The Roles of Orthostatic Hypotension, Orthostatic Tachycardia, and Subnormal Erythrocyte Volume in the Pathogenesis of the Chronic Fatigue Syndrome

In 1995, Rowe et al 1 and Bou-Holaigah et al 2 published convincing evidence that delayed orthostatic hypotension was followed by syncope or presyncope after head-up tilt for an average of 20.9 minutes in 22 of 23 patients with well-documented chronic fatigue syndrome (CFS). In these patients, treatment with fludrocortisone and other medications commonly, although not invariably, reduced the orthostatic hypotension and significantly ameliorated the fatigue. These findings have been confirmed by other groups of investigators. 3-6 Acute experimental blood loss by venesection has been shown to cause orthostatic hypotension in proportion to the volume of blood removed 7,8 and orthostatic hypotension, which is commonly seen in patients with subnormality of plasma volume (after fluid loss from diuresis or diarrhea), erythrocyte mass (in severe anemias), or total blood volume (after hemorrhage), can be overcome consistently by correction of the specific (plasma, erythrocyte, or whole blood) volume deficit. 9,10 There is also evidence that many patients with idiopathic orthostatic hypotension of both the hyper- and the hypoadrenergic types have subnormal circulating red cell volume 9,11 and that restoration of the subnormal red cell mass with erythropoietin therapy often improves the orthostatic hypotension significantly. 11,12

These facts led us to compare changes in blood pressure during prolonged standing (up to 89 minutes in these studies) in patients with CFS and for 60 minutes in healthy control subjects. We also studied the effects of lower body compression on orthostatic changes in blood pressure, heart rate and symptoms, and measured the circulating erythrocyte and plasma volumes by standard isotopic dilution methods in this group of patients.

Fifteen subjects were tested, all of whom met the diagnostic criteria of the Center for Disease Control for CFS. 13 The patients had been followed by one of us (DSB) for several years and had undergone extensive medical evaluations. All had completed depression questionnaires and were questioned about their mood at each follow-up visit. Although some subjects had experienced periods of depression during their illness, none were depressed at the time of the orthostatic testing, and no subject was considered to have major depression as an important factor restricting his/her activity. The patients were randomly selected from those who happened to return for follow-up as the current studies were in progress. None had been found to have blood pressure abnormalities on routine physical examination in the sitting or reclining posture prior to the studies here reported.

Twelve subjects were women, with an average age of 34.3 (range, 18-48) years. In the 3 men, mean age was 34.6 (range, 16-49) years. All of these persons completed a questionnaire related to their activity and severity of symptoms before the study. Six patients had severe restriction of activity because of fatigue and/or orthostatic intolerance that had reduced their upright activity to 2 hours or less on most days. The average duration of daily activity for the 12 women was 2.0 hours and for the 3 men, 6.2 hours. Visual analog scores (0 for none, 10 for severe) were totaled for the 12 symptoms commonly seen in CFS, and averaged 77 (range, 43-115). Fisk Fatigue Impact Scale 14 ranged from 73 to 157, with an average of 122.4. A modification of the Karnofsky performance score 15 averaged 27.7. This would correspond to: (a) moderate to severe symptoms at rest, (b) inability to leave the house except rarely, (c) very severe symptoms with exertion, and (d) inability to perform strenuous tasks. All of these questionnaire parameters at the time of the orthostatic testing would classify this group of patients as having severe CFS.

Fifteen age and gender matched control subjects [12 women and 3 men, aged 19 to 41 (mean, 27.8) years] were healthy medical students, nurses, and laboratory technicians. They underwent blood pressure and heart rate measurements every minute for 30 minutes in recumbency and for 60 minutes in the standing posture, contemporaneously with or a few months before the studies on the patients.

