myers ont bp survey 2008

7/24/2019 Myers Ont Bp Survey 2008

1/4

AMERICAN JOURNAL OF HYPERTENSION 1

ARTICLESnature publishing group

BACKGROUND

An automated sphygmomanometer, the BpTRU, was used in a blood

pressure (BP) survey of 2,551 residents in the province of Ontario.

Automated BP readings were compared with measurements taken

by a mercury sphygmomanometer under standardized conditions

in a random 10% sample.

METHODS

BP was recorded in 238 individuals in random order using

both a standard mercury device and an automated BP recorder,

the BpTRU. All subjects rested for 5 min prior to the first

BP reading, which was then discarded. The mean of the

next three readings was obtained using the mercury device

whereas the BpTRU was set to record a mean of five readings

taken at 1 min intervals with subjects resting alone in a

quiet room.

RESULTS

The mean s.d. BP with the automated device was 115 16/71

10 mm Hg compared to 118 16/74 10 mm Hg for the manual BP

(P< 0.001). A systolic BP 140 mm Hg was present for 16 automated

and 19 manual readings. Similarly, the diastolic BP was90 mm Hg

for 9 automated and 14 manual readings. Linear regression analysis

showed that automated BP was a significant (P

< 0.001) predictor ofboth manual systolic and diastolic BP.

CONCLUSION

Conventional manual BP readings can be replaced by readings taken

using a validated, automated BP recorder in population surveys.

The slightly lower readings obtained with the BpTRU device (in the

context of reduced observersubject interaction) may be a more

accurate estimate of BP status.

Am J Hypertens2008; xx:xxx-xxx 2008 American Journal of Hypertension, Ltd.

Comparison Between an Automated and ManualSphygmomanometer in a Population SurveyMartin G. Myers1, Natalie H. McInnis2, George J. Fodor2and Frans H.H. Leenen2

1Division of Cardiology, Schulich Heart Centre, Sunnybrook Health Sciences

Centre, Department of Medicine, University of Toronto, Toronto, Ontario,

Canada; 2University of Ottawa Heart Institute Ottawa, Ontario, Canada.

Correspondence: Martin G. Myers ([email protected])

Received 10 September 2007; first decision 7 October 2007; accepted 17 November

2007. doi:10.1038/ajh.2007.54

2008 American Journal of Hypertension, Ltd.

Blood pressure (BP) surveys in the community have gener-

ally been perormed using mercury sphygmomanometry. Tisapproach requires extensive training or health proessionalsin order to standardize procedures or BP measurement, sothat accurate readings can be taken or thousands o subjectsby different study personnel. Rigorous programs or monitor-ing quality control also need to be initiated in order to ensurecontinued compliance with the study protocol or BP mea-surement until the survey is completed. Te net result is anestimate o the BP status o a population based upon conven-tional BP readings, but obtained at great expense with exten-sive human resource efforts in training and execution.

Te Ontario Survey on the Prevalence o High BloodPressure (ON-BP) decided to consider alternatives to the con-ventional measurement o BP using the mercury sphygmoma-nometer, in order to simpliy the process o evaluating the BPstatus o a large number o individuals in the community. Anattractive alternative was the use o a validated automated BPrecording device, which is designed specifically or the office

or clinic setting. Te most widely studied device to date is the

BpRU (BpRU Medical Devices, Coquitlam, BC, Canada)which is able to perorm five readings with subjects rest-ing quietly and alone in an examining room, thus decreasingsubjectobserver interaction, reducing anxiety and decreasingobserver measurement error, all o which would tend to givea better estimate o an individuals BP status. In hypertensivepopulations, the BpRU has been shown to reduce the white-coat effect with readings exhibiting a significantly higher cor-relation with the mean waking ambulatory BP, the currentgold standard or assessing BP status, compared to BP readingsobtained in routine clinical practice.1,2Mean BpRU readingshave exhibited a close approximation to the manual BP in bothvalidation studies3,4and in clinical practice.5

Te ON-BP was undertaken to record the BP o about 3,000randomly selected subjects residing in the community in theprovince o Ontario. In a random sample o 10% o participants,BP readings were taken with both the BpRU and a mercurysphygmomanometer under similar standardized conditions inorder to compare the automated BP readings with the conven-tional standard or BP surveys, the mercury device.

METHODS

Patient population. Adult residents o the province o Ontario,aged 2079 years, were randomly selected or enrollment inthe ON-BP survey using census data rom Statistics Canada.

