hypertension: arm position important

 A recent article affirmed that the arm position was important in measuring blood pressure, in a crossover trial (see htn arm position important JAMA2024 in dropbox, ordoi:10.1001/jamainternmed.2024.5213)

 

Details:

-- 133 participants aged 18-80yo in Baltimore Maryland from 2022-2023

-- individuals excluded: those with rashes, gauze dressings, casts, edema, paralysis, tubes, open sores or wounds, or AV shunts on both arms, “mental impairment”, pregnancy, or mid to upper arm circumference of more than 55 cm

-- in order to replicate a more typical clinical scenario, participants walked for two minutes and then underwent a five-minute seated rest period with their back and feet supported prior to having their blood pressure measured by an automated device.

-- patients were randomly assigned to a set of triplicate blood pressure measurements, with the following measurements performed in random order:

    -- arm supported on the desk, with midcuff at heart level

    -- hand supported on lap

    -- arm unsupported at the side

    -- a subsequent assessment of the arm supported on the desk, to account for intrinsic blood pressure variability

-- mean age 57 (56% were at least 60 years old), 53% female, 77% Black/16% white/97% non-Hispanic

-- 59% were prescribed antihypertensive meds, 48% had taken their medications the day of the study, 33% had a history of hypertension or MI, 82% had chronic care visits in the past year

-- baseline mean systolic blood pressure (SBP) 126 mmHg and mean diastolic blood pressure (DBP) 74 mmHg; 36% had systolic blood pressure of at least 130 mmHg and 41% had a BMI of 30 or more

-- results were stratified by hypertension status, age, obesity status, and access to healthcare within the past year

  

-- primary outcome: differences in mean SBP and DBP between the reference blood pressure (desk) and the two other positions (lap and side)

    -- calculation: lap or side BP minus the first assessment of desk BP minus the difference between the two desk BP measurements

-- subgroup analysis: hypertension above or below SBP of 130 mmHg, age >60 vs <60, BMI >30 vs >30, and access to healthcare within the past year

   

Results:

-- lap blood pressure as compared to desk blood pressure:

    -- difference in systolic blood pressure: 3.9 mmHg (2.5-5.2)

    -- difference in diastolic blood pressure: 4.0 mmHg (3.1-5.0)

-- side blood pressure as compared to desk blood pressure:

    -- difference in systolic pressure: 6.5 mmHg (5.1-7.9)

    -- difference in diastolic blood pressure: 4.4 mmHg (3.4-5.4)

 

-- the main difference between the two blood pressures done using the desk orientation above: -0.21 mmHg for SBP and 0.09 mmHg for DBP

-- SBP was overestimated in the group of hypertensive individuals by about 9 mmHg when their arms were positioned at the side 

-- these results were roughly similar for all the subgroups, including systolic blood pressure at least vs below 130 mmHg, age younger vs at least 60 years old, BMI at least 30 vs less than 30, or healthcare utilization of at least once in the past year

 

Commentary:

-- hypertension is the leading cause of cardiovascular disease and preventable mortality globally

-- clinical practice guidelines have emphasized utilizing a standardized way of assessing blood pressure in order to have a more reliable/consistent results: making sure the right blood pressure cuff size is selected, back is supported, feet flat on the floor with legs uncrossed, and with appropriate arm position with the mid-cuff positioned at heart level, as well as having the patient rest for 5 minutes in a chair prior to the BP measurement

-- the issue here is that nonstandard positions are used frequently in measuring blood pressure in clinical practice, including the arm resting on the lap or unsupported on the side

    -- for example, patients often get their blood pressure checked when they are sitting on an exam table without any support for their arm or feet on the floor

    -- or they have  their recorded blood pressure be the one done without resting, as may be done when patients are escorted from the waiting room (perhaps after a long, frustrating wait) to the office, or when in the emergency room (where, in my experience, the blood pressures are often bimodal, being very high from the anxiety or very low from the overwhelming sepsis.....)

 

-- we do know a couple of things about hypertension measurement:

    -- a study of medical students, assessing the 11 elements identified as necessary to get an accurate blood pressure measurement, found that on the average only 4 of the 11 elements were actually performed,  and only 1 of the 159 students performed all 11 elements correctly: https://gmodestmedblogs.blogspot.com/2017/09/the-most-important-hypertension-blog.html . The point here was that medical students do learn how to take blood pressure correctly in their training, and it was likely that their observations in clinical practice when shadowing clinicians reinforced incorrect ways to check blood pressure (eg in a hectic emergency room...)

