Freidenberg, Donald L. DO*; Shaffer, Lynn E.T. PhD†; Macalester, Shawn DO‡; Fannin, Elizabeth A. BA§
Reader Benefit: Orthostatic hypotension in patients with dementia is less likely to present with dizziness than with mental fluctuations, confusion, drowsiness, and slow falls; distinguishing this treatable condition from dementia can greatly improve patients’ safety and quality of life.
Cognitive and Behavioral Neurology publishes this article as a rare combination of a retrospective study and personal opinion, based on Dr Donald Freidenberg’s experience in a quarter century of clinical neurology practice. The retrospective study has been peer-reviewed; I am allowing Dr Freidenberg’s personal opinions to be intermixed, because it is precisely his kind of careful, thoughtful observation and investigation that was once the foundation of our field, and that remains a critical component of the medical-scientific process. In publishing this article, we are not necessarily endorsing Dr Freidenberg’s opinions, but we are endorsing the courage to make new connections, to marshal the evidence, and to argue that “attention must be paid” to conditions that we might otherwise overlook.
Barry Gordon, MD, PhD
AD=Alzheimer disease; BP=blood pressure; OH=orthostatic hypotension; OHSS=Orthostatic Hypotension Suspicion Set (signs and symptoms).
Orthostatic hypotension (OH) is the body’s inability to maintain systemic blood pressure (BP) against gravity’s effects on blood volume. OH manifests as a drop in BP when a person moves from lying down to an upright position, whether sitting or standing. The literature extensively describes the mechanisms that maintain systemic BP (Hall, 2011) and those that lessen a change in BP when a person shifts posture (Freeman, 2008).
In 1996, the American Academy of Neurology and the American Autonomic Society developed consensus criteria for the diagnosis of clinical OH (Kaufmann, 1996): OH is a systolic drop of 20 mm Hg or a diastolic drop of 10 mm Hg from lying to standing within the first 3 minutes of standing. It has been argued that the criteria should be changed because some patients develop symptoms after less severe drops of BP (Braam et al, 2009; Wieling and Schatz, 2009), and some patients’ BP starts falling >3 minutes after they stand up (Cheshire and Phillips, 2006; Gibbons and Freeman, 2006; Mader, 2012; Streeten and Anderson, 1992).
Whatever the disagreements about the diagnostic criteria, what matters clinically is patients’ signs and symptoms. Many physicians base a diagnosis of OH on patients’ complaints of dizziness when they change posture. However, patients with OH may not feel dizzy (Arbogast et al, 2009; Aung et al, 2012; Kaufmann et al, 2012; Mader, 2012). Other signs and symptoms that have been reported in association with OH include light-headedness, vertigo, tremulousness, pre-syncope, syncope, generalized weakness, fatigue, nausea, cognitive slowing, leg buckling, posterior headache (“coat hanger” headache), blurred vision, postural shortness of breath, and postural angina pectoris (Arbogast et al, 2009; Ejaz et al, 2004; Freeman, 2008; Gibbons and Freeman, 2005; Grubb et al, 2008; Humm et al, 2011; Kaufmann et al, 2012; Low et al, 1995; Lyons and Pahwa, 2011; Mader, 2012; Mathias et al, 1999; Mehrabian et al, 2010; Robertson et al, 1994; van Nieuwenhuizen et al, 2010).
One of the problems in diagnosing OH is that it can come and go. After a postural shift, people vary in the magnitude of a BP drop that produces symptoms. Some people are hardly affected by a change in posture. For others, a small drop in orthostatic BP can cause extensive symptoms (Low et al, 1997). Possible explanations for these variations are differences in plasma volume, sympathetic supply to the arterioles, the person’s venous capacity, cerebral vasoregulation, and the splanchnic-mesenteric bed (Low et al, 1997). Further, the extent of orthostatic BP change can vary substantially within an individual at different times of day (Ooi et al, 1997; Ward and Kenny, 1996).
The prevalence of OH worldwide has been reported to be 6% at age 40 to 45, rising to 14% to 20% at age 70 to 75 and to 30% beyond age 75 (Lipsitz, 1989; Low et al, 1997; Rutan et al, 1992; Tilvis et al, 1996). The reported worldwide prevalence of OH in the elderly general population varies from 5% to as high as 30% (Feldstein and Weder, 2012; Lipsitz, 1989; Mader, 1989; Masaki et al, 1998; Rutan et al, 1992; Tilvis et al, 1996). This wide variation may be a function of differences in patient populations, definitions of OH, and methods for determining OH (Feldstein and Weder, 2012; Lipsitz, 1989; Mader, 1989; Masaki et al, 1998; Rutan et al, 1992; Tilvis et al, 1996). Shibao et al (2007) reported an OH prevalence of 6% in the general elderly US population, but 54% to 68% among elderly people living in long-term care facilities.
While more research is needed, recent studies have linked OH to brain atrophy (Jochemsen et al, 2012; Muller et al, 2012), especially in combination with reduced cerebral blood flow (Muller et al, 2010). OH has also been linked to cognitive impairment (Kim et al, 2012; Muller et al, 2012) and memory deterioration (Glodzik et al, 2013), notably in patients with a history of midlife hypertension (Glodzik et al, 2013; Muller et al, 2012). OH has also been reported as a risk factor for stroke (Eigenbrodt et al, 2000; Feldstein and Weder, 2012) and as a predictor of mortality in middle-aged adults (Fedorowski et al, 2011; Rose et al, 2006). In a study of elderly men, the worse the OH, the greater the risk of mortality within 4 years (Masaki et al, 1998).
More patients with than without OH are reported to have brain dysfunction, cerebral hypoperfusion on single photon emission computed tomography (Siennicki-Lantz et al, 1999), white matter low attenuation on computed tomography scans of the brain, and silent infarction on magnetic resonance imaging scans of the brain in the elderly (Kario et al, 2002; Raiha et al, 1993).
Missing the diagnosis of OH leaves patients at risk. The most severe consequences of OH are falling, syncope, confusion, ischemic brain injury, and death. Falls related to syncope account for 40 deaths per 100,000 by age 75, and 189 deaths per 100,000 by age 85 (Baker and Harvey, 1985; Low et al, 1997). Less severe injuries associated with OH include skin injuries, sprains, fractures, and subdural hematomas (Aung et al, 2012; Low et al, 1997).
Patients with some dementias and other neurodegenerative diseases (Allan et al, 2007; Fedorowski et al, 2011) are at greater-than-normal risk of developing OH and suffering related injury (Fedorowski et al, 2011) and probable cognitive decline (Maule et al, 2008). High risk for OH has been found especially in elderly people with Alzheimer disease (AD), dementia with Lewy bodies, Parkinson disease dementia, and vascular dementia (Allan et al, 2007; Mehrabian et al, 2010; Sonnesyn et al, 2009). In a prospective study of autonomic functions, Allan et al (2007) found OH in only 5/38 (13%) elderly controls, but in 13/38 (34%) patients with AD, 10/29 (34%) with vascular dementia, 18/37 (49%) with Parkinson disease dementia, and 14/27 (52%) with dementia with Lewy bodies.