[Q1]


2/4

2 AMERICAN JOURNAL OF HYPERTENSION

ARTICLES Automated Recording Devices for BP Surveys

Both men and women with and without high BP were eligi-ble or the survey. Potential subjects were contacted by homevisits to see i they were interested in participating in the sur-vey and attending a BP clinic (F.H.H.L., J. Dumais, N.H.M.,P. urton, L. Stratychuk, K. Nemeth, MM. Lum-Kwong, G.J.F.,unpublished data). A total o 2,551 individuals attended the

special clinic or BP measurement. One o every 10 partici-pants was selected to have their BP recorded using both theBpRU device and a standard mercury sphygmomanometer.Te order in which the devices were used was determined bythe individuals study number; those subjects with even lastdigits were measured with BpRU first and those with odd lastdigits were measured with the mercury device first. Te lefarm circumerence was measured and the appropriate cuff sizewas used or each device.

BP measurement. Manual BP was recorded by a registerednurse or registered nurse practitioner using a Baumanometersphygmomanometer (W.A. Baum. Copiague, NY) according

to the procedures o the Canadian Hypertension EducationProgram.6Subjects remained seated with the back supportedand lef arm at heart level. Tey were allowed to rest or 5 minin a quiet room beore the first reading was taken; this wasthen discarded. Tree more readings were taken, each 1 minapart and the mean value and heart rate were recorded or eachindividual. Automated BP readings were obtained using theBpRU (model BPM 300) afer patients rested alone in a quietroom or 5 min. Te device was set to take six readings at 1 minintervals with the first reading being discarded. Te meanautomated BP reading and heart rate were noted or each sub-ject. Te device was positioned in such a way as to prevent the

subject rom seeing the readings as they were taken. Tere wasa 2-min break between automatic and manual BP readings. Allmeasurements were obtained under similar conditions exceptor the two different BP recording techniques used.

Data analysis. Mean manual and automated systolic and dia-stolic BP values were tabulated as histograms showing the num-ber o participants or each 5 mm Hg interval o BP. A pairedt-test was used to assess the differences between the manualBP and automated BP readings. Delta values were calculatedas manual systolic BPautomated systolic BP and manual dia-stolic BPautomated diastolic BP. A one way analysis o vari-ance was used to analyze the differences between delta systolicand delta diastolic by age and cuff size. o assess the differencesbetween males and emales an independent t-test was used toanalyze the delta systolic and delta diastolic values. A linearregression analysis was perormed to examine the relationshipbetween the automated and manual BP readings with the auto-mated systolic and diastolic BPs as the independent variables.All data was analyzed using SPSS version 15.0.

RESULTS

A 10% sample (n= 238) o the entire survey population par-ticipated in this substudy. Te subjects in this substudy com-prised 96 males and 142 emales, mean (s.d.) age 50 15 and

47 15 years, respectively. Fify-nine subjects rememberedbeing told by a health proessional that they had hyperten-sion and, o these, 42 were receiving antihypertensive drugtherapy.

Individual readings or systolic and diastolic BP or each othe two methods were tabulated and displayed as requency

histograms (Figures 1 and 2). Diastolic BP was 90 mm Hgor 14 manual and 9 automated BP readings. Systolic BP was140 mm Hg or 19 manual and 16 automated measurements.Each data set was normally distributed.

Mean (s.d.) BP taken with the automated BpRU device was115 16/71 10 mm Hg compared to 118 16/74 10 mm Hgor the manual BP readings (P < 0.001). Parameters such asage, sex, and cuff size did not predict the differences in systolicand diastolic BP between the automated and manual measure-ments. For the automated/manual order o readings, the auto-mated and manual values (mm Hg) were 116 14/72 9 and118 15/75 10, respectively. Te corresponding results orparticipants having manual BP ollowed by automated BP read-

ings was 113 17/70 11 or the automated versus 117 17/74 10, or the manual BP. Te differences between systolicmanual readings and systolic automated readings were signifi-cantly (P< 0.01) smaller when automated was used first.

Te comparison o the manual versus the automated readingsshowed a high coefficient o correlation (r2) or systolic BP (r2=0.84) and diastolic BP (r2= 0.70) readings (Figure 3). Linearregression analysis showed that the automated systolic BP is asignificant predictor ( = 0.93, P< 0.001) o manual systolic BP

800

10

20

30

Frequency

40

0

10

20

30

Frequency

40

100 120

Automated systolic BP (mm Hg)

140 160 180 80 100 120

Manual systolic BP (mm Hg)

140 160 180

Figure 1 | Histogram representing number of participants (n= 238) for each

5 mm Hg intervals of automated and manual systolic blood pressure (BP).

400

10

20

30

Frequency

40

0

10

20

30

Frequency

40

60 80

Automated diastolic BP (mm Hg)

100 40 60 80 100

Manual diastolic BP (mm Hg)

Figure 2 | Histogram representing number of participants (n= 238) for each

5 mm Hg intervals of manual and automated diastolic blood pressure (BP).

