You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.

Primary Care Corner with Geoffrey Modest MD: USPSTF Recs on Ambulatory Blood Pressure Monitoring

15 Jan, 15 | by EBM

By: Dr. Geoffrey Modest 

The US Preventive Services Task Force (USPSTF) published a draft statement recommending ambulatory blood pressure monitoring (ABPM)….finally!!! (see here). If you have seen my frequent blogs on this over the past few years, I have been a strong advocate for using ABPM, as reinforced in the NICE guidelines (Natl Institute for Healthcare and Clinical Excellence, the UK organization which reviews and establishes guidelines for the UK). The NICE guidelines were, I think, really thoughtful, thorough, and well-referenced, and, among other things, promoted the use of ambulatory or home-based monitoring (as well as lots of information and guidelines about workup of hypertension, nonpharmacologic and pharmacologic management, special cases – including a great section on resistant hypertension).

The background rationale for doing ABPM, to me, is:

–An old study in NEJM (the first I saw, which was actually looking at “masked hypertension”, where people actually have lower BP in the office!) found minimal correlation between office-based blood pressure and clinical events, but a profound one with ambulatory monitoring (see Clement DL et al. N Engl J Med 2003;348:2407-15.). In this study of 1963 patients followed 5 years, there were 157 cardiovascular events, and, controlling for the routine risk factors, they found that the 24-hour ABPM correlated strongly with cardiovascular events (see graph below: for each office systolic blood pressure noted on the bottom of the graph, there was a significant difference in cardiovascular events per the ABPM systolic with the cutpoint of 135 mmHg. there was a much less impressive relationship with the office based systolic and cardiac events.) Of note, controlling for office-based blood pressure, there was a continuing profound association with ABPM.


–A systematic review in 2011 found that 8 of 9 studies done found ABPM was superior to clinic-based blood pressure in predicting clinical outcomes (see Hodgkinson J et al. BMJ 2011;342:d3621)

–22 studies reviewed by USPSTF have found that about 30% (range of 5% to 65%) of people with elevated office-based hypertension do not have hypertension confirmed by ambulatory monitoring (leading to inappropriate medicalization and exposure to meds).

–A study of 556 patients with resistant hypertension similarly found that, controlling for other risk factors, both the ambulatory systolic and diastolic blood pressures were associated with cardiovascular events, with no association with the office based blood pressures. In fact 40% of those with “resistant hypertension” based on office readings did not have elevated blood pressure based on ABPM… (Arch Intern Med. 2008;168(21):2340-2346). This study reinforces that the issue of “over-reading” hypertension applies to those with very high initial blood pressures as well as those with only low levels of hypertension per office-based readings and questionable diagnoses of hypertension in the first place.

–The data on home-based monitoring — HBPM — (which is mostly what I suggest to patients) is less robust, though studies have suggested it is a reasonable proxy (one study comparing ABPM with HBPM found that office-based blood pressure was only marginally significantly related to target-organ damage, but either HBPM or ABPM were dramatically related, and there was only a small trend in favor of ABPM over HBPM (see G Mule. J Cadiovasc Risk 2002). I make sure the patient has a correctly-sized cuff (not the one on the wrist, which is notoriously inaccurate) and that they bring it in so that I can check readings with my manual cuff on one arm and the patient does their own ambulatory reading simultaneous on the other to make sure they are doing it correctly (I then usually switch and check the other arms).

–So, when should one do ABPM or HBPM? It is difficult to make real evidence-based medicine type decisions because of lack of all of the necessary data. On the one hand, as noted in USPSTF, “good quality evidence that screening for hypertension has few major harms and provides substantial benefits”, and I would add, these older studies are uniformly based on office-based (or more accurately, study-based) measurement of blood pressure. But, on the other hand, ” the USPSTF considers ambulatory blood pressure monitoring to be the reference standard confirming the diagnosis of hypertension.” So, my best guess in terms of how to put these apparently diverse findings together is: hypertension matters a lot, but the sensitivity and specificity of office-based readings is much less than ABPM, reflecting the fact office-based blood pressure is a composite of real hypertension in about 70% of people with its attendant clinical consequences combined with the non-significant high BP readings in about 30%, which serves to dilute the effect of the real hypertension in the overall analysis. This conclusion is reinforced by the fact that of the 7 studies which have looked at it, there is a consistent finding that elevated ABPM significantly confers increased cardiovascular and mortality risk after controlling for office-based blood pressure.

My approach is:

 –I get an ambulatory or home-based reading if the level of hypertension if the office-based reading is only a little elevated after several checks over time, making sure that I have an accurate reading (patient sitting quietly for 5 minutes, where I typically turn off the light, go into another room to write my note, then return to check the pressure). The cutpoint suggested by USPSTF of 180/110 — see below — is probably reasonable, though this is not rigorously evidence-based. Sometimes the patient can go to the local pharmacy to get a check, though I advise them to wait 5 minutes sitting down comfortably prior to the check. [By the way, I pretty much always check the blood pressure myself in the clinic manually, as above. I often find huge discrepancies with the automated readings of the medical assistants (eg up to 30-40 mmHg). I think some of the discrepancies are related to issues where the ambulatory readings are less accurate (eg atrial fibrillation), though most are because many of my patients are pretty deconditioned, and the medical assistants are not able to bring in a patient, wait 5 minutes, then get the automated blood pressure.]

The bottom line advantage of getting an accurate blood pressure reading (for which I rely heavily on HBPM or ABPM readings) is to avoid labeling patients inaccurately with the attendant consequences of medicalization, and avoid giving medications if unnecessary. That being said, I think it is still very important to stress healthy lifestyle, with appropriate weight, diet, exercise (which is reinforced by some old studies finding that those with white-coat hypertension are at increased risk of developing sustained hypertension, and may therefore benefit from the lifestyle changes, as found in the hypertension prevention trials).

–If the non-office based reading is consistently in the goal range, I usually ignore the office-based reading. This can be hard to do if the patient pretty consistently has office-based readings which are 180/95 yet 140/75 at home with a validated cuff.  But the consequence of over-treating, especially in the elderly, I think far outweighs my discomfort in just following the HBPM readings.

 These USPSTF recommendations are still in a draft form, open to comment until 1/26/15, and they may be changed prior to their final release.  But I do think the data are pretty strong and suspect they will not be changed significantly. I should also add that these draft recommendations include new recommendations on the screening interval:

–Annual screening after age 40 and those at risk for hypertension (those with high normal BP of 130-9/85-89), overweight/obese, or African-American

–Those 18-39 with normal blood pressure and no risk factors should be rescreened every 3-5 years. If blood pressure is elevated, “confirming the diagnosis with ambulatory blood pressure monitoring”

–They suggest immediate treatment in those with very high BP (>180/110), if there is end-organ damage, or in those with secondary hypertension.

–I should also add the somewhat contradictory and more aggressive recommendations of the NHBLI/NIH on pediatric screening, which is to check all kids >3 years old “at least once during every health care episode”, and also supports the use and increased accuracy of ABPM (see the 2005 guidelines).​ Though I should further add that USPSTF does not recommend screening kids and adolescents for hypertension because of inadequate evidence. As you might have gathered, I am in favor of more aggressive screening, such as for each regular health visit.

By submitting your comment you agree to adhere to these terms and conditions
You can follow any responses to this entry through the RSS 2.0 feed.
EBM blog homepage

Evidence-Based Medicine blog

Analysis and discussion of developments in Evidence-Based Medicine Visit site

Creative Comms logo

Latest from Evidence-Based Medicine

Latest from EBM