mercury versus non-mercury sphygmomanometers: a review of ... · mercury leak, which is a major...

Disclaimer: The Rapid Response Service is an information service for those involved in planning and providing health care in Canada. Rapid responses are based on a limited literature search and are not comprehensive, systematic reviews. The intent is to provide a list of sources and a summary of the best evidence on the topic that CADTH could identify using all reasonable efforts within the time allowed. Rapid responses should be considered along with other types of information and health care considerations. The information included in this response is not intended to replace professional medical advice, nor should it be construed as a recommendation for or against the use of a particular health technology. Readers are also cautioned that a lack of good quality evidence does not necessarily mean a lack of effectiveness particularly in the case of new and emerging health technologies, for which little information can be found, but which may in future prove to be effective. While CADTH has taken care in the preparation of the report to ensure that its contents are accurate, complete and up to date, CADTH does not make any guarantee to that effect. CADTH is not liable for any loss or damages resulting from use of the information in the report. Copyright: This report contains CADTH copyright material. It may be copied and used for non-commercial purposes, provided that attribution is given to CADTH. Links: This report may contain links to other information available on the websites of third parties on the Internet. CADTH does not have control over the content of such sites. Use of third party sites is governed by the owners’ own terms and conditions.

TITLE: Mercury versus Non-Mercury Sphygmomanometers: A Review of the Comparative Clinical Effectiveness and Guidelines

DATE: 25 September 2012 CONTEXT AND POLICY ISSUES A mercury manual sphygmomanometer is a device attached to the forearm to record arterial pulsations. Basic components include an inflatable cuff to put pressure on the artery, a pressure-measuring scale, and a bulb to inflate the cuff. As the cuff is inflated, the mercury in the manometer rises through the column, and the pressure is displayed on the scale. A stethoscope is placed over the brachial artery of the elbow to detect the systolic and diastolic pressure during the slow deflation of the pressure of the cuff. Since the instrument relies on the human judgment to detect systolic and diastolic pressure, it is termed an auscultatory device.1 Accuracy of mercury sphygmomanometers depend on proper validation, calibration, and maintenance. Several factors may affect the accuracy including measurement technique, proper cuff size, proper placement of the cuff, accurate recording of the systolic and diastolic endpoints, patient’s level of anxiety, and patient-physician interaction (e.g. “white coat response”).1 Poor maintenance of the mercury sphygmomanometer may result in loss of accuracy due to equipment deterioration such as an obscure mercury column due to dirt or mercury oxidation, faded scales that are hindering the readings of mercury meniscus, and mercury leak, which is a major health concern.2 Despite its limitations, the mercury sphygmomanometer is still considered “gold standard” in terms of accuracy.3 There are two common types of non-mercury sphygmomanometers: aneroid and oscillometric devices.1 The aneroid devices are also operated manually and require a normal blood pressure cuff and a stethoscope, but are liquid-free and use mechanical parts to relay the blood pressure to the gauge.1 The oscillometric devices are operated automatically, and the inflation and deflation of the cuff are controlled electronically.1 Based on the pressure-wave changes, an algorithm in the oscillating device is used to calculate the systolic and diastolic pressures, which are displayed on a digital readout.1 Accuracy of mercury-free sphygmomanometers also depends on proper maintenance, calibration and validation.4 The perceive benefits of oscillometric devices are that they are more accurate, less time consuming, easier to operate, and safer in terms of being mercury-free.5,6 Oscillometric devices may result in greater “within-subject” reliability than manual readings, by eliminating digit preference (the preference to round a value to a lower or higher one), observer bias and the white coat effect.6 However, the

Mercury versus Non-Mercury Sphygmomanometers 2

accuracy of oscillometric devices remains questionable and the use of oscillometric devices is not clinically appropriate when patient has an arrhythmia.7 The aim of this report is to review the comparative accuracy, safety and guidelines regarding the use of mercury versus non-mercury sphygmomanometers for the measurement of blood pressure. RESEARCH QUESTIONS

1. What is the evidence regarding the comparative accuracy of mercury versus non-mercury sphygmomanometers for the measurement of blood pressure?

2. What are the evidence-based guidelines regarding the use of mercury versus non-mercury sphygmomanometers in higher risk patient groups such as pregnancy and pediatric specialty clinics?

KEY MESSAGE Automated (oscillometric) office blood pressure devices, with proper maintenance, calibration and validation, can be used as a replacement for the mercury sphygmomanometer in office settings for the measurement of blood pressure. In pregnant women, children, patients with arrhythmia, and hypertensive or hypotensive patients, identified guidelines recommend that the measurement readings of automated devices should be validated with a mercury sphygmomanometer. There is no evidence on the comparison of aneroid devices with mercury sphygmomanometer. METHODS Literature Search Strategy A limited literature search was conducted on key resources including PubMed, The Cochrane Library (2012 Issue 8), Canadian and major international health technology agencies, as well as a focused Internet search. Methodological filters were applied to limit retrieval to health technology assessments, systematic reviews, meta-analyses, randomized controlled trials, non-randomized studies containing safety data, and guidelines. The search was also limited to English language documents published between August 27, 2007 and August 27, 2012. Selection Criteria and Methods

One reviewer screened the titles and abstracts of the retrieved publications and evaluated the full-text publications for the final article selection, according to selection criteria presented in Table 1. Table 1: Selection Criteria

Population

Q1: Adult and pediatric patients in hospital or medical clinic

Q2: High risk patients

Intervention

Non-mercury sphygmomanometers

Comparator

Mercury sphygmomanometers (gold standard)


Outcomes

Accuracy and calibration needs for non-mercury sphygmomanometers

Possible harms (acute inhalation or chronic exposure) due to mercury spill with mercury sphygmomanometers

Guidelines regarding the use of non-mercury sphygmomanometers

Study Designs

Health technology assessments, systematic reviews, meta-analyses, randomized controlled trials, non-randomized controlled trials (for safety only), guidelines

