![ORCID Icon]({"type":"image","src":"/templates/jsp/_style2/_tandf/images/orcid.svg"} "ORCID Icon"){kind=small}
http://orcid.org/0000-0002-4605-5344
Abstract
Abstract
Purpose: Orthostatic hypotension (OH) is a common disorder, especially among hospitalised patients. Classic OH is defined as occurring 3 or less minutes of orthostatic stress, and delayed OH as occurring after 3 min of stress. We aimed to compare clinical characteristics and prognosis between inpatients with classic vs. delayed OH.
Methods: We performed a retrospective analysis of data from 358 inpatients, aged ≥60 years, who were evaluated for the occurrence of OH at the initial phase of ambulation in four previous prospective studies in our department. Demographic, clinical and prognostic data were compared between patients with (n = 191) vs. without (n = 167) OH, classic (n = 138) vs. delayed (n = 53) OH and seated (n = 115) vs. standing (n = 76) OH.
Results: Demographic characteristics, duration of bed rest, the main reasons for admission and the use of offending medications were comparable between the delayed and classic OH groups. Mean maximal postural diastolic (p < .001) and systolic (p = .063) blood pressure falls were higher among patients with classic v. delayed OH. No statistically significant difference between the patients with classic and delayed OH were observed in the occurrence of OH-related symptoms (62.3 vs. 69.8%, p = .42). During a median follow-up of 5.5 years, no statistically significant differences in survival were observed between patients with vs. without OH (p = .14), classic vs. delayed OH (p = .68) and seated vs. standing OH (p = .067). On multivariate analysis, these variables remained not significantly associated with decreased survival.
Conclusions: Among inpatients, delayed OH is associated with a lesser magnitude of orthostatic blood pressure fall than classic OH. However, rates of symptomatic OH and long-term mortality were comparable between the groups. Thus, among hospitalised patients, delayed OH should be considered as posing the same severity as classic OH.
Introduction
Orthostatic hypotension (OH) is a common disorder. Among community-dwelling persons, the prevalence of OH ranges from 5 to 30% [1–6]. In a hospitalised patient population, the prevalence of OH is higher, with rates in the range of 22–75% [7,8]. Compared to those without OH, patients with OH have greater risk of falls, by 1.7–2.5 times [6,9,10]; cardiovascular disorders, by 1.4–2.2 times [2–6,11]; and all-cause mortality, by 1.4–1.6 times [1–6,11].
Three clinical variants of OH are currently recognised, according to the timing of OH occurrence during orthostatic stress: initial (within 15 s), classic (within 30–180 s) and delayed (after 3–45 min) [4–6,12–14]. Initial OH is diagnosed using continuous beat-to-beat finger blood pressure (BP) monitoring and cannot be assessed with routine BP equipment [14–16]. Most clinical studies of OH have focussed on classic OH. Less information is available regarding delayed OH. Delayed OH is considered a milder form of OH than classic OH and may progress to classic OH [6,13,17]. Indeed, delayed OH was reported to occur in younger individuals, and to demonstrate a lesser magnitude of orthostatic BP fall compared to classic OH [17].
Differences in characteristics between delayed and classic OH have been studied only in patients with neurogenic OH [17–22]. Neurogenic OH is chronic and caused by primary neurodegenerative disorders and secondary polyneuropathies [4–6]. Non-neurogenic OH is more common than neurogenic OH, may be acute or chronic, and results from volume depletion, venous pooling, comorbidities and medications [4–8,14]. Among elderly hospitalised patients, OH is largely acute and non-neurogenic [7,8,14]. In this patient population, the possible differences in delayed vs. classic OH have not been investigated. Thus, the main aim of this study was to compare demographic, clinical and prognostic data between inpatients, aged 60 years and older, with delayed and classic OH occurring at the initial phase of ambulation. We also aimed to compare data between patients with vs. without OH, and with seated vs. standing OH.
Methods
Study population and design
The study was carried out in accordance with the Declaration of Helsinki and was approved by the institutional Ethics Committee (approval number 0160-19-ASF). The study was designed as a retrospective single-centre comparative investigation. Included were 358 patients who participated in four prospective investigations conducted in our internal medicine department during 1999–2005 [23–26]. The subjects were unselected from patients who were consecutively admitted. Informed consent was obtained from each patient. In these studies, patients aged ≥60 years and hospitalised for a variety of acute medical disorders, were prospectively evaluated for the occurrence of OH during the initial phase of ambulation, using similar standardised protocols [23–26]. Exclusion criteria were: previously known OH, non-cooperation, hemodynamic or respiratory instability and other medical conditions that contraindicate ambulation. OH was determined once, on the same occasion: in the morning while fasting, after bed rest that lasted at least 6 h. Further, the assessment followed first mobilisation by rapid active standing [25] or passive seating by two personnel members, in an armchair with legs in the vertical position [23,24] or in bed with legs bent at the knee and hanging over the side of the bed [26]. BP and heart rate measurements were recorded by a semiautomatic device (DINAMAPTM XL vital sign monitor; Johnson & Johnson Medical Incorporation, Arlington, TX, USA) that was regularly calibrated by qualified electronic personnel according to the manufacturer’s instructions. To avoid the effect of first BP measurement stress, BP was determined in the supine position at 3-min intervals, until the differences in systolic and diastolic BP values in 3 sequential measurements did not exceed 8 and 3 mmHg, respectively. The last values measured were recorded as the supine baseline. BP, heart rate and the appearance of associated symptoms were registered at 1, 3 and 5 min following the change of posture. Appropriate measures were undertaken for the prevention of eventual falls. In patients with severe symptoms or low BP, the orthostatic test was interrupted, with immediate return to the supine position. The study protocol was completed for 356 of 358 patients. The orthostatic test was interrupted after 3 min in two patients with OH who manifested severe dizziness and low BP. No other serious adverse events (syncope, fall or life-threatening arrhythmia) were observed.