Despite this risk, until now no systematic study has reported the clinical features of OH in patients with dementia, or examined how these manifestations are modified by treatment. Few studies of OH features have been restricted to participants with dementia, and few studies of dementia and OH have detailed the clinical features of OH in this population. No studies have evaluated how the signs and symptoms respond to treatment.
Author D.F. is a solo neurologist/neurobehaviorist specializing in geriatric neurology, primarily evaluating and treating patients with dementia. For about the past 23 years, he has had a large clinical community practice covering much of central Ohio. At the time of this study, in addition to his office practice he was caring for patients at >40 long-term care facilities.
Over the years, D.F. became aware that many of his patients had OH and that it was a common cause of delirium and disorders of gait and behavior. He also saw that family physicians, internists, cardiologists, psychiatrists, emergency room physicians, and even other neurologists were missing the diagnosis, even when patients had frank syncope. Not recognizing that OH could be causing their patients’ signs and symptoms, these physicians were prescribing more psychotropic medications and the patients were only getting worse. Yet, when suspected, OH is diagnosable and patients can improve dramatically with treatment.
D.F. and his colleagues theorized that better recognizing the clinical features of OH in elderly patients, particularly in those with dementia, might lead to earlier and more appropriate treatment, thereby preventing injuries and improving quality of life. With this thought in mind, the authors reviewed the charts of some of D.F.’s patients with dementia, with and without OH. The goals were to help determine the prevalence of OH, risk factors and symptom profile, and clinical response to treatment.
We analyzed the charts of 188 patients whom D.F. had treated for dementia between 2004 and 2007. The patients had been referred for evaluation of dementia symptoms, neuropsychiatric and behavioral issues, confusion, gait disturbances, and/or falling. We reviewed charts in alphabetical order, studying as many as time permitted over 1 month. We did initial reviews of 215 charts, but excluded 27 patients because they had not been given a diagnosis of dementia.
For the 188 qualified patients, author S.M. evaluated 148 of the charts and directed medical students in evaluating the other 40. In an Excel spreadsheet created by author L.S., the evaluators recorded the patients’ (Table 1):
- Age, sex, and marital status
- Living situation: at home, assisted living, or traditional nursing home (for purposes of the study, we grouped patients in assisted living and traditional nursing homes as being in “long-term care”)
- Dementia type
- Medical and social history
- Physical examination
- Detailed BP information
- Signs and symptoms from the Orthostatic Hypotension Suspicion Set (OHSS) reported by the patient, family, D.F., nurses, and other caregivers (the OHSS is shown in Appendix 1 and discussed in “Diagnosis of OH” below)
Of the 188 patients reviewed, 96 had OH at some point when they were under D.F.’s care, and 92 did not. For the patients who had OH, we also recorded:
- All available BP information recorded before and at diagnosis, and during and after treatment
- Signs and symptoms that matched features in the OHSS
- Medications that the patient was taking at any evaluation point
- Nondrug OH treatments that the patient was receiving at any evaluation point, eg, extra drinking water, compression stockings
- Response to treatment, as reported by D.F., family, and other caregivers (we did not measure response with a rating scale because of the retrospective nature of the study and because patients’ clinical improvement was already obvious to D.F. when he treated them)
The Institutional Review Board at Riverside Methodist Hospital in Columbus, OH, exempted the study from review. Thus, individual informed consents were not required.
Because the analysis was retrospective and we did not follow a study protocol, in the next sections we will describe D.F.’s clinical approach to diagnosing and treating OH before, during, and since the study.
General Clinical Procedures
D.F. sees patients with neurologic and neurobehavioral conditions at his office and in their homes, assisted-living facilities, and traditional nursing homes. He sees office patients every 1 to 3 months, and most long-term care patients on a regular monthly schedule, until their condition improves or another cause is found for their signs and symptoms. As he diagnoses and treats patients, he records all of their demographic and clinical information on standard paper forms that he created for intake, history, and neurologic and neurobehavioral examinations.
Between visits, he may need to follow patients’ progress, especially if he is adjusting their medications. He asks their family members and other caregivers to give him regular updates. To ensure that he receives all the information he needs to monitor the patients and make treatment decisions, he created paper forms for families and staff to use to record, among other things, patients’ BPs, signs and symptoms, and behavioral information. Patients living at home have their BP tested by a spouse, adult child, or visiting nurse. Office patients or their caregivers are asked to complete the report forms and phone in weekly updates to D.F.’s office; D.F.’s staff use a special list of questions to collect the information over the phone. For patients in long-term care, the facility staff are asked to test BP, fill out the forms each week, and fax them to D.F.’s office.
D.F. reviews all reports daily. Generally, the next morning his staff contact family by phone and long-term care staff via a standard faxed form, with requests for further needed information about signs and symptoms (eg, “Has the patient’s confusion improved?”), instructions to repeat BP measurements, and orders for medication changes. In the long-term care facilities, where several different caregivers might be measuring BP and filling out forms, each person receives the relevant faxes.
D.F. manages and prescribes medications for neurologic and neuropsychiatric conditions, including OH. When he believes that a medication prescribed by another physician may be contributing to a patient’s condition, he asks the physician to alter the dose or the drug. Most physicians are receptive to his requests.
Diagnosis of Dementia
About 75% of D.F.’s patients have dementia. He bases his diagnosis of dementia on the criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (American Psychiatric Association, 2000).
Patients whom he diagnoses with AD during an office visit must also meet the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS/ADRDA) criteria for probable AD (Lopez et al, 1990). Patients in long-term care who are diagnosed with AD must have slowly progressive amnesia and fluent aphasia, with a normal or nearly normal neurologic examination.
D.F. gives patients a diagnosis of dementia with Lewy bodies if they meet the consensus criteria for dementia with Lewy bodies (McKeith et al, 1996).
He diagnoses Parkinson disease dementia if patients developed dementia >2 years after the onset of Parkinson disease.
He diagnoses vascular dementia if patients have vascular risk factors and exhibit subcortical dementia (Albert, 2005) with associated early gait disorders and urinary incontinence. Available computed tomographic and magnetic resonance imaging scans for patients with vascular dementia show multiple focal areas or patches of lesions in the deep or subcortical white matter and/or the deep gray matter.
When D.F. diagnoses patients in the office, he can take as long as he needs to talk with the patient and family, do needed tests, and reach a confident diagnosis of the type of dementia. When he visits patients in long-term care, however, he does not have the luxuries of time and information. The chart lists the patient’s dementia type but gives minimal history. If D.F. questions the dementia type, he can usually correct it by evaluating the patient. If the rediagnosis is not straightforward, he tries to obtain all available test results. Assigning the correct type becomes more challenging the more advanced the patients are in their disease course. Endstage dementia can be so undifferentiated that the type is all but impossible to discern. Thus, for the study, a few diagnoses had to be educated guesses.
Diagnosis of OH
Orthostatic BP Testing Procedure
For nearly all of his patients, D.F. routinely takes orthostatic BP measurements at each visit, whether or not the patients have OH signs and symptoms. He wants to learn how far the patient’s BP drops from lying down to standing, and then how quickly and how completely it returns to baseline, or whether it keeps dropping.
