Which Blood Pressure Measurement Best Predicts Cardiovascular Outcomes?

Robert M. Carey, MD; Thomas H. Marwick, MBBS, PHD, MPH


J Am Coll Cardiol. 2020;76(25):2923-2925. 

Hypertension is the world's leading risk factor for cardiovascular disease (CVD), ranks first in disability-adjusted life-years and predicts approximately 1,000 deaths daily, comparable to the risk of malignant disease.[1] The diagnosis and management of hypertension are critically dependent on accurate blood pressure (BP) measurement. Although office BP measurement is standard for medical practice and is the basis of the vast majority of available evidence on efficacy and safety of antihypertensive agents in randomized clinical trials, out-of-office BP measurement (home blood pressure monitoring [HBPM] or ambulatory blood pressure monitoring [ABPM]) is now recommended for both the diagnosis and management of hypertension.[2] Among the reasons for this recommendation, out-of-office BP is required for the diagnosis of white coat and masked hypertension in persons not taking antihypertensive agents as well as the white coat effect and masked uncontrolled hypertension in patients taking antihypertensive drug therapy.[2] White coat hypertension is associated with minimal excess CVD risk, whereas masked hypertension is linked to detrimental cardiovascular outcomes similar to those in sustained hypertension.[2]

Despite the widespread use of office BP, measurements by ABPM or HBPM have a stronger association with hypertension-related target organ damage and clinical CVD outcomes compared with office-based BP measurement.[3] Although prospective head-to-head comparative outcome studies are currently unavailable, guidelines generally recommend ABPM over HBPM to rule out white coat hypertension, at least in part due to the larger number of published studies demonstrating the superiority of ABPM, rather than HBPM, over office BP measurement.[3] Thus, in current guidelines, HBPM is recommended as an acceptable alternative to ABPM, if ABPM is unavailable.[2] A recent systematic review and meta-analysis, however, found little evidence supporting the superiority of ABPM over HBPM in predicting cardiovascular events.[4] Without question, then, clarification of the ability of out-of-office BP measurements by ABPM or HBPM to predict CVD outcomes is a high priority to provide confidence for the critical BP measurement phase of hypertension management.

In this issue of the Journal, Schwartz et al.[5] report a small (400 participants) but well-performed prospective cohort study of the reliability (intraclass correlation coefficient) and predictive value of BP measured in the office, at home (HBPM), and by ABPM on left ventricular mass index (LVMI). The study population was relatively young (mean age 41.2 years) and excluded adults with severe hypertension (BP ≥160/105 mm Hg) as well as participants taking antihypertensive or other medications, which would be likely to alter BP. Although the mean BP values were low, 30% to 50% of the population would have been classified as having hypertension by the 2017 American College of Cardiology/American Heart Association BP guideline.[2] Surprisingly, HBPM demonstrated the best reliability and strongest association with LVMI among the 3 BP measurement modalities. The results were not due to number of BP measurements and remained significant after adjustment for age, sex, race/ethnicity, body mass index, diabetes, and the other modalities of BP measurement.

LVMI is a validated measure of CVD risk stratification and an established marker for subclinical heart disease. Sufficiently increased LVMI defines left ventricular remodeling and hypertrophy (LVH), an intermediate form of cardiac target organ damage. In addition to being an end-organ response, LVH is an independent risk factor for CVD morbidity and mortality and a fingerprint of sustained exposure to elevated BP over time.[6] LVH is a marker of high risk, and therefore was proposed to be of value in the 2018 European Society of Cardiology/European Society of Hypertension hypertension guideline for facilitating treatment initiation decisions for people with high normal blood pressure and grade 1 hypertension.[7] In addition to CVD and all-cause mortality, LVH is a risk factor for heart failure with preserved ejection fraction, myocardial ischemia, atrial and ventricular arrhythmias, cerebrovascular disease, and dementia.[6] However, LVMI is not a simple measurement. A controversy continues about how best to index this parameter, and the indexation of LV mass by body surface area (as used in this paper and most large population studies) carries the risk of correcting the measurement for the effects of obesity. Indexing to height raised to the power of 1.7 has advantages over indexing to body surface area, especially for the prediction of events in obese patients.[8] Moreover, the use of 2-dimensional echocardiography for the measurement of LV mass has limitations due to the need for geometric assumptions and image quality,[9] and the use of 3-dimensional echocardiography alleviates the former but not the latter. Therefore, cardiac magnetic resonance imaging is the optimal technique for LV mass calculation. Although the greater association of home BP with LVMI demonstrated in this study is likely to be clinically important, the level of association with other means of assessing or indexing LV mass may not be the same.

In the United States, ABPM is largely unavailable to primary care practitioners, who manage the vast majority of patients with hypertension; it is poorly reimbursed and rarely used. Thus, the demonstration that HBPM is more reliable and associates more closely with LVMI than ABPM, if confirmed, would carry the potential to change clinical practice. HBPM conveys a different set of information than ABPM. HBPM measures resting daytime BP, whereas ABPM measures dynamic daytime BP, including moment-to-moment variations induced by postural, postprandial, and drug-induced BP changes, among others, and nocturnal BP, including nondipping, reverse dipping, and morning BP surge. The results of the study by Schwartz et al.[5] suggest that measurement of resting daytime BP may be relatively more important than dynamic daytime and/or nocturnal parameters in predicting subclinical cardiac target organ damage, but this requires further study. The results of the present study, however, suggest that HBPM could be especially important for detecting elevated BP and hypertension early in life, when adults are relatively healthy, but those with hypertension have a high lifetime risk of CVD.

The realities of clinical practice today require optimization of resources, careful time management, and cost-effectiveness in the diagnosis and treatment of disease. Use of ABPM entails availability of special equipment, knowledgeable personnel, time for interpretation, and increased cost compared with HBPM. In addition, simplification of patient tasks/instructions increases adherence and facilitates the clinician-patient relationship. In the study by Schwartz et al.,[5] HBPM was performed by participants who had only 5 min of in-office training and an instruction sheet for reference, consistent with the constraints of practice, and HBPM was conducted for only 1 week; yet, the results demonstrated the highest degree of reliability and association with target organ damage among the tested modalities. Further studies are needed in older participants, those taking antihypertensive medications, and those with other comorbidities. Subject to confirmation, however, the findings of this study encourage the routine implementation of HBPM as the preferred method of out-of-office BP monitoring.