A full history was obtained and a physical examination was performed in the Clinical Research Unit of the University Hospital between 8:30 am and 9:30 am, at a room temperature of 20 to 22°C. An indwelling intravenous cannula, kept patent with 0.9% saline, was placed in a forearm vein at about 9:30 am. The patients' blood pressures (BP) and heart rates (HR) were then measured with a Dinamap (Critikon Co., Tampa, FL) every minute in recumbency for 30 minutes and while standing at the bedside with the arm on which the BP was measured resting on a support at the level of the second intercostal space. Leg movements were discouraged but occasional movements were not forbidden. Because of the previous finding that a delayed form of orthostatic hypotension (after standing for more than 10 minutes) occurred in patients complaining of severe fatigue, 16 the standing posture was maintained until the mean of measurements during one of the successive 10-minute periods of standing revealed at least one of the changes previously shown to be abnormal in single measurements 17 comprising (1) a fall in systolic BP ≥ 20 mm Hg, (2) a fall in diastolic BP ≥ 10 mm Hg, (3) an increase in heart rate by ≥ 27 or to ≥ 108 bpm, (4) a fall in systolic BP to below 90 mm Hg, (5) presyncopal symptoms or actual syncope occurred, or (6) at least 60 minutes had elapsed. During the measurements conversation was limited to spontaneous, brief descriptions of untoward symptoms as they occurred. Before resuming recumbency the color of the patients' legs was compared in a good light with that of a healthy subject (physician or nurse) who had remained standing at least as long as the patient.

Blood was drawn for measurements of plasma catecholamines after recumbency for 30 minutes and after standing for 10 minutes and 30 minutes in both groups of subjects. After the conclusion of the orthostatic measurements, the patients lay down for approximately 10 minutes while the lower limbs, pelvis and abdomen were enclosed in a military antishock trousers (MAST) suit (David Clark Co, Worcester, MA). The patients then stood again at the bedside, with the suit deflated for 10 minutes, after which the suit was inflated to 45 mm Hg for another 10 minutes, during BP and HR measurements every minute by Dinamap. Heparinized blood (4 ml) was drawn after 30 minutes of recumbency and at least 10 minutes of standing for determinations of plasma norepinephrine and epinephrine concentrations by HPLC separation and electrochemical measurements. 18

The clinical studies were followed about 2 hours later by measurements of circulating red blood cell volume and plasma volume. Five milliliters of each patient's heparinized blood, collected in a sterile vacutainer tube, was tagged with ⁵¹CrCl₃ (40 μCi). The radioactivity levels in aliquots of the blood before and 20 minutes after injection of the labeled blood were compared in a Volemetron for calculation of red blood cell volume. This determination was followed by calculation of the plasma volume, also with the Volemetron, on blood drawn before and 20 minutes after intravenous injection of a known amount of ¹²⁵I-labeled human serum albumen (2.5 μCi). 19,20

Statistical Methods included student's t tests, paired and unpaired, and comparisons of 95 or 99% confidence intervals. 21 The research procedures were performed after the subjects had signed a written consent document approved by the SUNY Syracuse Institutional Review Board for the Protection of Human Subjects.

In the healthy control subjects, there were small (usually <10 mm Hg) fluctuations in BP during successive 10-minute periods in the standing posture. The mean (and SD) of BP for the 15 subjects changed from 116.4/66.8 (SD 13.1/8.7) during the last 10 minutes of recumbency to 121.8/76.4 (SD 9.0/7.5) mm Hg in the last 10 minutes of standing (P > 0.05), with no evidence of a consistent upward or downward trend (Table 1). The heart rate showed a consistent, slight rise from recumbency (mean, 71.9; SD, 8.4 bpm) to 82.0 (SD 12.8) bpm in the last 10 minutes of orthostasis (P < 0.01 by t-test for paired variables) and no evidence of a progressive tachycardia as the subjects continued to stand. It is evident that 2 normal women experienced transiently abnormal reductions in BP during the first and second 10-minute periods of standing (patient 4), and during the second to fifth 10-minute periods of standing (patient 9), with improvement toward the end of the study in both subjects. However, neither of these subjects was symptomatic before or throughout the study.

