Orthostatic blood pressure adaptations, aortic stiffness, and central hemodynamics in the general population: insights from the Malmö Offspring Study (MOS)

The Malmö Offspring Study (MOS) is a population-based study consisting of children and grandchildren to the index participants in the well-established Malmö Diet and Cancer Study (MDCS) launched in the early 1990s with over 20 years of follow-up [14]. The data collection within MOS started in 2013 and was completed by the end of 2021. To ensure that participants in the different generations were directly related, a national taxation authority register in Sweden (NAVET) was used. A total of 5259 individuals were included in MOS. Of these, 3966 individuals had complete recordings of measurements of orthostatic BP reactions, aortic stiffness, and central hemodynamics, and were thus included in the final study population.

The primary outcome was increased aortic stiffness (c-f PWV) in relation to orthostatic BP differences. Secondary outcomes included increased aortic stiffness and BP reaction on standing in relation to sex and increasing age, respectively.

Supine systolic (SBP) and diastolic blood pressure (DBP) were reported after 5 min rest as the mean of two readings, using an automatic device (Omron M5-1 IntelliSense, the Netherlands). Subjects were thereafter asked to stand up, and the mean SBP and DBP of two consecutive readings performed after 3 min of standing was reported as the orthostatic blood pressure. Aortic stiffness was directly assessed noninvasively after 5 min of supine rest using the gold-standard measurement of c-f PWV by Sphygmocor (AtCor, Australia) [15]. The distance from the carotid to femoral artery was measured directly between each artery and the suprasternal notch. PWV was calculated by measuring the time delay between two characteristic timepoints on two pressure waveforms at a known distance apart. The SphygmoCor method uses the foot of the waveform as an onset point for calculating the time differences between the R wave of the electrocardiogram and the pulse waveforms at each site. PWV was automatically generated as the carotid–femoral artery distance divided by the wave traveling time between the above two sites.

Aortic augmentation index (AIx), an indirect measurement of aortic stiffness, was together with variables reflecting central aortic hemodynamics (central aortic SBP and central aortic DBP) acquired by radial applanation tonometry using PWA Sphygmocor (AtCor, Australia). SphygmoCor also measures and calculates the AIX standardized to a heart rate of 75 beats per minute (AIx@75), by adjusting the AIX by −4.8 for each 10 bpm above and +4.8 for each 10 bpm below a resting heart rate of 75 bpm.

Brachial SBP and DBP were used to calibrate the radial and aortic pressure waveform. This technology has good reproducibility under major hemodynamic changes, analyzing a central (ascending aortic) pressure waveform from the radial pressure waveform using a validated generalized transfer function [16].

Data on prevalent diseases at baseline were collected from The Swedish National Patient Register. “Orthostatic BP difference” (mmHg) was defined as [standing BP − supine BP], i.e., a positive value denotes an increase in BP upon standing. Antihypertensive treatment at baseline was defined as self-reported intake of antihypertensive drugs. Estimated glomerular filtration rate (eGFR) was calculated using the CKD-EPI Creatinine Equation [17]. Current smoking (defined as regular or occasional smoking) and educational level were self-reported in the questionnaire.

The MOS study was approved by the Regional Ethics Committee at Lund University (Dnr. 2012/594). All participants provided written informed consent.

Continuous data (of which all were normally distributed) are shown as mean ± standard deviation, whereas frequencies are used to describe categorical data. We first assessed aortic stiffness and central hemodynamic measurements by quartiles of orthostatic BP differences using one-way analysis of variance (ANOVA). Then, a linear regression model was constructed to assess the association of standing BP differences and subclinical orthostatic hypotension with aortic stiffness and central hemodynamics after different adjustments. The basic model was adjusted for age and sex. The multivariable model was adjusted for age, sex, body mass index (BMI), current smoking, fasting glucose, eGFR, antihypertensive medication, and supine SBP/DBP, as appropriate. Orthostatic diastolic BP reaction and aortic stiffness displayed a linear relationship, and was additionally adjusted for supine diastolic BP. In contrast, orthostatic systolic BP reaction displayed a U-shaped association with the outcome variables and was entered as mean-centered orthostatic systolic BP and adjusted for mean-centered supine systolic BP in addition to remaining covariates. For the secondary outcomes, analyses were performed separately according to sex and median age 44 years, respectively.

