Masked Hypertension: Fragile in More Ways Than One

In-office blood pressure has been the cornerstone of hypertension screening for several decades. As guidelines recommend increasingly stringent blood pressure thresholds for the diagnosis of hypertension,¹ we require increasingly accurate approaches to measure blood pressure to avoid misdiagnosis and over or undertreatment of hypertension. However, many individuals have discordant blood pressure readings in the office compared to outside of the office setting, including masked hypertension (normal office blood pressure with elevated out-of-office blood pressure) and white coat hypertension (elevated office blood pressure with normal out-of-office blood pressure). Since individuals spend the majority of their time outside of the doctor’s office, it is not surprising that blood pressures obtained outside of the office are more strongly linked with long-term cardiovascular risk than those obtained in the office.² Growing evidence supporting the prognostic value of out-of-office blood pressure monitoring has culminated in several national and international guidelines calling for the widespread use of out-of-office blood pressure monitoring for the diagnosis of hypertension.^{1, 2}

Twenty-four hour ambulatory blood pressure monitoring (ABPM) is often described as the “reference standard” of blood pressure measurement due to its distinctively strong association with longitudinal cardiovascular outcomes. However, there are several provider-level barriers very likely preventing wider adoption of 24-hour ABPM.³ Notably, 24-hour ABPM devices are expensive, and require purchasing proprietary software and investing time in patient education, application, and interpretation. For many years, the United States Center for Medicare Services (CMS) only consistently reimbursed for 24-hour ABPM performed for suspected white coat hypertension in individuals without a previous diagnosis of hypertension.⁴ In 2019, CMS announced that it would expand coverage of 24-hour ABPM to include suspected masked hypertension.⁵ This expansion of coverage was a welcome improvement, but was limited to a single 24-hour ABPM session per year. In contrast, internationally (including in most European and many Asian countries), 24-hour ABPM has been widely recommended and reimbursed for many years, without being limited to a single 24-hour ABPM session per year. With mounting evidence suggesting that 24-hour ABPM profiles may only be moderately (68% to 88%) reproducible in the short-term,^6–8 a single ABPM session may be inadequate.

In the current study,⁶ Cohen, et al. performed a rigorous evaluation of the short-term reproducibility of masked hypertension using successive 24-hour ABPMs, collected approximately one month apart, in a community-based sample of 254 adults who were not on antihypertensive medications and had normal office BPs. For their primary analyses, the authors used the 2017 American Heart Association/American College of Cardiology Blood Pressure Guideline’s definition of masked hypertension (mean office blood pressure <130/80 mmHg and mean awake ABPM blood pressure ≥130/80 mmHg).¹ Eligible participants were relatively young (mean age 38 years), 25% African American, 59% Hispanic, with low rates of comorbidities (<1% diabetic, 5% chronic kidney disease), and the majority endorsed moderate alcohol use (73%). Office blood pressures were on average lower than expected in patients with masked hypertension (mean 109/71 mmHg).^{1, 9}

The authors evaluated the reproducibility of masked hypertension across successive ABPM sessions by calculating the proportion of participants who consistently were classified as having (or not having) masked hypertension across the two ABPM sessions, described as agreement. The authors also calculated the kappa statistic, a formal metric of agreement across repeated measurements; a kappa statistic of 1 constitutes perfect agreement, a kappa statistic of 0 constitutes agreement that would happen by chance alone, and a kappa statistic of 0.5 represents moderate agreement. The authors observed that 24%−26% of participants had masked hypertension, with 81% agreement in the diagnosis of masked hypertension between the first and second ABPM sessions (Kappa statistic 0.50, 95% confidence interval [CI] 0.38–0.62). The results were similar using 24-hour and asleep ABPM, and oscillometric, as opposed to manual, office blood pressure measurements to define masked hypertension. In sensitivity analyses applying mean office blood pressure <140/90 mmHg and mean awake ABPM blood pressure ≥135/85 mmHg for the definition of masked hypertension, there was slightly improved reproducibility (88% agreement between sessions, Kappa statistic 0.62, 95% CI 0.51–0.72).

A small number of limited studies have previously evaluated the reproducibility of masked hypertension, using older diagnostic thresholds, and generally showing similar or worse short-term reproducibility of masked hypertension compared to the current study.^{7, 8} Cohen, et al. improved upon prior studies by restricting the cohort to those individuals who were eligible to have masked hypertension (i.e., those with normal office blood pressures), consistent with guideline recommendations for screening for masked hypertension. The authors observed that the newer thresholds were associated with slightly poorer reproducibility of masked hypertension on ABPM than older, less stringent thresholds. These findings support the need for greater precision in blood pressure measurement as diagnostic thresholds for hypertension decrease.

There remain important gaps in our understanding of the short-term reproducibility of masked hypertension. Twenty-four-hour ABPM is particularly useful for blood pressure measurement, in part, because it captures blood pressures during day-to-day activities and during sleep. ABPM, in essence, serves as a sensitive source of data collection of our blood pressure responses to routine daily exposures. For these same reasons, ABPM may be uniquely susceptible to inconsistencies across measurements. Factors that could feasibly contribute to short-term discordance in blood pressure readings include differences in activity level, drug exposures (e.g., caffeine, alcohol, tobacco, nonsteroidal anti-inflammatory drugs), sleep duration or quality, stress levels, and mood between ABPM sessions. However, there is a dearth of data on the reasons and risk factors for short-term irreproducibility of masked hypertension by 24-hour ABPM. Acknowledging the multitude of potential contributors to differences in ABPM readings across sessions, we have little guidance on which patients may require repeated out-of-office blood pressure measurements to appropriately diagnose masked hypertension.

The current study contributes important information to our knowledge of the limitations of ABPM. While the authors’ findings likely do not impact the prognostic implications of identifying masked hypertension by ABPM, they highlight the need to improve upon our current screening practices, in order to avoid misdiagnosis and over or undertreatment of hypertension. With the recent necessary expansions of telehealth amidst the novel coronavirus pandemic, out-of-office blood pressure monitoring is undoubtedly being used with accelerated frequency. We should be strongly considering self-blood pressure measurement at home (using appropriate technique, with a validated device) to aid in the diagnosis of hypertension, due to greater tolerability of repeated, longitudinal measurement and similar prognostic value to 24-hour ABPM.¹⁰ ABPM has a distinctive role in identifying high-risk blood pressure phenotypes (particularly nocturnal hypertension), but self-blood pressure measurement can be a valuable supplement to those readings obtained by 24-hour ABPM. Furthermore, while also susceptible to irreproducibility,⁷ self-blood pressure measurement can be more readily repeated in the short-term.

In conclusion, as healthcare providers, we need to more broadly adopt the notion that the diagnosis of hypertension should be thought of as an iterative process. No single method of blood pressure measurement is without flaws, including our reference standard, ABPM. Several factors have been associated with greater likelihood of masked hypertension and may be helpful to guide screening practices, including male sex, higher body mass index, current smoking, and higher office systolic blood pressure (e.g., close to, but not above, the office threshold for hypertension).⁹ However, we have no clear way to identify which patients require repeated out-of-office readings to determine the presence of masked hypertension. Furthermore, CMS currently only covers a single 24-hour ABPM session annually, and patients may not be willing to undergo repeated ABPM in the short-term, regardless of insurance coverage, due to poor tolerance and loss of productivity. In order to accurately diagnose, and thus appropriately treat, hypertension, we must rely upon multiple data points. We should be building a culture of blood pressure measurement that integrates blood pressures obtained from several, complimentary sources, including office visits, ABPM, and self-blood pressure measurement at home.