Menopause, Hormone Replacement Therapy, and Sleep-disordered Breathing
Are We Ready for the Heat?

Terry Young

Department of Preventive Medicine, University of Wisconsin-Madison, Madison, Wisconsin

	ARTICLE

TOP ARTICLE REFERENCES

The essential feature of obstructive sleep apnea syndrome is repetitive episodes of breathing pauses (apnea) and hypoventilation (hypopnea), often referred to as sleep-disordered breathing (SDB). SDB is prevalent in the general population, but less than 10% of clinically significant cases have been diagnosed (1). Population-based studies have linked even mild SDB with cardiovascular disease (2) and other adverse health outcomes (3), and so the high SDB prevalence poses a public health burden. Prevention and early intervention may be the most feasible ways to reduce this burden, and therfore high priority should be given to identifying causes of SDB that can be eliminated or modified. In this issue of the Journal (pp. 608- 613), the report of Bixler and colleagues (4) provides epidemiologic data pertinent to menopause as a risk factor and hormone replacement therapy (HRT) as a protective factor for SDB.

There is widespread belief that menopause is an established risk factor for SDB, but evidence to support this is lacking. The scarcity of data is not surprising in view of the methodological challenges in testing this hypothesis. Foremost is untangling the tight correlation of menopausal status and age to determine the effect of menopause apart from other aspects of midlife aging. Bixler and colleagues have dealt with some of these difficulties and in so doing provide the first epidemiological data to shed light on the association of menopausal status and SDB.

The investigators performed polysomnography on a population-based sample of 1,000 women aged 20-100 yr, and found that SDB prevalence, regardless of the definition used, was higher in the group of postmenopausal women compared with premenopausal women. However, these comparisons were not adjusted for confounding by age and body mass index (BMI). Confounding was handled in a multiple regression model with menopausal status categorized into premenopause (reference group), postmenopause with HRT, and postmenopause without HRT.

The adjusted odds ratios for SDB and postmenopause without HRT versus premenopause ranged from 2 to 4, depending on the SDB definition. In addition, the odds for SDB were not higher in postmenopausal women who were taking HRT versus premenopausal women. It is hoped that these results truly do support the hypothesis that HRT can lower SDB risk. But, before discussing what insight can be gained from the report, I would like to put the findings in the context of other research on HRT and chronic disease.

Whether long-term HRT should be used to reduce the risk of chronic disease is an emotionally charged question and much in the limelight of women's health (6). In 1976, the Framingham Heart Study showed that postmenopausal women, compared with premenopausal women, had four times the incidence of coronary heart disease (7). This hypothesis and the question of whether HRT reduces cardiovascular disease (CVD) risk have been intensely pursued for 20 years, with mixed results (8). Controversy over the benefit versus risk of long-term HRT use has been fueled further by studies showing, in addition to a protective effect of HRT, a lack of evidence for natural menopause as a significant risk factor for CVD, increased risks of breast cancer with long-term HRT use, and negative findings from a large randomized trial of HRT and CVD (9). My point here is that studies suggesting a causal role of menopause in chronic disease and a protective effect with HRT use are not taken lightly!

Shall we add SDB to this heated arena as another chronic condition related to menopause that may be prevented by HRT? The strongest results of the study by Bixler and colleagues show an odds ratio of 4.3 for having SDB, defined as 15 or more apnea and hypopnea episodes per hour, for postmenopausal women not using HRT versus premenopausal women. This is a substantial odds ratio, but we must consider first the degree to which study biases may account for the association. The authors found no evidence of participation bias. They also addressed confounding by BMI, age, alcohol, and smoking. But do the variables used adequately capture the true confounding factors? Perhaps the essence of "obesity" is only partially captured by BMI, leaving residual confounding and an overestimation of risk. Have important confounders been overlooked? It has been shown that women who choose to use HRT have a healthier profile to begin with, so HRT use is a surrogate for a myriad of factors other than estrogen exposure (10). By separating HRT users from nonusers, Bixler and investigators may have created categories that also reflect these HRT correlates, resulting in a less healthy group of postmenopausal women not using HRT. Measurement error causes serious bias if it differs by the study factors. Differential misclassification on menopausal status may have occurred because "possibly perimenopausal" women were classified as premenopausal only if they did not use HRT. Although not obvious, this misclassification could spuriously contribute to the finding of "no SDB risk" for HRT users. Most importantly, the ability of cross-sectional data to address relationships of time-dependent factors such as menopausal status, HRT use, and SDB is limited. Longitudinal data on SDB in women over their menopausal transitions will be necessary to address critical aspects of the temporal role of menopause and HRT in SDB development and progression.