Table 1. Healthy Subjects: Orthostatic Changes in BP and HR

The 15 patients with CFS were all normotensive while reclining (Table 2). It is evident from Table 2 that systolic BP fell within the 60 minutes or less of standing by ≥ 20 mm Hg below the same individual patient's recumbent value in 7 patients (patients 1, 2, 3, 5, 7, 13, and 15) and to below 90 mm Hg in another 2 patients (patients 8 and 14), while diastolic BP fell excessively in 1 more (patient 10), a total of 10 out of the 15 patients. Heart rate rose excessively in 3 of these (patient 2, 3, and 5) and in one other patient (patient 9) whose BP did not fall excessively. Thus, 11 of the 15 patients (73%) experienced excessive BP or HR changes while standing for up to 60 min.

Table 2. Patients with CFS: Orthostatic Changes in BP and HRLast Standing Individual Measurements and Final Changes in bold indicate (1) abnormal orthostatic reductions in systolic BP (by ≥20 mm Hg) or to below 90 mm Hg in the absence of a ≥20 mm Hg reduction from baseline, or (2) abnormal orthostatic reductions in diastolic BP (by ≥10 mm Hg), or (3) abnormal increases in HR (by ≥27 bpm or to ≥108 bpm). BP in pt 1 fell to 82/40 after 9 min of standing; pt #11 refused to continue the measurements in the standing posture beyond 30 min, despite absence of abnormal symptoms.

Although no direct comparison can be made with the findings in our healthy control subjects who were studied while standing for a maximum of 60 minutes, 2 of the remaining 4 patients, after standing for more than 60 minutes, did experience changes that cannot be considered normal. Thus, patient 4 experienced excessive tachycardia with an increase in heart rate from 72 to 100 bpm associated with presyncopal symptoms after standing for 62 minutes, whereas in patient 6, SBP fell from 131/66 to 94/59 associated with severe lightheadedness after standing for 89 min. Patient 11 felt unable to stand for longer than 35 minutes because of increased fatigue and requested that the study be terminated, despite orthostatic BP and HR changes that did not exceed normal limits, and patient 12 was able to stand for 70 minutes without abnormal orthostatic changes.

Comparison of the final BP and HR changes in the patients and these changes after 60 minutes of standing in the healthy subjects showed significantly excessive orthostatic reductions in systolic BP (P < 0.001) and diastolic BP (P < 0.001) and increases in HR (P < 0.01) in the patients. Presyncopal manifestations requiring resumption of the supine posture occurred in 11 of the 15 (73%) patients when orthostasis had continued for between 9 and 60 minutes and in none of the control subjects. Inspection and comparison of the lower limbs of the patients with those of a normal subject at the conclusion of the period of orthostasis revealed, in every patient, mild to moderate cyanosis that disappeared within about 2 minutes of the resumption of recumbency. These changes were not looked for in the 15 control subjects.

During the second period of recumbency and orthostasis, the patients were invited to describe any symptoms that they might experience both during the initial deflation and the later inflation of the MAST. The MAST was not available because of the need for repairs to a leak when 4 of the patients were studied.

As is evident from Table 3, during the initial 10-minute period of orthostasis with deflation of the MAST, the mean standing systolic BP fell by at least 20 mm Hg or to below 90 mm Hg in 6 of the 11 patients, whereas HR rose by > 27 or to at least 108 bpm 17 in 2 of these 6 patients (#2 and 6) and in 3 other patients (patients 3, 4, and 7). Thus, nine of the 11 patients experienced excessive orthostatic hypotension and/or excessive tachycardia during only 10 minutes of standing. These changes are reflected in the significant mean orthostatic changes in the group as a whole in systolic BP (P < 0.001), diastolic BP (P < 0.02), and heart rate (P < 0.001).

Table 3. Effect of Lower Body Compression with MAST on Orthostatic BP and HR Changes in PatientsSignificance of differences from recumbent values by paired t test:a P < 0.001;b P < 0.02;c P < 0.05).L, lightheadedness; F, fatigue; N, nausea; SOB, short of breath; Y, yawning repeatedly; BV, blurred vision; CT, can't think clearly.

