The Association Between Orthostatic Symptoms and Orthostatic Hypotension: Does it Really Matter?

Orthostatic hypotension (OH) is a clinical condition arbitrarily defined by consensus as a fall in systolic/diastolic blood pressures (BPs) of 20/10 mm Hg passing from the supine to the upright position during the first 3 minutes. The observation of a BP drop during standing is a frequent condition, mainly in elderly people, associated with different pathophysiological mechanisms and with potential prognostic value in several population settings. There are not univocal data from the literature about the prognostic role of OH, mainly concerning cardiovascular risk.1 Some studies found a significant association between OH and the risk of fall,2,3 syncope,4,5 mortality and cardiovascular disease,4,6–8 or risk of dementia,9 but this association is not always confirmed.1,6,10,11 In general, there is some agreement about the fact that OH could be more a marker of frailty and underling disease leading to increased mortality or cardiovascular risk than a causal condition.4,6,7,12,13 The association between symptoms and OH has been investigated in literature. A study systematically testing for OH found that most of the patients presenting OH are asymptomatic.11 Indeed, even the presence of asymptomatic OH has been shown to be independently associated with mortality and cardiovascular disease, but again, this seems linked to the presence of underlying disease.4 Data from literature suggest that patients who experienced hospital admission for symptomatic OH had increased risk of cardiovascular disease. Indeed, these patients were more frail and with more comorbidities, indicating a higher cardiovascular risk a priori.8 The most frequent symptom associated with OH is syncope. Many patients presenting with syncope have also OH, and vice-versa. In literature, many studies found a relationship between OH and syncope, but in most of the cases it was not possible to correctly identify the real cause of syncope. For example, 1 study showed that the presence of OH was not related with the self-reported symptom of syncope.11 In another study, by Atkins et al.,5 syncope was diagnosed based on clinical and instrumental examination. Even though they found a high prevalence of OH in patients with syncope, only in few cases OH was the probable cause of the syncope. This highlights that it is not trouble-free to establish a causal relationship between OH and syncope. In fact, in literature, the prevalence of this relationship is estimated with a very wide range at 4–24%.14 Even if dizziness is considered a classical symptom related with OH, it has been shown that the prevalence of dizziness was equally distributed between OH and non-OH patients.5,15 Furthermore, it has been shown by Rutan et al. that self-reported dizziness was not associated with symptomatic OH during the study protocol.11 In particular, the prevalence of self-reported dizziness and “dizzy when stand up quickly” was 19.5% and 18.8%, respectively, while the prevalence of symptomatic OH during the examination was 2.0%.11 The study from Juraschek et al. in this issue of the Journal16 investigated the association between self-reported symptoms related to OH and the office determination of OH in a large cohort of hypertensive African Americans with chronic kidney disease attributed to hypertension (The AASK trial). After excluding patients with diabetes, urinary protein/creatinine ratio >2.5, malignant or secondary hypertension, and heart failure, they randomized patients to intensive or standard BP control (with target MAP ≤92 or 102–107, respectively). Monthly or 2 month apart, BP and heart rate (HR) measures were obtained during the follow up. At each visit, patients were asked about the presence of 3 symptoms: syncope, dizziness, and lightheadedness. Orthostatic BPs were measured after 2 minutes and 45 seconds of standing, from the seated position. Orthostatic HRs were obtained after 2 minutes of standing. BP falls greater or equal to 20/10 mm Hg (for systolic BP and diastolic BP, respectively) and HR rise higher or equal to 20 bpm were used to define the “consensus OH definition” (for either systolic BP or diastolic BP) and the “clinical OH definition” (for HR). The results highlight that OH was a frequent clinical condition with a prevalence of 5.3% (for at least 1 of the 3 OH definitions, systolic BP or diastolic BP or HR). This is in line with the estimated prevalence in general population.1 There is no information about the symptoms during the OH test at each visit. The major finding is that the consensus OH definition was associated with symptoms showing great specificity and very low sensitivity. The Authors also found that the cutoff of the consensus definition approximates well the best association with the presence of symptoms. The Authors’ conclusion is that, in the evaluation of clinical events, testing for OH is useful when positive for confirming orthostatic symptoms; at the contrary, a negative test does not rule out the presence of orthostatic symptoms. Now, even if the arguments raised by the paper are methodologically solid, it is noteworthy to remember that the relationship between OH and orthostatic symptoms (and even their implication for the clinical risk identification) has not been elucidated yet. From a pathophysiological point of view, neither orthostatic symptoms nor OH are a clinical diagnosis identifying a specific disease. They should better be considered as manifestations of underlying clinical conditions potentially leading to events. Consequently, much attention should be paid to the interpretation of the symptoms and of the positive test for OH. It has been shown by Atkins et al.5 that self-reported symptoms like syncope or dizziness are not well correlated with symptoms during testing of OH, and that the accurate clinical/instrumental determination of the causes of syncope did identify OH as the probable cause of syncope in very few cases. Consequently, it seems noteworthy that the presence of orthostatic symptoms prompts an accurate workup to identify the possible explanation. The same is valid concerning OH. In this regard, it is useful to note that a much broader classification of orthostatic changes is proposed in literature. In fact, depending mainly on the time of BP drop, 6 clinical presentations of orthostatic changes have been proposed.17 For example, when BP drops more than 20/10 mm Hg after 3 minutes of standing delayed BP recovery is diagnosed.14 This pattern could be associated with mild noradrenergic failure or impaired neurohumoral responses.17 Of note, delayed BP recovery has been related with unexplained falls and syncope and has shown a prognostic role similar to OH.14 Interestingly, the figure 1 of the paper by Juraschek16 suggests that absolute orthostatic changes in BP in either direction (increase or decrease) are associated with higher prevalence of self-reported syncope, without significant thresholds. In the graph, it appears that the usual cutoff of −20 mm Hg selects a very small population with the highest prevalence of symptoms, which justifies its very high specificity and very poor sensitivity. It is not established whether this result has clinical implication, anyway the U-shaped curve presented in the figure 1 recalls the paper from Kario et al. where they found in elderly asymptomatic hypertensive subjects in primary prevention that orthostatic BP dysregulation (in either direction) was associated to higher risk of silent cerebrovascular disease.18 Likewise, in a recent work our team demonstrated that for very old institutionalized subjects both variations of standing BP (i.e., orthostatic hypertension and hypotension) were associated to increased risk of morbidity and mortality.19 These data strongly suggest that patients presenting either orthostatic symptoms or orthostatic BP variations deserve a broad clinical and instrumental workup aiming to possibly identify the underlying pathophysiological alterations. This could also lead to personalize the treatment and clinical follow up. The author declared no conflict of interest.