The patients included in the present study (n = 358) were categorised as: with and without OH. OH was subsequently categorised as classic vs. delayed, and as seated vs. standing.
Data collection
Demographic and clinical data were collected from patients’ research charts and files. The follow-up period ended on July 2019. Mortality data were obtained from the registry of the Internal Affairs Ministry or from hospital records. The following variables were registered: age, sex, duration of bed rest, the main reason for admission (infection, acute coronary syndrome, exacerbated heart failure, stroke and another disorder), comorbidities (hypertension, coronary artery disease, heart failure, anaemia, diabetes mellitus, renal failure, cerebrovascular disease and neurodegenerative disorder/polyneuropathy) and the use of relevant medications (diuretics, renin-angiotensin system antagonists, beta-receptor blockers, calcium antagonists, nitrates, alpha-receptor blockers and psychotropic agents). Additional data recorded were: values of heart rate, systolic BP and diastolic BP in the supine position and at 1, 3 and 5 min following the change of posture, occurrence of dizziness or palpitations at 1, 3 and 5 min following the orthostatic stress, vital status and date of death.
Definitions
According to the current guidelines, OH was defined as systolic OH or diastolic OH occurring within 5 min of bedside orthostatic stress [12–14]. Systolic OH was defined according to the following criteria: a postural drop of systolic BP ≥20 mmHg or under 90 mmHg in normotensive patients, and a systolic BP fall of ≥30 mmHg in patients with supine hypertension (systolic BP ≥160 mmHg) [12–14]. Diastolic OH was diagnosed according to a decline of ≥10 mmHg in diastolic BP during orthostatic stress [12–14]. According to the relation of OH to different parts of the cardiac cycle, the following definitions were used: isolated systolic OH (systolic OH without diastolic OH), isolated diastolic OH (diastolic OH without systolic OH) and combined OH (concomitant systolic and diastolic OH). Classic OH was defined as OH determined within 3 min of an orthostatic test [12–14]. A diagnosis of delayed OH was based on the occurrence of OH only after 5 min of postural change [12–14]. Standing OH and seated OH were diagnosed on lying-to-standing and lying-to-sitting tests, respectively, according to the aforementioned postural drops of BP values [14]. Impaired consciousness, fainting, light-headedness, and temporary visual or auditory disturbances were included in the definition of dizziness.
A diagnosis of hypertension was based on data from previous medical records and current hospitalisation, or the use of anti-hypertensive medications. Diabetes mellitus was diagnosed according to persistent fasting hyperglycaemia (≥126 mg/dl) or chronic anti-hyperglycemic therapy. The World Health Organisation criteria were used for the definition of anaemia: a haemoglobin concentration of <12 g/dl in women and <13 g/dl in men. Renal failure was defined as any value of estimated glomerular filtration rate <60 ml/min/1.73m2 during the current hospitalisation. Coronary artery disease, heart failure, cerebrovascular disease and neurodegenerative disorder/polyneuropathy were diagnosed according to data from the present hospitalisation or previous medical records.
Statistical analysis
The results were expressed as means and standard deviations for quantitative data and as numbers (percentages) of presented cases for qualitative data. Statistical comparisons were performed between the data obtained for groups of patients with vs. without OH, with classic vs. delayed OH and with seated vs. standing OH. Categorical variables were compared using Pearson’s chi-square or Fisher’s exact test. Analysis of variance (ANOVA) was applied for comparison of continuous variables. Variables without normal distributions were analysed using square root transformation. ANOVA with repeated measures was used to determine changes over time of continuous variables. Survival estimates were provided using the Kaplan–Meier method. Differences between the curves were evaluated by Mantel–Cox and Breslow tests. Two-sided p values <.05 were considered to indicate statistical significance. Variables that were found to be associated with decreased survival on univariate analysis were re-evaluated by a Cox proportional-hazards model to determine those most significantly associated with poor survival. For the variables most significantly associated with decreased survival, hazard ratios (HR) and 95% confidence intervals (CI) were calculated. The statistical analysis was performed using the Biomedical Package software [27].
Results
Baseline characteristics and orthostatic test data
The entire sample
In the entire cohort (n = 358), the mean age was 75.9 ± 8 years (range 60–97 years) and 58.1% were women. The most common reasons for admission were infection and decompensated heart failure. OH was diagnosed in 191 (53.4%) patients. OH was evaluated on lying-to-sitting in 237 (66.2%) patients and on lying-to standing in 121 (33.8%) patients. OH-associated symptoms occurred in 197 (55.0%) patients.
Comparisons between patients with and without OH
Demographic characteristics, the main reasons for admission and comorbidities were comparable between patients with vs. without OH. Duration of bed rest prior to the orthostatic test was longer in the OH group. Patients in the OH group were more often treated with alpha-receptor blockers (Table 1).
Table 1. Comparison of baseline characteristics between patients with and without OH.
Compared to patients with OH, patients without OH were more likely to be evaluated on lying-to-sitting than on lying-to-standing (Table 2). Comparing the former to the latter, the cumulative rate of occurrence of OH-related symptoms, and the mean supine values of heart rate and BP were higher. Moreover, during orthostatic stress, the mean heart rate was higher and the mean level of BP was lower in patients demonstrating OH (Table 2).
Table 2. Comparison of orthostatic test data between patients with and without OH.
Comparisons between patients with delayed and classic OH
Among patients with OH, 53 (27.7%) had delayed OH and 138 (72.3%) had classic OH. Demographic characteristics, duration of bed rest, the main reasons for admission and the use of offending medications were comparable between the two groups (Table 3). Hypertension was more common in the classic OH group, while primary neurodegenerative disorders/secondary polyneuropathies were more frequently observed in the delayed OH group (Table 3).
Table 3. Comparison of baseline characteristics between patients with delayed and classic OH.
