 |
| Body mass index (kg/m2) | 37.2 ± 11.0 | 39.1 ± 12.1 | .66 |
| Obesity (BMI ≥ 30) | 69% | 92% | .20 |
| Education | |||
| High school graduate or higher | 53%* | 53%* | .95 |
| Marital status | |||
| Married | 47%* | 67% | .30 |
| Duration of edema | |||
| > 2 years | 64%* | 55%* | .70 |
| Pulmonary artery pressure (mm Hg) | 37.3 ± 6.0 | 25.4 ± 4.2 | .001 |
| Obstructive sleep apnea | 63% | 75% | .48 |
| Apnea-hypopnea index | 32.3 ± 28.5 | 31.8 ± 23.4 | .96 |
| Systemic hypertension | 38% | 33% | .82 |
| Asthma | 6% | 17% | .56 |
| Spirometry data | |||
| FVC (% predicted) | 77.4 ± 17.6 | 70.3 ± 14.2 | .27 |
| FEV1(% predicted) | 82.8 ± 18.6 | 73.7 ± 15.1 | .17 |
| FEV1/FVC (%) | 106.9 ± 11.0 | 105.2 ± 6.5 | .64 |
| Oxygen saturation (%) | 96.7 ± 1.4 | 97.7 ± 1.7 | .31 |
| Epworth sleepiness scale score | 10.3 ± 4.9 | 8.0 ± 4.7 | .24 |
| Date presented as mean ± SD unless otherwise noted. | |||
| *Slightly reduced sample size due to occasional missing data. | |||
| FVC, forced vital capacity; FEV 1 , forced expiratory volume in 1 second. | |||
Discussion
We found a high prevalence of obstructive sleep apnea (68%) in patients with bilateral leg edema, most of whom were obese. The proportion of obstructive sleep apnea was high whether or not pulmonary hypertension was present. Our findings suggest that bilateral leg edema, but not pulmonary hypertension, may be a useful marker for underlying obstructive sleep apnea, especially in obese patients. Moreover, if the data are generalizable, many individuals with bilateral leg edema and normal left ventricular systolic function may be misdiagnosed or underdiagnosed as having idiopathic edema, venous insufficiency,1 or diastolic dysfunction.10 The finding that subjects with pulmonary hypertension were older than those with normal pulmonary artery pressures suggests that either patient age or the duration of the obstructive sleep apnea may be important variables in the development of pulmonary hypertension in edematous patients with obstructive sleep apnea.
Because of the small sample, a type II error might be the explanation for the lack of difference between the pulmonary hypertension and nonpulmonary hypertension groups. Because of the small sample size and the possibility of selection bias, the results of this study should be interpreted with caution. These findings need to be replicated with a larger sample to confirm the association. In addition, further research is necessary to clarify whether leg edema, obesity, or a combination thereof is the most useful marker for obstructive sleep apnea.
If our patients are typical of those in other practices, we estimate that leg edema associated with obstructive sleep apnea occurs frequently compared with other cardiovascular diseases. In both the inner city and suburban family practices of one of the authors (R.P.B.), leg edema associated with obstructive sleep apnea is the third most common cardiovascular condition, occurring less often than systemic hypertension and coronary artery disease but more frequently than congestive heart failure, cerebrovascular accidents, or cardiac arrhythmias.
Because our experience represents primary care rather than tertiary or specialty care, and because our experience is similar in inner city and suburban settings, we believe that our experience may be generalizable to a variety of practice settings. We now practice according to the clinical dictum that for patients without symptoms or signs of congestive heart failure and without overt lung disease, bilateral leg edema represents obstructive sleep apnea until proven otherwise.