He can test most patients for OH in the usual way: first supine (lying face up), then standing. As soon as patients lie down with their arm abducted 90 degrees, he begins measuring their BP repeatedly, waiting only until the cuff deflates all the way before starting the next reading. When 2 readings are the same, he records that measure as the supine BP. Then he asks patients to stand up and abduct their arm 90 degrees. Again, he immediately begins measuring their BP repeatedly, waiting only until the cuff deflates fully before starting the next reading. He stops measuring when 2 BP measurements are the same, and he records the last of the matching measurements as the standing BP. If patients’ legs begin to buckle and give way before the BP stabilizes, he stops testing, helps them lie down, and does not use the measurement.
In the few patients who are unable to stand, D.F. measures BP from supine to sitting with legs dangling. In the few patients who are unable to lie down, D.F. measures BP from sitting to standing. (A “BP set” is defined as “1 measurement down and 1 measurement up.” For most patients, this means 1 measurement supine and 1 standing. For the less able patients, a BP set is 1 measurement supine and 1 sitting, or else 1 sitting and 1 standing.)
Patients who have modified testing are diagnosed by the same criteria for OH as the patients who are evaluated from supine to standing. However, measurements taken from supine to standing are the most sensitive in detecting OH (Rutan et al, 1992). D.F. has found that when a symptomatic patient has a normal BP on sit-to-stand testing and then is tested from supine to standing, even during the same examination, the second measurement may well be orthostatic.
For family members and long-term care staff who are testing and reporting symptomatic patients’ BPs between visits, D.F. gives these instructions: The best times of day to test are just before the patient gets up in the morning, just after a meal, or when the patient is having symptoms. Practically, however, the BP can be measured whenever the caregiver is available and the patient is not too agitated to allow it. The patient should lie supine for 3 minutes, the tester should measure the supine BP, the patient should stand up, and the tester should immediately measure the BP again. (If needed, the tests can be modified to “lying to sitting” or “sitting to standing.”)
If caregivers report readings that do not match the patient’s symptoms, D.F. asks the tester to repeat the measurements, but now with 2 standing readings, the first immediately upon standing and the second after 1 minute of standing. These readings should be done once a day on each of 3 days, which need not be consecutive.
Even if the readings remain nondiagnostic, D.F. may still suspect OH based on the clinical features, history, and laboratory results. He remeasures BPs at the next visit, and oftentimes finds marked orthostatic drops.
If D.F. tests heart rates at all, he records them when he measures the BP, using radial artery palpation. He has not tested heart rates routinely enough for them to be evaluated in the chart review. This is a limitation of the study.
Clinical Clues to OH
D.F. suspects OH clinically when patients have any of the signs and symptoms that he calls the Orthostatic Hypotension Suspicion Set (OHSS) (Appendix 1). Although many of the OHSS signs and symptoms are not specific to OH, they are very sensitive in raising suspicion of OH. D.F. developed the OHSS based on his observations of thousands of patients with dementia and OH
. In most of these patients, the OH features begin acutely or subacutely, and they fluctuate rather than representing slow, permanent deterioration from baseline function.
For example, D.F. suspects OH if a caregiver sometimes reports that a given patient is doing well and functioning at baseline, and at other times reports that the same patient seems markedly confused, drowsy, and glassy-eyed (signaling mental fluctuations) or is much less steady, with leaning, dysarthria, and/or falling (signaling motor fluctuations) (Low et al, 1995). Another common tipoff to OH is delirium and other OHSS features in a patient who has been falling, complains of leg weakness, and wants to stay in bed.
As noted, the extent of orthostatic BP change can vary during the day (Ooi et al, 1997; Ward and Kenney, 1996). Although we could not find formal studies and would have found it difficult to assess ourselves, D.F.’s clinical experience is that the signs and symptoms of OH can also vary in type and severity—from none to incapacitating—throughout the day and over days. For some patients, the symptoms seem to start, worsen, and subside along with the severity of the BP drop. For others, the symptoms do not mirror the BP at all.
Symptoms do not seem to follow a person’s circadian rhythm, although they may worsen after a large meal or after a prolonged period of being supine.
OH symptoms disappear when patients lie down. However, if patients have been up for a long time, even after falling asleep while sitting in a chair, they may be confused when they lie down, and they may stay confused, restless, and agitated for the next few hours, and may endanger themselves by trying to get out of bed.
An episode of OH can last for days or weeks. In some patients, the BP improves spontaneously; these people are likely to have more episodes. In other patients, the BP remains orthostatic until it is treated.
Manual Versus Automated Sphygmomanometer
D.F. always measures BP with a manual sphygmomanometer. He prefers a manual cuff because he can repeat BP measurements quickly just after the patient stands up. Thus, he can watch the progressive signs of OH appear as the BP falls, and he can track the patterns of fall and recovery. A manual cuff is also preferable for testing uncooperative patients, and for testing patients whose Korotkoff sounds cannot be heard; as D.F. slowly inflates the cuff, he can palpate a BP by feeling the radial or brachial pulse.
Because manual sphygmomanometers require considerable training to use properly, D.F. recommends that patients’ families use an automated cuff.
Challenges of Monitoring Patients in Long-Term Care Facilities
Many patients’ BP measurements come from long-term care staff, and D.F. has to depend on whatever methods they use. Long-term care facilities tend to use automated more than manual cuffs, primarily to compensate for less skilled staff testers. There are different brands and types, some meant for the arm and some for the wrist, and all intended to be used only on bare skin. Automated cuffs are slow, and if the patient moves during the test, it has to be started over.
At the time of the study, a more serious problem than the type of cuff used was that long-term care facilities resisted measuring orthostatic BPs at all because they are time-consuming and difficult, the staff may have lacked the needed skills, and patients may have fought the test. If certain staff members refused to do the testing, D.F. might ask the director of nursing to ensure that qualified and willing nurses be assigned to make the measurements, or else D.F. did them himself during his next visit. This problem has eased as more nurses have become convinced of the value of testing for and treating OH.
When we reviewed the charts for the study, we could not find all of the BP measurements, especially from the follow-up phase of patients’ treatment. We know that these measurements were taken, because D.F. would not have changed a treatment without having BP results to guide his decision. We just could not access the numbers.
All of these factors constitute further limitations of the study.
Confirming the Diagnosis
Orthostatic BP testing can have 3 results:
- If patients have normal BPs and do not report symptoms, and if D.F. does not suspect OH, he follows them routinely, at least every 3 to 6 months.
- If patients have BPs in the orthostatic range but no symptoms, D.F. does not treat. He tells patients and their caregivers to try to avoid situations that can cause symptoms, such as standing still or being exposed to heat, alcohol, or OH-inducing medicines. He also explains the OH symptoms and signs to look for. Then he follows the caregivers’ weekly updates and reevaluates as needed or in 3 to 6 months.
- If patients have OHSS features that have no other cause, no matter what their BPs are, D.F. asks the caregivers to take an orthostatic BP reading on each of 3 days, and then he reevaluates.