Q2]

Q3]


3/4

AMERICAN JOURNAL OF HYPERTENSION 3

ARTICLESAutomated Recording Devices for BP Surveys

with a regression equation (Adjusted systolic BP = 11.4 + 0.9 automated systolic). Automated diastolic BP was also a signifi-cant predictor ( = 0.83, P< 0.001) o manual diastolic BP witha regression equation (adjusted diastolic BP = 15.6 + 0.83 automated diastolic). At lower BP values, readings taken with

the manual BP recorder were higher than corresponding read-ings obtained using the automated device, with the differencediminishing as BP values increased.

DISCUSSION

Te automated BpRU device was designed to minimize theimpact o observersubject interaction on the measuremento BP in the office/clinic setting.7 Tis approach removesseveral aspects o bias associated with conventional BP mea-surement using mercury sphygmomanometry.8 Te role othe observer in recording the BP is eliminated and replacedwith a validated, accurate, digital device programmed to

take readings at specific time intervals, thus eliminatingimprecision due to actors such as digit preerence, too rapiddeflation o the cuff, or reading up or down to influence thepatients BP status. Te absence o the observer rom theroom during readings also precludes conversation betweenthe subject and the observer, which is a actor known toincrease the BP.9Many individuals exhibit a all in BP withina minute or two afer being lef alone in a quiet room espe-cially in the context o a treatment setting such as a doctorsoffice or clinic.5

Te benefits o taking readings using an automated devicesuch as the BpRU are generally applicable to populationsurveys such as the ON-BP. Tus, one would anticipate lowermean BP values or the automated device compared to con-ventional manual readings obtained by a nurse, which is whatwas observed in the Ontario survey. In this instance, manualreadings exceeded automated readings by 3/3 mm Hg.

Tis difference represents the white-coat effect or a ran-dom sample o adult residents residing in Ontario. A similardifference was observed in hypertensive patients when man-ual readings recorded by a research technician outside o thetreatment setting were compared with the mean automatedreadings taken using the BpRU device.1In this instance, themanual readings were 3/2 mm Hg higher with the mercurysphygmomanometer.

Tese differences between automated and manual readingsare substantially less than reported or other hypertensivepopulations, when the manual readings were taken by physi-cians. Under these circumstances, the white-coat reactiontends to provoke a greater pressor response. In a series o50 hypertensive patients reerred to a specialty hypertension

centre or their management, the mean manual office BP was20/5 mm Hg higher than the mean o five automated BpRUreadings.5 Similarly, Beckett and Godwin2 noted a differ-ence o 11/3 mm Hg between the last routine office BP takenby a patients own amily physician and the mean automatedBpRU value. In a population survey o persons 65 years o ageand older with hypertension, Kaczorowski and colleagues10ound systolic BP readings (taken in a community pharmacyusing the BpRU device), produced a mean value 9 mm Hglower than the last routine office BP taken by the subjects ownamily doctor.

In the above studies, the automated readings were taken withthe BpRU device with subjects resting alone in a quiet room

in order to minimize actors that tend to provoke a white-coatreaction. However, i one perorms the automated readingsand manual BP measurements under standardized conditions,the mean values are quite similar. In a ormal validation studyreported by Wright et al.,4 mean BpRU values or systolicand diastolic BP differed rom reerence readings taken with astandard mercury device by only 0.2 4.3/1.4 4.2 mm Hg,respectively. In a study in clinical practice,5 the mean s.d.o two readings taken using a mercury sphygmomanometer(163 23/86 12) was similar to the first BpRU reading takenin the presence o the observer (162 27/85 12). Tus, auto-mated readings taken with the BpRU device closely approxi-

mate conventional measurements recorded with a mercurysphygmomanometer when taken under similar conditions.

In the ON-BP, the mean BP o the subsample was substan-tially lower than in previous studies involving hypertensivepatients with only 8.0/7.6% exhibiting high systolic/diastolicreadings using the manual recorder. Estimates o the cutoffpoints or normal versus automated BpRU values are still pre-liminary but the available data suggest conventional office BPat 140/90 mm Hg is equivalent to an automated BpRU read-ing o 135/85 mm Hg taken in the office with subjects restingalone in a quiet room.2In this population, 8.0/7.6% would bedesignated as being hypertensive according to a cutoff value oBP 140/90 mm Hg, or manual BP and 9.7/8.0% or BpRU135/85 mm Hg. Te number o subjects in this subset o theON-BP was too small to provide reliable comparative esti-mates o normal versus abnormal cutoff points or a diagnosiso hypertension with the two methods o measurement.

