Postprandial Hypotension: Simple Treatment But Difficulties With the Diagnosis

Over the past two decades, a growing amount of work on postprandial hypotension has been published. Postprandial hypotension has now been recognized as a very common clinically relevant disorder in elderly persons (1,2). In healthy elderly persons, meal-induced decreases in blood pressure are common but mostly asymptomatic (2). Elderly persons with hypertension have an increased frequency of postprandial hypotension. Patients with heart failure, syncope, or Parkinson's disease; patients on dialysis; and persons with autonomic dysfunction tend to have worse postprandial hypotension (2–5). In these patients, the prevalence of postprandial hypotension ranges from 40% to more than 80%. Postprandial hypotension was also highly prevalent in very old depressed patients. Treatment with nortriptyline or paroxetine did not affect postprandial blood pressure responses and symptoms (6). In hospital and institutionalized populations, the prevalence of postprandial hypotension in elderly persons is higher than that in the community because of the higher frequency of comorbid conditions and diseases and the increased number of medications, which all may have effects on blood pressure regulation. Nearly all elderly nursing home residents experience postprandial hypotension (7). In almost 40% of these residents, systolic blood pressure decreases more than 20 mmHg within 75 minutes of eating a meal (7,8). Postprandial hypotension is probably more common than is orthostatic hypotension. In this issue of the Journal, Vloet and colleagues provide further evidence that in elderly patients admitted to hospitals postprandial hypotension is extremely common. In 85 consecutive frail geriatric patients, 67% had postprandial hypotension and 52% were diagnosed with orthostatic hypotension (9). Eighty-one percent of the patients had either postprandial hypotension or orthostatic hypotension. It has been reported that orthostatic hypotension and postprandial hypotension occur frequently but not necessarily together in geriatric patients suggesting different pathophysiological mechanisms (2). The basic mechanisms by which postprandial hypotension is produced remain uncertain (2,10). Clearly the extent of postprandial splanchnic hyperemia is an important factor (11,12). The other major factors appear to be impairments in baroreflex function, inadequate postprandial increases in cardiac output, and inadequate sympathetic nervous system compensation (2). Our current understanding of postprandial hypotension is limited by the absence of a standardized clinically meaningful definition. Jansen and Lipsitz (2) defined postprandial hypotension, analogous to orthostatic hypotension, as a decline of 20 mmHg or more in systolic blood pressure within 2 hours following the start of a meal or when the absolute level of systolic blood pressure following a meal falls below 90 mmHg, and preprandial systolic blood pressure is over 100 mmHg. Postprandial hypotension occurs both in the sitting and supine position and can be found at all meal times. Although not found in all studies, it appears that postprandial hypotension is more prevalent and more profound in the morning. In this issue of the Journal, Vloet and colleagues (13) show that postprandial hypotension was most prevalent in the morning and least prevalent in the evening. In addition, test meals at dinnertime induced significantly shorter postprandial hypotensive periods, and these patients had almost no symptoms compared to those at breakfast- or lunchtime. Also, Puisieux and colleagues (14) showed that elderly persons admitted to a geriatric short stay department for evaluation of falls and syncope had a higher prevalence of postprandial hypotension at breakfast time than at other times. Because it is most likely that postprandial hypotension occurs after breakfast, evaluation of meal-related blood pressure decreases should be performed in the morning. The timing of the fall in postprandial systolic blood pressure is variable for each person. The time period of measuring blood pressure to detect postprandial hypotension should be at least 90 minutes. Apparent postprandial decreases in blood pressure can be found almost immediately after a meal with a nadir in blood pressure as early as 15 minutes following the meal in approximately 15% of patients. Postprandial blood pressure usually reaches a nadir within 30–60 minutes in 70% of patients. However, in the remaining 15% of patients, decreases in systolic blood pressure are apparent as late as 75 minutes following the meal (2). Including the rest period of at least 10 minutes for preprandial blood pressure measurements, the total time of a meal test takes at least 2 hours, making these tests quite time-consuming. This might be a reason that meal tests for the evaluation of postprandial hypotension are not always included in the geriatric assessment. Postprandial hypotension is easily detectable by manual or automatic blood pressure measurements. Ambulatory blood pressure monitoring is a valuable and useful method to investigate meal-related blood pressure changes in elderly patients (9,14). Only a few studies have used ambulatory blood pressure monitoring. There is no standard definition for postprandial hypotension using these devices. Grodzicki and co-investigators defined postprandial hypotension as a decline in the average systolic blood pressure of 20 mmHg or more during the 2 hours following a meal (15). The intervals between blood pressure recordings were not greater than 30 minutes. The average systolic blood pressure during the 2-hour period could underestimate the real decline in blood pressure. I recommend measuring blood pressure and heart rate at intervals of 10 minutes until 2 hours after the start of the meal. Postprandial hypotension has been associated with falls, syncope, dizziness, instability, coronary events, stroke, and total mortality at long-term follow-up (2,8,12,16–19). The presence of symptoms depends on a reduction in the blood supply to a specific organ. For instance, ischemic chest pain may be caused by postprandial hypotension or may be attributable to an increased cardiac output compensating for the hypotension. In addition to symptomatic postprandial hypotension, asymptomatic cerebrovascular damage has been related to postprandial hypotension (8). In most elderly persons, falls and syncope are multifactorial, and their resolution requires a careful evaluation. As a consequence, postprandial hypotension represents an important medical and economic problem. Postprandial hypotension is more common and more pronounced in elderly patients who have experienced a fall (18), and it accounts for 6%–8% of syncopal episodes (9,19). In a group of patients with unexplained syncope by conventional in-hospital evaluations, half had postprandial hypotension (12). Surprisingly, recently published guidelines on syncope did not address the blood pressure lowering effects of meals in older persons (20). Postprandial hypotension is mentioned as a possible cause or circumstance in syncope, as part of orthostatic hypotension, in the European guideline. Although several authors indicated the association of postprandial hypotension, falls, and syncope, no advice on blood pressure measurement around meals to diagnose postprandial hypotension or on treatment to prevent postprandial hypotension is given in these guidelines. A high percentage of elderly patients experiencing postprandial hypotension were also found in a group admitted to a hospital for evaluation of falls and syncope (14). Postprandial decreases in cerebral cortical oxygenation, as determined by near-infrared spectroscopy, were remarkably larger in syncope patients; these changes in oxygenation explain why these patients are more vulnerable to ischemic cerebral symptoms (21). In the study of Vloet and colleagues (9), two-thirds of the patients with postprandial hypotension were symptomatic. Postprandial hypotension symptoms included sleepiness, nausea, headache, and chest pain. Five patients developed postprandial syncope. Remarkably, dizziness was not reported in the patients with postprandial hypotension, whereas this is the most important symptom in orthostatic hypotension. These investigators reported that there was little overlap in symptoms between postprandial hypotension compared to orthostatic hypotension. The time period of hypotension might play a role in the type of symptoms. Orthostatic hypotension is more a short-lived phenomenon, whereas postprandial hypotension in general lasts longer. Another factor might be that the cerebral symptoms depend on the extent to which cerebral perfusion is compromised, regardless of whether it is caused by orthostatic or postprandial hypotension. Although older persons may have dramatic postprandial drops in their blood pressure, the majority of them have no symptoms at all at the time of the test. So, which elderly patients with postprandial hypotension are at risk for falls and syncope? It would be helpful in daily practice when we are able to target our patients at risk for symptomatic postprandial hypotension. In the study by Le Couteur and coworkers (17), postprandial hypotension was not associated with a risk of falling. Only in those persons who had a postprandial systolic blood pressure below 115 mmHg did the risk of falling significantly increase. These investigators suggested that the absolute level of postprandial blood pressure is more relevant to the risk of symptoms than to the magnitude of the postprandial decline in blood pressure. In 10% of nursing home residents, a reduction in postprandial systolic blood pressure to less than 100 mmHg was found (7). Two percent of these residents became symptomatic at their blood pressure nadirs; one resident developed ischemic chest pain, and the other resident showed right-sided weakness and global aphasia. In our own experience, we could not confirm the relation between symptoms and the decline in blood pressure below 115 mmHg. In contrast, some elderly patients can tolerate extremely low levels of postprandial systolic blood pressure, especially when they are chronically exposed to low blood pressures. In recent studies by Vloet and colleagues (9,13), the absolute level of postprandial systolic blood pressure was not indicative of the presence of symptoms during the test meal. However, these patients in the studies of Vloet and colleagues were not followed for a longer time period. Elderly patients with hypertension are more susceptible to cerebral underperfusion due to the reduction in blood flow associated with both age and hypertension (2). In these hypertensive patients, a shift in the threshold for cerebral autoregulation might explain cerebral postprandial symptoms at higher blood pressure levels. We assume that both large falls in postprandial blood pressure and low absolute levels of systolic blood pressure put frail elderly patients with comorbidity at high risk for symptomatic postprandial hypotension. The frequent lack of symptoms associated with a decline in blood pressure of 20 mmHg or more following meal ingestion makes the value of this definition uncertain. Also, symptomatic postprandial hypotension may arise without a decline in systolic blood pressure of at least 20 mmHg. If a given meal-related decline in systolic blood pressure exceeds the threshold for cerebral autoregulation, patients may become symptomatic. Therefore, more studies are needed to identify the factors resulting in symptomatic postprandial hypotension. Apart from the magnitude of postprandial hypotension, from a clinical perspective the question remains why most patients do not have symptoms following each meal. Assuming that an older person with postprandial hypotension has at least two meals a day, certainly not every meal is accompanied by symptoms, such as falls or syncopal episodes. Apparently, multiple factors have to interact together to compromise cerebral perfusion to such an extent that symptoms will occur. Several studies failed to demonstrate an additive effect of orthostatic hypotension. The well known variability of orthostatic hypotension can, however, play a role at certain moments (18). Attention should be paid to a possible association with dehydration, anemia, carotid sinus syndrome, and cardiovascular medications with hypotensive effects. Overall, we need more insight into postprandial hypotension and its effects on cerebral blood flow and oxygenation. Novel methods, such as near-infrared spectroscopy, enable easy-to-use noninvasive bedside monitoring for continues assessment of changes in cerebral oxygenation and blood volume (21–23). Management of postprandial hypotension includes a combination of pharmacologic and nonpharmacologic interventions. Adequate treatment of hypertension and limitation of the use of diuretics and nitrates may improve postprandial hypotension and ameliorate symptoms. One of the most important suggested measures in the treatment of postprandial hypotension is limitation of the size of a meal. There is evidence that lowering the carbohydrate amount in meals reduces the magnitude and duration of postprandial hypotension (11,24). In addition, elderly patients with postprandial hypotension experienced fewer and less severe symptoms after a small meal compared to test meals with normal or high amounts of carbohydrates (24). Thus, reducing the size and increasing the frequency of meals is an easy, cost-effective, and successful intervention in the management of postprandial hypotension in elderly patients. Another clinically important measure for the treatment of postprandial hypotension is walking exercise after a meal (2,25). Frail elderly patients with postprandial hypotension increased their postprandial blood pressure and heart rate while walking. This effect is not sustained when they stop walking, so patients with symptomatic postprandial hypotension should walk and continue to move on or to sit down. Postprandial hypotension is a prevalent condition in the elderly population and should be considered in any elderly patient with falls, syncope, dizziness, or cardiac or cerebral ischemic symptoms. Evaluation for postprandial hypotension should be included in the workup. Screening is also important in asymptomatic elderly patients at risk for postprandial hypotension. Further research is needed to identify the pathophysiology, in particular the factors resulting in cerebral hypoperfusion. Further epidemiologic studies are needed to identify the relevance of postprandial hypotension in relation to the occurrence of symptoms. Finally, as pointed out by Fisher and colleagues (26), we need guidelines for defining symptomatic postprandial hypotension and for how and when to measure blood pressure following the meal. Decision Editor: John E. Morley, MB, BCh