Hypertension: office BP not correlate with clinical trial BP

 A recent study found a lack of concordance between blood pressure measurements made in the SPRINT hypertension trial vs those found in routine clinical practice, with striking variability between sites (see htn SPRINT vs office JAMAintmed2020in dropbox, or doi:10.1001/jamainternmed.2020.5028) 

 

Details: 

-- the SPRINT trial took place in 102 clinical sites with a total of 9361 participants, evaluating whether intensive BP treatment (target SBP <120 mmHg) vs standard treatment (target SBP <140 mmHg) had better clinical outcomes. Trial terminated after 3.3 years (vs anticipated 6 yrs) because of lower rates of cardiovascular disease events and all-cause mortality in the intensive treatment group. Diabetics were excluded in the trial

    -- Achieved blood pressures were: 121/69 vs 136/76 mmHg [see http://gmodestmedblogs.blogspot.com/2015/11/tighter-blood-pressure-control-sprint.htmlfor review of trial] 

-- the BP measurement technique used in the SPRINT trial was: BP was based on 3 automated readings using a properly sized cuff while the patient was seated, readings taken 1 minute apart after 5 minutes of quiet rest, and without speaking with the patient. cuff size was determined by measurement of arm circumference at the midpoint of the upper arm

-- 49 out of the total of 102 sites were used (with 3074 participants), since these sites had electronic health record (EHR) data recording at least 3 outpatient and at least 3 trial BP measurements from the SPRINT trial in the period within 6 months until the end of the trial 

-- 81% men, mean age 69 (30% at least 75yo), 65% white/30% black/4% Hispanic, baseline blood pressure 138/77 

-- mean eGFR 71, urine microalbumin 10, number of antihypertensive agents 1.9, 50% on a statin, 57% on aspirin 

    -- as compared to the overall SPRINT group, those included in the trial were less likely to be female and more likely to be white, had a lower baseline blood pressure, and were more likely to be treated with a statin or aspirin 

--the median number of outpatient BP measurements was 17, and 16 were taken during the SPRINT trial 

 

Results:  

-- intensive treatment group: mean SBP was 7.3 mmHg (-7.6 to -7.0) lower at SPRINT trial visits vs corresponding outpatient visits (128.2 mmHg) 

-- standard treatment group: mean SBP difference was 4.6 mmHg (-4.9 to -4.4), with a mean SBP at the trial visits being 134.6, and 139.3 mmHg in the outpatient setting 

 -- there were blood pressure reading differences depending on some demographic variables:  

    -- women: intensive group - 9.5 mmHg (-10.1 to -8.8); -6.2 mmHg in the standard treatment group (-6.9 to -5.5), vs

        --men: -6.7 mmHg in intensive group (-7.0 to -6.4); -4.2 mmHg in standard group (-4.5 to -3.9) 

    -- age: no significant difference between those <65 years old or 65-80 (both were -7.4 mmHg in the intensive group and -4.5 mmHg in the standard treatment), but for those > 80yo, there was a nonsignificant difference of -6.9 mmHg in the intensive group (-7.8 to -6.0) and -5.8 in the standard group (-6.7 to -4.9) 

 -- there was also a difference by site: those at the VA sites had a smaller mean difference in the intensive treatment group: -6.9 mmHg in the VA vs -8.0 mmHg other sites in the intensive group, though no significant difference in the standard group (this difference in the VA was independent of the fact that more men are in the VA system) 

    -- in the intensive treatment group, 2 of 46 sites (4%) had a median mean SBP difference of 2 mm or less, 16 of the 46 sites (35%) had a difference of 5 mm or less. 

    -- In the standard treatment group, 10 of 46 sites (22%) had a median mean SBP difference of 2 mm or less and 26 of 46 sites (57%) had a difference of 5 mm or less 

--but, the overall analysis of the spread of blood pressure differences showed a range from -30 mmHg to +45 mmHg in both the intensive and standard treatment groups[!!!!!] 

 

Commentary: 

-- the value of this study is that outpatient blood pressure readings do tend to differ from those found in clinical trials, from which guidelines arise. 