For D.F. to give patients a diagnosis of OH, they must meet the American Academy of Neurology/American Autonomic Society consensus criteria for OH established in 1996. These criteria define OH as “a reduction of systolic blood pressure of at least 20 mm Hg or diastolic blood pressure of at least 10 mm Hg within 3 min of standing” (Kaufmann, 1996). The diagnosis holds even if the BP returns to baseline.
For D.F., the diagnosis of OH depends on both the orthostatic BP measurements and clinical features. He considers patients to have OH if, over 3 days of testing, at least 1 of their orthostatic BP measurements meets the diagnostic criteria and if the patients have OHSS features that began either acutely or subacutely.
After establishing OH, D.F. must exclude contributing medications, volume depletion, anemia, and other treatable causes. He assesses volume depletion by taking a history of fluid intake and medicines that might affect volume, by examining the mucous membranes of the eyes and mouth, and by testing blood urea nitrogen, hemoglobin, and, if needed and available, electrolytes.
Treatment of OH
D.F. treats patients if they have diagnostic BP readings and at least 1 OHSS feature. He begins by treating medical causes of OH and, when feasible, taking patients off drugs that might be contributing to it. If a drug was prescribed by another physician, he asks that physician to stop it. Then D.F. remeasures the orthostatic BP on at least 3 days, consecutive or nonconsecutive. If patients still have diagnostic BPs and OH signs and symptoms, he treats their OH directly, customizing treatment to their individual needs.
Whenever possible, he begins with nondrug measures, alone or in combination. He might prescribe extra drinking water, compression stockings, and, rarely, an abdominal binder. (Most patients with dementia resist wearing stockings and binders.) The head of patients’ bed is elevated to 30 degrees to raise their renin levels and reduce supine hypertension.
D.F. prescribes OH medications only if patients have potentially serious or life-threatening signs and symptoms. Immediately before starting medicines, he reconfirms the OH by taking 3 more days of orthostatic BP measurements. For patients with OH whose underlying BP is normal or low, he prescribes sodium chloride tablets, fludrocortisone, and/or midodrine. For patients whose underlying BP is high, he prescribes pyridostigmine.
He prescribes sodium chloride tablets, fludrocortisone, and midodrine only to normotensive or hypotensive patients with OH. These agents can be used alone or in combination. If the first one prescribed is helpful but insufficient at the highest safe dose, another can be added.
More than half of D.F.’s patients with OH—and almost half of the study patients—also have chronic hypertension. Some have supine hypertension, and some have hypertension when both supine and standing. He uses the generally accepted definitions of hypertension as a systolic BP >140 mm Hg or a diastolic BP >90 mm Hg. Prehypertension is defined as a systolic BP of 120 to 139 mm Hg and diastolic BP of 80 to 89 mm Hg. Normotension is a systolic BP of <120 mm Hg and diastolic BP of <80 mm Hg. Hypotension is a systolic BP of <90 mm Hg. For purposes of the study, we considered a “normal” BP to be in the normal or prehypertension range.
D.F. always addresses the hypertension before the OH, especially in patients who have marked supine hypertension. Because the patients’ primary care physician manages their hypertension drugs, D.F. begins by asking the physician to manage any uncontrolled hypertension. When the hypertension is controlled, D.F. asks the physician to adjust current antihypertension drugs that might be contributing to the OH. For example, he might suggest that the physician lower the dose of lisinopril or switch to low-dose amlodipine. He also recommends that the physician try to avoid prescribing diuretics, alpha antagonists, and anything more than low-dose calcium channel antagonists.
In treating the OH directly, D.F. tries not to worsen patients’ hypertension. For patients with hypertension and symptomatic OH, he prescribes pyridostigmine. Before starting pyridostigmine, patients must stop taking acetylcholinesterase inhibitors (eg, donepezil, rivastigmine, galantamine). He believes that the clinical benefit of treating symptomatic OH far exceeds any subtle benefit that could be derived from acetylcholinesterase inhibition.
D.F. considers patients with underlying hypertension to be normotensive after OH treatment if their systolic
BP is below 140 and their diastolic BP is below 90. If OH treatment causes or exacerbates hypertension, he asks their primary care physician to adjust their medicines until they become normotensive and free of OH symptoms. If their BP drop continues but their symptoms improve sufficiently, their regimen is kept unchanged. If both their BP drop and their symptoms continue, D.F. and the physician reevaluate. OH treatment can cause hypertension, but for patients whose hypertension is at most mild to moderate, D.F. feels that the risks are offset by the seriousness of OH symptoms and the clear benefits of treatment.
When D.F. prepares to treat patients with Parkinson disease, he pays extra attention to their current non-OH medications because several drugs commonly used to treat Parkinson disease can cause or worsen OH (Hohler et al, 2012; Lyons and Pahwa, 2011) (Table 2). D.F. tries to reduce doses of these drugs as low as possible or wean patients off them entirely, while monitoring to ensure that their Parkinson disease does not worsen. He then rechecks the patients’ BP. If they still have OH and signs or symptoms, he tries treating them with the same OH medicines and procedures used for patients who do not have Parkinson disease. However, because patients with parkinsonism may have either a central or a peripheral mechanism for their OH, the adjustment of OH medicines is necessarily slow, requiring frequent reevaluations of BP and symptoms.
Almost all patients with symptomatic OH can be treated. There are only a few exceptions. A rare patient or health care surrogate refuses treatment. More often, the obstacle is an underlying medical condition. Patients should not take pyridostigmine if they have asthma or bradycardia. They should not take sodium chloride tablets or fludrocortisone if they have a history of congestive heart failure. They should not take midodrine if they have peripheral artery disease or unstable angina pectoris. These patients need to have their OH managed by their regular specialist. D.F. confers with the cardiologist who is treating patients’ congestive heart failure, unstable angina, or another unstable cardiopulmonary condition. As the cardiologist alters the drug regimen, D.F. monitors the BP and symptoms. He remeasures BPs over 3 days, usually starting 5 days after a dose adjustment.
For all patients, D.F. gives new treatments time to take effect and then monitors patients’ initial response. He remeasures BP over 3 days beginning 3 days after patients start wearing compression stockings or taking midodrine, and 5 days after starting extra water, sodium chloride, fludrocortisone, or pyridostigmine. He always obtains more BP readings before altering a drug.
He continues to adjust treatment until patients are clearly much improved in both BP and symptoms, they have become symptom-free, or they no longer meet the BP criteria for OH. It can take days to as long as several weeks for treatment to end an OH episode. Until patients respond fully, their BPs and symptoms may fluctuate markedly. Most patients’ symptoms do not improve until their BP improves. However, some patients show substantial clinical improvement while their BP remains somewhat orthostatic; for these patients, D.F. may try holding the medication doses steady and following closely.
During treatment and follow-up care, D.F. continues to request a weekly update on the patient’s signs and symptoms in the form of a phone call from each office patient’s home caregivers and a fax from long-term care staff. During follow-up, he sees office patients every 3 to 6 months and long-term care patients every month, until their symptoms have substantially improved. Then he sees them every 3 months.
