| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
1 Department of Geriatric Medicine, University Medical Center Nijmegen, The Netherlands.
2 Department of Geriatric Medicine, University Medical Center Utrecht, The Netherlands.
Address correspondence to René W. M. M. Jansen, MD, PhD, Department of Geriatric Medicine 318, University Medical Center Nijmegen, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. E-mail: r.jansen{at}ger.umcn.nl
|
A |
|---|
|
|
|---|
Methods. During 9 months, hemodynamic changes were measured with Spacelab 90207 after standing and after meals in all eligible patients. PPH is defined as a meal-related decline in systolic blood pressure (SBP)
20 mmHg, OH after standing up.
Results. Eighty-five patients (44 men), mean age 80 ± 7 years (range 6098 years), with 4 ± 2 diseases and 6 ± 3 prescriptions, were included. PPH was present in 57 patients (67%) with a significant postmeal SBP decrease of 34 ± 4 mmHg. OH was present in 44 patients (52%) with a mean SBP decline of 44 ± 4 mmHg after standing. Thirty-two patients (37%) had OH and PPH. Only 16 patients (19%) had neither OH nor PPH. Symptoms of PPH were present in 65% of patients, with syncope (in five patients) and sleepiness as the most common symptoms. OH was symptomatic in 61% of patients, with dizziness and risk for falls as the most common symptoms.
Conclusions. PPH and OH are more common in geriatric patients than was previously appreciated, with a high statistical probability that OH and PPH occur simultaneously. There is little overlap in symptoms of OH (dizziness, fall risk) versus PPH (sleepiness, syncope), which can play an important role in diagnosis. Because of the high prevalence of symptomatic PPH and OH, blood pressure measurements for diagnosing hypotensive syndromes should be part of a comprehensive geriatric assessment.
The prevalence of both PPH and OH is known to be high (2,10,1418) and to increase with age (4), with the number or the type of drugs used (19), with certain diseases such as Parkinson's disease (10,15,17,18,20), and in circumstances such as activity (1,21,22). We therefore hypothesized that PPH and OH occur even more often in geriatric patients with old age, high level of comorbidity, and use of multiple drugs than has been described so far in elderly populations (6,13,14). Detailed information about the prevalence of PPH and presentation of symptoms in a large group of frail geriatric patients admitted to hospitals is very limited. Furthermore, because PPH and OH are both very common in elderly persons, it might be expected that these syndromes frequently occur together in the same patient. However, in a small study in nursing home residents, it appeared that PPH occurred more frequently than OH, and PPH and OH were infrequently found together in the same resident (23). The present study was designed to establish the prevalence of PPH (in comparison to OH) and symptoms of hypotension in patients admitted to the geriatric ward of Dutch hospitals and to determine to what extent PPH and OH occur simultaneously.
| METHODS |
|---|
|
|
|---|
Preset exclusion criteria for this study were: infectious diseases with fever, delirium, insulin-dependent diabetes mellitus, fixed pacemaker, intravenous medication, problems with oral food ingestion, and/or inability to cooperate with the study protocol. Before participation all patients gave their informed consent.
Instrumentation and Procedure
Blood pressure (BP) was measured with a validated ambulatory automatic BP device for elderly persons (Spacelab 90207; Spacelabs Medical Inc, Redmond, WA) (24). All tests took place in the morning in a quiet room at an ambient room temperature of 2124°C. Before all measurements, the patients fasted overnight and refrained from taking medication from midnight the night before to avoid acute (side-) effects of medication on BP. Before the start of the test, the patients voided then were familiarized with the study protocol, and the BP equipment was applied to the upper arm. To ensure correlation with the ambulatory equipment, BP was first measured with a mercury device. The BP measurement recorded was that from the arm with the higher BP; if there was no difference between measurements, that from the nondominant arm was used. To diagnose OH, baseline supine BP was measured in the morning after at least 5 minutes of supine rest. BP in standing position was measured exactly 1 and 3 minutes after the patients had taken the upright position. OH was defined as a decline in systolic BP (SBP)
20 mmHg within 3 minutes after standing up (25). Because OH is highly variable over time, the BP measurements were repeated the next morning (1).
