Pulse pressure and blood pressure components:. Is the sum more than the parts?

Pulse pressure (PP) is the arithmetic difference between systolic blood pressure (SBP) and diastolic blood pressure (DBP), commonly measured as brachial cuff pressures. There is a progressive increase in blood pressure (BP) with ageing in industrialized societies, beginning in childhood and continuing into adulthood. This trend is associated with a greater increase in SBP than diastolic DBP during the middle adult years. Whereas SBP continues to rise until the eighth or ninth decade, DBP tends to remain constant or decline after the fifth or sixth decade; as a consequence, PP increases progressively with age and the rate of rise accelerates after age 50 years. There also is a sex difference in BP trends; women’s BP starts lower than men’s, catches up by the sixth decade and frequently becomes slightly higher thereafter.1,2 These natural trends are increasingly modified by contemporary treatment of hypertension. The prognostic importance of PP has been widely described in various patient populations, starting with the seminal Framingham cohort.3 Elevated DBP has long been associated with increased cardiovascular risk (CVR), albeit in younger populations.4 SBP is directly and linearly related to CVR over a wide range of ages. With advancing age, there is a decline in DBP and in the role of DBP in predicting coronary heart disease risk. DBP becomes inversely related to CVR, especially in multivariate models including SBP.3 With increasing age, there is a gradual shift from DBP to SBP and then to PP as predictors of CVR. In patients <50 years of age DBP is the strongest predictor. Age 50 to 59 years is a transition period when all three BP indices are comparable predictors, and from 60 years of age on, DBP is negatively related to CVR so that PP becomes superior to SBP as a risk predictor.5 These Framingham data were largely confirmed in the huge Multiple Risk Factor Intervention Trial (MRFIT) database.6 In men screened for MRFIT (age 35–57), PP was not as strong a predictor of cardiovascular mortality as SBP and DBP. In contrast to DBP, which displays a quadratic relation with cardiovascular risk, both mean BP and PP have a linear, independent relation with risk.7 The best predictor function among the possible linear combinations of SBP and DBP is similar to that of PP, suggesting that their association is real and not merely a statistical artefact caused by the correlation between SBP and PP.8 PP therefore combines prognostic information related to increased SBP and decreased DBP. What is the underlying mechanism of increased BP components? Haemodynamics tell us that an increase in SBP with fixed PP occurs when there are parallel increases in SBP and DBP; this is a consequence of a rise in mean arterial pressure and peripheral vascular resistance. This typically occurs in younger hypertensives. An increase in SBP together with an increase in PP occurs as a consequence of a rise in large artery stiffness. DBP may slightly increase, remain unchanged or even decrease.9 This typically occurs in aged hypertensives. Increased SBP and PP may also reflect increased stroke volume. This sometimes happens in young hypertensives. Physiologically, PP is determined by arterial stiffness, ventricular ejection, and by wave reflections.8 Arterial stiffness. During systole, the aortic wall expands, limiting the rise in SBP. In diastole, the aorta contracts, due to elastic recoil. This contributes to elevating DBP and perfusion pressure. The pulsatile flow therefore is transformed in a more continuous perfusion flow.9 With age and hypertension, these reciprocal properties are progressively lost. A stiffer aortic wall and loss of elasticity contribute to increased PP and to increased pulsatility at the target organs. Ventricular ejection. It was long believed that ventricular ejection would not affect age-related changes in PP. However, with age and hypertension, systolic ventricular elastance increases in order to match the increased arterial elastance, accordingly reduced arterial compliance, thereby preserving ventricular-arterial coupling, stroke volume and ejection fraction.10 In the Framingham offspring cohort, an age-related increase in aortic stiffness, as opposed to peripheral arterial stiffness, was associated with increasing forward wave amplitude and PP. This phenomenon may facilitate forward transmission of potentially deleterious pressure pulsations into the periphery, thereby providing a pathophysiological link between PP and target organ damage.11 Wave reflection. In young and healthy subjects, the forward travelling pressure wave is reflected at the sites of impedance mismatch and returns to the heart, contributing to diastolic perfusion pressure. With progressing aortic stiffening, the forward wave travels faster and returns earlier to the heart, boosting late-systolic pressure.12 These reflected waves are stronger, due to the impact of stronger forward wave on backward wave pressures. Backward waves contribute to age-related increases in aortic PP across the adult lifespan.13 The bulk of wave reflections arise at the lower body, travel back to the heart and penetrate the carotid artery as a forward wave, which increases pulsatile pressure and flow in the