    -- my own quite consistent finding is that the blood pressure I check with the patient waiting a couple of minutes in a dark room, breathing slowly in and out, and imagining a very restful scenario has found systolic blood pressure differences pretty often 30mmHg or so lower than when the patient is walking to the nursing station to get their blood pressure checked prior to my seeing them
    -- there is even question about how long a patient should be in a quiet situation prior to having their blood pressure checked, with some patients requiring a resting time of up to 25 minutes: https://gmodestmedblogs.blogspot.com/2022/07/hypertension-resting-time-of-25min.html

      

    -- ambulatory blood pressure is the preferred measurement to assess blood pressure, since those results are the most highly correlated with cardiovascular events, and performing ambulatory assessment is the priority of the USPSTF: https://gmodestmedblogs.blogspot.com/2015/01/uspstf-recs-on-ambulatory-blood.html . Several studies have found that it was only ambulatory blood pressures that correlated with future adverse clinical outcomes, and not clinic-based readings

     --  this result was also confirmed in the SPRINT trial: https://gmodestmedblogs.blogspot.com/2020/10/hypertension-office-bp-not-correlate.html

     -- there were pretty compelling data from several studies done several years earlier documenting  the benefit of ambulatory blood pressure monitoring (ABPM) vs clinic-based monitoring, leading to the European hypertension guidelines highlighting the large benefit of home-based or ambulatory blood pressure monitoring https://gmodestmedblogs.blogspot.com/2013/12/hypertension-guidelines-european.html

 

-- this current study did show that there were important differences in blood pressure evaluations depending on the patient’s arm position at the time, and that the quantity of these differences was potentially enough to change clinical management

    -- the recorded blood pressures were significantly higher in the lap or side position than in the standard desk position

-- prior studies have confirmed an overestimation of SBP by 4 to 23 mmHg and DBP by 3 to 12 mmHg in patients having their blood pressure checked when their arm was either unsupported or below heart level

-- the researchers in the current study, on applying data from the National Health and Nutrition Examination Survey, calculated that improper arm position would result in 22% of US adults (54 million individuals) being misclassified as hypertensive when using an SBP cutoff of 130 mmHg or higher

 

Limitations:

-- the randomization done in the study by chance did not actually equalize participants by their 4 main separate categories: systolic blood pressure above or below 130 mmHg, age greater than or less than 60, BMI greater than or less than 30, and last healthcare utilization being more less than one year

    -- in terms of healthcare utilization, only 8 of the 133 participants did not have care in the last year, a number too low to yield useful results

-- there are concerns about the baseline demographics and assessments: not a very diverse population, no information about important variables that might affect blood pressure (smoking and drinking coffee, both affect blood pressure if checked 30 minutes or so prior to the blood pressure recording; important psychosocial data that affect blood pressure such as general living conditions, stressors, etc). these issues could possibly affect the results and their generalizability to other populations

-- these subgroup analyses were broken down as binary variables where those 59 versus 60 years old, or BMI 29 versus 30, or systolic blood pressure 130 versus 129 mmHg would all be separated into different buckets in the analysis

-- there was no gold standard in this study: the current gold standard for hypertension in terms of the predictive value for clinically adverse events is ambulatory or home-based blood pressure monitoring. So, though it is important to have reliably consistent BP measurements, and that would support using the single desk position, it would be helpful to know how any of these measurements correlated with the actual gold standard of ABPM

 

So,

-- my guess is that the vast majority of clinicians do base their decisions around hypertension management according to clinic-based blood pressure assessments, per my anecdotal experience in 2 community health centers as well as in some hospital-based outpatient clinics

-- in this context, in terms of comparing blood pressure readings from one clinic visit to the next, it would be important to have a consistent, standardized approach to determining the blood pressure and utilizing the above desk method: arm supported on the desk, with midcuff at heart level measurement after at least several minutes of rest (preferably 5 minutes, as recommended)

    -- however, I think it should be emphasized that the best predictor of adverse clinical events is ambulatory blood pressure monitoring, though this is difficult to follow repeatedly in patients. As the poor cousin of this, I do feel that home-based blood pressure monitoring is a reasonable substitute, one that can be repeated regularly. The issue here is strongly encouraging patients to assess their BP when they are relaxed and sitting down in a quiet room for a few minutes, best with the lights lowered and no distractions, prior to measuring their blood pressure, and then remeasure it two minutes later and record this number. And it is useful to get some readings at different times of the day. I also make sure that they bring in their blood pressure cuffs to assure that they accurately reflect the blood pressure as compared to the BP at the clinic, which presumably has an accurate measurement. If they are equivalent, even with patients with clinic systolic blood pressures in the 150 mmHg plus range but with home blood pressures at goal, it does make sense to me to use the home blood pressure recordings to determine adequacy of blood pressure control (even some patients i have known for decades still have "white-coat hypertension" even though i do not wear a white coat). One advantage to this approach is that home blood pressure measurements empower patients to take more control over their health and also will give them direct feedback as to their blood pressure at different times (e.g. stressful times may increase the pressure, losing weight may decrease it, etc.). There was also a recent article suggesting that blood pressure self-monitoring with self-titration of antihypertensive medication according to algorithms worked out by clinicians did lead to improved long-term blood pressure control (see htn home based bp and med self-titration works JAMA2024 in dropbox, or doi:10.1001/jamanetworkopen.2024.10063)

 

geoff

-----------------------------------

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

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

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