Diastolic Blood Pressure, Not Just Systolic Blood Pressure, Is Related to Cerebral Measures in Middle Age: Implications for Prospective Studies

The two fundamental components of blood pressure (BP) are systolic BP (SBP) and diastolic BP (DBP). Debate over the prognostic importance of each component at middle age represents an important period in the history of hypertension, as it relates to the prediction of cardiovascular outcomes and brain injury. In recounting this history, Schillaci et al.1 point out that the emphasis on SBP was supported by literature indicating that it is a stronger predictor of cardiovascular morbidity and mortality than DBP in mature adults and elderly persons. They describe the emphasis on SBP that followed in the literature and how it led to a rather extreme movement to discard DBP as prognostically important in individuals 50 years of age or older. The pendulum has swung from DBP to SBP and finally, back to the importance of both as predictors of cardiovascular disease and altered brain function and structure. This history is reflected in the Joint National Committee (JNC) reports on the detection, evaluation, and treatment of high blood pressure as assessed by Pogue et al.2 Prior to the 5th JNC report in 1993,2 there was a long-standing conviction that the cardiovascular risk associated with arterial hypertension was mostly derived from DBP, not SBP. In the 5th report,2 hypertension was defined on the basis of both DBP and SBP, with recognition of the importance of isolated SPB. From that perspective, it is surprising to find that some papers promoted the argument that DBP should be discarded as a prognostic indicator in older adults.3 The argument did not prevail. Obviously, there are a number of common sense reasons for keeping DBP assessments. They are important in the calculation of mean arterial pressure and pulse pressure, and the data are available at little, if any, extra cost. Aside from logical reasoning, the argument that DBP should be kept as a prognostic indicator was dramatically and elegantly established in a study of untreated hypertensive persons who were early participants of the Framingham Heart Study. Franklin et al.4 included 9,657 untreated participants of the Framingham Heart Study with a mean (±SD) age of 42 ± 11 years. SBP was a stronger risk factor for cardiovascular disease. However, the combined prediction afforded by SBP and DBP was stronger than the prediction afforded by either BP component alone. Elias et al.5 extended these findings to cognitive functioning using a much smaller sample (n = 1,702) of the same Framingham study participants. SBP and DBP during the surveillance period, employed as separate predictors, were inversely associated with multiple measures of cognitive functioning assessed 22 years later. In subsequent research, the importance of both SBP and DBP as predictors of brain structure and function5–8 and accelerated decline in cognitive function6,8,9 was confirmed. The aforementioned history regarding the importance of SBP and DBP as predictors of brain structure and function provides an important framework for our comments on a paper published by Shokouhi et al.10 in this issue of the American Journal of Hypertension. We argue that it provides a new set of variables and a research paradigm with which to study the role of DBP and SBP in brain function and structure. Moreover, it highlights the need to shift the surveillance period (baseline) from midlife to young adulthood in prospective studies of hypertension and decline in cognitive performance over the lifespan. Shokouhi et al.10 examined cross-sectional associations between BP and measures of brain structure and function (cerebral blood flow, cortical, and white matter lesions). The sample consisted of 265 adults (>50 years of age), free from dementia. Analyses were conducted in individuals with normal cognitive functioning and mild cognitive impairment (70% of the sample). Cognitive determinations were based on a battery of cognitive tests and clinical diagnosis by a study physician or neuropsychologist. In multiple regression analyses adjusted for multiple clinically and theoretically relevant confounders, DBP and SBP were positively associated white matter lesion volume. However, only DBP was inversely associated with gray matter cerebral blood flow and cortex-to-brain volume ratio. Surprisingly, results did not differ between the normally functioning group and the group with mild cognitive impairment. In their discussion of the similarity of results for these two groups, Shokouhi et al.10 argue that their research paradigm may provide a tool for detecting the earliest biological signs of conversion from normal functioning to MCI. However, the lack of significant differences in results for normal functioning and MCI groups suggests that the set of relations reported for DBP and SBP may not presage conversion from normal-to-mildly impaired cognition. However, the hypothesis is worthy of pursuit with larger, more representative samples. The estimated prevalence of normal cognitive functioning vs. mild cognitive functioning in the general population is 3% to 19%.11 This should be reflected in future studies, as it would yield a larger sample size and a more proportional representation of normally functioning individuals. It would also be interesting to repeat the Shokouhi et al.10 study with a more stringent operational definition of MCI. In a paper on MCI methods, Schinka et al.11 make it clear that the way in which MCI is defined often determines the results obtained when comparing normal functioning and MCI groups. In our studies, we like to see clear evidence, based on data from informants and relatives, that the individual diagnosed with MCI cannot perform multiple routine activities of daily living without assistance. These recommendations for future work within the cross-sectional framework used by Shokouhi et al.10 are typical for a “initial study.” The Shokouhi et al.10 study has important implications for future prospective studies with baseline data collection moving backward in time from MCI. Inspection of Figure 1, reproduced from Sperling et al.,12 will be helpful in presenting our argument for these studies. This figure contains two lines beginning early in life that illustrate: (i) the progression of normal changes in cognitive functioning into old age and (ii) the progression from normal functioning to preclinical cognitive impairment (i.e., the period between chronic exposure to hypertension and the appearance of symptoms of cognitive impairment). As we move from normal function into MCI and dementia, cognitive changes manifest more clearly in parallel with changes in brain function and structure. A conceptual plot showing the progression of normative and non-normative change in cognitive performance over the course of the lifespan. Figure reused with permission and recaptioned by the authors of this comment. Given the importance of DBP, we need to move backward in time toward the preclinical and incubation period for significant cognitive deficit. We need more prospective studies with younger individuals, in whom cognitive function has been shown to differ by hypertensive status.13,14 Moreover, we need a better understanding of relations between isolated DBP, a known risk factor for reduced cognitive function, cardiovascular morbidity and mortality in late life,15,16 and brain structure and functioning. Middle age has been heralded as the critical point in time for collecting baseline data for prospective studies of brain injury and cognitive decline, and with good reason. Relations between these predictors and outcomes are weaker when we perform cross-sectional studies with demented individuals.6 Moreover, relations between BP and dementia may be confounded by bi-directional relations, as the dementing process can affect BP and BP regulation.17 Moving baseline observations back to middle age avoids this confounding, but why not move even closer to the beginning of cognitive, structural, and functional changes that occur in the brain? We would like to see more studies with younger adults with DBP, especially isolated DBP. We know that isolated DBP raises the risk for SBP in adulthood and old age. Moreover, it is related to increased cardiovascular mortality and morbidity15 and lower cognitive performance in early life.13,14,16 Thus, recognizing the importance of the isolated diastolic phase of hypertension could take our investigations backward in time to the beginning of BP-related brain changes in young adults and adolescents that ultimately may lead to the MCI and dementia trajectory as shown in Figure 1. With adequate statistical power, negative results could help define the incubation period between exposure to chronic hypertension and the appearance of changes in the brain. The study by Shokouhi et al.10 gives us an important set of variables with which to pursue these goals. The authors declared no conflict of interest.