Isolated systolic OH occurred more frequently with delayed than classic OH (p = .002), while combined OH prevailed in the classic OH group (p = .001, Table 4). The rates of isolated diastolic OH, seated OH and standing OH were similar in the two groups. The occurrence of OH symptoms during the orthostatic test was common and comparable between the groups. However, following 5 min of the test, patients with delayed OH more often complained of dizziness than did those with classic OH (p = .007, Table 4)). Compared to patients with asymptomatic OH, patients with symptomatic OH demonstrated a higher mean maximal heart rate (92.8 ± 17 vs. 82.6. ±18 beats/min, p = .001) during the orthostatic test, and tended to demonstrate lower mean nadir values of systolic (124.2 ± 27 vs. 131.8 ± 22 mmHg, p = .072) and diastolic (60.0 ± 13 vs. 63.7 ± 16 mmHg, p = .10) BP. The respective magnitudes of systolic (22.7 ± 14 vs. 19.9 ± 19 mmHg, p = .24) and diastolic (14.6 ± 9 vs. 14.0 ± 8 mmHg, p = .67) BP fall were comparable in patients with symptomatic and asymptomatic OH.
No statistically significant difference was observed between the delayed and classic OH groups in the mean values of heart rate. In the supine position (p = .031) and following 5 min of postural change (p = .007), mean systolic BP values were significantly higher in the classic than delayed OH group. Moreover, in the classic OH group, the mean values of diastolic BP were significantly lower following 1 min (p = .001) and 3 min (p = .004) of orthostatic stress, compared to the delayed OH group. Patients with classic OH demonstrated a higher magnitude of diastolic BP drop after the change in posture (p< .001) and tended to manifest a higher magnitude of postural systolic BP fall (p = .063). Among the patients with delayed OH, compared to supine position, the mean systolic and diastolic BP levels following 1 and 3 min of orthostatic test remained practically stable, with abrupt decline in BP following 5 min of the test. This observation contrasts with the persistent postural decline in the mean values of systolic and diastolic BP among the patients with classic OH (Figure 1).
Published online:
04 March 2020Figure 1. Mean values with standard errors of the mean (SEM) for systolic and diastolic BP prior to and 1, 3 and 5 min following change in posture among hospitalised patients with delayed and classic OH. Systolic and diastolic BP levels differed significantly between the groups over time (by ANOVA with repeated measures). Abbreviations: BP: blood pressure; OH: orthostatic hypotension; ANOVA: analysis of variance.
![]({"type":"image","src":"/na101/home/literatum/publisher/tandf/journals/content/iblo20/2020/iblo20.v029.i04/08037051.2020.1733389/20200803/images/medium/iblo_a_1733389_f0001_b.jpg"})
Figure 1. Mean values with standard errors of the mean (SEM) for systolic and diastolic BP prior to and 1, 3 and 5 min following change in posture among hospitalised patients with delayed and classic OH. Systolic and diastolic BP levels differed significantly between the groups over time (by ANOVA with repeated measures). Abbreviations: BP: blood pressure; OH: orthostatic hypotension; ANOVA: analysis of variance.
Table 4. Comparison of orthostatic test data between patients with delayed and classic OH.
Comparisons between patients with seated and standing OH
In the OH group, seated OH and standing OH were diagnosed in 115 (60.2%) and 76 (39.8%) patients, respectively. Patients who were evaluated on lying-to-standing were more likely to demonstrate OH than those investigated on lying-to sitting (the respective rates of standing and seated OH were 62.8 and 48.5%, p = .010). Compared to standing OH, seated OH was associated with female gender and longer bed rest. Patients with seated OH were more often admitted for exacerbated heart failure, presented with renal failure, and treated with diuretics and psychotropic agents than those with standing OH (Table S1).
The rates of combined OH and symptomatic OH were significantly higher among patients with standing than seated OH. The rates of isolated systolic and diastolic OH, and of classic and delayed OH, were comparable in patients with seated vs. standing OH. No significant differences were observed between patients with seated and standing OH, regarding the mean values of heart rate and BP (Table S2).
Survival
The entire sample
During a median follow-up period of 5.5 years, 292 (81.6%) patients died. The mean survival duration, and 1-year and 5-year mortality rates were 7.2 ± 6 years, 21.2 and 47.8%, respectively. On univariate analysis, no statistically significant difference in survival was observed between patients with vs. without OH (Figure 2(A), p = .14); the respective mean survival durations, and 1-year, 5-year and total mortality rates were 7.6 ± 7 vs. 6.8 ± 6 years (p = .21), 24.6 vs. 17.4% (p = .12), 44.5 vs. 51.5% (p = .20) and 78.0 vs. 85.6% (p = .076). For the entire cohort, variables associated with decreased long-term survival were: older age; longer bed rest; OH testing by lying-to-sitting; infection, acute coronary syndrome and exacerbated heart failure as a reason for hospitalisation; and the following comorbidities: chronic heart failure, coronary artery disease, diabetes mellitus, anaemia and renal failure. These variables were re-evaluated by a Cox proportional-hazards model, together with gender and the occurrence of OH. The results showed that the following variables were most significantly associated with shorter survival: age (for each 10-year increment, HR 1.48, 95% CI 1.27–1.73, p < .001), chronic heart failure (HR 1.71, 95% CI 1.33–2.21, p < .001), diabetes mellitus (HR 1.38, 95% CI 1.09–1.75, p = .004), renal failure (HR 1.36, 95% CI 1.06–1.76, p = .021), infection (HR 1.34, 95% CI 1.02–1.77, p = .029) and bed rest (for each 12-h increment, HR 1.03, 95% CI 1.00–1.06, p = .049).
Published online:
04 March 2020Figure 2. (A, B) Kaplan–Meier estimates for the various study groups. (A) Association between OH and survival. (B) Association between classic vs. delayed OH and survival. Abbreviation: OH: orthostatic hypotension.