If symptoms reemerge with no other clear cause, D.F. performs or orders repeat BP sets for 3 days, reassesses all of the patient’s illnesses and medicines, and alters the OH medications. Again, he asks the primary care physician to change the patient’s BP and other medications that might be contributing to OH. He alters treatment similarly if symptoms fail to respond or if they worsen, or if the orthostatic drop increases. Most patients who respond to OH treatment remain on the treatment to prevent their symptoms from returning, as often happens when treatment is stopped by other out- or inpatient physicians. For the study, we did not collect data when patients off treatment had a return of symptoms.
We used the Fisher exact test to compare the prevalence of OH between patient subgroups. We used the McNemar test to determine statistically significant changes in OH signs and symptoms after treatment. We used logistic regression to evaluate which symptoms and other factors (eg, age, smoking habit, comorbidities) might have an independent relationship with OH.
We reviewed the charts of 188 patients, all with dementia: 56 (29.8%) men and 132 (70.2%) women. The average age was 80.8 years, with a standard deviation of 7.7 (range: 55 to 101 years). Ninety-six (51.1%) of the patients had OH, defined by 1 BP set meeting the criteria for OH and at least 1 of the OHSS symptoms (Appendix 1); the other 92 (48.9%) patients did not have OH.
Table 1 shows demographic and baseline clinical information for the non-OH and OH groups. The 2 groups were similar in age; sex; history of cardiovascular disease, stroke, and diabetes mellitus; and drugs known to cause or exacerbate OH. The 2 groups had no statistical differences between numbers of patients living at home and those in long-term care facilities. Patients with AD had a lower prevalence of OH than did those with non-AD dementia: We found OH in 40% of patients (40/100) with AD versus 63.6% of patients (56/88) with other types of dementia (P=0.0014).
Of the 96 patients with OH, 71 (74.0%) were taking 1 or more of over 50 medications (Table 2) known to cause or exacerbate OH. In particular, 44 (45.8%) were on antihypertensives. Of the 92 patients without OH, 59 (64.1%) were taking 1 or more of the medicines known to cause or exacerbate OH; 50 (54.3%) were on antihypertensives. The differences in medicines between the 2 groups were not statistically significant (P=0.1579), suggesting that drug type alone did not account for the OH.
We found the acute or subacute onset of mental and motor fluctuations in a majority of our patients with OH, but in a minority of those without OH. Of the 96 patients with OH, 69 (71.9%) exhibited mental fluctuations; 27 (28.1%) did not. In contrast, of the 92 patients without OH, only 29 (31.5%) exhibited mental fluctuations; 63 (68.5%) did not (P<0.0001). Of the 96 patients with OH, 40 (41.7%) had motor fluctuations; 56 (58.3%) did not. In contrast, of the 92 patients without OH, only 12 (13%) exhibited motor fluctuations; 80 (87%) did not (P<0.0001).
The logistic regression that simultaneously examined the factors of AD, mental fluctuations, and motor fluctuations for their independent associations with OH revealed that mental and motor fluctuations had the strongest relationship with OH (P<0.0001 and P=0.0231, respectively). The strength of the association between mental fluctuations and OH depended on whether the patient also had AD (P=0.0205 for the interaction term). For patients with AD, the association between mental fluctuations and OH had an odds ratio of 2.5, while for patients with another form of dementia, the odds ratio was 13.1. Other factors, including drugs known to cause or exacerbate OH, were not significantly associated with OH when we took mental and motor fluctuations and AD into account.
Table 3 shows how many of our patients with OH had particular signs and symptoms. Most of our patients’ clinical features match those in the OHSS (Appendix 1), although our group did not display everything in the OHSS. Most common in our group were mental fluctuations, sleeping in one’s chair, slow falls without loss of consciousness (adagio falls), lethargy or fatigue, and dizziness upon standing.
We compared the frequency of these 5 most common signs and symptoms before and after our patients were treated for OH. Before treatment, only 11 (11.5%) patients with OH had none of the signs and symptoms shown in Table 4, while 46 (47.9%) had 1 of the signs and symptoms and 39 (40.6%) had 2 or more (Table 3). These numbers declined significantly after treatment (Table 4).
Although our documentation of BP was incomplete, we found from the available data on the OH group that the orthostatic drop for both systolic and diastolic BP improved significantly with treatment. Table 5 shows the combined BP information. Table
6 shows that clinical improvement tended to be greater when the systolic BP improved.
Prevalence of OH
OH affected just over half of our elderly population with dementia. Showing similar trends to Allan et al (2007), we found OH in 40 of our 100 (40%) patients with AD (vs 34% in Allan) and in 7 of our 9 (77.8%) patients with parkinsonian dementias (vs 49% in Allan). Our latter percentage proved not to be statistically significant, perhaps because we had few patients with parkinsonian dementias and because our retrospective chart review may not have provided enough detail to let us document whether our patients had orthostatic BPs with compatible clinical features.
Our reported percentages of dementia types may have had some imprecision. For the patients whom D.F. diagnosed in the office, he was able to follow the most rigorous criteria; however, as noted in the Diagnosis of Dementia section above, he could not be completely sure of some diagnoses for patients in long-term care.
It is also possible that some of our patients had mixed dementias. For example, some of the reported “non-AD dementias” might have been AD mixed with other dementia types. For another example, the patients classified as having vascular dementia had presented with primarily subcortical dementia and their scans had shown significant cerebrovascular disease, but we did not collect chart information on their specific types of strokes. Thus, their vascular dementia could potentially have included cortical, subcortical, or mixed cortical-subcortical dementias.
Although our study concurred with others in finding the prevalence of OH to vary among patients with different types of dementia, we diverged from other reports in our results for prevalence in different living situations. Shibao et al (2007) reported OH in 54% to 68% of elderly Americans living in long-term care facilities, but we did not find more OH in our patients in long-term care. We see 2 possible explanations for this disagreement. The prevalence in our outpatients may have been higher than Shibao’s because D.F.’s office practice does not cover a general population of elderly people with neurologic problems. Most of his patients have dementia and/or movement disorders, and thus would be more likely to have OH. The OH prevalence in our long-term care group may have been lower than Shibao’s because we included patients in both assisted living facilities and nursing homes. Two thirds of our patients with OH had AD; in our experience, the risk for OH rises as AD becomes more advanced. If this clinical impression is true, our presumably earlier-stage patients in assisted living were less likely to have OH.
Medications have been shown to affect nondemented people’s risk for OH (Grubb and Karas, 1999; Kamaruzzaman et al, 2010; Mosnaim et al, 2010; Rutan et al, 1992); the risk in people with dementia has not been studied. Our experience is that elderly people, particularly those with dementia, may have other medical conditions and/or be on medicines that can cause or exacerbate OH. The medicines that our study patients were taking (Table 2) cover most of the main categories of drugs found in people with dementia and OH. Among the commonly implicated medicines not listed in Table 2 are the antispasticity drug tizanidine and the alpha-1 antagonist terazosin.
Our patients with and without OH took the Table 2 medications in fairly equal numbers. This finding would seem to contradict the general consensus that medicines can cause and worsen OH. Our result may reflect our small patient sample, our retrospective analysis, and our analyzing our patients’ drugs but not their dosing regimens. Further, we could not determine which of our patients might have been especially resistant or susceptible to a drug’s OH-inducing effects, and which patients might already have had OH before they started taking a drug that then worsened their orthostasis.