All statistical analyses were performed in IBM SPSS Statistics 27 (IBM Corporation, Armonk, NY, USA).

We observed a U-shaped association between orthostatic systolic BP differences, aortic stiffness, and central hemodynamics, i.e., subjects in the lowest and highest quartiles of orthostatic systolic BP differences demonstrated the highest degree of aortic stiffness and central aortic BP (all p < 0.001 in ANOVA; Fig. 2, Supplementary Fig. S1, Table 2). In contrast, orthostatic diastolic BP reaction displayed a linear association, with a gradual reduction in aortic stiffness and central aortic BP across all four quartiles of increasing orthostatic diastolic BP reaction (all p < 0.001 in ANOVA; Fig. 3, Supplementary Fig. S2, Table 2). No difference was observed in the sensitivity analysis after exclusion of subjects ≥ 60 years old (n = 372, 9.4%) (Supplementary Table S1).

After full adjustments, orthostatic diastolic BP remained significantly associated with aortic stiffness (c-f PWV p = 0.001, Aix p = 0.005, and Aix@75 p = 0.007) and central aortic BP (p < 0.001), whereas orthostatic systolic BP was significantly associated only with indirect aortic stiffness adjusted for HR (Aix@75 p = 0.04) and central aortic systolic BP (p = 0.009) (data not shown). No significant associations were found between subclinical manifest orthostatic hypotension, aortic stiffness, and central hemodynamics (Supplementary Table S2).

We performed fully adjusted subgroup analysis of aortic stiffness and central hemodynamics stratified by median age (Table 3). Overall, orthostatic diastolic BP was inversely associated with indirect aortic stiffness (AIX@75) (p = 0.003 and p < 0.001, Table 3), and central aortic systolic as well as diastolic BP in both age groups (p < 0.001). Orthostatic systolic BP was associated only with indirect aortic stiffness in older individuals (Aix p = 0.008, Aix@75 p = 0.008), whereas it was inversely associated with central aortic systolic BP in both age groups (p = 0.008 versus p = 0.049). Manifest subclinical OH was associated with central aortic systolic BP in the above-median-age group only.

Following sex stratification, orthostatic diastolic BP was inversely associated with direct aortic stiffness (c-f PWV p = 0.03 versus p = 0.006, Table 4), and central aortic systolic and diastolic BP (all p < 0.001) in both sexes. Only women displayed an inverse association between aortic stiffness and orthostatic diastolic BP (AIx p = 0.03 and p = 0.04), and a positive association between orthostatic hypotension and central aortic systolic BP (p = 0.02). Likewise, central aortic systolic and diastolic BP was significantly associated only with orthostatic systolic BP in women (all p = 0.02).

A distinct increase in diastolic BP of approximately 10 mmHg upon standing is considered the normal reaction [20], and accordingly we observed that persons with a diastolic BP increase on standing of ≥ 14 mmHg had the lowest aortic stiffness. This is in line with previous data from the Framingham Heart Study Third Generation cohort, even though mean arterial pressure (MAP) was used in that study [8]. For the systolic BP increase upon standing, an U-shaped association with aortic stiffness was noted, which is in line with data suggesting that both marked systolic blood pressure decrease [5] and increase [6] upon standing is associated with adverse cardiovascular outcomes [21].

Aortic stiffness is an important risk factor/marker for cardiovascular disease (CVD) and a key element in the pathogenesis of CVD. Growing evidence shows that increased aortic stiffness can predict both cardiovascular mortality and all-cause mortality [22, 23].