    -- overall, the average SBP measured in the trial was 5 to 8 mmHg lower than in the clinic visits 

    --though, the 2017 AHA/ACC guidelines for hypertension suggest a systolic blood pressure target of 130 mmHg, in part because of an anticipated 10 mmHg difference between study BP readings and those in routine clinical care 

-- but, most notably, the spread of the variability of pressure readings was quite dramatic, with pretty shocking differences from one site to the next 

    -- So, assuming that the standardized approach in the SPRINT trial was more accurate and more consistent than at the many different clinical sites, this huge deviation reinforces the often found of lack of clinical utility of office-based blood pressure readings: there been several blogs showing effectively no correlation between office-based blood pressure readings and actual clinical events, eg see http://gmodestmedblogs.blogspot.com/2015/01/uspstf-recs-on-ambulatory-blood.html 

    --a disappointing study found that medical students did pretty miserably in measuring blood pressure accurately: see http://gmodestmedblogs.blogspot.com/2017/09/the-most-important-hypertension-blog.html , and that this might well reflect that, despite how they were taught to measure blood pressure in school, their actual role models in their clinical rotations reinforced an incorrect approach (eg, quick measurements in the ED, or in ambulatory clinics, where patients were likely stressed and not given any time to relax)

-- the findings in this study supports the need for a more consistent approach to blood pressure measurement. Perhaps the most real-world accurate approach is through ambulatory blood pressure monitoring (ABPM), which quite consistently has proven a better marker of future clinical cardiovascular events than office-based measurements. And home-based blood pressure readings are a close second. see http://gmodestmedblogs.blogspot.com/2015/01/uspstf-recs-on-ambulatory-blood.html, and http://gmodestmedblogs.blogspot.com/2018/04/ambulatory-blood-pressure-monitoring.html 

    --another options with some appeal (though less data): using automated office BP machines (which function by themselves: eg after one puts the patient in a quiet room, a few minutes later the machine takes the BP a few times at intervals of 1 minute, then report results; these machines have somewhat differing algorithms). the results correlate with day-time ambulatory BP recordings (and not with office-based ones). see https://gmodestmedblogs.blogspot.com/2019/02/automated-office-bp-as-good-as-abpm.html 

 

-- also it is notable that a sizable minority of the participants (22%) in the above study had outpatient SBPs in their EHR from the clinics that was lower than their SPRINT trial measurements. This is sometimes referred to as masked hypertension (and, actually has a poor prognosis), see http://gmodestmedblogs.blogspot.com/2016/12/masked-hypertension.html. Masked hypertension does suggest that the clinic setting for some people may be less stressful than in other settings in which they are taking their blood pressure (e.g. home, pharmacy). Or that at home they are not relaxing for a few minutes before checking their blood pressure 

 

Limitations of study:  

-- they did not have a simultaneous comparison of blood pressure measurements recorded at the outpatient sites and by the SPRINT protocol. And there may have been significant differences because of this, even though they tried to use blood pressure values near the time of the trial. But it is certainly possible that diet, exercise, stress levels, etc were different during the different times of measurement, and that may have unduly influenced the results

 

So, 

--Given the large variability in blood pressure measurements at outpatient sites as compared to the standardized measurement in the SPRINT study, perhaps it makes sense to regularly compare values generated in a clinic setting vs another more accurate reading (e.g. ambulatory blood pressure monitoring or home-based blood pressure readings), or perhaps have some internal mechanism in the clinic where another person checks the blood pressure after a 5 minute rest in a quiet setting, without direct communication with the patient, then measuring the blood pressure every minute for 3 minutes and averaging the results), or trying the automated office BP machines. This would be a great quality assessment/improvement initiative, given how important it is to have accurate blood pressure readings in terms of improving future morbidity and mortality. 

-- the real take-home message here: hypertension is remarkably common in the US (and elsewhere), it is associated with major morbidity and mortality for huge numbers of people, we do have better and better meds to treat it (as well as non-med treatments), our blood pressures goals are determined by rigorous clinical studies, but our most common measurement tools (office-based blood pressure) hails from the stone age (no offense to those of us still living in the stone age)....

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list