Clinicians need to be aware of possible medication effects. If a patient taking any of the OH-inducing drugs shows OH signs and symptoms, the clinician should review the patient’s risk factors for OH and test orthostatic BPs.
Clinical Features of OH
We used the OHSS to help confirm OH in patients whose signs and symptoms appeared and fluctuated acutely or subacutely and had no other identifiable causes.
Interestingly, only 14 of our 96 patients with OH demonstrated, reported, or acknowledged having either dizziness or swaying without dizziness. It is certainly possible that language or other impairment could limit demented patients’ ability to complain of or answer “yes” to a question about whether they have a subjective feeling like dizziness. The difficulty of detection may mean that dizziness affects many more patients than we realize. However, undetected dizziness may be of less consequence than the fact that symptomatic OH may be underrecognized in people with dementia because physicians are looking only for complaints of dizziness. Clinicians may be missing readily detectable serious and sometimes life-threatening manifestations like confusion, drowsiness, weakness on standing, and falling (Tables 3 and 4).
In patients with dementia, OH manifestations like mental and motor fluctuations, lethargy, drowsiness, and falls point to OH at least as much as dizziness does. These features are consistent with an acute or subacute confusion or delirium, and OH should be included in the differential diagnosis.
We have seen patients with OH misdiagnosed as having transient ischemic attacks, seizures, depression, and manipulative behavior. OH should be part of the differential diagnosis of these conditions, as well as the differential of what has been termed the nonfocal transient neurologic attack (Bos et al, 2007). Patients having an episode of OH can be thought to be suffering from such an attack, based on shared clinical features and a similar clinical course. After the subacute or acute onset of their OH, these patients are taken to the emergency department, where they improve after being kept supine and getting volume expansion with intravenous fluids. In fact, we believe that OH is an underrecognized cause of nonfocal transient neurologic attacks.
Patients with dementia are also put at risk if their physician recognizes their mental and motor fluctuations and other features that are common to both OH and dementia, but if the physician assumes that the symptoms represent no more than progression of the dementia, and neither suspects nor treats the OH.
Some of our patients’ clinical features can be seen in other chronic conditions, such as sleep apnea, chronic kidney or liver disease, and long-term use of medications. Unlike OH, however, the signs and symptoms of these conditions tend to develop slowly.
All such misdiagnoses can lead to costly unnecessary tests (Mendu et al, 2009) and treatments that do not help the real problem and that may accomplish little more than creating anxiety for both patient and family.
Treatment of OH
Treatment of OH brought a majority of our patients significant improvement in their signs and symptoms (Table 4). Most of the treated patients had a corresponding reduction in their measured orthostatic BP drop: 38/53 patients (71.7%) improved their diastolic BP drop and 46/55 patients (83.6%) improved their systolic BP drop.
Seven treated patients whose signs and symptoms improved did not have improvement in their orthostatic BP drop. The explanation may be 1 or more of these mechanisms: the elimination of medicines that could have been contributing to the signs and symptoms, differences in the degree and type of orthostatic drops and the timing and degree of returns to higher BP (Novak, 2011), differences in patients’ autoregulatory capacity (Low et al, 1997), and the fluctuating nature of OH (Ooi et al, 1997; Ward and Kenny, 1996).
In 16 treated patients, the orthostatic BP drop lessened but the signs and symptoms did not resolve. This finding may be explained by mechanisms similar to those noted above for patients who had clinical but not BP improvement, and by the possibility that other concurrent disorders may have contributed to these patients’ symptoms. Further, categorizing symptoms as “resolved” or “not resolved” may not have captured patients whose signs and symptoms improved but did not disappear completely.
It is our experience that treatment can shorten both the duration and severity of episodes, but we did not collect data on this, nor did we find any in the literature.
D.F. videotapes some of his patients during visits. Supplemental Digital Content videos 1 through 4 (http://links.lww.com/CBN/A35, http://links.lww.com/CBN/A36, http://links.lww.com/CBN/A37, http://links.lww.com/CBN/A38) show a patient whom D.F. treated after this study was completed. Appendix 2 describes each of 4 video clips that illustrate the patient’s dementia and clinical features of OH, and his response to treatment.
Limitations of the Study
We have already discussed most of the study’s limitations in context: lack of heart rate data, imprecise diagnosis of dementia, inconsistency in BP measurements, and missing BP recordings. The lack of a complete and fully reliable data set limited our ability to interpret our results.
Future studies will need better ways to evaluate and track OH and its response to treatment, to rule out other contributory factors. A prospective study with scaled measurements and independent observers could confirm observations and could measure improvement and its impact on patients’ quality of life.
Our purpose in this study was not to examine our patients’ baseline cognitive function or the severity of their dementia in relation to their OH, or to make any statement about whether patients with OH have consistently lower cognitive function. Instead, we tried to learn whether OH might be related to certain patterns of clinical decline, including the acute or subacute presentation of delirium or confusion. Because of our goals, as well as the retrospective nature of the study, we did not collect our patients’ neuropsychometric test data, either before or after treatment, or staging data on the overall severity of their dementias. Therefore, we cannot comment on the stage of dementia at which OH is most likely to start or on the relationship of OH signs and symptoms to any neuropsychological profile.
To our knowledge, ours is the first study both to delineate the signs and symptoms of OH systematically in multiple types of dementia, and to determine whether these signs and symptoms respond to treatment. Our results show that when patients with dementia present with acute or subacute clinical declines, manifesting any of the OHSS features or the signs and symptoms of a nonfocal transient neurologic attack, OH should be considered in the differential diagnosis. This is particularly important if the patients have marked fluctuations of mental and motor function, including falling. Our findings also suggest that carefully chosen and monitored treatment of OH can prevent serious injuries and can benefit both patients’ health and their quality of life.
Although there are legitimate concerns about OH treatment causing hypertension and incurring its risks, we found these concerns offset by the seriousness of the patients’ OH symptoms and the significant clinical improvements brought by treatment. With careful monitoring, both the hypertension and the OH can be managed successfully.
Our study raises an interesting possible correlation for dementia with Lewy bodies. The high rates of fluctuations of mental and motor function in our population, and the overlap of our patients’ signs and symptoms with those reported to be characteristic of the fluctuations of dementia with Lewy bodies (Ferman et al, 2004), suggest that the Lewy body fluctuations may in part be caused by OH. This idea is supported by the observations that fluctuations in dementia with Lewy bodies do not appear to correlate directly with the other core clinical features of the disorder (Ferman et al, 2004), and the pathophysiology of the fluctuations seems to be independent of the basal ganglia and brainstem disease that leads to the other signs and symptoms (Cummings, 2004).
Although our findings need to be confirmed with prospective longitudinal studies, we believe that orthostatic BP measurements should become a routine part of the medical evaluation of the elderly. OH is common in older people, particularly those with dementia, and BP measurements are inexpensive and often diagnostic (Mendu et al, 2009). Clinicians should be aware of the full range of OH signs and symptoms, and should evaluate them in patients with dementia, especially when the changes begin acutely or subacutely. Clinicians should also be aware of medications that can contribute to OH. Because OH may be intermittent, several sets of orthostatic BP measurements should be taken from supine to standing.