After standing, patients were seated for 20 minutes of rest and were then given a standardized liquid test meal to ingest within 10 minutes which consisted of a mixture of 100 ml of liquid glucose-syrup (Nutrical; Nutricia, Zoetermeer, The Netherlands) and 100 ml of lactose-free whole milk, containing a total amount of 65 g carbohydrate, 2 g fat, and 4 g protein. Jansen and Lipsitz (13) have previously suggested the use of this test meal instead of a regular meal. The test meals were served at room temperature to avoid the potential influence of temperature on BP (13). The patients remained in a sitting position for 90 minutes after the start of the meal. Movement was limited as much as possible during the test.
SBP, diastolic BP (DBP), and heart rate (HR) were measured every 10 minutes, from 20 minutes before until 90 minutes after the start of the meal. Baseline BP was defined as the measurement taken just before meal ingestion (which occurred after 20 minutes of rest). PPH was defined as a meal-induced decline in SBP
20 mmHg compared to baseline SBP (13) within 90 minutes after eating. Symptoms after standing up and after meal ingestion as well as changes in the patients' baseline condition in relation to BP changes were observed continuously by the researcher, unaware of the hemodynamic changes during the measurement, because the display of the BP device was covered. The patient was asked for symptoms every 15 minutes, or when the researcher noticed changes in the patient. Symptoms were classified and coded in a 4-point scale of severity; 0 = absence of symptoms, 1 = mild, 2 = moderate, and 3 = severe.
Statistical Analysis
Statistical analysis was performed with SPSS for Windows 10.0 (SPSS Inc., Chicago, IL). A p value <.05 was taken as the level of significance. The results are expressed as mean ± standard deviation (SD).
Patient characteristics were compared by means of one-way analysis of variance (ANOVA). Two-way repeated-measures ANOVA was applied to examine the effect of time on the BP and HR changes versus baseline. The correlation between BP responses and patient characteristics was determined by Pearson's correlation test, and for dichotomous variables with the chi-square test. In addition, paired t tests were used to examine the similarity of the individual maximum variable changes during the two orthostatic tests, and unpaired t tests to compare the means of selected variables in the subgroups with or without OH and PPH. Fisher's exact test was used to calculate the relative risk for patients with certain characteristics such as heart failure and the occurrence of OH or PPH. The relations between patient characteristics and postprandial changes in BP were calculated using multiple regression analysis. The explanatory variables were chosen based on expected influence on BP (6,13,22,26). To reduce the number of variables and therefore the risk of colinearity, data were combined (e.g., height and weight to body mass index).
| RESULTS |
|---|
|
|
|---|
|
|
|
|
20 mmHg after standing up at the first test, 33 patients at the second test. Seventeen patients had OH in one measurement, and 27 patients on both measurements. In all 85 geriatric patients, mean BP decreased at the first test from 153/82 mmHg to 137/80 mmHg after both 1 and 3 minutes of standing, and from 154/84 mmHg to 140/80 mmHg after both 1 and 3 minutes of standing at the second test. In patients diagnosed with OH (n = 37, 44%) BP decreased from 161/84 mmHg to 131/77 mmHg after both 1 and 3 minutes of standing at the first test. HR increased from 78 bpm to 80 bpm and 83 bpm after, respectively, 1 minute and 3 minutes of standing. The mean maximum decrease in SBP in the OH group at the first test was 36 ± 25 mmHg. In seven patients, OH was established only after 1 minute of standing. Twenty-three patients had OH at both 1 and 3 minutes after standing up. In seven patients OH was determined only after 3 minutes of standing.
At the second measurement, in the patients with OH (n = 33, 39%) BP declined from 162/86 mmHg to 132/77 mmHg and 134/77 mmHg after, respectively, 1 minute and 3 minutes of standing. The mean maximum decrease in SBP at the second test was 33 ± 30 mmHg. HR increased from 71 bpm to 84 bpm after 1 minute of standing, and to 74 bpm after 3 minutes standing. In 6 patients, OH was determined only after 1 minute of standing, in 4 patients just after 3 minutes of standing, and in 23 patients at both time points. Comparison of those patients who had OH after 1 minute with those patients who had OH only after 3 minutes for the test on the study day and test 2 the next day showed no difference in patients' characteristics between these small groups. The maximum SBP decrease after standing up at 1 and 3 minutes was significantly correlated with the number of prescriptions (p =.027) and with baseline SBP (p =.006).
