Evaluation of performance of selected devices for measuring blood pressure

Standardization of blood pressure measurement in an international comparative study

In the context of a collaborative study on the epidemiology of hypertension in populations of West African origin procedures for standardization of the measurement of blood pressure were evaluated. Comparisons of mean levels of blood pressure, which in large part determine prevalence rates, are highly sensitive to differences in technique. While rotating a single field team may be the ideal approach to multisite studies, it is not practical in international collaborative research. Appropriate techniques to standardize multiple teams over a long period of time have not been well developed, however. In the present study 8981 individuals were examined in eight sites in six countries with the standard mercury sphygmomanometer. An evaluation of the effectiveness of central training, site visits, monitoring of digit preference, and the use of an electronic device for internal standardization is described. In all but one of the sites reliability was high and comparable to the observers at the Coordinating Center. Digit preference for the entire set of measurements was limited (frequency of terminal zero = 23.5% for systolic and 28.9% for diastolic readings) and could be shown to have virtually no effect on prevalence rates or correlation estimates. Mean differences among observers within a given site and between sites were small (+/- 0-5 mmHg). While logistically complex, these methods can provide the basis for standardization in international comparative blood pressure surveys.

State of the market. A review of ambulatory blood pressure monitoring devices

The introduction of 24-hour ambulatory blood pressure measurement into clinical practice created a large market for ambulatory blood pressure measurement devices. Forty-three such devices from 31 manufacturers or suppliers are now available to satisfy a market demand that is likely to increase. The aim of this article is to identify the devices available and then to examine critically any validation studies assessing accuracy and performance. Of the 43 devices available 18 have been validated according to the protocols of the Association for the Advancement of Medical Instrumentation (AAMI) or the British Hypertension Society (BHS) in 25 reported studies. In 9 of these studies the protocol was not adhered to, and the results, which are therefore questionable, are noted but not considered further. Fourteen devices were evaluated according to the accuracy criteria of both protocols, and of these 9 fulfilled the requirements. From this review of 43 devices on the market it may be concluded that, at the time of writing, there is published evidence for only 9 devices meeting the generally accepted AAMI and BHS criteria for accuracy and performance; these are the A&D TM-2420 models 6 and 7 and TM-2421, CH-Druck, Nissei ABPM DS-240, Profilomat, QuietTrak, and SpaceLabs SL-90202 and SL-90207.

Potential problems with the random-zero sphygmomanometer

The random-zero sphygmomanometer is frequently used in epidemiologic and clinical research to increase the reliability and validity of blood pressure readings. However, recent reports have suggested that there is a correlation between the zero values of a random-zero sphygmomanometer and the zero-corrected blood pressure readings obtained. The design of the random-zero sphygmomanometer would suggest that the zero values and zero-corrected blood pressures should be uncorrelated. Hence, residual correlation might be of importance in determining the utility of this device. We have explored this relation in the Middlesex County Blood Pressure Survey, which collected data on 2725 randomly selected adults. Each person had three blood pressure readings taken with a random-zero sphygmomanometer operated by trained interviewers. There was a very weak but statistically significant correlation between the zero values and zero-corrected systolic blood pressure (r = .034, regression slope = -.10), but there was no statistically significant relation with zero-corrected diastolic blood pressure (r = .0003, slope = .0006). Both the correlations and regression slopes were higher for subjects over age 65 years. These data fail to confirm the observed correlations found by Kronmal et al. This discrepancy might be explained by differences in measurement technique, which could introduce a blood pressure-dependent skewing of the range of zero values. If confirmed, this effect would have no effect on the validity of the final blood pressure readings and hence would not need to be considered in decisions about the use of the random-zero sphygmomanometer.

Evaluation of blood pressure measuring devices

Validation of blood pressure measuring devices is a relatively new field of research. There are two national protocols for validating blood pressure measuring devices: the protocol of the American Association for the Advancement of Medical Instrumentation (AAMI) and the protocol of the British Hypertension Society (BHS), each of which has recently been revised. 19 blood pressure measuring devices have been validated according to one or both of these protocols. These protocols have been beneficial in drawing attention to the potential inaccuracy of blood pressure measuring systems, they permit comparison between devices and they have brought manufacturers of blood pressure measuring devices into closer contact with the profession. There are some inherent weaknesses in both protocols which include the fallibility of the 'gold standard', the lack of provision for validation in special circumstances and in special groups, such as the elderly and pregnant women, and failure to allow for deteriorating accuracy with higher pressure levels. The revised BHS protocol attempts to redress these deficiencies.

Bias and variability in blood pressure measurement with ambulatory recorders

This study sought to determine whether patient characteristics such as age, sex, blood pressure, and pulse pressure differently affect the accuracy of an oscillometric (SpaceLabs 90207) and a microphonic (TM2420 version 7) blood pressure monitor. Blood pressure recorded by two oscillometric and two microphonic ambulatory monitors was compared with simultaneous readings by two pairs of trained, blinded observers using random-zero sphygmomanometry. One hundred and eighteen subjects (53 men and 65 women, aged 17 to 94 years; systolic pressure, 89 to 211 mm Hg; diastolic, 44 to 116 mm Hg) were studied. There were no significant differences within each observer pair or between the two observer pairs as well as no correlation between interobserver differences and patient characteristics. The differences between the monitor and trained observers' readings were 2.8 +/- 9.9 mm Hg systolic and 3.9 +/- 6.8 mm Hg diastolic for the SpaceLabs and 5.0 +/- 5.2 mm Hg systolic and 3.4 +/- 6.1 mm Hg diastolic for the TM2420. Patient characteristics that predicted measurement error were defined by multiple regression. For oscillometry, systolic measurement error was highly correlated with systolic pressure, pulse pressure, and subject age. The diastolic error was significantly correlated with pulse pressure, diastolic pressure, and subject sex. For the oscillometric monitor, patient characteristics accounted for 36.6% of the variation of the systolic error and 34.7% of the variation of the diastolic error. For the microphonic monitor, only age correlated with diastolic error, and no significant correlations were seen with systolic error. Patient characteristics accounted for only 1.2% of the systolic and 8.9% of the diastolic error.(ABSTRACT TRUNCATED AT 250 WORDS)

Evaluation of the Colin STBP-680 at rest and during exercise: an automated blood pressure monitor using R-wave gating

The application of automated blood pressure measurement during exercise has been limited by inaccuracies introduced by the effects of accompanying motion and noise. We evaluated a newly developed automated blood pressure monitor for measuring exercise blood pressure (Colin STBP-680; Colin, San Antonio, Texas, USA). The STBP-680 uses acoustic transduction with the assistance of the electrocardiogram R-wave to trigger the sampling period for blood pressure measurement. The automated monitor readings were compared with simultaneous technician mercury sphygmomanometric readings in the same arm. Blood pressure was measured in 18 men at rest and during exercise at 40% VO2 peak, (low intensity), 70% VO2 peak (moderate intensity) and VO2 peak (high intensity) on the cycle ergometer. Mean(s.d.) systolic blood pressure difference between the automated monitor and mercury manometer readings at rest and during exercise at low, moderate and high work intensities were 3(0) mmHg, 3(2) mmHg, 1(1) mmHg, and 0(11) mmHg respectively (analysis of variance; P > 0.05). Resting diastolic blood pressure obtained with the STBP-680 was similar to the mercury manometer readings (78(10) versus 81(7) mmHg (P > 0.05). Exercise diastolic pressure at the low level of work intensity was almost identical between the automated monitor and mercury manometer readings (64(8) versus 65(10) mmHg (not significant)). Diastolic blood pressure readings between the STBP-680 and mercury manometer showed a greater difference at the moderate and high workloads (11 mmHg and 9 mmHg, respectively), but this difference was not significant (P > 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

Measurement error in the Hawksley random zero sphygmomanometer: what damage has been done and what can we learn?

The Hawksley random zero sphygmomanometer is used in all aspects of blood pressure research, from clinical trials to evaluation of new blood pressure recorders. It is designed to reduce observer bias in blood pressure measurement. The problem is that it also underestimates blood pressure. Furthermore, this was first reported more than two decades ago. In this paper Rónán Conroy and colleagues explore the consequences of using an inaccurate instrument for important research and why prestigious organisations like the World Health Organisation continue to use it.

Properties of the random zero sphygmomanometer

The random zero sphygmomanometer is widely used in studies involving blood pressure measurement because it is believed to eliminate digit preference and reduce measurement error. We performed blood pressure measurements sequentially using random zero and standard sphygmomanometers in random order in 1,356 participants in the Cardiovascular Health Study. Despite adherence to the manufacturer's instructions, we observed a substantially nonuniform distribution of zero levels generated by the random zero sphygmomanometer and a disturbing correlation between the zero level and blood pressures taken with the standard sphygmomanometer. With the random zero device, the pooled estimated slopes for the regression of standard systolic and diastolic pressures on the zero level were -0.71 and -0.17, respectively (both p < 0.0001). The only plausible explanation for this relation between the random zero device and the standard device is that by some unknown mechanism the subject's blood pressure is influencing the zero level. Systolic and diastolic blood pressures measured with the random zero device were, respectively, 1.65 and 1.84 mm Hg lower (both p < 0.0001) than standard blood pressures. Digit preference was detectable in the uncorrected blood pressure and zero level measured with the random zero device but was eliminated after calculation of the corrected blood pressure. For most epidemiological studies, the random zero sphygmomanometer offers no significant advantage over the standard sphygmomanometer. It may still be useful in those epidemiological studies and clinical trials where blinding is important.

Accuracy of ambulatory blood pressure determination: a comparative study

This study was designed to discriminate, according to their accuracy, between three ambulatory pressurometers (Diasys 200R, Novacor; P IV, Del Mar Avionics; SpaceLab 90202, SpaceLab). The evaluation was performed against invasive arterial reference measurements. Accuracy was assessed by calculating the error on pressure (EOP) as the difference between invasive and non-invasive measurement of arterial blood pressure. For the systolic values, accuracy (mean of EOP differences) and uncertainty (SD of these differences) were -0.9 +/- 9.7, -4.3 +/- 10.1 and -16.7 +/- 10.1 mmHg for, respectively, Diasys, PIV and SpaceLab. For diastolic values, they were, respectively, 5.9 +/- 6.7, 6.8 +/- 8.5 and 9.1 +/- 6.6 mmHg. EOP was then separated in two different types of errors: (i) the error of dispersion appreciated by the index of homogeneity calculated by a Lehmann analysis and leading to a statistical classification (ii) the error due to the drift of EOP with the reference value, this last error being easier to correct. Two different behaviours were observed for the EOP: (i) the drift of EOP of systolic values was significantly larger for the oscillometric (SpaceLab) than for the auscultatory (Diasys and P IV) method, with no difference between Diasys and P IV (ii) the homogeneity index was not statistically different among these three devices. These data suggest that, in case the correction of the drift of EOP is carried out, there is no statistical significant difference in accuracy between these three pressurometers. However, in our experimental conditions, the two ambulatory pressurometers recording the Korotkoff sounds have a better accuracy than the one using the oscillometric approach.

Blood pressure changes associated with tilting in normotensive subjects: differences in response pattern as measured by oscillometry and auscultation

Three non-invasive instruments were used to measure blood pressure in the supine position and on tilting--a conventional and a random-zero sphygmomanometer, and an oscillometric device (Accutorr 1A). Twenty normotensives volunteered for the study. There was no statistically significant difference in systolic blood pressure and diastolic blood pressure measured by the conventional and random-zero sphygmomanometers in the supine position. There was a difference between these recordings and those of the Accutorr, with the Accutorr giving higher readings of systolic blood pressure (p less than 0.001), analysis of variance, 95% confidence interval of the difference between the Accutorr and the random-zero was 5.1-15.7 mmHg) and lower readings of diastolic blood pressure (p less than 0.0001, analysis of variance, 95% confidence interval of the difference between the Accutorr and the random-zero was -12.2- -2.2 mmHg). On tilting, the Accutorr showed an increase in systolic blood pressure while the other two machines did not (p less than 0.01, analysis of variance). By contrast, the Accutorr detected a smaller rise in diastolic blood pressure than with the other two instruments (p less than 0.05, analysis of variance). The difference between blood pressure measurements made in the supine position by the two different techniques, auscultation and oscillometry, might be expected. However, the two different techniques do not detect the same blood pressure responses to a change in posture.

Why do lean hypertensives have higher mortality rates than other hypertensives? Findings of the Hypertension Detection and Follow-up Program

Specific causes of death were analyzed for 10,908 participants in the Hypertension Detection and Follow-up Program, to explore possible explanations for the observed excess 8.3-year mortality from all causes in hypertensives with low body mass. Although the cardiovascular mortality rate among men in the lowest decile of body mass (body mass index 21.96 or less) was 50% higher than that of men in the median class (body mass index 26.4-28.8), death rate for noncardiovascular deaths was more than 2 1/2 times higher in men with lean versus median body mass. The pattern was similar among women. Among noncardiovascular causes, striking differences in mortality rates between lean hypertensives and those of average body mass were observed for cirrhotic death (relative risk of 12+ in men and 11+ in women), for nonmalignant respiratory disease in men (relative risk of 7+), for violent death (both sexes), and for malignant neoplasms in men. Prevalence of smoking was almost twice as high in the lowest compared with the median body mass group; among the lean, excess deaths, particularly noncardiovascular deaths, were concentrated among smokers. Thus, male smokers in the lowest decile of body mass constituted only 3% of the study population, but contributed 8% of all deaths, 11% of all noncardiovascular deaths, and 22% of all cirrhotic deaths. A larger proportion of deaths occurred early in follow-up in the lean versus other hypertensives, suggesting occult disease among the lean at baseline. There was no evidence that more severe or treatment-resistant hypertension was present in or could explain excess mortality among the hypertensives with low body mass. The inference from the findings is not that overweight is protective for hypertensives nor that excess risk is due to leanness per se. Rather, a reasonable hypothesis, particularly from findings on specific causes of death, is that excess mortality in lean hypertensives is due to deleterious lifestyles, particularly smoking and excess alcohol intake, contributing to both leanness and risk of death.

Inaccuracy of the Hawksley random zero sphygmomanometer

To examine the accuracy of the Hawksley random zero sphygmomanometer two studies were done with subjects with a wide range of blood pressure. When readings made by one observer on the UK model of the Hawksley sphygmomanometer were compared with readings by two independent observers on separate mercury sphygmomanometers, the Hawksley device underestimated systolic readings by a mean (SD) of 2.0 (2.4) and 0.5 (3.6) mm Hg and diastolic readings by a mean of 3.7 (2.7) and 2.8 (2.9) mm Hg. When readings made on the UK and US models of the Hawksley sphygmomanometer were compared with those made on mercury sphygmomanometers, with observers exchanging devices half way during the experiment, the UK Hawksley device underestimated systolic pressure by a mean of 3.8 (SD 3.5) mm Hg and diastolic blood pressure by 7.5 (3.8) mm Hg; and the US model by 2.6 (3.4) mm Hg for systolic pressure and 6.2 (3.7) mm Hg for diastolic pressure. There was better agreement between two observers using standard sphygmomanometers than between an observer using the Hawksley random zero sphygmomanometer and an observer using a standard sphygmomanometer. Thus, the quantitative aspects of blood pressure in epidemiological and intervention studies in which the Hawksley random zero sphygmomanometer was used need re-evaluation. Moreover, the Hawksley random zero sphygmomanometer, in its present design, should not be used in hypertension research.

Parkinson's disease and hypotension: 24-hour blood pressure recording in ambulant patients

Continuous intra-arterial blood pressure was recorded in 5 ambulatory patients with Parkinson's disease and in 5 control subjects. The 24-h mean systolic blood pressure was 135 +/- 7.6 mmHg in controls and 123.8 +/- 8.1 mmHg (p less than 0.01) in the parkinsonian group. Similarly, diastolic blood pressure was 89 +/- 8.1 mmHg in the control group while in the parkinsonian patients it was lower, 69.4 +/- 5.8 (p less than 0.01). Averages were also calculated for 8-h periods, the results of which indicate that both systolic and diastolic blood pressure were significantly lower in patients than in controls in all three 8-h periods of the day. We normalized the blood pressure curve to mealtimes and arousal times and did not observe any difference between parkinsonian patients and controls. This first study reporting continuous intra-arterial blood pressure measurements in ambulant parkinsonian patients demonstrates that blood pressure in such patients is lower than the mean for their age group. This finding may be of direct relevance in the management of idiopathic parkinsonism.

Once daily nisoldipine in hypertension: cuff and ambulatory intra-arterial blood pressure

Cuff blood pressure data has suggested that the calcium channel antagonist nisoldipine has full twenty four hour efficacy. To test this, 24 h ambulatory intra-arterial blood pressure monitoring was performed on 18 untreated hypertensive subjects (12 men, 6 women) (cuff blood pressure greater than 150/95 mm Hg) before and after chronic treatment with 10-20 mg oral nisoldipine taken daily at 08.00 h. Twelve patients completed the study, six being withdrawn, four because of side-effects. After baseline intra-arterial monitoring patients were started on 10 mg nisoldipine daily. Response was assessed by cuff pressures taken 24 h after dosing at fortnightly intervals, and if not controlled (less than 150/95 or at least 10 mm Hg reduction in diastolic BP) the dose was increased to 20 mg. All patients received at least six weeks' therapy before the second intra-arterial blood pressure monitoring. There was a slight but insignificant reduction in mean daytime heart rate of 3 beats.min-1. Mean significant reduction in daytime systolic and diastolic BP was 19 mm Hg and 13 mm Hg respectively but there was no significant mean night-time reduction. By comparison 8 out of 12 patients were apparently controlled more than 24 h post dose according to cuff pressures. This study suggests that this formulation of nisoldipine does not control blood pressure over a full 24-h period, and emphasises the importance of 24 h ambulatory monitoring in assesing the efficacy of once-daily antihypertensive agents.

Quality assurance and monitoring in the Hypertension Prevention Trial. Hypertension Prevention Trial Research Group

The Hypertension Prevention Trial (HPT), was a randomized, controlled, multicenter (four clinics, four resource centers) trial designed to test the feasibility of achieving and sustaining dietary changes in the intake of calories, sodium, and potassium and to assess the effect of those changes on blood pressure in a normotensive population. The trial involved 841 men and women (plus a test cohort of 78) who, at the first baseline (BL) examination were in the age range of 25-49 years and had diastolic blood pressure (DBP) greater than or equal to 76 but less than 100 mm Hg (average of two readings), and at the examination prior to randomization (BL 2) had DBP greater than or equal to 78 but less than 90 mm Hg (also averaged). Participants were randomly assigned to a control treatment group (no dietary counseling) or to one of four dietary treatment groups involving counseling designed to reduce calorie intake, reduce sodium intake, reduce sodium and calorie intake, and reduce sodium and increase potassium intake. This chapter describes HPT procedures for training and certifying clinic staff, for data entry checks and data audits of its distributed data entry system, and for inspecting clinical equipment. Replicate analyses were performed regularly by the two arms of the Data Coordinating Center. The Food Coding Center and the Central Laboratory were evaluated by both internal and external monitoring techniques. The performance monitoring report, prepared semiannually for the governing committees of the HPT, reported on the effectiveness of quality assurance procedures and served to alert staff to developing problems.

Field evaluation of the Copal UA-231 automatic sphygmomanometer

Automated sphygmomanometers should be evaluated in field studies before being recommended for epidemiological use. In this study an automated sphygmomanometer, the Copal UA-231, was evaluated on the basis of duplicate blood pressure measurements, one with this machine and one with a Hawksley random zero machine, taken on 1536 participants in a health survey. The Copal-random zero differences had a mean of 2.3 mm Hg for systolic blood pressure and 1.9 mm Hg for diastolic pressure, the Copal reading higher than the random zero. The standard deviations of these differences were similar to the standard deviations of the differences between two random zero measurements taken on subjects in a separate survey. However, Copal-random zero differences varied systemically with blood pressure levels. In this study a 10 mm Hg increase in systolic blood pressure was associated with a 1.22 mm Hg increase in the Copal-random zero difference and a 10 mm Hg increase in diastolic blood pressure was associated with a 1.15 mm Hg increase in the Copal-random zero difference. The Copal UA-231 is recommended for use in field surveys, but it should be recognised that a small systematic increase in Copal-random zero difference with increasing blood pressure may complicate interpretation of epidemiological associations.

A comparison of the random-zero and standard mercury sphygmomanometers

Both the standard mercury sphygmomanometer and the random-zero sphygmomanometer have been used in epidemiological studies and clinical trials. Problems arise in comparing studies since, in addition to other methodological differences, the readings obtained with the random-zero sphygmomanometer have been found to be lower than those obtained with the standard mercury sphygmomanometer. In the present study, blood pressures were measured in 66 subjects to examine the comparability of findings with the two instruments. Trained observers measured blood pressures simultaneously using a double-headed stethoscope and one cuff connected to the two sphygmomanometers. Use of instrument was randomly assigned for each blood pressure measurement; each observer was unaware of the other's blood pressure reading. Readings were lower with the random-zero sphygmomanometer; mean difference ranged from 2.5 to 3.3 mm Hg for systolic pressure and 1.9 to 2.7 mm Hg for diastolic pressure. Digit distributions recorded by the two observers for the standard mercury sphygmomanometer and the random-zero sphygmomanometer were not significantly different for either systolic or diastolic blood pressure. Intraindividual variation was greater with the random-zero sphygmomanometer than with the standard mercury sphygmomanometer. These data do not indicate that one instrument is clearly superior to the other, although in studies where the observer seeks to reduce the bias of multiple readings per person, the random-zero sphygmomanometer may be the more appropriate instrument. Critical to the use of either instrument are careful training, standardization, certification, and periodic recertification of observers.

Comparison of a new portable electronic sphygmomanometer (Copal UA251) with the Hawksley random zero machine

The performance of the Copal Digital Sphygmomanometer UA251, a new semi-automatic blood pressure recorder, was evaluated by comparing the results obtained simultaneously and on the same arm using the standard zero-muddler sphygmomanometer. The study was performed on 67 hypertensive patients displaying a wide range of blood pressure and arm circumference. The agreement was acceptable with a mean difference, using the first set of readings of -0.45 (S.D. 2.9, range 8 to -5) and -0.95 (S.D. 2.6, range 6 to -4) mmHg for the systolic and diastolic pressures respectively. The repeatability of the Copal machine was also good; the mean difference between first and second measurements was -0.21 (S.D. 3.4, range 11 to -11) and -0.69 (S.D. 1.9, range 5 to -8) mmHg for systolic and diastolic readings respectively. Since this electronic machine has an acceptable accuracy, is readily portable, simple to use and relatively inexpensive, it seems suitable for use both in clinics and for more frequent monitoring of blood pressure throughout the day.

CARDIA: study design, recruitment, and some characteristics of the examined subjects

In 1984, a prospective cohort study, Coronary Artery Risk Development in Young Adults (CARDIA) was initiated to investigate life-style and other factors that influence, favorably and unfavorably, the evolution of coronary heart disease risk factors during young adulthood. After a year of planning and protocol development, 5,116 black and white women and men, age 18-30 years, were recruited and examined in four urban areas: Birmingham, Alabama; Chicago, Illinois; Minneapolis, Minnesota, and Oakland, California. The initial examination included carefully standardized measurements of major risk factors as well as assessments of psychosocial, dietary, and exercise-related characteristics that might influence them, or that might be independent risk factors. This report presents the recruitment and examination methods as well as the mean levels of blood pressure, total plasma cholesterol, height, weight and body mass index, and the prevalence of cigarette smoking by age, sex, race and educational level. Compared to recent national samples, smoking is less prevalent in CARDIA participants, and weight tends to be greater. Cholesterol levels are representative and somewhat lower blood pressures in CARDIA are probably, at least in part, due to differences in measurement methods. Especially noteworthy among several differences in risk factor levels by demographic subgroup, were a higher body mass index among black than white women and much higher prevalence of cigarette smoking among persons with no more than a high school education than among those with more education.

Clinical and hormonal conditions associated with sodium retention in cirrhotic patients with ascites. Evaluation by univariate and multivariate analyses

Using multiple regression analysis, we have evaluated the clinical and hormonal conditions associated with impaired urinary sodium excretion in normoazotemic patients with cirrhosis and ascites. We retrospectively identified 13 patients with a urinary sodium excretion lower than 15 mmol/day and 13 patients with a sodium excretion higher than 15 mmol/day. Using univariate analysis, all the patients with poor sodium excretion had abnormally high levels of plasma renin activity, plasma aldosterone, and arginine vasopressin. In addition, they had a diastolic blood pressure lower than patients with high urinary sodium excretion, although otherwise were comparable as regards clinical and biochemical data. The consistency of the above associations was then tested by multiple-regression analysis in an attempt to control for potentially confounding factors and to identify only true, independent associations. After a discriminant stepwise procedure, we found that low diastolic blood pressure (P less than 0.01) and high plasma aldosterone levels (P less than 0.05) were the only two conditions independently associated with abnormally low urinary sodium excretion. These findings are consistent with the view that sodium retention in decompensated cirrhosis results from a concomitant severe contraction in the effective blood volume and an increased production and/or retention of aldosterone. The concordance between our results and several pathophysiological findings supports the validity of this statistical approach to confirm physiological and/or clinical predictions.

IgE and cardiovascular disease. Results from a population-based study

Because previous reports have suggested that IgE-mediated events may lead to both platelet activation and arterial spasm, a population-based study of 262 men and 315 women, aged 38 to 82, was conducted to investigate the association of serum IgE levels with myocardial infarction, stroke, and noninvasively diagnosed large-vessel peripheral arterial disease. In men with previous myocardial infarction, previous stroke, or current large-vessel peripheral arterial disease, geometric mean serum IgE levels were increased 119 percent, 164 percent, and 78 percent, respectively. These associations were statistically significant (p less than 0.05). Because IgE was positively or inversely correlated with several traditional cardiovascular disease risk factors, logistic regression was used to evaluate the independent association of IgE with any cardiovascular disease (myocardial infarction, stroke, or large-vessel peripheral arterial disease). In a model including age, cigarette smoking, fasting plasma glucose level, diastolic blood pressure, and low-density lipoprotein cholesterol level as covariates, IgE was positively and independently associated with any cardiovascular disease (p = 0.03). Similar evaluations in women revealed no correlation between IgE and cardiovascular disease by either univariate or multivariable analysis. These data indicate that IgE may be an independent marker for cardiovascular disease in men, and thus suggest IgE-mediated events may play a role in the pathogenesis of cardiovascular disease.

Potassium and norepinephrine- or angiotensin-mediated pressor control in pre-hypertension

Blood pressure (BP), plasma electrolytes, renin, aldosterone, angiotensin II (AII) or catecholamines, the chronotropic effects of intravenous isoproterenol, norepinephrine (NE) or AII, the pressor responses to NE or AII, and the relationship between plasma AII and aldosterone concentrations were studied before and after 10 days of dietary supplementation with potassium 100 mmol/day, in normotensive members of normotensive (N = 12) or hypertensive (N = 12) families, and 11 patients with borderline essential hypertension. Under control conditions, the pressor responsiveness to NE was significantly enhanced in normotensive with positive family history for hypertension and hypertensive subjects; the other variables were comparable in the groups. After potassium supplementation, plasma potassium, renin, aldosterone or AII, and the relationship between AII and aldosterone levels increased significantly, while body weight, plasma catecholamines, the chronotropic effects of isoproterenol, AII or NE, the pressor effects of AII and plasma clearance of AII or NE were unchanged in all groups. In normotensive members of hypertensive families and patients with hypertension, BP was decreased and the exaggerated pressor responsiveness to NE was normalized; these variables were not modified in normotensive members of normotensive families. These observations are consistent with a potassium-remediable disturbance in NE- but not AII-dependent regulation of BP in the pathogenesis of essential hypertension.

Accurate blood pressure measurement

The most critical requirement for obtaining accurate blood pressure measurements is that the Korotkoff sounds be loud. Loudness can be enhanced by various techniques of cuff inflation and chest piece placement. The type of manometer, cuff size, and cuff placement are also important factors in obtaining accurate blood pressure readings. Correct systolic pressure measurement depends on proper inflation and deflation of the cuff. True diastolic pressure is usually closer to the disappearance point of Korotkoff sounds than to the muffling phase. Blood pressure should be recorded to the nearest 5 mm Hg because measurement to the nearest 2 mm Hg is not meaningful and is too difficult and time-consuming.

Increasing daytime vascular resistance with progressive hypertension in ambulant patients

Twenty-four-hour recordings of intra-arterial blood pressure, heart rate, and calculation of the index of peripheral vascular resistance were obtained in 20 untreated ambulant hypertensive patients during their routine activities outside the hospital. Blood pressures and heart rates were higher during the day, with a mean night-to-day increase of 22%, 34%, and 26%, for systolic, diastolic, and mean pressures, respectively. The average increase in the heart rate was 32%. Overall, the index of resistance was lower (12%) during the day. However, eight patients (40%) demonstrated no change or an increase in the peripheral vascular resistance. Six of these eight patients had an average daytime mean pressure above 120 mm Hg. There was a significant decrease in the index of resistance during exercise in all but one of the 16 subjects. Hemodynamic parameters of 1 hour before and 1 hour after awakening were similar to the values obtained during the entire night and the rest of the day, respectively. Our data suggest that in patients with "progressive" hypertension, the ability to decrease vascular resistance from night to day is compromised.

Quality control aspects of blood pressure measurements in the Multiple Risk Factor Intervention Trial

Quality control of an ongoing study of blood pressure must be an active, dynamic process in which attempts are continually made to reduce error and bias and maintain comparability among clinics. Training is an important means of initiating and maintaining proper procedures. In addition, changes in procedures during the trial must be properly evaluated so that their impact upon the conduct of the trial is known and, if necessary, quantifiable. The monitoring of quality measurement of blood pressure in a large clinical trial such as MRFIT must be a two-pronged effort that considers observation and evaluation of methods of measuring blood pressure in actual clinical practice and evaluation of data resulting from these procedures.

Home blood pressure recording

Blood pressure is an inherently variable phenomenon. In addition to this the level rises in the clinical setting; a response which is extremely variable. Problems therefore arise in the diagnosis and treatment of hypertension. Studies have shown that blood pressure can reliably be measured at home either with a stationary or ambulatory apparatus. The readings which are reproducible are generally lower than those obtained in the clinic. Ambulatory recordings are a more reliable guide to prognosis and to the response to hypotensive agents. Home recordings should therefore be more widely used in diagnosis and treatment of the disorder.

Noninvasively diagnosed peripheral arterial disease as a predictor of mortality: results from a prospective study

Intermittent claudication has been reported in previous studies to approximately double the risk of subsequent mortality. However, a history of claudication is often present in the absence of significant peripheral arterial disease (PAD) and absent in the presence of PAD. For this reason we evaluated the association between large-vessel and small-vessel PAD, measured by highly reliable and valid noninvasive tests, and mortality in 567 older subjects from a defined population followed-up for an average of 4 years. Large-vessel PAD was strongly and significantly predictive of all-cause mortality in both men and women with a relative risk of 4 to 5, and this finding was independent of other cardiovascular disease risk factors in multivariable analysis. In addition, this finding persisted after exclusion of subjects with extant cardiovascular disease at baseline. The associations of both claudication and abnormal peripheral pulses with mortality were weaker than the large-vessel PAD association. Isolated small-vessel PAD was unrelated to subsequent mortality. These findings suggest older subjects of both sexes at a high risk of impending mortality can be identified through noninvasive testing for large-vessel PAD.

Comparison of two automatic devices and the standard mercury sphygmomanometer in hypertensive patients

The standard mercury sphygmomanometer (SMS) and two automatic blood pressure recording devices, Dinamap 845 (D) and Sentron (S), were compared by means of a randomized 3-period cross-over experiment. Both devices recorded diastolic BP lower than SMS, on average and for most individual patients. Systolic BP was similar for SMS and S, and slightly lower for D, with variations for individual patients. A second study failed to detect effects of the physician's presence when BP was measured, whereas the difference between D and SMS was substantially confirmed.

Assessment of Hitachi HME-20 pulse and blood-pressure monitor

The performance of the Hitachi HME-20 pulse and blood-pressure (BP) monitor in comparison with direct intra-arterial BP recording and electrocardiographic monitoring is described. Highly significant (p less than 0.001) correlations were found between intra-arterial systolic and diastolic pressures and pressures recorded by the Hitachi monitor. Similarly, the electrocardiographically computed heart-rate, and that given by the Hitachi monitor were significantly correlated (p less than 0.001). Systolic blood-pressure was underestimated by a mean of -12 mmHg and tended to become more erroneous when intra-arterial pressure was greater than 150 mmHg. These results are comparable to more expensive pulse and blood-pressure monitors. We conclude that the instrument can reproduce a satisfactory estimate of heart-rate and blood-pressure and may be of particular use when a change of blood-pressure is of prime importance, rather than an absolute measurement.

Who should code cause of death in a clinical trial?

Clinical trials of intervention in chronic disease often use cause-specific mortality as a principal outcome variable. Surprisingly, there has been little standardization of the approach to determining cause of death. Some studies use standard nosological coding based on the International Statistical Classification of Diseases while others rely on panels of physicians. Some studies utilize autopsy findings; others do not. There is a clear need for standardization, and a unified approach is suggested. In this approach, panels of physicians prepare death certificates and are trained and standardized to generate reproducible information. A system for adjudication of differences is a part of the trial's design. Death certificates are then transmitted to panels of nosologists who assign cause of death. Again, the nosologists are standardized, and an adjudication system for resolving differences is developed. This two-stage system takes advantage of strengths of the two types of cause of death coding now in use in clinical trials and should produce results permitting cross-trial and cross-time comparisons.

Evaluation of the Remler M2000 blood pressure recorder. Comparison with intraarterial blood pressure recordings both at hospital and at home

The Remler M2000 is a semiautomated device that has been used to collect epidemiological data and assess blood pressure variability. It has been subjected to limited evaluation in operation, however, and no studies of its accuracy away from the hospital or office environment have been undertaken. We recruited a group of 28 patients with essential hypertension who were undergoing intraarterial ambulatory blood pressure monitoring and compared the intraarterial recordings with those made with the Remler instrument both at home and in the hospital. The Remler recordings were also compared with simultaneous indirect blood pressure measurements made with the random zero sphygmomanometer. The mean difference between the Remler and intraarterial blood pressure recordings was -3/7 in the hospital and 7/0 at home. All standard deviations were greater than 10 mm Hg, indicating large between-subject variability. Overall, the relationship of the Remler M2000 readings to intraarterial pressures was as close if not closer than standard indirect sphygmomanometry and thus might provide useful data for epidemiological surveys or drug trials. It would appear that for accurate measurement of short-term blood pressure variation and 24-hour recording, intraarterial recording is the method of choice.

Concurrent validity studies of blood pressure instrumentation. The Philadelphia Blood Pressure Project

There is increasing evidence that blood pressure (BP) screening during adolescence may detect early evidence of hypertensive disease. Choice of instrumentation becomes important to maximize the accuracy and interpretive value of several serial readings. We therefore conducted two studies of adolescents, comparing auscultatory (Baumanometer and Random-zero), ultrasonic (Arteriosonde 1216), and infrasonic (Physiometrics SR-2) devices. These instruments were compared serially (Study 1) for older adolescents (n = 48). The Baumanometer, Arteriosonde 1216, and Physiometrics SR-2 were compared both serially and simultaneously (Study 2) for younger and older adolescents (n = 24). In both studies, the order of device presentation was completely counterbalanced. In Study 1, reliability estimates were high (r greater than 0.65), with the exception of the Arteriosonde 1216 diastolic readings, and the Physiometrics SR-2 yielded significantly lower mean diastolic BP readings than the other devices. Diastolic Phase V readings taken with the Random-zero were significantly higher when the Random-zero was presented at the end of the sequence of instrument presentation. This order of presentation effect is probably attributable to the need for overinflation when using the Random-zero device and may cause an overestimation of diastolic BP in the young. In Study 2, again the Arteriosonde 1216 did not perform reliably in recording diastolic BP. Further, the Physiometrics SR-2 device, although reliable, yielded diastolic readings significantly lower (p less than 0.001) than diastolic Phase IV readings taken both serially and simultaneously.(ABSTRACT TRUNCATED AT 250 WORDS)

Training and certification of blood pressure observers

Accurate, reproducible measurements of blood pressure (BP) were central to the goals and objectives of the Hypertension Detection and Follow-Up Program (HDFP), a multicenter clinical trial on the efficacy of pharmacological treatment of individuals with elevated BP. All potential BP observers with or without previous experience in measuring BP were required to undergo a defined training program and meet set performance criteria to be certified to take HDFP BP. Recertification was required twice a year. Originally an audiotape test was used to measure accuracy of BP readings. This approach was later replaced by a videotape test, which proved more realistic and an equally effective tool for long-term quality control. With this technique of certifications, 75% of the individuals taking the test passed on the first attempt and more than 95% passed with one or two attempts. Although agreement for blinded BP duplicates was generally good, the appearance of sound (systolic BP) was identified with greater reproducibility than was the disappearance (diastolic BP). These recertification procedures were of great value in assuring the continued high quality of our BP data.

Continuous vs intermittent blood pressure measurements in estimating 24-hour average blood pressure

In the past few years noninvasive automatic blood pressure (BP) recorders have been increasingly used to estimate patients' 24-hour BP more accurately than by one or few isolated measurements. However, these recorders only allow BP to be intermittently measured at intervals between 5 to 30 minutes, which means that the number of values collected over 24 hours (10 to 100) remains a tiny fraction of the thousands of values that occur during the same period. To determine whether this represents a limitation to this approach, BP was recorded intraarterially for 24 hours (Oxford method) in 20 ambulant hypertensive patients. A beat-to-beat analysis of the BP recording was provided by a computer, and the average 24-hour systolic, diastolic, and mean BP values were compared with those obtained by analyzing single BP waves of the same recording at intervals of 5, 10, 15, 30, and 60 minutes. In each subject the average 24-hour BP values obtained by the beat-to-beat analysis closely corresponded to those obtained by the analysis performed at 5-, 10-, 15-, or 30-minute intervals. In most subjects, this was the case also when the analysis was performed at 60-minute intervals. These findings demonstrate that intermittency of measurements does not limit the accurate assessment of true average BP. Indeed, accurate assessment can be achieved at intervals as much as 30 or 60 minutes apart.

Changes in blood pressure in the lying and sitting positions in normotensive, borderline and hypertensive subjects

Blood pressure of hypertensive subjects reacts differently from that of borderline or normal subjects at rest in either sitting or lying positions. In the hypertensive group the diastolic pressure tends to rise more on consecutive measurements and the systolic blood pressure tends to fall less than in the two other groups. Similarly, on going from the lying to the sitting position, diastolic blood pressure of hypertensives rises less than that of normotensive or borderline hypertensive individuals. Thus, people with 'casual' hypertension differ in their reaction to rest and changes in body position in their reaction to rest and changes in body position with respect to normotensives or borderline hypertensives. This behavior could lead to labelling of subjects as 'normal' or 'hypertensive' depending on position and timing of the blood pressure measurement.