The supplemental videos show a patient with dementia who illustrates important signs and symptoms of orthostatic hypotension (OH) and a good response to treatment. D.F. saw this patient after the study was completed.
“Pete” was an 87-year-old man with a 5-year history of Alzheimer disease. He also had hypertension and a left pleural effusion, but no history of congestive heart failure. He had undergone a colon resection for colon cancer.
On June 2, Pete had a spell of confusion and weakness. The next day, a magnetic resonance scan of his brain showed multiple small-vessel cerebrovascular infarcts in hemispheric white matter, and a small area of diffusion restriction in the right pons, consistent with a more recent small-vessel infarct. A magnetic resonance angiogram performed at the same time did not show significant large-vessel intracranial disease.
On June 22, he
fell at home and was hospitalized briefly. Computed tomography scans on June 22 and 24 showed a left-sided subdural hematoma that grew to a width of 8 mm, with a mild midline shift. The medical staff decided to let the hematoma reabsorb. In the hospital Pete was confused and combative, but his behavior improved enough that he was allowed to go home.
Once home, he had an episode of worse confusion, accompanied by drowsiness, bilateral leg weakness, and slurred speech. On June 26, he was admitted to a nursing home.
The nursing home asked D.F. to evaluate Pete for confusion, sedation, restlessness, worsening unsteadiness of gait, falling, and increasing incontinence. At that time, Pete’s medications were quetiapine 50 mg in the morning and 75 mg at bedtime, memantine 10 mg twice a day, a rivastigmine patch 4.6 mg/day, sertraline 100 mg/day, clonazepam 0.5 mg every 8 hours as needed, amlodipine 10 mg/day, combined hydrocodone-acetaminophen (5 mg and 500 mg, respectively) 1 tablet/day, and folic acid 1 mg/day.
During the first evaluation, on July 16, D.F. diagnosed OH by measuring blood pressures (BPs) of 150/60 mm Hg supine and 90/40 mm Hg standing.
As a first step in trying to control the OH, D.F. weaned Pete off quetiapine over the next 6 days, weaned him off memantine over 3 days, stopped his clonazepam, and halved his sertraline dose to 50 mg/day. D.F. started Pete on fludrocortisone 0.1 mg every morning, and asked his primary care physician to try to reduce the amlodipine dose.
On July 20, D.F. started Pete on midodrine 2.5 mg at 6:00 am and 11:00 am. The primary care physician did not reduce the amlodipine dose but did prescribe olanzapine 2.5 mg twice/day as needed. Pete turned out not to need the olanzapine.
On July 22, D.F. saw Pete for the second time. His BPs had improved to 136/60 mm Hg supine and 100/60 mm Hg standing.
D.F. increased Pete’s midodrine regimen to 2.5 mg 4 times/day, at 6:00 am, 10:00 am, 2:00 pm, and 6:00 pm, and arranged with the primary care physician to withdraw the order for olanzapine.
On July 29-31, the nursing home staff measured Pete’s BP and faxed the results to D.F.:
Because of the continuing OH, on August 3 D.F. increased the daily midodrine regimen to 5 mg at 6:00 am and 2:00 pm, while continuing the 2.5 mg doses at 10:00 am and 6:00 pm. The fludrocortisone dose stayed at 0.1 mg every morning.
By August 7, when D.F. next visited Pete, his orthostatic BPs had improved. (The actual measurements have been lost.) His medications were maintained at current levels.
On August 27, D.F. saw Pete for the fourth time and again found his orthostatic BPs improved (actual measurements lost). Again, his medications were continued at current doses.
Albert ML .Subcortical dementia: historical review and personal view.Neurocase. 2005; 11:243–245.
Allan LM, Ballard CG, Allen J, et al .Autonomic dysfunction in dementia.J Neurol Neurosurg Psychiatry. 2007; 78:671–677.
. American Psychiatric Association .Diagnostic and Statistical Manual of Mental Disorders. 2000; :4th ed, text rev.Washington, DC:American Psychiatric Association; 147–165.
Arbogast SD, Alshekhlee A, Hussain Z, et al .Hypotension unawareness in profound orthostatic hypotension.Am J Med. 2009; 122:574–580.
Aung AK, Corcoran SJ, Nagalingam V, et al .Prevalence, associations, and risk factors for orthostatic hypotension in medical, surgical, and trauma inpatients: an observational cohort study.Ochsner J. 2012; 12:35–41.
Baker SP, Harvey AH .Fall injuries in the elderly.Clin Geriatr Med. 1985; 1:501–509.
Bos MJ, Van Rijn MJ, Witteman JC, et al .Incidence and prognosis of transient neurological attacks.JAMA. 2007; 298:2877–2885.
Braam EA, Verbakel D, Adiyaman A, et al .Orthostatic hypotension: revision of the definition is needed.J Hypertens. 2009; 27:976–982.
Cheshire WP, Phillips LH .Delayed orthostatic hypotension: is it worth the wait? Neurology. 2006; 67:8–9.
Cummings JL .Fluctuations in cognitive function in dementia with Lewy bodies.Lancet Neurol. 2004; 3:266
Eigenbrodt ML, Rose KM, Couper DJ, et al .Orthostatic hypotension as a risk factor for stroke: the atherosclerosis risk in communities (ARIC) study, 1987-1996.Stroke. 2000; 31:2307–2313.
Ejaz AA, Haley WE, Wasiluk A, et al .Characteristics of 100 consecutive patients presenting with orthostatic hypotension.Mayo Clin Proc. 2004; 79:890–894.
Fedorowski A, Hedblad B, Melander O .Early postural blood pressure response and cause-specific mortality among middle-aged adults.Eur J Epidemiol. 2011; 26:537–546.
Feldstein C, Weder AB .Orthostatic hypotension: a common, serious and underrecognized problem in hospitalized patients.J Am Soc Hypertens. 2012; 6:27–39.
Ferman TJ, Smith GE, Boeve BF, et al .DLB fluctuations: specific features that reliably differentiate DLB from AD and normal aging.Neurology. 2004; 62:181–187.
Freeman R .Clinical practice: neurogenic orthostatic hypotension.N Engl J Med. 2008; 358:615–624.
Gibbons CH, Freeman R .Orthostatic dyspnea: a neglected symptom of orthostatic hypotension.Clin Auton Res. 2005; 15:40–44.
Gibbons CH, Freeman R .Delayed orthostatic hypotension: a frequent cause of orthostatic intolerance.Neurology. 2006; 67:28–32.
Glodzik L, Rusinek H, Pirraglia E, et al .Blood pressure decrease correlates with tau pathology and memory decline in hypertensive elderly.Neurobiol Aging. 2013; 10.1016/j.neurobiolaging.2013.06.011.
Grubb BP, Kanjwal Y, Karabin B, et al .Orthostatic hypotension and autonomic failure: a concise guide to diagnosis and management.Clin Med Cardiol. 2008; 2:279–291.
Grubb BP, Karas B .Clinical disorders of the autonomic nervous system associated with orthostatic intolerance: an overview of classification, clinical evaluation, and management.Pacing Clin Electrophysiol. 1999; 22:798–810.