Combined Occurrence of PPH and OH in Geriatric Patients
The combined occurrence of PPH and OH is shown in Table 4. Thirty-eight percent (n = 32) of the geriatric patients were diagnosed with both OH and PPH. Baseline SBP in this group was significantly higher than in the other groups (p =.002 between groups). In only 16 patients (19%), neither OH nor PPH were present. Baseline SBP in the patients without PPH and OH, and in patients with OH, was significantly lower than that in patients with PPH or with PPH combined with OH (p <.05). Other outcome measures were not significantly different between the four groups.
|
|
| DISCUSSION |
|---|
|
|
|---|
In contrast to previous studies, this study does not show that any specific disease or medication is indicative for developing PPH or OH (4,11,13,18,20,25,26,28). The small number of patients in the subgroups and characteristics such as high comorbidity and use of medication in all geriatric patients can explain this. Systolic hypertension is a risk factor for OH and PPH, confirmed by our results (4,13,18,27).
We have presented our results in two ways, as the mean of the postprandial BP levels and as the mean of the maximum postprandial BP declines (Table 3). The mean of the maximum BP declines might explain the high percentage of PPH in this study. The individual postprandial BP changes are variable. In previous studies, postprandial BP has been shown to reach a nadir within 3060 minutes (13). The systolic BP nadir occurred as early as 15 minutes after the meal in approximately 15% of the patients and as late as 75 minutes after the meal in 11%13% of the patients. Presenting the results as the mean of postprandial BP levels would make the effect of postprandial changes in BP unclear.
In this study, PPH and OH occurred simultaneously in one third of the geriatric patients. Jansen and colleagues (23) found OH and PPH together in only 2 of 22 nursing home residents. Puisieux and colleagues (10) found a slightly higher frequency of OH in elderly patients with PPH compared to patients without PPH. Other researchers found in hypertensive middle-aged patients that those with OH were more at risk to develop PPH (10,29,30). In our patients, we observed a nonsignificant difference between the occurrence of OH in patients with PPH (56%) and that in patients without PPH (43%).
It has been suggested that the effects of OH and PPH can be cumulative (13). In patients with autonomic disorders it has been suspected that PPH and OH have similar mechanisms (30). We conclude that OH and PPH occur frequently but not necessarily together in geriatric patients. In our total population, 52% of the patients had OH, so the presence of OH in slightly more than half of the patients with PPH could also be expected by a distribution of probability and argues for independent underlying mechanisms for OH and PPH. Our results are in agreement with others suggesting that the underlying mechanisms are at least partly different (10,23). The high frequency of the simultaneous occurrence of OH and PPH in geriatric patients could be explained by the large analogy of causes and circumstances influencing OH and PPH such as autonomic failure, hypertension, age, (cardiovascular) medication, and comorbidity (13,23,26,29,30).
The patients with OH could have different medical conditions with different HR responses to postural changes. However, none of our patients were diagnosed with autonomic failure syndromes, such as primary autonomic failure or multiple system atrophy. In addition, none of the patients were hypovolemic on the day of testing. Of the OH group, 10 patients had Parkinson's disease; of the group without OH, 5 patients had Parkinson's disease. Subgroup analyses of these small groups of Parkinson's patients revealed no difference in HR response to postural changes.
The third finding was that both PPH and OH were often symptomatic in the geriatric patient with different presentation. The number of geriatric patients with symptomatic OH was unexpectedly high (3,26), with dizziness as the most pronounced symptom. In PPH, other and more severe symptoms such as sleepiness, nausea, headache, chest pain, and syncope were present (Table 4). The postprandial symptoms in the symptomatic patients were related to the changes in BP. When postprandial BP recovered to preprandial BP levels, the symptoms disappeared. Remarkably, dizziness is absent in PPH, whereas this is the most important symptom in OH. Tiredness, which is the most common symptom in PPH, is absent in OH. The cerebral symptoms of PPH and OH probably depend on the extent to which cerebral perfusion is compromised, leading to postprandial syncope in five cases (13,31). However, the large differences in presentation of symptoms between OH and PPH could not be explained by the BP declines alone. Probably the duration of the BP decline in PPH and the duration of the test also play a role in differences of symptoms. The variance in symptoms of PPH and OH suggests that these phenomena have different pathophysiological backgrounds and, therefore, a distinct presentation (13,23). These striking differences can be important in a patient's history when diagnosing OH or PPH.
Surprisingly, in the recently published clinical guidelines on syncope, PPH is only mentioned as a possible cause or circumstance in syncope as part of OH (32). Although several authors have indicated the association of PPH, falls, and syncope, no advice on BP measurement around meals to diagnose PPH or on treatment to prevent hypotension is given in the guidelines (6,10,32,33). Because of the high prevalence of symptomatic PPH and OH in geriatric patients and the seriousness of consequences, it is of clinical importance for all geriatric patients to track their BP after meals or when standing up to diagnose PPH and OH, especially in case of symptoms such as dizziness, falls, and syncope.
This study has a few limitations. First of all, our study group was, as can be expected in geriatric patients, very heterogeneous. Because of the diversity of the characteristics of the patients, the numbers of patients in the subgroups were too small to demonstrate significant relations between patient characteristics and the presence of PPH or OH. Second, we measured PPH in the sitting position. Although orthostatic changes in BP during prolonged sitting might contribute to the postprandial decreases in BP, we gave patients their meal while seated so that we could more closely duplicate their usual eating posture. The glucose component in meals is responsible for the BP decrease, and not the rebound hypoglycemia (13,34,35). Xylose, fructose, water, fat, and protein do not show an effect on BP (13,34). Third, some of the symptoms of PPH were general (e.g., tiredness could also be due to the duration of the test), but all symptoms were related to declines in SBP. Fourth, there is no uniform definition or BP measurement procedure of PPH and OH which could affect the measurement and lead to overestimation or underestimation of prevalence (10,36). Although we measured OH twice, we cannot exclude the possibility that the number of patients with OH is even higher, because of the large day-to-day variability of OH (1). To exclude observer-related problems in BP measurement and the duration of the measurement, we note the potential role of noninvasive ambulatory BP monitoring to identify PPH and OH in geriatric patients (10,36).
Conclusion
We have identified geriatric patients as an important group at risk for serious and symptomatic PPH and OH. In contrast to previous findings, we have found that PPH and OH occur frequently together in patients admitted to geriatric wards of two Dutch hospitals and that PPH and OH are both highly prevalent, and often symptomatic, with striking differences between symptoms of PPH or OH. These findings emphasize the importance of early screening for OH and PPH in frail geriatric patients at risk, to reduce serious complications due to hypotensive syndromes. OH is highly variable during the day and from day to day, and is most prevalent in the morning before eating breakfast and taking medication (1,2,7,10,14,27). PPH also shows variability during the day, with frequent absence of PPH in the evening (37). Because of the high prevalence of OH in the morning, and the fact that postprandial BP responses have shown to be reproducible in the morning, we recommend measuring BP for diagnosing both OH and PPH in the morning, before eating a meal or taking medication (1,2,7,10,14,23,27,37). Because of the high within-day variability of OH, BP measurements were repeated the next morning (1,2,7,10,14,27).
|
A |
|---|
|
|
|---|
We thank Rita Smits for her assistance in performing this study and the entire staffs of the Departments of Geriatric Medicine, University Medical Centers Nijmegen and Utrecht, The Netherlands, for their participation in this study.
|
F |
|---|
|
|
|---|
Received April 27, 2004
Accepted September 22, 2004
| References |
|---|
|
|
|---|
70 years of age with heart failure. Am J Cardiol. 2002;90:596.[Medline]
This article has been cited by other articles:
![]() |
R. W. M. M. Jansen Postprandial Hypotension: Simple Treatment But Difficulties With the Diagnosis J. Gerontol. A Biol. Sci. Med. Sci., October 1, 2005; 60(10): 1268 - 1270. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|