An epidemiological study such as ours cannot prove causality but may generate hypotheses for mechanism-oriented investigations. While we cannot discern whether altered orthostatic blood pressure control promoted vascular damage or vice versa, our study provides an impetus to investigate mutual interactions between cardiovascular autonomic control and vascular structure in more detail. For example, baroreflex counter-regulation, which stabilizes blood pressure upon standing, could be impeded by changes in vascular structure. In a previous study, we found that proteins associated with atherosclerosis were also related to impaired blood pressure control [19, 24]. Diminished baroreflexes often occur in patients with arterial stiffness due to impaired stretch of the baroreceptors and reduced neural input to the brain stem’s autonomic control centers, causing decreased output to the cardiovascular system. Another study demonstrated that asymptomatic subjects with subclinical coronary atherosclerosis display severely blunted baroreflexes and may also have advanced coronary atherosclerosis [25]. Conversely, blood pressure swings elicited through poor cardiovascular autonomic control could negatively affect vascular structure [7].

The current study, Malmö Offspring Study, is unique since it contains a wealth of data on younger and rather healthy individuals (mean age of 41 years) with few comorbidities. For example, the hypertension prevalence (based on use of antihypertensive drugs) was only 5%. In addition, this is the first population-based cohort to investigate the association of orthostatic BP reactions with well-established and validated assessments of aortic stiffness and central hemodynamics, including the gold-standard method c-f PWV, as well as central aortic hemodynamic measurements such as AIX and central aortic BP [26‐28]. Previous studies have only assessed aortic stiffness by c-f PWV in this context [8, 21, 29‐31] or brachial-to-ankle PWV [32]. Moreover, the abovementioned studies have focused on older individuals (mean age ranging from 44 to 80 years) with higher prevalence of established comorbidities such as type 2 diabetes or hypertension, ranging between 20% and 60%.

To our knowledge, this is the first and largest observational study investigating the association between orthostatic blood pressure reactions and markers of aortic stiffness in nearly 4000 individuals with a mean age of 41 years. The Malmö Offspring (MOS) cohort was recently completed at the end of December 2021, providing unique data on study participants from national Swedish registers with complete coverage of public healthcare in Sweden, making our findings reliable and robust. Moreover, we assessed aortic stiffness by robust and validated methods with over 1000 peer-reviewed studies published reporting data derived from the well-validated tonometry-based SphygmoCor devices [33]. Our findings provide additional knowledge related to the potential importance of evaluating orthostatic BP adaptations at an early adult age in terms of risk assessment and in relation to markers of vascular aging. Future investigations in MOS will enable us to assess novel aspects of vascular aging, with the unique possibility of national register-based follow-up for morbidity and mortality.

However, our study has also some important limitations that need to be addressed. Study participants were predominantly young- to middle-aged individuals of white European ancestry, and therefore, the results of this study may not be generalizable to other racial/ethnic groups. Likewise, information on BP levels was obtained at a single visit, meaning that we were not able to assess the day-to-day reproducibility of our measurements. Furthermore, we could not draw any conclusions on causality from the cross-sectional design of the study, i.e., whether increased aortic stiffness leads to abnormal orthostatic BP reactions or vice versa, or if there is a common cause contributing to both aortic stiffness and abnormal orthostatic blood pressure reactions, for example, the influence of genetics or early life programming [34]. Since the MOS cohort was recently completed (end of 2021), we currently lack longitudinal data to perform prospective analyses.

The co-existence of orthostatic hypotension, increased BP variability, and arterial stiffness represents a hemodynamic aging syndrome [10] with important prognostic implications for public health. These three entities are independent risk markers for CVD, and their confluence, therefore, is of impactful significance. In addition, increasing data indicate that also orthostatic hypertension confers an increased risk of CVD [6].

We demonstrate here that even subtle, subclinical abnormalities in orthostatic BP regulation are associated with changes in central hemodynamics and arterial stiffness—a marker of early vascular aging (EVA) [35]—in a population-based cohort of predominantly young adult and middle-aged healthy subjects, whereas the prevalence of subclinical OH was low. Only individuals above the median age of 44 years showed a positive correlation between a subclinical OH diagnosis and worsened central aortic hemodynamics, i.e., aortic stiffness, in the fully adjusted model, suggesting that clinical OH is likely a marker of more advanced vascular aging [30].