Hall JE .Guyton and Hall Textbook of Medical Physiology. 2011; :12th ed.Philadelphia:Saunders/Elsevier.
Hohler AD, Zuzuárregui JR, Katz DI, et al .Differences in motor and cognitive function in patients with Parkinson’s disease with and without orthostatic hypotension.Int J Neurosci. 2012; 122:233–236.
Humm AM, Bostock H, Troller R, et al .Muscle ischaemia in patients with orthostatic hypotension assessed by velocity recovery cycles.J Neurol Neurosurg Psychiatry. 2011; 82:1394–1398.
, Muller M, Visseren F, et al .Blood pressure and progression of brain atrophy—lower is not always better: the SMART-MR study.Alzheimers Dement. 2012; 8:suppl P89
Kamaruzzaman S, Watt H, Carson C, et al .The association between orthostatic hypotension and medication use in the British Women’s Heart and Health Study.Age Ageing. 2010; 39:51–56.
Kario K, Eguchi K, Hoshide S, et al .U-curve relationship between orthostatic blood pressure change and silent cerebrovascular disease in elderly hypertensives: orthostatic hypertension as a new cardiovascular risk factor.J Am Coll Cardiol. 2002; 40:133–141.
Kaufmann H .Consensus statement on the definition of orthostatic hypotension, pure autonomic failure, and multiple system atrophy.Clin Auton Res. 1996; 6:125–126.
Kaufmann H, Malamut R, Norcliffe-Kaufmann L, et al .The Orthostatic Hypotension Questionnaire (OHQ): validation of a novel symptom assessment scale.Clin Auton Res. 2012; 22:79–90.
Kim JS, Oh YS, Lee KS, et al .Association of cognitive dysfunction with neurocirculatory abnormalities in early Parkinson disease.Neurology. 2012; 79:1323–1331.
Lipsitz LA .Orthostatic hypotension in the elderly.N Engl J Med. 1989; 321:952–957.
Lopez OL, Swihart AA, Becker JT, et al .Reliability of NINCDS-ADRDA clinical criteria for the diagnosis of Alzheimer’s disease.Neurology. 1990; 40:1517–1522.
Low PA, Bloomfield DM, Gilden JL, et al. 1997;
Advances in the diagnosis and treatment of orthostatic hypotension and selected syncopal conditions [continuing medical education learning kit sponsored and accredited by the Dannemiller Memorial Educational Foundation, San Antonio, TX; produced by American Medical Communications, Morganville, NJ; and funded by Roberts Pharmaceuticals, Eatontown, NJ]
Low PA, Opfer-Gehrking TL, McPhee BR, et al .Prospective evaluation of clinical characteristics of orthostatic hypotension.Mayo Clin Proc. 1995; 70:617–622.
Lyons KE, Pahwa R .The impact and management of nonmotor symptoms of Parkinson’s disease.Am J Manag Care. 2011; 17:suppl 12 S308–S314.
Mader SL .Orthostatic hypotension.Med Clin North Am. 1989; 73:1337–1349.
Mader SL .Identification and management of orthostatic hypotension in older and medically complex patients.Expert Rev Cardiovasc Ther. 2012; 10:387–395.
Masaki KH, Schatz IJ, Burchfiel CM, et al .Orthostatic hypotension predicts mortality in elderly men: the Honolulu Heart Program.Circulation. 1998; 98:2290–2295.
Mathias CJ, Mallipeddi R, Bleasdale-Barr K .Symptoms associated with orthostatic hypotension in pure autonomic failure and multiple system atrophy.J Neurol. 1999; 246:893–898.
Maule S, Caserta M, Bertello C, et al .Cognitive decline and low blood pressure: the other side of the coin.Clin Exp Hypertens. 2008; 30:711–719.
McKeith IG, Galasko D, Kosaka K, et al .Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): report of the consortium on DLB international workshop.Neurology. 1996; 47:1113–1124.
Mendu ML, McAvay G, Lampert R, et al .Yield of diagnostic tests in evaluating syncopal episodes in older patients.Arch Intern Med. 2009; 168:1299–1305.
Mehrabian S, Duron E, Labouree F, et al .Relationship between orthostatic hypotension and cognitive impairment in the elderly.J Neurol S ci. 2010; 299:45–48.
Mosnaim AD, Abiola R, Wolf ME, et al .Etiology and risk factors for developing orthostatic hypotension.Am J Ther. 2010; 17:86–91.
Muller M, Gudnason V, Aspelund T, et al .The combined effect of midlife hypertension status and late-life blood pressure on brain volumes: the AGES-Reykjavik Study.Alzheimers Dement. 2012; 8:suppl P88
Muller M, van der Graaf Y, Visseren FL, et al .Blood pressure, cerebral blood flow, and brain volumes: the SMART-MR study.J Hypertens. 2010; 28:1498–1505.
Novak P .Assessment of sympathetic index from the Valsalva maneuver.Neurology. 2011; 76:2010–2016.
Ooi WL, Barrett S, Hossain M, et al .Patterns of orthostatic blood pressure change and their clinical correlates in a frail, elderly population.JAMA. 1997; 277:1299–1304.
Raiha I, Tarvonen S, Kurki T, et al .Relationship between vascular factors and white matter low attenuation of the brain.Acta Neurol Scand. 1993; 87:286–289.
Robertson D, Kincaid DW, Haile V, et al .The head and neck discomfort of autonomic failure: an unrecognized aetiology of headache.Clin Auton Res. 1994; 4:99–103.
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; 114:630–636.
Rutan GH, Hermanson B, Bild DE, et al .Orthostatic hypotension in older adults: the Cardiovascular Health Study: CHS Collaborative Research Group.Hypertension. 1992; 19:508–519.
Shibao C, Grijalva CG, Raj SR, et al .Orthostatic hypotension-related hospitalizations in the United States.Am J Med. 2007; 120:975–980.
Siennicki-Lantz A, Lilja B, Elmstahl S .Orthostatic hypotension in Alzheimer’s disease: result or cause of brain dysfunction? Aging. 1999; 11:155–160.
Sonnesyn H, Nilsen DW, Rongve A, et al .High prevalence of orthostatic hypotension in mild dementia.Dement Geriatr Cogn Disord. 2009; 28:307–313.
Streeten DH, Anderson GH Jr .Delayed orthostatic intolerance.Arch Intern Med. 1992; 52:1066–1072.
Tilvis RS, Hakala SM, Valvanne J, et al .Postural hypotension and dizziness in a general aged population: a four-year follow-up of the Helsinki Aging Study.J Am Geriatr Soc. 1996; 44:809–814.
van Nieuwenhuizen RC, van Dijk N, van Breda FG, et al .Assessing the prevalence of modifiable risk factors in older patients visiting an ED due to a fall using the CAREFALL Triage Instrument.Am J Emerg Med. 2010; 28:994–1001.
Ward C, Kenny RA .Reproducibility of orthostatic hypotension in symptomatic elderly.Am J Med. 1996; 100:418–422.
Wieling W, Schatz IJ .The consensus statement on the definition of orthostatic hypotension: a revisit after 13 years.J Hypertens. 2009; 27:935–938.