Why the study?
Are automated oscillometric home blood pressure monitors accurate compared to a reference standard in patients with hypertension?
Are automated oscillometric home blood pressure monitors accurate compared to a reference standard in patients with hypertension?
Answers are grounded in this paper and cite their sources.
A significant proportion of automated home blood pressure monitors used by patients are clinically inaccurate, highlighting the urgent need for routine accuracy testing of these devices in clinical practice.
In this issue of the American Journal of Hypertension, Padwal et al. report data that is extremely relevant clinically.1 A significant number of automatic oscillometric blood pressure (BP) monitors, owned by Albertans with hypertension, are inaccurate. In particular, in this well-designed study, the difference between systolic or diastolic BP taken simultaneously by the patient’s own automated oscillometric home BP device and by a 2-observer auscultatory reference standard BP measurement (mercury sphygmomanometer) was more than 5, 10, and 15 mm Hg in 69%, 29%, and 7% cases, respectively.1 These data are truly concerning given the fact that guidelines on the diagnosis and management of hypertension issued by major national professional organizations such as Hypertension Canada, American Society of Hypertension, Heart Foundation of Australia, and National Institute for Health and Care Excellence all endorse home BP monitoring.2–5 Accordingly, data on home BP readings do play a significant role in the decision making process for management of hypertension in the individual patient. In agreement with these trends, the market for oscillometric automated home BP monitors has expanded dramatically over the last decade or so, reaching over 1 billion US dollars.6 It is expected to increase by 13% annually and reach about 2 billion US dollars in 2022, mostly related to aging population, the high prevalence of hypertension in the elderly, and last but not least, by the ongoing emphasis on home BP monitoring.6 Hence, the accuracy of these monitors is crucial since both under- and over-treatment of hypertension could have adverse outcomes ranging from symptomatic hypotension, fall-related injuries, and hypoperfusion-induced cerebrovascular and cardiovascular adverse events on the one end of the spectrum, and hypertension related stroke, left ventricular hypertrophy, heart failure, and myocardial infarction on the other end. It is somehow perceived, however, that home BP monitors which are validated and endorsed by guidelines issued by the national professional organizations on hypertension is the same as accuracy. Padwal et al. dispel us of this belief. Both, certified and noncertified devices showed a very similar rate of clinically relevant inaccuracy. Furthermore, this is not the first study pointing out discrepancy between “perceived/believed versus real life accuracy” of automated oscillometric home BP devices. As a matter of fact, results of Padwal et al. are in agreement with studies by others including data from our Comprehensive Renal Hypertension Program in Ottawa and others.7,8 Given the fact that methods used by Padwal et al. were highly appropriate and that the scientific interpretation of the results is also sound, we are left with several questions. First, does validation and certification process include accuracy assessment, and if so, are methods for the assessment of accuracy of these devices rigorous and independent enough? Home BP monitors vary with respect to physical components (e.g., cuff, transducer) and the electronic algorithms; hence, many different standards and protocols have been developed over the years for validating these devices. The American Medical Instrumentation (AAMI) came out with standards in 1987, followed by the British Hypertension Society in 1990, the European Society of Hypertension (ESH) in 2002.9–11 The ESH developed a simplified validation procedure in 2010.12 Finally, there were revisions to the AAMI standard (with additional involvement of the International Organization of Standardization [ISO] and the American National Standards Institute [ANSI]) in 2009, with a second edition in 2013.9,10,13 In the United States (Food and Drug Administration [FDA]) and Europe (CE-certification), the second edition of the AAMI/ANSI/ISO standard is followed. These standards are all quite rigorous, prescriptive, and exhaustive in the order of procedures (see Table 1 for some salient features). Altogether, clearly the validation and certification of these devices includes rigorous criteria for an assessment of accuracy. Despite these standards, then, why do we see the results, reported by Padwal and others? Selected features of the existing Blood Pressure validation protocols Selected features of the existing Blood Pressure validation protocols There are several possible reasons (see Figure 1). Though the standards for validation are quite prescriptive, there is insufficient data to support that these procedures are actually carried out as they should be. As an example, a systematic review, which examined accuracy of validated ambulatory BP devices, reported that only 21% of published validation studies adhered to the established protocols.14 Second, one could speculate that there might be quality control issues with the manufacturing process that might potentially be a source of error in individual devices, despite the model/series itself being validated. Another source could be the “human–device interface”, i.e., the device itself is accurate in the population, but for some biological or algorithmic reason, it fails to produce accurate results in a particular individual. As Padwal and others have found no difference in accuracy rates between validated and certified devices as compared to noncertified ones, this certainly could be an issue. Algorithms from which systolic and diastolic BPs are derived vary from one brand to another and may suit more or less an individual patient. We are not aware of any study that tested an “inaccurate” oscillometric home BP monitor on a different patient(s). Finally, are these devices showing decay in accuracy from simply “old age” or inappropriate use? In other words, they were once accurate, but no longer are. Though the algorithm itself should not change with time, the mechanical components could potentially undergo damage or wear and tear, resulting in erroneous results. However, it appears that the age of the device is not an independent predictor of accuracy.1,15 On the other hand, these studies did not compare performance of these patient-owned devices to brand new equivalents. This was indeed assessed in a study by Akpolat et al., also suggesting that the age of the device is not an independent predictor of poor accuracy.7 Having said that, we believe that ongoing validation of these devices is an important issue perhaps not properly recognized by all national guidelines on diagnosis and management of hypertension. The British Hypertension Society and the National Institute for Health Research have the most rigorous advice for ongoing review of home BP devices.16 First, it recommends that BP monitors should be maintained and calibrated according to manufacturers’ instructions. It suggests that one should check when the monitor was purchased and/or last calibrated, and should not be used if outside the recommended timeframe for calibration/recalibration until it is recalibrated according to the manufacturer’s instructions or replaced. Furthermore, it also endorses as a good practice that the patients bring their BP machine with them to reviews so that an appropriately trained healthcare professional can check that a validated device and the right sized cuff are being used, and also that the device is being used in the correct way. Approach to Inaccurate Home Blood Pressure Monitor. We salute the study by Padwal et al. which together with other studies highlights the potential limitations of using home BP monitoring in the management of hypertension. There are many clinically relevant questions still left on the floor. These should engage national professional organizations, in the process leading to issuance of specific guidelines on when, where, how and how often should these devices be tested for accuracy. Given our current knowledge, we would endorse testing for accuracy of new devices as soon as possible after the purchase so that an inaccurate device can be replaced without cost incurred. We are less sure on how often these devices should be rechecked subsequently, perhaps annually and after any situation potentially leading to malfunction. Even more important is a clear message on who and how should one test accuracy in these devices for patients, in a way that is both practical and does not incur extra cost. This is all urgently needed to avoid confusion among patients and physicians regarding the endorsement of home BP devices on one side and more and more publications on poor accuracy of these same devices on the other side. The authors declared no conflict of interest. Dr Marcel Ruzicka and Dr Swapnil Hiremath receive research award from the Department of Medicine, University of Ottawa.
No takes yet. Share an insight, caveat, or question.
Ruzicka et al. (2017) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: