Blood Pressure—The Better Biomarker: Delay in Clinical Application


Thomas D. Giles, MD, 109 Holly Drive, Metairie, LA 70005


“The truly great advances in our understanding of nature originated in a way almost diametrically opposed to induction. The intuitive grasp of the essentials of a large complex of facts leads the scientist to the postulation of a hypothetical law or laws. From these laws, he derives his conclusion.”—Albert Einstein, “Induction and Deduction in Physics,” Berliner Tageblatt, December 25, 1919, CPAE 7:28.

One widely accepted definition of a biomarker is a characteristic that is objectively measured and evaluated as an indicator of normal biologic or pathogenic processes or pharmacologic responses to a therapeutic intervention. A plethora of “new” biomarkers have been suggested for the detection of cardiovascular (CV) disease, such as abnormalities in blood lipids (eg, particle size), measures of systemic inflammation (eg, high-sensitivity C-reactive protein), indicators of thrombogenic potential (eg, plasminogen activator inhibitor 1), or indices of oxidative stress (eg, urinary isoprostanes).1 When evaluated against the usual markers of CV risk, however, new biomarkers do not contribute greatly to assessment.2

Blood pressure (BP), on the other hand, is not often described as a biomarker. Either it is linked to a particular level of increase at which the term hypertension is used or it is viewed as a therapeutic target to be lowered. With the evolving appreciation of BP as a signal representing a physical force that may reflect abnormalities in the CV system, however, the utilization of BP as a biomarker, in the true sense, is compelling.


There is, of course, an optimal BP range for individuals at rest. For general purposes, this level of BP was described as ≤120/80 mm Hg in the Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC VI).3 Indeed, data from the Framingham Study4 demonstrated that individuals with optimal BP fared better than those with “normal” or “high-normal” BP levels in progressing to higher BP levels. Eventually, prospective data indicated that the ability of BP to detect CV risk begins at approximately 115/75 mm Hg and that the risk doubles for each 20/10-mm Hg increase.5 These data indicate not just risk for CV events but the actual presence of disease (no events without disease).


The International Database on Ambulatory BP Monitoring in Relation to Cardiovascular Outcomes (IDACO)6 performed 24-hour ambulatory BP monitoring (ABPM) in 5682 participants (mean age, 59 years; 43.3% women) enrolled in prospective studies in Denmark, Belgium, Japan, and Sweden and followed for 9.7 years. Utilizing multivariate analysis, ABPM thresholds yielded 10-year CV risks similar to those associated with optimal, normal, and high (>140/90 mm Hg) BP on office measurement. Diagnostic thresholds for ABPM were obtained elegantly in 5 steps ultimately utilizing the bootstrap point estimates and 95% confidence intervals of the ambulatory thresholds as the mean ± SEs of the bootstrap distribution.

Data are shown in the Figure. For optimal BP, 24-hour BP was 116.8/74.2 mm Hg, 121.6/78.9 mm Hg for daytime, and 100.9/65.3 mm Hg for nighttime. The authors emphasized that recognition of the importance of lower optimal BP is needed.

Figure Figure.

The association between ambulatory blood pressure (BP) monitoring and cardiovascular risk and the number of events contributing to the incidence rates are shown. Incidence rates were standardized by the direct method for cohort, sex, and age (≤40, 40–60, and ≥60 y). Of particular interest is the steep increase associated with systolic blood pressure levels <140 mm Hg systolic. Adapted from Kikuya et al.6

ABPM is of course a technique that allows recording of BP throughout the entire day and is highly reproducible.7 The technique eliminates the white-coat effect and, in clinical trials, the placebo effect. The IDACO supports the concept that the incidence of CV events did not differ between a “normotensive” control group and a “hypertensive” group with the white coat effect. The circadian patterns of BP are important in certain disease states (eg, the absence of a nocturnal decline [“nocturnal dip”] indicates higher CV risk).


Recognition of phenotypes among the patterns of systolic, diastolic, mean, and pulse blood pressures are a good beginning.8 These phenotypes are similar to the phenotypes of cardiomyopathy (ie, dilated, hypertrophic, and restrictive in that each phenotype suggests a differential diagnosis).9 For example, disproportionate elevation of systolic BP indicates a loss of proximal arterial compliance or an increase in left ventricular stroke volume (eg, aortic insufficiency, fever, hyperthyroidism) easily discerned at the bedside.

Estimation of central aortic pressure is of importance, particularly in determining the effects of certain antihypertensive drugs.10 Measurements of augmentation index, reflected waves, and pulse wave velocity, as well as using mathematic models to derive compliance from the pulse wave contour, are valuable in assessing the contribution of vascular stiffness to central aortic pressure increases.

Other analyses of the BP signal may utilize mathematic techniques, such as Fourier or Haar dyadic wavelet transforms, to decompose the signal into useful information on the state of the CV system. When combined with analysis of heart rate variability, an assessment of integrated CV control is improved.


It is time for state-of-the-art BP analysis to become part of a program in the prevention of CV disease. BP analysis is a sensitive biomarker for the detection of early CV disease. In this regard, it resembles mammography and colonoscopy in the detection of cancer. The magnitude of CV disease, however, is much greater.

It is time to apply what we have learned about BP analysis to the assessment of global CV risk. Thus, at certain milestones during the life of an individual, analysis of systemic arterial BP utilizing advanced techniques should be carried out. The frequency of such an analysis should be performed depending on the presence or absence of other risk factors and biomarkers. Such assessments are similar to those recommended for other serious biomarkers. This assessment should be performed by a physician with expertise in BP analysis and risk factor assessment. And, just as with colonoscopy and mammography, experts performing such an assessment should be reimbursed accordingly. Depending on the findings, interventions may consist of everything from watchful waiting to pharmacologic therapy.

Obviously, periodic assessment will permit the detection of trends in BP abnormalities. Tools to assist in following longitudinal trends in BP and other biomarkers are being developed. Also, a greater number of individuals will be performing CV risk self-assessment using out-of-office BP measurements in the future that will permit earlier contact with physicians to discuss the results of their assessment.

The naysayers will comment that even using conventional casual office BP assessment, we are not doing a good job of reducing morbidity and mortality from hypertension and increased BP and that sophisticated analysis by an expert will only increase cost. Armed with the understanding that a 2-mm Hg decrease in BP is associated with a 7% reduction in ischemic heart events and a 10% reduction in stroke, however, the return on investment would be enormous. In addition, to accomplish this task physicians will require additional education to qualify as hypertension specialists.

It is time to quit “dumbing down” the concept of BP analysis and treatment and to keep pace with other disease specialties. We now perform routine colonoscopy instead of flexible sigmoidoscopy and mammography to supplement breast self-examination. Chronic increases in BP should be taken seriously. We have lived through the concept of “essential hypertension” (the classic oxymoron) and that systolic BP should be (mm Hg) = 100 + age (years). After all, should our patients not benefit from the dedicated efforts of those contributing to hypertension research? We have observed the natural history of hypertension for too long.