Exclusion Criteria Studies were excluded if they did not satisfy the selection criteria in Table 1, if they were published prior to 2007, duplicate publications of the same study, or included in a selected health technology assessment or systematic review. Critical Appraisal of Individual Studies The methodological quality of the systematic reviews were assessed using AMSTAR checklist.8 The quality of the RCTs and observational studies was assessed using Downs and Black checklist.9 The Appraisal of Guidelines Research & Evaluation (AGREE) instrument10 was used to evaluate the quality of the included guidelines. For the critical appraisal of studies, a numeric score was not calculated. Instead, the strength and limitations of the study were described. SUMMARY OF EVIDENCE Quantity of Research Available The literature search yielded 55 citations. Upon screening titles and abstracts, eight potential relevant articles were retrieved for full-text review. Ten additional relevant reports were retrieved from other sources. Of the 18 potentially relevant articles, nine were included in this review. They are two systematic reviews,11,12 three RCTs,13-15 and four guidelines.16-19 The study selection process is outlined in a PRISMA flowchart (Appendix 1). Summary of Study Characteristics A summary of study characteristics can be found in Appendix 2. Systematic reviews and meta-analyses:

The systematic review and meta-analysis by Stergiou et al. 201211 assessed the accuracy of automated blood pressure measurement in patients with sustained atrial fibrillation. Eight observational studies (12 validations: four ambulatory, five home, and three office devices) published between January 1986 and October 2011 with a total of 566 participants (57% men, mean age 73 years) with sustained atrial fibrillation were included. Sample sizes ranged from 20 to 255 participants. A standard mercury sphygmomanometer was used as reference method in five validations, a Hawksley random-zero method in five, and an aneroid sphygmomanometer in two. For comparison of the two methods, blood pressure was measured either sequential (nine validations) or simultaneous (three validations). Average values pooled by random-effects meta-


analysis and accounting for heterogeneity were estimated with 95% confidence intervals for correlation coefficients and for differences between automated and auscultatory blood pressure measurements (systolic and diastolic).

The systematic review by Skirton et al. 201112 compared the accuracy and appropriateness of auscultatory (manual) and oscillometric (automated) devices for measuring blood pressure in clinical settings. Sixteen studies (15 observational studies and 1 RCT) published in English between January 1997 and May 2009 were included. Quality of the evidence was appraised and the results were summarized in a narrative form. The score for quality appraisal ranged from 45 to 95%, covering study design, subject group, means of analysis, sample size and estimates of variance. In nine of the studies, the manual reading was done by “trained observers”, while in three studies the readings were made by clinical nurses. In other studies, measurements were taken by nurses with or without specific training or by medical residents. The practice settings for the studies included emergency care, hypertension clinics, general wards, or outpatient clinics. Sample sizes ranged from 33 to 997 participants. The way measurements were taken and recorded varied among studies. Randomized controlled trials:

The trial by Brown et al. 201213 was a prospective randomized open study with blinded endpoints design. The objective was to determine whether women diagnosed with hypertension in pregnancy, using traditional mercury sphygmomanometers, have equivalent pregnancy outcomes when blood pressure is measured thereafter by either an automated blood pressure recorder or a mercury sphygmomanometer throughout pregnancy. Pregnant women with hypertension (mean age 31 years, N=220) were randomized into one of the two groups: the mercury sphygmomanometer (n=110) or the automated (Omron HEM-705CP) blood pressure recording group. The primary outcome was the number of women in each group developing any episode of severe hypertension (≥170 mm Hg systolic and/or ≥110 mm Hg diastolic), which was used as a surrogate marker for detecting the risk of stroke and alerting the cases of severe preeclampsia or gestational hypertension. Secondary endpoints included gestational time at birth, small for gestational age (<10th centile) rates, caesarean section, and induction of labor rates.

The multi-site cluster randomized controlled trial by Myers et al. 201114 compared the quality and accuracy of manual office blood pressure and automated blood pressure using the awake ambulatory blood pressure as a gold standard. The setting was primary care practices in five cities in eastern Canada. The sites consisted of either one family physician or a group of two or three physicians sharing the same office space. Adult patients (over 45 years, N=555) with systolic hypertension (160 mm Hg / 95 mm Hg for untreated or 140 mm Hg / 90 mm Hg for treated) and no serious comorbidities were randomly assigned to either manual office blood pressure with continuing use of manual sphygmomanometer (control group) or automated office blood pressure (intervention group) using the BpTRU device. The baseline characteristics were balanced between groups in terms of age, gender, duration of hypertension, antihypertensive treatment and cigarette smokers. The BpTRU is a fully automated sphygmomanometer that records blood pressure by an oscillometric method. It was set to take readings at two minute intervals (from the start of one reading to the start of the next one) while the patients were left alone during five readings. For the manual office blood pressure control group, physicians continued recording blood pressure as before, without any additional instructions on proper blood pressure measurement technique. All patients were instructed to have 24 hour ambulatory blood pressure monitoring using a Spacelabs Model 90207 unit, which has been validated for accuracy. The device was set to record blood pressure at 15 minute intervals between 6:00 and 22:00 and at 30 minute intervals during the night. The primary outcome was


the mean estimated difference between the mean awake ambulatory blood pressure and the automated blood pressure (intervention group) or manual office blood pressure (control group) at the first office visit after enrolment.

Lamarre-Cliché et al. 201115 conducted a randomized, crossover study to compare the blood pressure measurement by mercury sphygmomanometer and other three blood pressure measurement modalities, namely the automated office blood pressure (AOBP) device (BpTRU), the ambulatory blood pressure measurement (ABPM) device (Spacelabs, model 90207-30), and self-measurement (Omron HEM-780CAN). Patients (mean age 58 years, N=101) from hypertension clinic with average blood pressure of 134 ± 12.7/83.6 ± 8.8 mm Hg were recruited and randomized to either the office or home group. The office group was then randomized to undergo an AOBP – Sphygmomanometer or Sphygmomanometer – AOBP sequence. The home group was randomized to undergo either a self-measurement – ABPM or ABPM – self measurement sequence. The total length of the study was 8 to 14 days. The primary study outcome was achievement of blood pressure goal.

Guidelines:

The characteristics on the grading of recommendations and levels of evidence used to develop the corresponding guidelines are summarized in Appendix 3. Summary of Critical Appraisal

The strengths and limitations of included studies are summarized in Appendix 4.

The methodological quality assessment of the systematic review by Stergiou et al. 201211 met 9 out of 11 items, while that of the systematic review by Skirton et al. 201112 met 8 out of 11 items of the AMSTAR criteria. In Stergiou, the scientific quality of the include studies was not assessed and considered in the conclusions. In Skirton, the likelihood of publication bias was not assessed, methods of study selection and data extraction were not described, and there was no pooling of the results from the included studies.

Overall, all three included RCTs13-15 met most items of the Downs and Black checklist for measuring study quality. A major limitation of all RCTs was the lack of identification of potential confounders. One RCT13 made attempt to blind the observers undertaking data collection and analyzing data.

Although the scope and purpose were explicit, the included guidelines16-19 suffered from multiple limitations, covering stakeholder involvement, rigour of development applicability, and editorial independence according to AGREE instrument. Methods used to search for the evidence were not reported in the identified guidelines. One guideline16 reported procedures for updating the recommendations. Clarity and presentation criteria were met in three guidelines.16,17,19 Summary of Findings The main study findings and authors’ conclusions from the clinical studies can be found in Appendix 5. Systematic review:

In the systematic review and meta-analysis by Stergiou et al.,11 automated measurements were higher than manual blood pressure measurements [pooled average systolic difference 0.5 mm Hg, 95% CI -0.9, 1.9; diastolic 2.5 mm Hg, 95% CI -0.6, 5.7). Pooled correlation coefficients between automated and manual bleed pressure measurements were stronger for systolic than


diastolic blood pressure (r=0.89 versus 0.76, P<0.001). Four of eight studies reporting data regarding the within 5 mm Hg ESH-IP criterion have passed for SBP (at least 65% of comparisons within 5 mm Hg) and two for DBP, suggesting better measurement accuracy for systolic blood pressure. Three studies showed no impact of heart rate on the blood pressure differences between automated and manual measurements. It was concluded that automated blood pressure monitors are accurate in measuring systolic but not diastolic blood pressure in elderly with atrial fibrillation.

In the systematic review by Skirton et al 2011,12 oscillometric (automated) devices were found less accurate than the mercury sphygmomanometer in 10 of the included studies. However, in most case the authors of the included studies concluded that the differences were not clinically significant, and the oscillometric devices could be used in the clinical settings, with the exceptions being used with hypertensive patients, patients with arrhythmia and patients with trauma. Randomized controlled trial:

In the trial by Brown et al. 2012,13 blood pressures obtained by six sequential recordings (alternating between mercury and automated) at randomization, according to the allocated group were similar. Both groups had similar maternal and fetal outcomes. The percentage of women with any episode of severe hypertension was similar between groups (39% for mercury vs. 44% for automated, P=0.5). The average number of episodes of severe hypertension per woman affected was also similar between groups (3 ± 2 for mercury vs. 4 ± 5 for automated, P=0.3). Small gestational age (<10th centile) occurred more frequently in those using automated blood pressure than mercury sphygmomanometer (17% vs. 12%), but the difference was not statistically significant (P=0.3). Birth weight and perinatal mortality were similar among groups. It was concluded that the use of automated blood pressure recording is associated with similar maternal and fetal outcomes compared with mercury sphygmomanometer.

In the trial by Myers et al. 2011,14 both groups showed a drop in mean office blood pressure between the pre-study office visit and blood pressure recorded after enrolment. The reduction in the automated group was numerically higher than that in the manual group, but the difference was not statistically significant. With respect to the primary outcome measure, the mean estimated difference between awake ambulatory blood pressure and office blood pressure after enrolment was -2.3 (95% confidence interval [CI] -0.3 to -4.3) / -3.3 (95% CI -2.2 to -4.4) (P=0.02 / P<0.001) for automated group and -6.5 (95% CI -4.3 to -8.6) / -4.3 (95% CI -2.9 to -5.8) (P<0.001 / P<0.001). The difference for systolic manual office blood pressure (-6.5) was significantly greater (P=0.006) than that for the systolic automated office blood pressure (-2.3). Automated office blood pressure showed a stronger (P<0.001) within group correlation (r=0.34 / r=0.56) with the awake ambulatory blood pressure compared with manual office blood pressure. The number of individual readings that showed digit preference, with readings being rounded off to the nearest zero, was significantly reduced (P<0.001) when automated office blood pressure readings were taken. It was concluded that the quality and accuracy of the automated blood pressure was better than the manual office blood pressure in relation to the awake ambulatory blood pressure. The automated office blood pressure measurement significantly reduced the white coat response compared with the manual office blood pressure measurement.

In the trial by Lamarre-Cliché et al. 2011,15 average blood pressures taken from all measurement methods were similar with the exception of daytime ambulatory blood pressure measurement (ABPM), which resulted in significantly greater systolic pressure compared with other methods. With the sphygmomanometer, blood pressure decreased by 1.2 ± 5.5 (P=0.029) / 0.4 ± 4.3 (P=0.40) mm Hg between the first and third measurements. With the automated


office blood pressure (AOBP) device, blood pressure decreased by 2.8 ± 7.5 (P=0.0004) / 1.1 ± 4.7 (P=0.40) mm Hg between the first and fifth measurements. According to cut-off values, target blood pressure of 140/90 mm Hg was reached in 67% of patients according to office sphygmomanometer and 70% according to the AOBP. Kappa coefficients measuring concordance showed fair agreement between two office methods (AOBP and sphygmomanometer; kappa coefficient 0.653; 95% CI 0.493 to 0.813). AOBP had the best diagnostic performance when using a sphygmomanometer as the gold standard but had variable results when choosing other standards. It was concluded that AOBP device can be used as a replacement for the sphygmomanometer in office settings for the measurement of blood pressure in hypertensive patients. Guidelines:

Three guidelines were identified that provided recommendations on the use of blood pressure measuring devices in hypertensive adults, in children, and in pregnant women. Recommendations stated that extra care should be given to the use of automated device in measuring blood pressure for patients having pulse irregularity, because the device may not measure blood pressure accurately. In pregnant women, mercury sphygmomanometer remains the gold standard for measurement of blood pressure. Automated or aneroid devices can be used to monitor blood pressure in women with preeclampsia. Because automated and aneroid devices are prone to error, it is recommended that a mercury sphygmomanometer should be in place for validation. In children, mercury or aneroid sphygmomanometer is recommended to be used for diagnostic evaluation of blood pressure elevation. However, automated devices may be acceptable for routine surveillance or in newborns and young infants, in whom auscultation may be difficult. Limitations

Clinical

Several limitations of the two systematic reviews11,12 included differences in patient populations among studies, differences in protocol design, the lack of information about sample selection, and small sample sizes in some included studies. Many included studies in the systematic reviews were of observational design, which suffered from potential biases in selection and measurement.

Due to the nature of the studies, it is difficult to conduct a blinded RCT in blood pressure measurement. All three RCTs used open-labeled design. However, one RCT13 reported an attempt to blind those measuring the outcomes. The use of one type of automated device (BpTRU in two trials14,15 and Omron in one trial13) in each trial makes it difficult to extrapolate the findings to other automated devices. The study population in each trial was selective and it is not known for certain if the findings could be applied to the general population.

Guidelines

Of the four identified guidelines, one was from Canada,19 which had recommendations for the type of device for blood pressure measurement in pregnant women. There was no specific guideline emphasizing the use of blood pressure monitoring devices, rather recommendations on the measurement of blood pressure were part of the management guidelines of specific individuals with hypertension. There were no tools or algorithms for the use of different type of devices in each situation.


CONCLUSIONS AND IMPLICATIONS FOR DECISION OR POLICY MAKING The collective current evidence suggests that automated office blood pressure monitoring devices, which are mercury-free, can be used in substitution of the mercury sphygmomanometer in clinical settings, although there are still some concerns about the accuracy of the devices with respect to specific populations such as those with hypertension, hypotension, arrhythmia, and children. Evidence from one RCT suggests that the maternal and fetal outcomes were comparable between automated and manual devices used to monitor throughout pregnancy. The mercury sphygmomanometer is still considered as the “gold standard” according to clinical practice guidelines for blood pressure measurement in patients with pulse irregularity, in children, and in pregnant women with hypertension. In those patients, it was recommended that blood pressure readings by automated devices should be verified by a mercury sphygmomanometer or an aneroid device. No evidence on the comparison of aneroid devices with mercury sphygmomanometers was identified. PREPARED BY: Canadian Agency for Drugs and Technologies in Health Tel: 1-866-898-8439 www.cadth.ca

http://www.cadth.ca/


REFERENCES 1. World Health Organization. Replacement of mercury thermometers and

sphygmomanometers in health care. Technical guidance [Internet]. Geneva: WHO; 2011. [cited 2012 Sep 14]. Available from: http://whqlibdoc.who.int/publications/2011/9789241548182_eng.pdf

2. Markandu ND, Whitcher F, Arnold A, Carney C. The mercury sphygmomanometer should be abandoned before it is proscribed. J Hum Hypertens. 2000 Jan;14(1):31-6.

3. Buchanan S, Orris P, Karliner J. Alternatives to the mercury sphygmomanometer. J Public Health Policy. 2011 Feb;32(1):107-20.

4. Scientific Committee on Emerging and Newly Identified Health Risks. Mercury sphygmomanometers in healthcare and the feasibility of alternatives [Internet]. Brussels: European Commission; 2009. [cited 2012 Sep 14]. Available from: http://ec.europa.eu/health/ph_risk/committees/04_scenihr/docs/scenihr_o_025.pdf

5. Medicines and Healthcare products Regulatory Agency. Minutes of the meeting of the Committee on Blood Pressure Monitoring in Clinical Practice, Tuesday, 4 November 2003. London (UK): The Agency; 2003.

6. Myers MG, Godwin M. Automated office blood pressure. Can J Cardiol. 2012 May;28(3):341-6.

7. Beevers G, Lip GY, O'Brien E. ABC of hypertension. Blood pressure measurement. Part I-sphygmomanometry: factors common to all techniques. BMJ [Internet]. 2001 Apr 21 [cited 2012 Sep 14];322(7292):981-5. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1120141

8. Shea BJ, Grimshaw JM, Wells GA, Boers M, Andersson N, Hamel C, et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Med Res Methodol [Internet]. 2007 Feb 15 [cited 2012 Aug 8];7:10. Available from: http://www.biomedcentral.com/1471-2288/7/10

9. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health [Internet]. 1998 Jun [cited 2012 Aug 8];52(6):377-84. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1756728/pdf/v052p00377.pdf

10. The AGREE Collaboration. Appraisal of guidelines for research and evaluation (AGREE) instrument [Internet]. London: The AGREE Research Trust; 2001 Sep. [cited 2012 Aug 8]. Available from: http://www.agreetrust.org/?o=1085

11. Stergiou GS, Kollias A, Destounis A, Tzamouranis D. Automated blood pressure measurement in atrial fibrillation: a systematic review and meta-analysis. J Hypertens. 2012 Aug 21. Epub ahead of print.

http://whqlibdoc.who.int/publications/2011/9789241548182_eng.pdf

http://ec.europa.eu/health/ph_risk/committees/04_scenihr/docs/scenihr_o_025.pdf

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1120141

http://www.biomedcentral.com/1471-2288/7/10

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1756728/pdf/v052p00377.pdf

http://www.agreetrust.org/?o=1085


12. Skirton H, Chamberlain W, Lawson C, Ryan H, Young E. A systematic review of variability and reliability of manual and automated blood pressure readings. J Clin Nurs. 2011 Mar;20(5-6):602-14.

13. Brown MA, Roberts LM, Mackenzie C, Mangos G, Davis GK. A prospective randomized study of automated versus mercury blood pressure recordings in hypertensive pregnancy (PRAM Study). Hypertens Pregnancy. 2012;31(1):107-19.

14. Myers MG, Godwin M, Dawes M, Kiss A, Tobe SW, Grant FC, et al. Conventional versus automated measurement of blood pressure in primary care patients with systolic hypertension: randomised parallel design controlled trial. BMJ [Internet]. 2011 [cited 2012 Sep 4];342:d286. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3034423

15. Lamarre-Cliche M, Cheong NN, Larochelle P. Comparative assessment of four blood pressure measurement methods in hypertensives. Can J Cardiol. 2011 Jul;27(4):455-60.

16. National Institute for Health and Clinical Excellence. Hypertension: clinical management of primary hypertension in adults [Internet]. London: NICE; 2011 Aug. 36 p. [cited 2012 Sep 4]. (Clinical Guidelines CG127). Available from: http://guidance.nice.org.uk/CG127/NICEGuidance/pdf/English

17. Cincinnati Children's Hospital Medical Center. Blood pressure measurement in children [Internet]. Cincinnati (OH): The Centre; 2009 Jan 9. 9 p. [cited 2012 Sep 4]. (Best evidence statement (BESt)). Available from: http://www.cincinnatichildrens.org/assets/0/78/1067/2709/2777/2793/9198/20912557-f3e9-49b6-9f30-0f66bb9c8efb.pdf

18. Lowe SA, Brown MA, Dekker G, Gatt S, McLintock C, McMahon L, et al. Guidelines for the management of hypertensive disorders of pregnancy 2008 [Internet]. Sydney: Society of Obstetric Medicine of Australia and New Zealand (SOMANZ); 2008. 31 p. [cited 2012 Sep 4]. Available from: http://www.somanz.org/pdfs/somanz_guidelines_2008.pdf

19. Magee LA, Helewa M, Moutquin JM, von DP, Hypertension Guideline Committee, Strategic Training Initiative in Research in the Reproductive Health Sciences (STIRRHS) Scholars. Diagnosis and classification. Diagnosis, evaluation, and management of the hypertensive disorders of pregnancy. J Obstet Gynaecol Can. 2008 Mar;30(3 Suppl):S9-S15.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3034423

http://guidance.nice.org.uk/CG127/NICEGuidance/pdf/English

http://www.cincinnatichildrens.org/assets/0/78/1067/2709/2777/2793/9198/20912557-f3e9-49b6-9f30-0f66bb9c8efb.pdf

http://www.cincinnatichildrens.org/assets/0/78/1067/2709/2777/2793/9198/20912557-f3e9-49b6-9f30-0f66bb9c8efb.pdf

http://www.somanz.org/pdfs/somanz_guidelines_2008.pdf


APPENDIX 1: Selection of Included Studies

47 citations excluded

8 potentially relevant articles retrieved for scrutiny (full text, if

available)

10 potentially relevant reports retrieved from other sources (grey

literature, hand search)

18 potentially relevant reports

9 reports excluded:

irrelevant comparator (3)

published in language other than English (1)

observational study with no safety data (1)

other (review articles, editorials)(4)

9 reports included in review

2 SR/MA

3 RCTs

4 guidelines

55 citations identified from electronic literature search and screened


APPENDIX 2: Characteristics of Included Clinical Studies

First Author, Publication Year, Country

Study Design

Patient characteristics, sample Size (n)

Intervention Comparators Clinical Outcomes

Systematic reviews and meta-analyses

Stergiou et al. 2012

11

Greece

Systematic review and meta-analysis

Individuals (57% men, mean age 73 years) with sustained atrial fibrillation from 8 studies, N=566

Oscillometric (automated) devices

Mercury sphygmomanometer

Mean BP

Skirton et al. 2011

12

UK

Systematic review

Participants (inpatients and outpatients) with or without elevated blood pressure from 16 included studies (15 observational studies and 1 RCT), N=4,429

Oscillometric (automated) devices

Ausculatory (manual) devices

Mean BP

Randomized controlled trials

Brown et al. 2012

13

Australia Source of funding: not reported

RCT open study with blinded end points Follow-up: remainder of pregnancy

Women (mean age 31 years) with hypertension in pregnancy (outpatients and inpatients), N=220

Automated (Omron HEM-705 CP) devices, n=110

Mercury sphygmomanometer, n=110

Number of women having episode of severe hypertension

Maternal outcomes

Fetal outcomes

Mean BP

Myers et al. 2011

14

Canada Public funding

Multi-site cluster RCT

Patients (>45 years) with systolic hypertension and no serious comorbidities, N=555

Automated office blood pressure device (BpTRU), n=303

Manual blood pressure device, n=252

Mean BP

Lamarre-Cliché et al. 2011

15

Canada Source of funding: not reported

RCT, crossover Length of study: 8-14 days

Hypertensive patients (mean age 58 years)

Automated office blood pressure device (BpTRU), n=101

Office mercury sphygmomanometer 24-hour ambulatory Daytime ambulatory Self-measurement

Mean BP

Target BP

BP=blood pressure; RCT=randomized controlled trial


APPENDIX 3: Grading of Recommendations and Levels of Evidence

Guideline Society or Institute

Recommendation Level of Evidence

National Institute for Health and Clinical Excellence (NICE), UK, 2011

16

The Center established a Guideline Development Group, which reviewed the evidence and updated the recommendations. An independent Guideline Review Panel oversaw the updating of the guideline. For more information, see WWW.nice.org.uk/HowWeWork

Cincinnati Children’s Hospital Medical Center. Best evidence statement, 2009

17

“Strongly recommended”: There is consensus the benefits clearly outweigh risks and burdens (or visa-versa for negative recommendations) “Recommended”: There is consensus that benefits are closely balanced with risks and burdens No recommendation made: There is a lack of consensus to direct development of a recommendation

1a or 1b: Systematic review, meta-analysis, or meta-analysis of multiple studies 2a or 2b: Best study design for domain 3a or 3b: Fair study design for domain 4a: or 4b: Weak study design for domain 5: Other: General review, expert opinion, case report, consensus report, or guideline a = good quality study; b = lesser quality study

Society of Obstetric Medicine of Australia and New Zealand (SOMANZ) 2008

18

Not indicated Not indicated

Society of Obstetricians and Gynaecologists of Canada Magee et al. 2008

19

A. There is good evidence to recommend the clinical preventive action

B. There is fair evidence to recommend the clinical preventive action

C. the existing evidence is conflicting and does not allow to make a recommendation for or against use of the clinical preventive action; however, other factors may influence decision-making

D. There is fair evidence to recommend against the clinical prevention action

E. There is good evidence to recommend against the clinical prevention action

I. There is insufficient evidence (in quantity or quality) to make a recommendation; however, other factors may influence decision-making

I: Evidence obtained from at least one properly randomized controlled trial II-1: Evidence from well-designed controlled trials without randomization II-2: Evidence from well-designed cohort (prospective or retrospective) or case-control studies, preferably from more than one centre or research group II-3: Evidence obtained from comparisons between times or places with or without the intervention. Dramatic results in uncontrolled experiments (such as results of treatment with penicillin in 1940s) could also be included in this category III: Opinions of respected authorities, based on clinical experience, descriptive studies, or reports of expert committees


APPENDIX 4: Summary of Study Strengths and Limitations

First Author, Publication Year

Strengths Limitations

Systematic Reviews and Meta-analyses


Scored “Yes” on 9 out of 11 items of the AMSTAR checklist (1-6, 9-11)

Scored “No” on 2 out of 11 items of the AMSTAR checklist (7, 8). List of excluded studies not provided

Skirton et al. 201112

Scored “Yes” on 8 out of 11 items of the AMSTAR checklist (1, 3-8, 11)

Scored “No” on one item (10), “Can’t tell” on one item (2), and “Not applicable” on one item (9)

AMSTAR check list 1. Was an “a priori” design provided? 2. Was there duplicate study selection and data extraction? 3. Was a comprehensive literature search performed? 4. Was the status of publication (i.e., grey literature) used as an inclusion criteria? 5. Was a list of studies (included and excluded) provided? 6. Were the characteristics of the included studies provided? 7. Was the scientific quality of the included studies assessed and documented? 8. Was the scientific quality of the included studies used appropriately in formulating conclusions? 9. Were the methods used to combine the findings of studies appropriate? 10. Was the likelihood of publication bias assessed? 11. Was the conflict of interest stated?


Brown et al. 201213

Hypothesis/objective, main outcomes to be measured, characteristics of included patients, interventions of interest and the main findings were explicit

The study provided mean and standard deviation or percentage for outcomes

All important adverse events were reported

The actual p-values were reported

The recruited subjects represented the entire population within a specified period of time

Staff, places and facilities where patients were treated, were representative of the treatment the majority of patients received

Attempt was made to blind those measuring the main outcomes of the intervention.

Follow-up was the same for all study patients

Statistical tests used to assess the main outcomes were appropriate

The outcome measures were clearly described

Patients in both groups were recruited from the same population, over the same period of time, and were randomized to the interventions groups

The trial was powered to detect the difference in primary outcome

A list of principal confounders was not provided

The characteristics of patients lost to follow-up were not described

Attempt was made to blind study subjects (not applicable in this study)

Unable to determine compliance with the interventions

Random assignment was not concealed from staff

Unable to determine if there was adequate adjustment for confounding in the analyses, or losses of patients to follow-up were taken into account

Myers et al. 201114

Hypothesis/objective, main outcomes to be measured, characteristics of





included patients, interventions of interest and the main findings were explicit

The study provided mean differences and 95% confidence intervals for outcomes









All important adverse events were not reported


No attempt was made to blind study subjects (not applicable in this study)

No attempt was made to blind those measuring the main outcomes of the intervention.





15

Hypothesis/objective, main outcomes to be measured, characteristics of included patients, interventions of interest and the main findings were explicit

The study provided mean differences and 95% confidence intervals for outcomes










All important adverse events were not reported


No attempt was made to blind study subjects (not applicable in this study)

No attempt was made to blind those measuring the main outcomes of the intervention.




Guidelines

National Institute for Health and Clinical

Objectives and target patients population were explicit.

The guidelines are not supported with application tools




Excellence (NICE), UK, 2011

16

A procedure for updating the guideline was provided

The recommendations were specific and unambiguous

Key recommendations were easily identified

The editorial independence of the guidelines was specified.

The cost implication of applying these guidelines was not evaluated and reported.

The recommendations were based on consensus of appraisal committee members, who considered evidence from a number of sources including systematic reviews, assessment reports and manufacturers documents.

Patient’s preferences and views were not explicitly taken into consideration.

Methods used for searching the evidence were not reported.


17





A procedure for updating the guideline was not provided


Formulation of the recommendations was based on expert consensus.



The editorial independence of the guidelines was not specified.


18






Formulation of the recommendations was based on literature reviews and expert consensus.




Key recommendations were not easily identified


19




Formulation of the recommendations was based on clinical evidence







Financial disclosures and conflict of interest were not stated.


APPENDIX 5: Main Study Findings and Authors’ Conclusions – Clinical


Main Findings

Systematic Review


11

Correlation coefficient between automated and manual SBP/DBP (4 studies):

For SBP: r=0.89, 95% CI 0.84, 0.94 For DBP: r=0.76, 95% CI 0.70, 0.81, p<0.001 for difference Average blood pressure differences between automated and manual method (6 studies):

SBP difference = 0.5 mm Hg, 95% CI -0.9, 1.9 DBP difference = 2.5 mm Hg, 95% CI -0.6, 5.7 Absolute blood pressure differences within 5 mm Hg

Four of eight studies reporting data regarding the within 5 mm Hg ESH-IP criterion have passed for SBP (at least 65% of comparisons within 5 mm Hg) and only two for DBP Impact of heart rate

Heart rate had no impact on the blood pressure differences between automated and manual measurements

Authors’ Conclusions: “this analysis suggests that the currently available automated blood pressure monitors that

have been validated in individuals with sinus rhythm appear to be accurate in measuring systolic blood pressure but not diastolic blood pressure in individuals with sustained atrial fibrillation” p.8

Skirton et al. 2011

12

UK

Study Intervention vs. control Population Main findings

Cross-sectional survey

Automated (BpTRU) vs. manual (mercury)

50 patients in a hypertension or internal medicine clinic

Results suggested that the automated device should replace manual instruments in clinical practice and research

Cross-sectional

Automated (Dinamap) vs. manual (mercury)

47 hospital patients

Dinamap demonstrably less accurate than mercury sphygmomanometer, especially systolic pressure (p<0.05)

Prospective Cross-sectional study


171 patients in emergency department

The automated device consistently overestimated systolic BP and underestimated diastolic BP

Cross-sectional

Automated BP wrist monitor (Omron Rx) vs. manual (mercury)

85 participants Wrist device readings are more closely correlated with mercury sphygmomanometer readings using appropriately sized cuffs.

Retrospective cross-sectional study


388 trauma patients

Manual BP readings were more reflective of patient condition. Automated BP measurements were higher than manual readings, especially in hypotensive patients

Prospective cohort study


85 participants (outpatients or staff of a large teaching hospital)

The automated device is acceptable for clinical use, according to BHS criteria

Cross-sectional quasi-

Automated (Omron R3 and Visomat OZ2) vs. manual (mercury)

85 patients attending a medical

All three types of devices showed good reproducibility. Automated devices performed



Main Findings

experimental outpatient clinic less well than mercury sphygmomanometer in hypertensive patients. Measuring BP on the wrist using automated devices was not found to be satisfactory

Cross-sectional

aneroid and oscillometric vs. manual (mercury)

500 participants from Hypertension Center, Preventive Medicine Research Center and local fairs

Oscillometric device slightly overestimated systolic BP and underestimated diastolic BP compared with mercury sphygmomanometer. Oscillometric device was comparable in performance to the aneroid device, and both could be used to replace the mercury manometer

Post-hoc analysis of observational

Automated (OMRON 705 CP) vs. manual (mercury)

313 untreated hypertensive patients

There are differences between readings obtained by both devices.

Prospective cohort

Automated (Dinamap Pro-care 400) vs. manual (mercury)

300 patients at one tertiary care teaching hospital

Authors state ‘reasonable good correlation’ between the two devices for systolic and diastolic pressure. Some observer bias noted when mercury manometer was used notably repeated measure and terminal digit bias. Inter-operator measurements were more consistent for the automated device than intra-operator measurements using the mercury device

Prospective cohort

Automated (TM-2564G) vs. sphygmomanometer

85 inpatients from one hospital

Diastolic readings taken with the device did not differ more than 5 mmHg from those taken with the sphygmomanometer. 99.2% of the systolic readings taken with the device were within 5 mmHg of sphygmomanometer

Prospective cross-sectional

aneroid Welch Allyn Tycos 767 vs. manual (mercury)

997 participants at high risk of diabetes

There were no clinically significant differences between readings according to the type of sphygmomanometer

Prospective cross-sectional

Automated (BpTRU) vs. manual (mercury)

238 Ontario residents

Mean BP taken with automated device was lower than that taken with mercury sphygmomanometer. Authors conclude the BpTRU able to be used because it reduced white coat effect

Prospective cohort

Automated (Dinamap Pro-care) vs. manual (mercury)

33 outpatients 79% of measurements with device were within 5 mm Hg of the systolic reading taken with



Main Findings

mercury, while 77% of diastolic readings were within 5 mmHg. The device is acceptable for clinical use, according to the Advancement of Medical Instrumentation

Prospective Automated (OMRON) vs. sphygmomanometer

907 hypertensive patients

99% readings by traditional nurses ended in zero. Using a level of plus and minus 5 mm Hg, traditional nurse over estimated systolic BP in 45% cases and underestimated systolic BP in 26.5% cases. Automated device overestimated systolic BP in 51.3% cases and underestimated systolic BP in 12% cases

Randomized controlled trial

Automated (IVAC 4200) vs. manual (mercury)

145 stable inpatients

IVAC measurements are less than satisfactory in both systolic and diastolic BP. Arrhythmia or low K5 reduces IVAC accuracy. IVAC should not be used for patients with hypertension

Authors’ Conclusions: “There are situation where the substitution of oscillometric for auscultatory devices could

have particularly serious repercussions for the patient, such as when the patient is either hypertensive or hypotensive. However, further research is required on the use of aneroid sphygmomanometers as a replacement for mercury devices” p.602


Brown et al. 2012

13

Blood pressures obtained by six sequential recordings (alternating between mercury and automated)

Mercury (n=110)

Automated (n=110)

Average BP using mercury manometer (mm Hg) 139 (9) / 90 (7) 137 (9) / 91 (7)

Average BP using automated device (m Hg) 136 (12) / 86 (8) 137 (11) / 87 (8)

Difference in average BP (mm Hg) 2 (8) / 3 (6) 1 (7) / 5 (5)

Average SBP difference between methods (%)

≤5 mm Hg 55 56

≤10 mm Hg 80 90

≤15 mm Hg 96 96

Average DBP difference between methods (%)

≤5 mm Hg 60 63

≤10 mm Hg 88 92

≤15 mm Hg 97 99

Maternal and fetal outcomes were similar between groups.

Mercury Automated p-value

Gestation at last antenatal visit (weeks) 37 (2) 37 (2) 0.1

Gestation at birth (weeks) 38 (3) 37 (3) 0.08

Diagnosis at birth (%)

Essential hypertension 15 14 0.96



Main Findings

Gestational hypertension 39 38

Preeclampsia 46 48

Onset of labor (%)

Spontaneous 63 51 0.052

Induction of labor 11 23

Nil 26 26

Mode of birth (%)

Normal vaginal birth 39 46 0.552

Instrumental 15 13

Lower segment caesarean section 46 41

Gestational diabetes (%) 12 7 0.3

Maternal antenatal transfer (n) 3 4 0.7

Antihypertensive medications at birth (%) 90 85 0.2

Women with any episodes of severe hypertension (%) 39 44 0.5

Average number of episodes of severe hypertension per woman affected

3 (2) 4 (5) 0.3

Birth weight (grams) 3014 (780) 2890 (712) 0.2

Small for gestational age (<10th centile) (%) 12 17 0.3

Premature (<37 weeks) (n) 30 38 0.1

Transfer to NICU (%) 4 4 1

Perinatal mortality (n) 1 1 1

Percentage of women with any episode of severe hypertension was similar between groups (39% for mercury vs. 44% for automated, p=0.5) Average number of episodes of severe hypertension per woman affected was similar between groups (3±2 for mercury vs. 4±5 for automated, p=0.3)

Authors’ Conclusions: “This study showed that there are similar maternal and fetal outcomes when BP is measured

by either routine mercury sphygmomanometer or the Omron HEM-705CP automated BP device after the initial detection of hypertension by mercury sphygmomanometer in pregnant women” p.117



Main Findings

Myers et al. 2011

14

Canada

Mean (SD) blood pressure (BP) taken in physicians’ office before and after enrolment into study and baseline mean awake ambulatory BP recorded between two office visits

Measurement Automated (n=299) Manual (n=249)

Last routine office BP 149.5 (10.8)/81.4 (8.3) 149.9 (10.7)/81.8 (8.5)

Office BP after enrolment 135.6 (17.3)/77.7 (10.9) 141.4 (14.6)/80.2 (9.5)