Although the new data suggest a link between the lack of HRT use in menopause and SDB, we do not want to get thrown prematurely into the hot spotlight of HRT intervention research. This is not a criticism of the report by Bixler and coworkers: their goal was not to provide estimates to justify planning a multimillion dollar trial. But, we still need citable, quantitative findings that menopause is an independent risk factor for SDB and that HRT use, apart from its healthy lifestyle correlates, modifies that risk. When we become armed with robust evidence, the next step could indeed be consideration of an intervention trial of HRT to reduce SDB.

The study by Bixler and colleagues is important in several ways. It provides the first epidemiological evidence suggestive of a link between postmenopause without HRT use and SDB. Equally important, the results suggest that premenopausal and postmenopausal women using HRT do not differ on SDB risk. This is a message of particular relevance to premenopausal women with undiagnosed SDB: neither their decision to seek help nor the decision by primary care providers to refer them for SDB evaluation should be negatively influenced by their premenopausal status.

The impressive work of Bixler and colleagues provides encouragement for a more intense focus on prevention and intervention as a means of reducing SDB prevalence. I look forward eagerly to further studies on menopause as a modifiable cause of SDBand to strong data that may lead to an intervention trial. Bixler and colleagues have provided a foundation to build on; this is clearly a significant step forward in improving the health of women.

	References

TOP ARTICLE REFERENCES

1. 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 [Medline].

2. Peppard P, Young T, Palta M, Skatrud J. Prospective study of the association of sleep-disordered breathing and hypertension. N Engl J Med 2000; 342: 1378-1384 [Abstract/Free Full Text].

3. Gottlieb DJ, Whitney C, 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 [Abstract/Free Full Text].

4. Bixler EO, Vgontzas AN, Lin HM, Ten Have T, Rein J, Vela-Bueno A, Kales A. Prevalence of sleep disordered breathing in women: effects of gender. Am J Respir Crit Care Med 2001;163:608-613.

5. Fitzpatrick L, Litin S, Bell M. The Women's Health Initiative: a heart to HRT conversation. Mayo Clin Proc 2000; 75: 559-561 [Medline].

6. Charney P, Walsh J, Nattinger AB. Update in women's health. Ann Intern Med 1998; 129: 551-558 [Free Full Text].

7. Kannel WB, Hjortland MC, McNamara PM, Gordon T. Menopause and risk of cardiovascular disease: the Framingham Study. Ann Intern Med 1976; 85: 447-452 .

8. Schairer C, Lubin J, Troisi R, Sturgeon S, Brinton L, Hoover R. Menopausal estrogen and estrogen-progestin replacement therapy and breast cancer risk. JAMA 2000; 283: 485-491 [Abstract/Free Full Text].

9. Nabulsi AA, Folsom AR, White A, Patsch W, Heiss G, Wu KK, Szklo M. Association of homone-replacement therapy with various cardiovascular risk factors in postmenopausal women. N Engl J Med 1993; 328: 1069-1075 [Abstract/Free Full Text].

10. Rodstrom K, Bengtsson C, Lissner L, Bjorkelund C. Pre-exisiting risk factor profiles in users and nonusers of hormone replacement therapy: prospective cohort study in Gothenburg, Sweden. Br Med J 1999; 319: 890-893 [Abstract/Free Full Text].