coronary and cerebrovascular bed.14 The magnitude of wave reflection appears to be independently associated with diastolic dysfunction, left ventricular hypertrophy and CVR (summarized in Nagueh et al.15). The prognostic value of wave reflection for incident heart failure is independent of, and as potent as, SBP and DBP.16 A paper in this issue of the EJPC17 analysed the prognostic value of high PP (HPP; PP ≥ 60 mmHg) in The Campania Salute Registry, a big cohort of treated hypertensive patients. A total of 7336 patients were included, with a mean age of 53.6 years. In this cohort 32% of the patients presented with HPP. Patients with HPP were older, were more likely to be women, had much higher BP at the start of the study (160/89 mmHg versus 135/89), more diabetes, lower glomerular filtration rate (eGFR), and were treated with more anti-hypertensive drugs. Patients with HPP had slightly greater left ventricle (LV) chamber dimension and stroke index, greater LV mass and relative wall thickness, and more prevalent carotid plaques. Major adverse cardiovascular end-points (MACE) at their first presentation were defined as fatal or non-fatal myocardial infarction or stroke, and sudden death. In the final Cox regression model, the condition of high PP was adjusted for age and sex, presence of diabetes, left ventricular hypertrophy, carotid plaque, classes of antihypertensive medications and mean arterial pressure. HPP patients had 57% increased hazard of MACE, compared with normal pulse pressure (NPP). A similar model was run changing PP with its two components, SBP and DBP. In this model, both SBP and DBP were independent predictors of cardiovascular events, but in opposite directions. As could have expected, SBP exhibited a direct relation with probability of events, whereas the relation of DBP was negative. It may appear as a semantic discussion, but it is not obvious to adjust one of the most readily available clinical parameters, for laboratory or imaging parameters. In the present study, these adjustments were done in order to show that PP is related to CVR, even independently of well-known structural markers of target organ damage. It is, however, unexpected that the authors did not adjust for eGFR. This inclusion could have weakened the association between PP and MACE.18 The study provides confirmatory evidence that PP is a superior predictor of CVR to SBP, DBP or mean BP. Some aspects are of particular interest. The first is the mean age of the population (54 years). At this age PP is not yet a strong prognosticator in the Framingham or MRFIT population.5,6 Albeit that in the MRFIT trial, men 45–57 years old, with higher SBP and lower DBP (discordant elevation) yielded a greater risk of cardiovascular mortality. It is obvious that the patients with HPP of the present investigation represent a high-risk population. It is of further interest to observe the huge difference in BP values between HPP and NPP patients at the start of the study. The authors redid their analysis using BP values during follow-up and kept the prognostic significance. The authors report an average BP on follow up of <140/90 mmHg, while initial average was 143/89 mmHg. It remains uncertain how well BP was controlled in this survey. An interesting additional analysis was performed on patients with adequately controlled BP during follow-up. In the corresponding adjusted Cox regression, PP came out as not significant. Uncontrolled SBP therefore appears to be a confounder of the present findings. Increased pulse pressure reflects aortic stiffening, increased wave strength (forward and backward), premature reflection and increased peripheral pulsatility of flow. These alterations affect target organ damage and independently predict cardiovascular complications. The implications of the study under scrutiny and the preceding studies on the prognostic value of PP may be easily derived. Any combination of two of the three BP measures provides more information about CVR than a single measure. Since pulse pressure is simply the difference between SBP and DBP, given predictive information from SBP and DBP considered together, PP cannot add further information.6 It, however, is a powerful predictor of CVR, especially in an older population. In order to use the prognostic information of BP measurements you should select two BP components, either SBP and DBP or PP with DBP or mean BP. The choice may depend on the population characteristics under scrutiny. It finally has to be noted that the present investigation showed that PP remained a significant predictor in a multivariate analysis. It, however, did not show that adding PP to a panel would significantly improve CVR prediction (receiving operating characteristic and net reclassification index). For any ‘newcomer’ measurement, closely related to an already settled one, the likelihood of improving prediction of the person-specific absolute cardiovascular risk in a clinically relevant way is low.19,20 It remains to be investigated whether substituting SBP by PP and adding it to DBP or mean BP enhances CVR prediction. The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author received no financial support for the research, authorship, and/or publication of this article. Thierry C Gillebert https://orcid.org/0000-0002-3832-919X