Accurate blood pressure (BP) measurement remains the cornerstone of diagnosing and managing arterial hypertension (HTN). Despite decades of clinical use and extensive guideline development, office BP measurement (OBPM) continues to suffer from significant variability related to technique and protocols. [1–8]
International hypertension guidelines uniformly emphasize the importance of proper BP measurement, with proper patient preparation including rest, repeated readings, and averaging values. However, disagreement persists regarding the number of measurements required, the duration of rest before measurement, and which readings should be used for clinical decision-making (See Table 1). [1–8] However, does this lack of harmonization contribute to inconsistent BP classification, potential misdiagnosis, or suboptimal treatment decisions?
Office blood pressure measurement recommendations – a comparison between different international guidelines on hypertension.
| GUIDELINE | Number of measurements | Time between measurements | Averaged measurements | Comments |
|---|---|---|---|---|
| ESC 2024[1] | 3 (BP1, BP2, BP3) | 1-2 min | BP2 and BP3 | Further measurements should be obtained if the readings differ by > 10 mmHg |
| ESH 2023 [2] | 3 (BP1, BP2, BP3) | 1 min | BP2 and BP3 | HTN diagnosis made during ≥ 2 separate office visits (within four weeks) unless |
| ACC/AHA 2025 [3] | ≥ 2 (BP1, BP2, ‥, BPx) | 1 min | Average all BPs available | • Average of ≥ 2 BP measurements obtained on ≥ 2 separate occasions |
| ISH 2020 [4] | 3 (BP1, BP2, BP3) | 1 min | BP 2 and BP 3 | If BP1 < 130/85 mm Hg, no other measurement is required. |
| Canada Primary Care 2025 [5] | 3 (BP1, BP2, BP3) | 1 min | BP1, BP2, BP3 | All values are recorded and averaged. |
| Japanese Society of Hypertension 2019 [6] | ≥ 2 (BP1, BP2, ‥, BPx) | 1-2 min | Average of 2 measurements | The mean value of two measurements that provides stable values (difference between values: <5 mmHg) should be used |
| Korean Hypertension Society [7] | ≥ 2 (BP1, BP2, ‥, BPx) | Not stated | Average all measurements | At least two measurements of BP are taken and the average of those readings is used. |
| Clinical practice guideline for the management of hypertension in China [8] | 3 (BP1, BP2, BP3) | Not stated | BP2 and BP3 | For practical implementation in a clinical setting, the guideline suggests using the average of the last two measurement readings as the BP level. |
ACC/AHA – Amercan College of Cardiology / American Heart Association, BP – blood pressure, CVD – cardiovascular disease, ESC – European Society of Cardiology, ESH – European Society of Hypertension, HMOD – hypertension mediated organ damage, HTN – hypertension, ISH – International Society of Hypertension, and OBPM – office BP measurement
Furthermore, in busy clinical settings, adherence to BP measurement protocols can be challenging, and important procedural steps are frequently omitted, including the repetition of measurements. Would this be remedied by a more rapid protocol measuring only one BP?
In this issue of the Romanian Journal of Cardiology, Nora and Hangouche [9] present evidence from their May Measurement Month 2024 study in Morocco indicating that different OBPM approaches yield significantly different results, which may impact individual patient care.
They compared four different measurement approaches—a single reading without rest (P-5), the first reading after five minutes of rest (P1), the mean of three readings at rest (P123), and the average of the last two readings after rest (P23). Significant differences between P5, P1 and P123 were found when compared with P23, which is considered the gold standard. However, the magnitude in mean differences between the measurements progressively decreased with rest allowance and averaging measurements. The mean systolic BP difference was only 0.75 ± 2.63 mmHg between P123 and P23, while bigger differences were detected between P5 and P23 (4.68 ± 9.26 mmHg) and P1 and P23 (2.25 ± 7.9 mmHg). Moreover, when using these measurements to classify participants as hypertensive or non-hypertensive, or to classify hypertensive patients as controlled versus uncontrolled, the level of agreement between measurements increased when using values obtained after rest, and after averaging P123 and P23.
Therefore, there is new evidence that averaging multiple OBPMs yields lower mean BP values and differences, though minor, do exist when including the first measurement in the BP mean. Moreover, BP taken without rest is generally higher and classifies more patients as hypertensive or having uncontrolled HTN.
