Hypertension affects an estimated 60-70 million Americans, predisposing them to potentially life threatening cardiovascular complications. Resistant hypertension, defined as uncontrolled blood pressure on 3 or more different antihypertensive agents, is common, affecting 15-20% of the entire hypertensive population or an estimated 12-14 million Americans. Although associated with obesity, increasing age, black race, and chronic kidney disease, mechanisms of treatment resistance remain obscure. The investigators' laboratory identified primary aldosteronism (PA) as a common cause of treatment resistance with a prevalence of 20% among subjects with resistant hypertension. This is clinically important because recognition of PA can lead to effective treatment with use of aldosterone blockers. Obstructive sleep apnea (OSA) is strongly associated with and predicts development of hypertension as demonstrated in landmark cohort studies including the Sleep Heart Health Study and the Wisconsin Sleep Cohort Study.

The investigators' laboratory has confirmed OSA to be extremely common in subjects with resistant hypertension, with a prevalence of approximately 85%. Recognizing that PA and OSA are exceptionally common in subjects with resistant hypertension, the investigators hypothesized that the 2 may be causally related. In testing this hypothesis, the investigators recently reported that plasma aldosterone levels are positively correlated with OSA severity in subjects with resistant hypertension but not in normotensive control subjects. This observation suggests that there is an important mechanistic interaction between untreated OSA and aldosterone excess in subjects with resistant hypertension. While the investigators' original hypothesis was that OSA stimulates aldosterone release, the investigators recognize that the opposite may also be true; that is, aldosterone excess in subjects with resistant hypertension worsens OSA. Distinguishing between these two possibilities has potentially far-reaching clinical implications. If the former hypothesis is true, effective treatment of OSA would be expected to suppress aldosterone release in subjects with resistant hypertension, thereby reversing the underlying cause of their treatment resistance. If the latter hypothesis is true, use of mineralocorticoid receptor antagonists would be expected to reduce OSA severity in subjects with resistant hypertension, thereby enhancing treatment of OSA. Either scenario would represent a new treatment approach for a highly prevalent and serious medical problem.

Inclusion Criteria:

• Resistant hypertension defined as office BP that is uncontrolled with 3 or more antihypertensive medications
• Moderate-severe OSA defined as AHI ≥15 events/hr
• Self-reported adherence >80% with prescribed antihypertensive medications.

Exclusion Criteria:

• Ongoing use of a potassium sparing diuretic
• History of congestive heart failure (ejection fraction of <40%)
• Chronic kidney disease (creatinine clearance <60 ml/min)
• History of cardiovascular disease (stroke, TIA, myocardial infarction, or revascularization procedure)
• Known or suspected history of secondary cause of hypertension other than primary aldosteronism
• Severe nocturnal hypoxemia (O2 desaturation nadir <60%)
• White coat hypertension defined as office BP >140/90 mm Hg and ambulatory daytime BP <135/85 mm Hg
• Central sleep apnea (defined as 5% or more of the apneas as central apneas) and/or the presence of any Cheyne-Stokes breathing
• Subjects working shift work or having other known circadian rhythm disorders such that their sleep-wake schedule is altered
• Excessive daytime sleepiness as indicated by an Epworth score of >10
• Pregnant Women

3 months

1. Unanticipated enrollment challenges led to small numbers of participants analyzed.
2. Open-label study design may have prevented a placebo effect.
3. An outlier in AHI was present in the control group.