197Data from the Sleep Heart Health Study has been less conclusive. This multi-
center longitudinal cohort study examining the cardiovascular consequences of
sleep-disordered breathing in over 6000 individuals also found an increased odds of
HTN in patients with sleep-disordered breathing in a dose-response manner [ 16 , 17 ].
Although some of this relationship was explained by body mass index (BMI), the
odds ratio for HTN after adjustment for possible confounders remained 1.37 com-
paring the highest and lowest categories of AHI (≥30 versus <1.5). It has been
postulated that the weaker association between obesity and HTN seen in this popu-
lation as compared with the Wisconsin Sleep Cohort may be related to the older age
of patients and the reference category of AHI 0–4.9, which may have included sub-
jects with very mild OSA.
Taken together, these data suggest that CPAP can be expected to result in a mod-
est reduction in blood pressure on the order of 2–3 mmHg for the population at
large. However, this degree of reduction should be considered significant, as it has
been shown to reduce cardiovascular mortality by 4–8% [ 18 ]. In patients with resis-
tant hypertension, the magnitude of BP reduction attributable to CPAP appears to be
even greater.
The strength of the association between OSA and hypertension may be modu-
lated by other factors including age and somnolence status. For example, the asso-
ciation between OSA and hypertension appears to be stronger in young to
middle-aged adults than older adults [ 17 , 19 ]. Haas et al. conducted a large cross-
sectional study of participants in the Sleep Heart Health Study stratified by age [ 20 ]
and found that an association between systolic and diastolic hypertension existed
only in those patients under the age of 60 years. However, it is possible that the
higher medical comorbidity that accumulates with aging has obscured an indepen-
dent association between OSA and HTN in such studies. Data are inconsistent
regarding the role of sex in mediating the relationship between OSA and HTN [ 19 ,
21 ]. Nonetheless, given evidence that OSA is often underdiagnosed and under-
treated in women, it is important to remain cognizant that the presence of OSA
confers a likely similar risk of subsequent hypertension as it does in males [ 10 ].
Circadian Variability of BP in Patients with OSA
Most patients with OSA lack the normal physiological reduction in blood pressure
(“nocturnal dipping”) that occurs during sleep [ 22 , 23 ]. The normal nocturnal dip-
ping phenomenon is defined as a reduction in BP by at least 10% during the night
as compared with the daytime. A “non-dipping” phenomenon has been observed
commonly in both normotensive and hypertensive patients with OSA [ 24 ] and is
thought to represent one mechanism by which OSA leads to an increased risk of
target organ damage and cardiovascular events, as compared with subjects who
experience the normal BP decline during sleep [ 22 , 25 , 26 ]. For example, left ven-
tricular hypertrophy appears more closely linked to HTN during sleep than during
wakefulness [ 27 ]. Data from the WSCS showed a dose-response relationship
between the severity of baseline OSA and odds of developing an incident non-
dipping profile in systolic blood pressure.
13 Obstructive Sleep Apnea and Resistant Hypertension