![]({"type":"image","src":"/na101/home/literatum/publisher/tandf/journals/content/iblo20/2020/iblo20.v029.i04/08037051.2020.1733389/20200803/images/medium/iblo_a_1733389_f0002_c.jpg"})
Figure 2. (A, B) Kaplan–Meier estimates for the various study groups. (A) Association between OH and survival. (B) Association between classic vs. delayed OH and survival. Abbreviation: OH: orthostatic hypotension.
OH group
The Kaplan–Meier estimates of survival showed no statistically significant differences between patients with classic and delayed OH (Figure 2(B), p = .68). The respective mean survival durations, and 1-year, 5-year and total mortality rates were: 7.8 ± 7 vs. 7.2 ± 7 years (p = .58), 23.9 vs. 26.4% (p = .71), 43.5 vs. 47.2% (p = .75) and 77.5 vs. 79.2% (p = .85).
Seated OH tended to be associated with poorer long-term survival than standing OH (p = .067). The mean survival durations, and 1-year, 5-year and total mortality rates were 7.0 ± 7 vs. 8.5 ± 6 years (p = .16), 30.4 vs. 15.8% (p = .021), 50.4 vs. 36.9% (p = .083) and 79.1 vs. 76.3% (p = .72), respectively. Survival did not differ significantly between patients with isolated systolic OH, isolated diastolic OH and combined OH (p = .77). Patients with symptomatic and asymptomatic OH had comparable survival curves (p = .91).
On multivariate analysis, classic vs. delayed OH and seated vs. standing OH were not among the variables that most significantly associated with decreased survival in the OH group. Predictors of survival were the same as those identified in the entire sample: older age (for each 10-year increment, HR 1.50, 95% CI 1.21–1.85, p < .001), renal failure (HR 1.42, 95% CI 1.00–2.01, p = .005), bed rest (for each 12-h increment, HR 1.04, 95% CI 1.00–1.07, p = .017), diabetes mellitus (HR 1.37, 95% CI 0.98–1.92, p = .032) and chronic heart failure (HR 1.53, 95% CI 1.07–2.19, p = .054).
Discussion
We analysed demographic, clinical and prognostic data of 358 patients hospitalised in an internal medicine ward, according to the development of OH and according to various forms of OH, diagnosed at a bedside orthostatic test at the initial phase of ambulation. The possible differences in clinical characteristics and prognostic significance between delayed and classic OH in this context have not been reported. Among hospitalised patients, OH is largely acute and non-neurogenic (functional), and relates to hypovolemia, venous pooling, comorbidities and treatment with certain medications [7,8,14]. Indeed, only 5.2% of our OH patients presented with chronic neurodegenerative disorders. Moreover, the development of OH was associated with longer bed rest and treatment with alpha-receptor blockers. Previous studies comprising patients with delayed OH focussed on population with chronic neurogenic (structural) OH, as diagnosed in a tilt laboratory [17–21]. A number of factors have been proposed as promoting the development of delayed OH. One factor is increased peripheral pooling during prolonged orthostatic stress. This results from increased transudation of fluid to the interstitial space and continuous relaxation of the low body capacitance vessels. A second factor is decreased total peripheral resistance, related to gradual impairment of neural and humoral compensatory mechanisms [17,18]. Consequently, during prolonged orthostatic stress, venous return to the heart and cardiac filling pressure are decreased, thus delaying the decline in cardiac output and BP [17,18].
We found that delayed OH was less common than classic OH (27.7 vs. 72.3%). A varying prevalence of delayed OH has been reported among patients with neurogenic OH, in the range of 16.3–54.0% [17,19–22]. The relatively lower rate of delayed OH reported in the present study is probably related to the evaluation of delayed OH during 5 min only. Extending the duration of the orthostatic test may have diagnosed additional persons with delayed OH.
In our cohort, gender and age distribution did not differ significantly between patients with delayed and classic OH. This finding concurs with that of another study [22]. However, others showed an association of delayed OH with younger age [17,19] and female sex [19]. The discrepancies between the studies may be explained by differences in study design and populations. Specifically, our study comprised acutely ill inpatients aged 60 or more years, while the studies of Gurevich et al. [19], and Gibbons and Freeman [17] were conducted in the ambulatory setting and included younger patients. In addition, in the present report and in the study of Byun et al. [22], the period of delayed OH assessment was limited to 5 and 10 min, respectively. This contrasts to the evaluation of OH for up to 40–45 min in two other studies [17,19]. Defining of delayed OH by a longer period would evidently have identified more cases in the current cohort. This could have affected comparisons of patient characteristics between those with classic vs. delayed OH.
A novel aspect of the present investigation is the detailed evaluation of clinical characteristics associated with delayed OH. The patients with delayed and classic OH were similar: in durations of bed rest prior to the orthostatic test, the reasons for hospitalisation and their receiving medications that affect OH. Of the relevant comorbidities examined, differences between the groups were found only for hypertension and neurodegenerative disorders. The former was more common in the classic OH group and the latter in the delayed OH group. Moreover, the mean supine systolic BP values were significantly higher in the classic than delayed OH group. Similar results were reported in one study [17]. However, two other investigations did not show a significant difference in supine BP levels between persons with classic and. delayed OH [19,22].
Another interesting finding of this study is the difference in prevalence of various OH modalities between patients with delayed and classic OH. Specifically, isolated systolic OH was more common in delayed OH, combined (systolic-diastolic) OH prevailed in classic OH, and the prevalence of isolated diastolic OH was comparable between the groups. To the best of our knowledge, this is the first report that compares patients with delayed and classic OH according to the relation of OH to different parts of the cardiac cycle. A more frequent appearance of combined OH among patients with classic than delayed OH may be explained by more severe impairment of vasoconstrictor compensatory mechanisms in the former, as has been previously reported [17].