Inflation of the pressure suit to 45 mm Hg increased the standing systolic BP and HR in every patient and statistically significantly in the group as a whole (P < 0.001), so that the mean orthostatic changes in systolic BP (-4.0 mm Hg) and heart rate (+4.9 bpm) while standing in the inflated MAST were no longer statistically significantly different from those of the untreated control subjects after standing for 10 minutes (ΔSBP, 0.8 mm Hg; ΔHR, +13.0 bpm). Symptoms described by the patients while standing with the MAST deflated (Table 3) included lightheadedness (6 of 11), fatigue (6 of 11), nausea (5 of 11), shortness of breath (6 of 11), and chest pain (2 of 11), and one patient experienced presyncope. When the MAST was inflated for 10 minutes, all symptoms were completely relieved within 3 to 5 minutes in 6 of 11 (55%) patients, improved in 4 of 11 (36%), although one patient experienced no noticeable symptomatic change. The one patient who had become presyncopal while standing with the MAST deflated recovered within 2 minutes of inflation of the suit while she continued to stand and felt considerably improved. Another patient who had complained of impaired capacity to think when the suit was deflated, volunteered that she could think clearly again when the suit was inflated. None of these comments were made in response to direct questioning; all were spontaneous except in one patient who volunteered no comments and stated that she had noticed no change when she was questioned after completion of the study.

An unexpected finding was made by calculation of the mean orthostatic changes during the 2 periods of standing for 10 minutes after recumbency (shown in Tables 2 and 3) in the 11 patients who participated in the MAST study. The orthostatic changes during the 10 minutes preceding the MAST inflation (shown in Table 3) were significantly larger than those changes in the corresponding first 10 minutes of standing in the measurements shown in Table 2, not only for systolic BP and heart rate (P < 0.01 for both) but also for the diastolic BP (P < 0.05).

Results of the measurements of RBC, plasma, and whole blood volumes reported in 101 healthy women 15 have been compared in Table 4 with the measurements in our 12 women with CFS. The RBC findings in 11 of the 12 patients fell below the mean of the measurements in the control subjects, and the 99% confidence intervals (CIs) of the 2 groups of subjects showed no overlap, indicating statistical significance (P < 0.01). In contrast with the measurements of RBC volume, both the plasma volumes and the whole blood volumes in the CFS patients revealed overlapping 95% CIs with the normal data. Our measurements in the 3 men with CFS showed that the RBC volume in 1 of the 3 patients (811 mL/m²) fell below the mean -2 SD in the 201 healthy men reported by Wennesland et al. 20

Table 4. Comparison of Mean RBC, Plasma and Whole Blood Volumes in Healthy Women and Women with CFS (all in mL/m2)

Compared with the plasma norepinephrine concentrations in 15 healthy subjects, these patients' concentrations were normal in recumbency but significantly elevated (P = 0.05) after standing for 10 minutes (Table 5). Plasma epinephrine measurements and changes (not shown) were all normal.

Table 5. Plasma Norepinephrine Concentrations (pg/mL)F, female; M, male.

Previous reports of the high prevalence of delayed orthostatic hypotension associated with orthostatic tachycardia or of excessive orthostatic tachycardia alone in tilt-table studies on patients with CFS 1-6 have been confirmed in this group of patients during the more physiological condition: standing. The finding of significantly elevated plasma norepinephrine concentrations after standing for 10 minutes confirms the measurements of Jacob and Biaggioni 22 in patients with chronic orthostatic intolerance, some of whom had CFS, and was associated with increasing orthostatic tachycardia as the blood pressure fell in the upright posture. These findings clearly indicate that the orthostatic changes were of the hyperadrenergic type and did not result from global autonomic insufficiency of the type seen in multiple system atrophy (the Shy-Drager syndrome). 23,24 The findings also indicate that the orthostatic fall in BP and rise in HR in CFS are seldom, if ever, associated with baroreceptor failure. Prevention of orthostatic venous pooling by compression of the lower body at 45 mm Hg restored the orthostatic changes in systolic BP and heart rate near the magnitude of these changes seen in healthy subjects-as it does in almost all patients with orthostatic hypotension of any type. 9,24,25 This consistent evidence and the orthostatic blue discoloration of the legs 4,17 strongly suggest that excessive orthostatic venous pooling was an important pathogenetic component in 73% of the patients we are reporting here and that we have reported in abstract in a larger group of patients. 12 The occurrence of orthostatic cyanosis exclusively of the lower limbs, also a common phenomenon in other patients with orthostatic hypotension, 17 is difficult to attribute to any disorder other than excessive accumulation of blood in the lower limbs. This interpretation is supported by measurements of radiation emanating from autologous erythrocytes labeled with 99mTc pertechnetate in the legs during recumbency and standing in a previously studied group of patients with orthostatic hypotension. 17

