The recent commentary on high rates of uncontrolled blood pressure (BP) and opportunities for health care providers in this Journal prompted an examination of treated but uncontrolled BP in relation to pulse wave velocity (PWV) in the Maine Syracuse Longitudinal Study (MSLS).[2, 3] Cohen and Townsend noted that a Center for Disease Control analysis of National Health and Nutrition Examination Survey (NHANES) data (2003–2010) indicated that the presence of uncontrolled hypertension (≥140 mm Hg systolic and ≥90 mm Hg diastolic) in US adults was 53.5%. They further noted that approximately 45% of these individuals with uncontrolled hypertension were aware of their hypertension and were receiving treatment.
In the community-based MSLS, beginning in 1975 and concluding in 2008, there were 7 serial waves of data collection.[2, 3] After each wave of data collection, participants were made aware of prevalent and incident cardiovascular risk factors, including hypertension, and were referred to their physicians for treatment as usual. The 140/90 mm Hg criteria were the prevailing standards for BP control for waves 4 to 7 of the study. Failure to achieve control was observed for 40%, 41%, and 41% of the participants at waves 5, 6, and 7, respectively; polypharmacy was the predominant treatment mode.
In the last wave of MSLS (wave 7) for a subset of individuals participating in PWV studies, 95% of individuals with hypertension were being treated. Of these, 46% had BP levels defined as uncontrolled using the 140/90 mm Hg criteria. Data for PWV (m/s) and BP (mm Hg) parameters for treated hypertensive MSLS participants at wave 7 are shown in Table 1 to illustrate differences relating to BP control status. PWV for persons with untreated normal BP was 8.9 m/s (not shown), which was significantly lower than values for either the controlled or uncontrolled BP groups (P<.001).
|Blood Pressure||No.||PWV||Systolic BP||Diastolic BP||Mean Arterial Pressure|
|Controlled||252||10.5 (2.8)||121.8 (11.1)||70.9(6.5)||87.9 (6.5)|
|Uncontrolled||218||11.9 (3.2)||147.1 (18.1)||84.5 (8.6)||105.4 (9.2)|
The controlled and uncontrolled groups differed for sex but not for the 4 cardiovascular risk factors, age (years), diabetes mellitus, body mass index (>30 kg/m2), and total cholesterol (mg/dL) (Table 2). Prevalence of all multiple cardiovascular diseases combined was 14% of the total sample. Table 3 shows categorical regression coefficients (b) relating controlled vs uncontrolled BP groups to PWV.
|Blood Pressure||No.||Age||Female Sex, %||Education||Diabetes, %||Body Mass Index||Total Cholesterol|
|Controlled||252||67.7 (10.7)||63||14.3 (2.8)||21.6||31.2 (7.4)||173.7 (36.7)|
|Uncontrolled||218||67.5 (12.4)||51||14.2 (2.8)||24.9||31.2 (7.2)||177.7 (38.1)|
|b||95% CI||b||95% CI||b||95% CI|
For all three covariate sets, significant associations remained with statistical adjustment for cardiovascular risk factors, although the magnitudes of the association were attenuated.
Our data are not original with respect to evidence that hypertension is the predominant risk factor for higher levels of PWV and, by inference, arterial stiffness. Clearly a causal association between BP control and PWV cannot be inferred; however, we present these data to illustrate an opportunity for further research. PWV is the gold standard indirect measure of arterial stiffness and there is growing evidence that it is very important in the early detection of vascular diseases.[5, 6] We especially need longitudinal studies and additional controlled clinical trials of medications that have been associated with reduction in PWV (eg, ACE inhibitors, calcium channel antagonists, and selective β-blocking agents). We also need trials with new candidate drugs that may facilitate the achievement of hypertension management goals without adding to what Cohen and Townsend describe as a “burdensome list of medications.”