We confirmed the supposition that patients with delayed and classic OH demonstrate two relatively distinct patterns of orthostatic response. Thus, in the delayed OH group, BP levels remained practically stable until an abrupt decline in BP occurred 5 min after the change of posture. In contrast, in the classic OH group, BP decreased persistently over time. We also confirmed data from previous studies on patients with chronic neurogenic OH, which demonstrated an orthostatic BP fall of lesser magnitude in patients with delayed than classic OH [17,20,22]. A novel observation in the present study is the common occurrence of OH-related symptoms, at similar rates in the delayed and classic OH groups (69.8 and 62.3%, respectively). This is despite the difference between the groups in maximal orthostatic BP drop. Possibly, in some of our patients, the appearance of dizziness or palpitations could be related to the nature of the acute illness and deconditioning rather than to a postural BP fall. Moreover, the occurrence of OH symptoms may be more affected by the individual hemodynamic response to orthostatic stress and by the absolute BP level than by the magnitude of the BP drop [5,8,13]. Indeed, among our OH patients, the occurrence of symptoms was associated with higher peak values of heart rate and lower levels of nadir BP, rather than with a greater BP drop.
An important novel aspect of the present investigation is the comparison of long-term survival between the various groups of patients with OH. During a median follow-up period of 5.5 years, long-term survival, and 1-year and 5-year mortality rates were comparable between patients with and without OH. Our results contrast with the data of large prospective longitudinal studies that demonstrated a strong association of OH with increased risk of death in general populations [1–6,11]. Solid information regarding the prognostic significance of OH in hospitalised patient populations is scarce. In a single available study that comprised 471 elderly inpatients, with a mean follow-up of 3.5 years, OH diagnosed before discharge was not associated with an increased risk of mortality [28]. Our findings concur with that study. Several reasons may explain the discrepancies between studies on OH performed in ambulatory and hospital settings. First, in an inpatient population, mortality risk may be more affected by older age, deconditioning from the severity of an acute illness, prolonged bed rest and comorbidities, rather than by OH. Indeed, advanced age, longer bed rest, admission for infectious disease and the number of comorbid disorders (chronic heart failure, diabetes mellitus and renal failure) were the factors demonstrating the strongest associations with mortality in the present study. Second, OH in hospitalised patients is often acute, highly variable and potentially reversible [7,8]. Mobilisation following prolonged bed rest and recovering from acute disease may result in the disappearance of OH after discharge from the hospital, and could thus affect the association of OH with long-term survival [28].
The most important novelty of the present study is the comparison of prognosis between patients with delayed vs. classic OH. We found comparable mean survival durations between those with delayed and classic OH (7.2 ± 7 vs. 7.8 ± 7 years, respectively). Our findings contrast with the results of a single available study on this issue [20]. In that investigation, Gibbons and Freeman evaluated 10-year all-cause mortality in patients with chronic neurogenic OH and found significantly lower mortality rates in 48 patients with delayed OH than in 42 patients with classic OH: 29 and 64%, respectively [20]. The discrepancy between their study and ours may have several explanations. First, in their study, OH was diagnosed in community-dwelling individuals with suspected orthostatic intolerance. In the present investigation, OH was diagnosed during the initial phase of ambulation of hospitalised patients with no history of prior OH. Second, in the former study, OH was determined by head-up tilt table testing, while we evaluated OH at a bedside orthostatic test. Moreover, for a considerable proportion of our patients, OH was evaluated on supine-to-sitting; this reflects the clinical reality of in-hospital patients. Patients with seated OH had more severe clinical characteristics than did those with standing OH. Thus, more frail patients were less likely examined for standing OH. However, the severity of orthostatic stress is not equal on standing and sitting. In fact, in the present study, the rate of OH occurrence was significantly higher among patients evaluated on lying-to-standing (62.8%) than in those investigated on lying-to-sitting (48.5%). The abovementioned difference in the means of OH testing could have affected the study results. Third, our patients were older (mean age 75.9 ± 8 years) than those in the former study (59.0 ± 11 years). Fourth, the period of delayed OH determination was longer in the earlier report (45 min) than in the present study (5 min). Taken together, the aforementioned differences in study design and population, and also in the pathophysiology of neurogenic and non-neurogenic delayed OH, may explain the difference in survival observed between the studies. Notably, our patients with either classic or delayed OH demonstrated similar long-term prognosis yet higher rates of symptoms than those without OH. Consistent with its acute and transient nature, non-neurogenic OH (whether classic or delayed) among in-patients may resolve upon hospital discharge and does not appear to influence long-term mortality. Thus, we suggest that in-hospital OH should be acutely managed for reducing OH symptoms, generally by non-pharmacologic measures, following repeated evaluation of OH in the community setting [7,8].
Our study has a number of limitations. First, the defined clinical setting of a single medical centre may limit its generalizability to other populations. Second, the study design was retrospective. However, the prospective evaluation of OH in the study participants mitigates the possible selection bias. Third, the number of patients with delayed OH was relatively small and this may affect statistical power for comparison of some relevant data. Still, our study represents one of the largest investigations that have been conducted on patients with delayed OH. Fourth, initial OH, which is a distinctive form of OH, was not determined. However, initial OH should be evaluated in two ways: lying-to-standing and standing-up from a squatting position [14,15]. These means of OH testing were inappropriate for a considerable proportion of our patients. Moreover, initial OH may be diagnosed only using non-invasive continuous beat-to-beat finger BP monitoring [14–16]. However, during the period of the study, this technique was considered experimental and available only in research laboratories. Fifth, the reporting of symptoms was quite narrow and included only palpitations and dizziness, with its variations. Occurrences of other common OH symptoms (coat-hanger pain and angina pectoris) were not registered. In addition, some relevant data, such as the prevalence of dementia and manifestations of autonomic failure other than OH (sweating abnormalities and gastrointestinal, erectile and bladder dysfunction), were not recorded. Finally, the major limitations of the study are the evaluation of OH once only, and the determination of delayed OH at 5 min following the change of posture. Additional cases of OH would presumably have been identified had orthostatic tests been repeated. Moreover, the extension of duration of orthostatic stress beyond 5 min would evidently increase the incidence of delayed OH. The abovementioned limitations may have influenced the study results. Thus, the impact of delayed OH on the occurrence of symptoms, magnitudes of BP fall and mortality cannot be completely ascertained from the present report and should be investigated in future prospective studies.