This effect is attributed to elevated initial readings related to the alerting response, as well as regression to the mean observed with subsequent measurements, among other factors. Furthermore, activities such as talking have been shown to induce transient increases in BP, with reported increases of approximately 9.1 mmHg in systolic and 4.5 mmHg in diastolic pressure. BP values typically require about five minutes to return to baseline levels following such stimuli. Accordingly, the implementation of a quiet rest period prior to OBPM appears to be a reasonable and methodologically sound practice. [10]
Despite the existing recommendations for OBPM, many clinicians report routinely performing a single measurement. Although this approach is endorsed by certain guidelines for patients with normal BP [4], Nora and Hangouche, [9] as well as other researchers, [10] have demonstrated that single measurements yield values that differ significantly from those obtained through repeated OBPM. For instance, when using P1, 40.5% of patients were classified as hypertensive, while using P23, only 30.7% of patients would receive this diagnosis.
Current guidelines strongly recommend the use of out-of-office BP measurements for both the diagnosis of HTN and the selection and monitoring of antihypertensive treatment, due to their ability to reveal true 24-hour patterns; capture nocturnal dipping; show day-to-day variability; and diagnose white coat/masked hypertension. Last but not least, ABPM (ambulatory blood pressure measurement) is also the best predictor of cardiovascular risk. [1–8] However, OBPM remains the cornerstone of current medical practice due to its greater availability and reduced cost. In this regard, increasing the number of OBPM performed seems to improve the agreement with ABPM. In a recent trial, no differences were noted between mean daytime BP with ABPM and the average of the second and third in-office measurement, while including the first BP measurement resulted in higher BP on average. [11]
Indeed, multiple OBPMs demonstrate stronger concordance with ABPM, but they also provide superior prediction of cardiovascular risk. Data from the CARTaGENE, a population-based survey comprising individuals aged 40 to 70 years, showed that obtaining three consecutive OBPMs at two-minute intervals was associated with improved prediction of cardiovascular outcomes. The third measurement, as well as its average with the second, demonstrated the strongest predictive performance. At a given BP level, the estimated 10-year cardiovascular risk was approximately two times higher when it was based on the third measurement compared with the first. [12]
It may be argued that, in the context of busy clinical practice, healthcare staff encounter practical difficulties in obtaining multiple BP measurements. Nevertheless, the use of automated office blood pressure (AOBP) devices warrants consideration, as exemplified by their implementation in the SPRINT trial. [13] With AOBP, a staff member initiates the device and then leaves the patient unattended in a quiet room. Following a predefined rest period, the device automatically records multiple measurements at fixed intervals and reports their average. This allows clinical staff to attend to other responsibilities in the meantime. Importantly, the mean of the multiple readings obtained via AOBP has been shown to be comparable to the mean awake ambulatory BP. [14] Moreover, the presence or absence of the medical stuff (attended versus unattended AOBP) does not seem to significantly impact the resulting average. [15] Moreover, with AOBP, it was recently shown that the rest period can be significantly reduced to as little as 30 seconds. [16]
An alternative to save time that has been suggested by some guidelines would be to do no further measurement if the first BP recorded is < 130/80 mmHg. [4] This practice could be misleading in some cases. For instance, a rare phenomenon known as short-term masked hypertension has been described, in which patients with a normal first OBPM have an elevated mean of the second to fourth OBPMs in subsequent measurement. [9] Cardiovascular risk — as well as vascular and all-cause mortality — also increases starting at values as low as 115/75 mmHg. [17] According to the European Society of Cardiology Guidelines on Hypertension, a BP of 120-139/80-89 mmHg is regarded as elevated BP and needs different management approaches according to baseline risk. [1] Therefore, it is important to acknowledge that even lower BPs, when measured with a standardized method, can provide valuable data about cardiovascular risk and facilitate the communication of preventive measures.
All international hypertension guidelines do recommend multiple OBPM; however, disagreement remains regarding which BP values and average should be taken into account. For instance, Nora and Hangouche’s finding [9] that even the inclusion or exclusion of the first reading significantly affects BP classification challenges us to reconsider whether our current approach to hypertension diagnosis is sufficiently rigorous. We would never accept hemoglobin or creatinine measurements that varied by 15-20% based on the laboratory’s protocol preference, yet we tolerate such substantial variability in BP assessment.
Although some differences reported in the paper [9] may appear minor and may not all be significant at the population level, they are clinically relevant at the individual level. Underestimation of BP increases cardiovascular risk, overestimation may lead to overtreatment and potential harm when BP drops below 120/70 mmHg. The diagnostic labelling of HTN carries additional adverse psychological and social consequences, including increased psychological distress and work absenteeism independent of BP severity or control.
Hypertension affects over 1.3 billion people worldwide [2] and is the leading preventable risk factor for cardiovascular disease and premature death. We cannot afford to have its diagnosis and management compromised by measurement variability. Harmonization of OBPM recommendations would probably enhance diagnostic accuracy, improve hypertension management, and strengthen the comparability of epidemiological data globally. Until then, clinicians and researchers alike should remain vigilant about the protocol used when interpreting BP values and making therapeutic decisions.