Both the relatively small number of patients studied and the moderate or severe illness level in our patients are distinct limitations of this study. Clearly, these findings cannot be extrapolated to all patients with CFS and need to be replicated in larger groups of patients, including subjects with less severe illness and with sedentary control subjects or persons whose activity level is measurably reduced. However, the present findings are probably relevant to the profound fatigue that is very common, if not invariable, in patients with chronic, severe hypocortisolism. This condition, in the experience of one of us, is almost always associated with striking orthostatic hypotension, and both the orthostatic hypotension and the profound fatigue are dramatically abolished by appropriate steroid replacement therapy. 26 It is also impossible to rule out the possibility that chronic deconditioning may well play a role in the pathogenesis of the orthostatic abnormalities in CFS, but Freeman and Komaroff 27 concluded that deconditioning was unlikely to be the sole pathogenetic process responsible for this disorder.

Somewhat against the possible role of deconditioning has been our finding that patients with CFS usually fail to adhere to conditioning programs because of the common experience that exercise consistently aggravates their symptoms for several days.

Subnormality of the volume of circulating erythrocytes has been shown to be another frequent but not invariable finding in patients with CFS. This abnormality, too, is common in patients with acute orthostatic hypotension (which is evident within 1 to 5 minutes of orthostasis). 9,28,29 The reduced volume of circulating erythrocytes in CFS patients is unlike that in simple iron-deficiency anemia but very much like the anemia of primary autonomic failure, 30 in that it was associated with no increase in plasma erythropoietin in our patients (data not shown). The combination of lower body venous pooling during orthostasis and lack of an appropriate erythropoietin response to reduced red cell mass, as in patients with diffuse autonomic insufficiency, suggests the possibility that impaired autonomic innervation, not only of the lower body veins, 23,24 but also of the kidneys, may be responsible for these abnormalities.

It is well known that patients with CFS experience impairment of cognitive function together with their fatigue. It seemed almost unbelievable to us that some of our patients, including one in this group, volunteered reduction or complete loss of fatigue as well as the ability to think clearly again, within 10 minutes of restoring the subnormal blood pressure and excessive tachycardia to normal by lower body compression. However, the frequent and rapid symptomatic improvement during lower body compression seems to indicate that excessive gravitational pooling of blood was causing not only the orthostatic hypotension and tachycardia, but also the impaired cognitive function and fatigue in at least some of these patients. Therefore, the rapid symptomatic improvement that resulted from prevention of excessive venous pooling in the lower body, strongly suggests that increases in cardiac output and cerebral blood flow were the mechanisms of the observed and described improvements. We have not yet been able to determine whether the use of custom-fitted elastic leotards-expensive items not usually covered by medical insurance, and rather intolerable in warm weather -might be therapeutically helpful in an air-conditioned environment, as our MAST results suggest.

Several authors have attributed CFS to clinical depression. It is likely that in some patients, depression may be the primary cause of symptoms expressed as fatigue. However, it is certainly reasonable to believe that if CFS were the primary disorder, it would secondarily result in severe depression in most people. It would be interesting to know how many healthy subjects would fail to experience depression when incapacitated by fatigue and by the consequent inability to work, to associate with colleagues and friends, and to enjoy an active life. A careful evaluation of autonomic and psychiatric components in the pathogenesis of CFS has led Freeman and Komaroff 27 to conclude that antecedent depression, anxiety, or dysthymia could not be correlated with any of the measures of autonomic dysfunction that they found in patients with CFS.

It is tempting to speculate from these and previous observations 1-6 that cerebral hypoxemia, consequent upon gradually progressive lower body venous pooling and a subnormal circulating red cell volume, may be the cause of symptoms in CFS in many patients. Confirmatory, direct evidence of orthostatic reduction in the flow of adequately oxygenated blood to the brain has recently been reported in abstract form. 31