Conclusions
This is the first published report to compare clinical characteristics and prognosis among hospitalised patients with delayed vs. classic OH. Patients with delayed OH demonstrated lower supine systolic BP levels and an orthostatic BP fall of lesser magnitude than patients with classic OH. However, no significant difference was found between the groups in frequency of orthostatic symptoms and in long-term mortality. Thus, in the studied population, delayed OH should not be considered as milder than the classic form of OH.
| Variable | Patients with OH (n = 191) | Patients without OH (n = 167) | p value |
|---|---|---|---|
| Age (years) | 75.9 ± 8 | 75.9 ± 8 | .99 |
| Female gender | 110 (57.6%) | 98 (58.7%) | .91 |
| Bed rest duration (h) | 52.9 ± 64 | 21.4 ± 29 | <.001 |
| Main reasons for admission | |||
| Infection | 50 (26.2%) | 40 (24.0%) | .71 |
| Acute coronary syndrome | 38 (19.9%) | 39 (23.4%) | .44 |
| Exacerbation of heart failure | 53 (27.7%) | 47 (28.1%) | .93 |
| Stroke | 29 (15.2%) | 18 (10.8%) | .27 |
| Other | 21 (11.0%) | 23 (13.8%) | .52 |
| Comorbid diseases | |||
| Hypertension | 158 (82.7%) | 137 (82.0%) | .89 |
| Heart failure | 99 (51.8%) | 89 (53.3%) | .83 |
| Coronary artery disease | 108 (56.5%) | 103 (61.7%) | .33 |
| Diabetes mellitus | 81 (42.4%) | 77 (46.1%) | .52 |
| Anaemia | 92 (48.2%) | 78 (46.7%) | .83 |
| Renal failure | 67 (35.1%) | 50 (29.9%) | .31 |
| Cerebrovascular disease | 72 (37.7%) | 50 (29.9%) | .15 |
| Neurodegenerative disorder/polyneuropathy | 10 (5.2%) | 8 (4.8%) | .99 |
| Medications used | |||
| Diuretics | 116 (60.7%) | 113 (67.7%) | .19 |
| Renin-angiotensin system antagonists | 99 (51.8%) | 90 (53.9%) | .75 |
| Beta-receptor blockers | 89 (46.6%) | 81 (48.5%) | .75 |
| Calcium antagonists | 66 (34.6%) | 57 (34.1%) | .99 |
| Alpha-receptor blockers | 44 (23.0%) | 20 (12.0%) | .008 |
| Nitrates | 64 (33.5%) | 60 (35.9%) | .66 |
| Psychotropic agents | 37 (19.4%) | 42 (25.2%) | .27 |
Data are expressed as means ± standard deviations or numbers (percentages) of presented cases. The bold entries in the table indicate p value of <.05.
| Variable | Patients with OH (n = 191) | Patients without OH (n = 167) | p value |
|---|---|---|---|
| OH evaluation | |||
| Lying-to-sitting/lying-to-standing | 115 (60.2%)/76 (39.8%) | 122 (73.1%)/45 (26.9%) | .014 |
| Occurrence of dizziness | |||
| Following 1 min | 66 (34.6%) | 38 (22.8%) | .015 |
| Following 3 min | 75 (39.3%) | 33 (19.8%) | <.001 |
| Following 5 min | 81 (42.4%) | 29 (17.4%) | <.001 |
| Following 1 − 5 min | 110 (57.6%) | 57 (34.1%) | <.001 |
| Occurrence of palpitations | |||
| Following 1 min | 44 (23.0%) | 24 (14.4%) | .043 |
| Following 3 min | 63 (33.0%) | 18 (10.8%) | <.001 |
| Following 5 min | 61 (31.9%) | 11 (6.6%) | <.001 |
| Following 1 − 5 min | 88 (46.1%) | 31 (18.6%) | <.001 |
| Occurrence of dizziness or palpitations | |||
| Following 1 min | 77 (40.3%) | 54 (32.3%) | .12 |
| Following 3 min | 97 (50.8%) | 44 (26.3%) | <.001 |
| Following 5 min | 96 (50.3%) | 34 (20.4%) | <.001 |
| Following 1 − 5 min | 123 (64.4%) | 74 (44.3%) | <.001 |
| Heart rate (beats/min) | |||
| Supine | 78.4 ± 16 | 73.6 ± 14 | .003 |
| Following 1 min | 84.5 ± 17 | 77.7 ± 15 | <.001 |
| Following 3 min | 84.6 ± 17 | 77.1 ± 14 | <.001 |
| Following 5 min | 85.7 ± 18 | 76.5 ± 15 | <.001 |
| Systolic BP (mmHg) | |||
| Supine | 149.5 ± 27 | 144.6 ± 25 | .072 |
| Following 1 min | 140.9 ± 28 | 151.5 ± 26 | <.001 |
| Following 3 min | 138.5 ± 26 | 149.4 ± 25 | <.001 |
| Following 5 min | 133.3 ± 26 | 147.3 ± 24 | <.001 |
| Maximal BP fall following 1 − 5 min | 22.6 ± 16 | 1.1 ± 13 | <.001 |
| Diastolic BP (mmHg) | |||
| Supine | 76.1 ± 15 | 70.8 ± 13 | <.001 |
| Following 1 min | 71.6 ± 16 | 76.6 ± 14 | .002 |
| Following 3 min | 68.4 ± 14 | 76.0 ± 14 | <.001 |
| Following 5 min | 66.5 ± 15 | 74.8 ± 14 | <.001 |
| Maximal BP fall following 1 − 5 min | 14.3 ± 8 | −0.7 ± 8 | <.001 |
Data are expressed as means ± standard deviations or numbers (percentages) of presented cases. OH: orthostatic hypotension; BP: blood pressure. The bold entries in the table indicate p value of <0.05.
| Variable | Delayed OH group (n = 53) | Classic OH group (n = 138) | p value |
|---|---|---|---|
| Age (years) | 75.7 ± 8 | 76.0 ± 8 | .7 |
| Female gender | 29 (54.7%) | 81 (58.7%) | .63 |
| Bed rest duration (h) | 64.3 ± 66 | 48.5 ± 63 | .087 |
| Main reasons for admission | |||
| Infection | 19 (35.8%) | 31 (22.5%) | .068 |
| Acute coronary syndrome | 10 (18.9%) | 28 (20.3%) | .94 |
| Exacerbation of heart failure | 12 (22.6%) | 41 (29.7%) | .37 |
| Stroke | 9 (17.0%) | 20 (14.5%) | .66 |
| Other | 3 (5.7%) | 18 (13.0%) | .20 |
| Comorbid diseases | |||
| Hypertension | 36 (67.9%) | 122 (88.4%) | .002 |
| Heart failure | 27 (50.9%) | 72 (52.2%) | .96 |
| Coronary artery disease | 30 (56.6%) | 78 (56.5%) | .99 |
| Diabetes mellitus | 21 (39.6%) | 60 (43.5%) | .74 |
| Anaemia | 22 (41.5%) | 70 (50.7%) | .26 |
| Renal failure | 15 (28.3%) | 52 (37.7%) | .24 |
| Cerebrovascular disease | 21 (39.6%) | 51 (37.0%) | .74 |
| Neurodegenerative disorder/polyneuropathy | 7 (13.2%) | 3 (2.2%) | .005 |
| Medications used | |||
| Diuretics | 31 (58.5%) | 85 (61.6%) | .74 |
| Renin-angiotensin system antagonists | 23 (43.4%) | 76 (55.1%) | .20 |
| Beta-receptor blockers | 23 (43.4%) | 66 (47.8%) | .63 |
| Calcium antagonists | 19 (35.8%) | 47 (34.1%) | .87 |
| Alpha-receptor blockers | 12 (22.6%) | 32 (23.2%) | .98 |
| Nitrates | 21 (39.6%) | 43 (31.2%) | .31 |
| Psychotropic agents | 10 (18.9%) | 27 (19.6%) | .94 |
Data are expressed as means ± standard deviations or numbers (percentages) of presented cases. OH: orthostatic hypotension; the bold entries in the table indicate p value of <.05.
| Variable | Delayed OH group (n = 53) | Classic OH group (n = 138) | p value |
|---|---|---|---|
| Isolated systolic OH | 16 (30.2%) | 15 (10.9%) | .002 |
| Isolated diastolic OH | 24 (45.3%) | 56 (40.6%) | .62 |
| Combined (systolic and diastolic) OH | 13 (24.5%) | 67 (48.6%) | .001 |
| Seated OH/standing OH | 30 (56.6%)/23 (43.4%) | 85 (61.6%)/53 (38.4%) | .62 |
| Occurrence of dizziness | |||
| Following 1 min | 17 (32.1%) | 49 (35.5%) | .74 |
| Following 3 min | 18 (34.0%) | 57 (41.3%) | .41 |
| Following 5 min | 31 (58.5%) | 50 (36.2%) | .007 |
| Following 1 − 5 min | 33 (62.3%) | 77 (55.8%) | .42 |
| Occurrence of palpitations | |||
| Following 1 min | 11 (20.8%) | 33 (23.9%) | .71 |
| Following 3 min | 13 (24.5%) | 50 (36.2%) | .17 |
| Following 5 min | 21 (39.6%) | 40 (29.0%) | .23 |
| Following 1 − 5 min | 27 (50.9%) | 61 (44.2%) | .42 |
| Occurrence of dizziness or palpitations | |||
| Following 1 min | 19 (35.8%) | 58 (42.0%) | .51 |
| Following 3 min | 24 (45.3%) | 73 (52.9%) | .42 |
| Following 5 min | 36 (67.9%) | 60 (43.5%) | .004 |
| Following 1 − 5 min | 37 (69.8%) | 86 (62.3%) | .40 |
| Heart rate (beats/min) | |||
| Supine | 78.7 ± 17 | 78.3 ± 16 | .86 |
| Following 1 min | 83.2 ± 17 | 85.0 ± 17 | .51 |
| Following 3 min | 83.2 ± 17 | 85.1 ± 18 | .50 |
| Following 5 min | 85.9 ± 20 | 85.7 ± 17 | .94 |
| Systolic BP (mmHg) | |||
| Supine | 142.8 ± 30 | 152.1 ± 25 | .031 |
| Following 1 min | 146.0 ± 29 | 138.9 ± 28 | .12 |
| Following 3 min | 142.4 ± 25 | 136.9 ± 26 | .19 |
| Following 5 min | 125.2 ± 28 | 136.5 ± 25 | .007 |
| Maximal BP fall following 1 − 5 min | 18.7 ± 13 | 23.5 ± 17 | .063 |
| Diastolic BP (mmHg) | |||
| Supine | 73.2 ± 15 | 77.3 ± 15 | .084 |
| Following 1 min | 77.6 ± 15 | 69.3 ± 16 | .001 |
| Following 3 min | 73.3 ± 15 | 66.6 ± 14 | .004 |
| Following 5 min | 63.6 ± 17 | 67.7 ± 14 | .081 |
| Maximal BP fall following 1 − 5 min | 10.3 ± 5 | 16.0 ± 9 | <.001 |
Data are expressed as means ± standard deviations or numbers (percentages) of presented cases. OH: orthostatic hypotension; BP: blood pressure. The bold entries in the table indicate p value of <.05.
Supplemental_Material_for_BP_Delayed_OH_Study.docx
Download MS Word (18 KB)Disclosure statement
There are no potential conflicts of interest.
Data availability statement
The data that support the findings of this study are available from the corresponding author, [O.G.], upon reasonable request.
References
- Rose KM, Eigenbrodt ML, Biga RL, et al. Orthostatic hypotension predicts mortality in middle-aged adults: The Atherosclerosis Risk In Communities (ARIC) Study. Circulation. 2006;114(7):630–636. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Fedorowski A, Stavenow L, Hedblad B, et al. Orthostatic hypotension predicts all-cause mortality and coronary events in middle-aged individuals (The Malmö Preventive Project). Eur Heart J. 2010;31(1):85–91. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Chou RH, Liu CJ, Chao TF, et al. Association between orthostatic hypotension, mortality, and cardiovascular disease in Asians. Int J Cardiol. 2015;195:40–44. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Fedorowski A, Melander O. Syndromes of orthostatic intolerance: a hidden danger. J Intern Med. 2013;273(4):322–335. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Ricci F, De Caterina R, Fedorowski A. Orthostatic hypotension: epidemiology, prognosis, and treatment. J Am Coll Cardiol. 2015;66(7):848–860. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Freeman R, Abuzinadah AR, Gibbons C, et al. Orthostatic hypotension: JACC state-of-the-art review. J Am Coll Cardiol. 2018;72(11):1294–1309. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Feldstein C, Weder AB. Orthostatic hypotension: a common, serious and underrecognized problem in hospitalized patients. J Am Soc Hypertens. 2012;6(1):27–39. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Tzur I, Izhakian S, Gorelik O. Orthostatic hypotension in internal medicine wards. Curr Med Res Opin. 2019;35(6):947–955. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
- Finucane C, O'Connell MDL, Donoghue O, et al. Impaired orthostatic blood pressure recovery is associated with unexplained and injurious falls. J Am Geriatr Soc. 2017;65(3):474–482. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Mol A, Bui Hoang PTS, Sharmin S, et al. Orthostatic hypotension and falls in older adults: a systematic review and meta-analysis. J Am Med Dir Assoc. 2019;20(5):589–597. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Ricci F, Fedorowski A, Radico F, et al. Cardiovascular morbidity and mortality related to orthostatic hypotension: a meta-analysis of prospective observational studies. Eur Heart J. 2015;36(25):1609–1617. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Freeman R, Wieling W, Axelrod FB, et al. Consensus statement on the definition of orthostatic hypotension, neurally mediated syncope and the postural tachycardia syndrome. Clin Auton Res. 2011;21(2):69–72. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Brignole M, Moya A, de Lange FJ, ESC Scientific Document Group, et al. Practical Instructions for the 2018 ESC Guidelines for the diagnosis and management of syncope. Eur Heart J. 2018;39(21):e43–e80. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Tzur I, Izhakian S, Gorelik O. Orthostatic hypotension: definition, classification and evaluation. Blood Press. 2019;28(3):146–156. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
- Wieling W, Krediet CT, van Dijk N, et al. Initial orthostatic hypotension: review of a forgotten condition. Clin Sci (Lond). 2007;112(3):157–165. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Van Wijnen VK, Finucane C, Harms MPM, et al. Noninvasive beat- to-beat finger arterial pressure monitoring during orthostasis: a comprehensive review of normal and abnormal responses at different ages. J Intern Med. 2017;282(6):468–483. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Gibbons CH, Freeman R. Delayed orthostatic hypotension: a frequent cause of orthostatic intolerance. Neurology. 2006;67(1):28–32. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Podoleanu C, Maggi R, Oddone D, et al. The hemodynamic pattern of the syndrome of delayed orthostatic hypotension. J Interv Card Electrophysiol. 2009;26(2):143–149. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Gurevich T, Machmid H, Klepikov D, et al. Head-up tilt testing for detecting orthostatic hypotension: how long do we need to wait? Neuroepidemiology. 2014;43(3-4):239–243. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Gibbons CH, Freeman R. Clinical implications of delayed orthostatic hypotension: a 10-year follow-up study. Neurology. 2015;85(16):1362–1367. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Roy AG, Gopinath S, Kumar S, et al. Delayed orthostatic hypotension: a pilot study from India. Ann Indian Acad Neurol. 2017;20(3):248–251. [PubMed], [Web of Science ®], [Google Scholar]
- Byun JI, Moon J, Kim DY, et al. Delayed orthostatic hypotension: severity of clinical symptoms and response to medical treatment. Auton Neurosci. 2018;213:81–85. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Cohen N, Gorelik O, Fishlev G, et al. Seated postural hypotension is common among older inpatients. Clin Auton Res. 2003;13(6):447–449. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Gorelik O, Fishlev G, Almoznino-Sarafian D, et al. Lower limb compression bandaging is effective in preventing signs and symptoms of seating-induced postural hypotension. Cardiology. 2004;102(4):177–183. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Gorelik O, Fishlev G, Litvinov V, et al. First morning standing up may be risky in acutely ill older inpatients. Blood Press. 2005;14(3):139–143. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
- Gorelik O, Almoznino-Sarafian D, Litvinov V, et al. Seating-induced postural hypotension is common in older patients with decompensated heart failure and may be prevented by lower limb compression bandaging. Gerontology. 2009;55(2):138–144. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Dixon WJ, editor. BMDP statistical software. Los-Angeles: University of California Press; 1993. [Google Scholar]
- Weiss A, Beloosesky Y, Kornowski R, et al. Influence of orthostatic hypotension on mortality among patients discharged from an acute geriatric ward. J Gen Intern Med. 2006;21(6):602–606. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]