Difference from last routine -13.9 (-11.8, -16.1)***/ -3.7 (-2.5, -4.8)***

-8.5 (-6.5, -10.4)***/ -1.6 (-0.4, -2.8)**

Awake ambulatory 133.2 (12.4)/74.4 (9.8) 135.0 (13.1)/75.9 (10.0)

Difference from last routine -16.3 (-14.5, -18.1)***/ -7.0 (-5.8, -8.1)***

-14.9 (-12.9, -17.0)***/ -5.9 (-4.6, -7.2)

Difference from post-enrolment -2.3 (-0.31 to -4.3)*/ -3.3 (-2.2 to -4.4)***

-6.5 (-4.3, -8.6)***/ -4.3 (-2.9, -5.8)***

*p=0.02 **p=0.01 ***p<0.001

Coefficients of correlation between office systolic/diastolic BP readings and mean awake ambulatory BP

Measurement Automated Manual

Last routine vs. awake ambulatory r=0.10 / r=0.40 r=0.04 / r=0.42

After enrolment vs. awake ambulatory

r=0.34 / r=0.56 r=0.22 / r=0.30

Estimated mean difference (95% CI) in r from before to after enrolment

0.24 (0.12, 0.36)*/ 0.16 (0.07, 0,25)*

0.18 (0.02, 0.32)**/ -0.12 (-0.01, 0.24)

*p<0.001 **p=0.03

Authors’ Conclusions: “In compliant, otherwise healthy, primary care patients with systolic hypertension,

introduction of automated office blood pressure measurement into routine primary care significantly reduced the white coat response compared with the ongoing use of manual office blood pressure measurement. The quality and accuracy of automated office blood pressure in relation to the awake ambulatory blood pressure was also significantly better when compared with manual office blood pressure.” p.1


15

Canada

Average Blood Pressure

Mean ± SD

Sphygmomanometer 129.9 ± 13.7 / 80.9 ± 9.3

AOBP (BpTRU) 128.4 ± 13.9 / 80.0 ± 9.4

24-hour ABPM 131.4 ± 11.7 / 78.7 ± 9.7

Daytime ABPM 135.5 ± 11.4 / 82.0 ± 11.9

Self-Measurement 131.0 ± 14.3 / 82.5 ± 8.2

Differences between methods

Sphygmomanometer vs. BpTRU 1.45 ± 7.59 / 0.84 ± 5.05

Sphygmomanometer vs. 24-hour ABPM -1.29 ± 11.85 / 2.30 ± 7.59**

Sphygmomanometer vs. daytime ABPM -5.43 ± 12.12*** / 0.98 ± 7.55

Sphygmomanometer vs. self-measurement -1.11 ± 14.05 / -1.56 ± 8.50

BpTRU vs. 24-hour ABPM -2.76 ± 12.56* / 1.39 ± 7.09

BpTRU vs. daytime ABPM -6.90 ± 12.86*** / -1.90 ± 6.90**

BpTRU vs. self-measurement -2.56 ± 15.25 / -2.41 ± 8.02**

24-hour ABPM vs. self-measurement 0.34 ± 10.09 / -3.66 ± 7.88***

Daytime ABPM vs. self-measurement 4.48 ± 11.24*** / -0.38 ± 8.30

ABPM=ambulatory blood pressure measurement; AOBP=automated office blood pressure *p<0.05 **p<0.01 ***p<0.001 Target BP 140/90 mm Hg: 67% with office sphygmomanometer; 70& with AOBP



Main Findings

Target BP 135/85 mm Hg: 52% with self-measurement; 43% with daytime ABPM readings Target 130/80 m Hg: 37% with 24-hour ABPM

Authors’ Conclusions: “This study shows that sphygmomanometer, AOBP, ABPM, and self-measurement provide

similar average blood pressure estimates but generate many discordant results on an individual basis. The AOBP is a new office measurement device that offers specific advantages but has diagnostic properties that are highly dependent on chosen diagnostic thresholds” p. 459


APPENDIX 6: Guidelines and Recommendations on Use of Device for Blood Pressure Measurement

Guideline Society, Country, Author, Year, Indication

Recommendations

National Institute for Health and Clinical Excellence (NICE), UK, 2011

16

Clinical Management of primary hypertension in adults

“Because automated devices may not measure blood pressure accurately if there is pulse irregularity (for example, due to atrial fibrillation), palpate the radial or brachial pulse before measuring blood pressure. If pulse irregularity is present, measure blood pressure manually using direct auscultation over the brachial artery” p.10

“Healthcare providers must ensure that devices for measuring blood pressure are properly validated, maintained and regularly recalibrated according to manufacturers’ instructions” p.10

“If using an automated blood pressure monitoring device, ensure that the device is validated and appropriate cuff size for the person’s arm is used” p.11


17

Blood pressure Measurement in Children

“It is recommended that auscultation with mercury or aneroid sphygmomanometer be used for diagnostic evaluation of blood pressure evaluation [5]” p.1

“Note 1: Blood pressure measurement with an oscillometric device may be

acceptable for routine surveillance [5]. An elevated blood pressure reading obtained with an oscillometric device should be repeated using auscultatory method [5]” p.1

“Note 2: Use of automated devices may be necessary for blood pressure

measurement in newborns and young infants, in whom auscultation may be difficult [5]” p.1


18

Management of Hypertensive Disorders of Pregnancy

“Mercury sphygmomanometers remain the gold standard for measurement of blood pressure in pregnancy however occupational health concerns are limiting their availability” p.4

“Each unit should maintain a mercury sphygmomanometer for validation of automated and aneroid devices” p.4

“All devices should be calibrated on a regular basis (ideally monthly), as recommended by the British Hypertension Society” p.4


19

Diagnosis, evaluation and management of the hypertensive disorders of pregnancy

“BP can be measured using a mercury sphygmomanometer, calibrated aneroid device, or an automated BP device that have been validated for use in preeclampsia. (II-2A)

Automated BP machines may underestimate BP in women with preeclampsia, and comparison of readings using mercury sphygmomanometer or an aneroid device is recommended. (II-2A)

Ambulatory BP monitoring (by 24-hour or home measurement) may be useful to detect isolated office (white coat) hypertension. (II-2B)” p.S9

mercury versus non-mercury sphygmomanometers: a review of ... · mercury leak, which is a major...

Documents