Linear regression analysis o the automated and manual BPdata provided a correction actor to convert the automatedreadings obtained in the survey into comparable manual BPreadings. Tis conversion makes it possible to compare dataderived rom BP surveys perormed using an automated BPrecorder with previous surveys that have employed manual BPmeasurement techniques. According to the linear regressionin this sample rom the ON-BP survey, automated systolic BP

180

160

140

ManualsystolicBP(mmH

g)

120

100

80

100

ManualdiastolicBP(mmH

g)

80

60

40

80 100 120

Automated systolic BP (mm Hg) Automated diastolic BP (mm Hg)

140 160 180 40 60 80 100

Figure 3 | A linear regression analysis was performed to examine the

relationship between the automated and manual blood pressure (BP)

readings with the automated systolic and diastolic BP as the independent

variables. The r2values for systolic and diastolic BP are 0.84 and 0.70,

respectively (P< 0.001).

[Q4

[Q5


4/4

4 AMERICAN JOURNAL OF HYPERTENSION

ARTICLES Automated Recording Devices for BP Surveys

readings o 120 and 140 mm Hg are equivalent to manual read-ings o 123 and 142 mm Hg, respectively. Automated diastolicBP readings o 80 and 90 mm Hg are equivalent to manualreadings o 82 and 90 mm Hg, respectively.

Te results o the ON-BP have demonstrated that conven-tional manual BP readings taken using mercury sphygmoma-

nometry can be replaced by a validated, automated recorder.I a major effect o using the BpRU is to minimize the white-coat effect it would not be surprising to see lower BP readingsin participants with high normal or mild hypertensive read-ings, possibly indicating a lower prevalence o hypertension.Instead o underestimating hypertension, the automated read-ings may actually reflect the true hypertension status in thepopulation, by minimizing the white-coat effect when onetakes into effect the close relationship between automated BPreadings and mean waking ambulatory BP, the current goldstandard or assessing cardiovascular risk.

Acknowledgment:This survey was supported by a Heart and Stroke

Foundation of Ontario contract awarded to F.H.H.L. and G.J.F. F.H.H.L. holds

the Pfizer Chair in Hypertension Research, an endowed chair, supported

by Pfizer Canada, University of Ottawa Heart Institute Foundation, and

Canadian Institutes of Health Research.

Disclosure: The authors declared no conflict of interest.

1. Myers MG, Valdivieso MA. Use of an automated blood pressure recording device,the BpTRU, to reduce the white coat effect in routine practice.Am J Hypertens2003; 16:494497.

2. Beckett L, Godwin M. The BpTRU automatic blood pressure monitorcompared to 24 hour ambulatory blood pressure monitoring in the assessmentof blood pressure in patients with hypertension. BMC Cardiovasc Disord2005; 5:18.

3. Mattu GS, Perry TL Jr, Wright JM. Comparison of the oscillometric blood pressuremonitor (BPM-100

) with the auscultatory mercury sphygmomanometer.Blood Press Monit2001; 6:161165.

4. Wright JM, Mattu GS, Perry TL Jr, Gelfer ME, Strange KD, Zorn A, Chen Y. Validationof a new algorithm for the BPM-100 electronic oscillometric office blood pressuremonitor. Blood Press Monit2001; 6:161165.

5. Myers MG. Automated blood pressure measurement in routine clinical practice.Blood Press Monit2006; 11:5962.

6. Hemmelgarn BR, McAlister FA, Grover S, Myers MG, Mckay DW, Bolli P, Abbott C,Schiffrin EL, Honos G, Burgess E, Mann K, Wilson T, Penner B, Tremblay G,Milot A, Chockalingam A, Touyz RM, Tobe SW; Canadian HypertensionEducation Program: The 2006 Canadian Hypertension Education Programrecommendations for the management of hypertension: Part 1Blood pressuremeasurement, diagnosis and assessment of risk. Can J Cardiol2006; 22:573581.

7. Myers MG, Godwin M. Automated measurement of blood pressure in routineclinical practice.J Clin Hypertens2007; 9:267270.

8. Reeves RA. Does this patient have hypertension? How to measure bloodpressure. JAMA1995; 273:12111216.

9. Le Pailleur C, Helft G, Landais P, Montgermont P, Feder JM, Metzger JP, Vacheron A.The effects of talking, reading, and silence on the White Coat phenomenon inhypertensive patients.Am J Hypertens1998; 11:203207.

10. Sullivan SM, Kaczorowski J, Myers MG, Karwalajtys T, Chambers LW. Use ofautomated blood pressure measurement to reduce white coat response in apharmacy setting. Can J Cardiol2007; 23(Suppl C): 85C.Q6]

myers ont bp survey 2008

Documents