Orthostatic intolerance is a common medical problem in Japanese children and adolescents (1, 2). Attention to this problem has also been paid recently in Western countries (3), but prevalence of this illness seems to be much higher in Japanese children than in their Caucasian counterparts probably because of the racial difference in cardiovascular control (4).

Patients with orthostatic intolerance have a variety of symptoms, such as recurrent dizziness, chronic fatigue, headache, and syncope, that result in impairment of quality of life in proportion to the severity of the illness, but a significant decrease in BP during standing is often not detected by conventional measurements as defined by Robertson (5).

Dambrink et al. (6) and Wieling and Shepherd (7) reported that a large initial decrease of BP upon standing was observed in healthy individuals with initial orthostatic dizziness. We have previously reported that 21 (62%) of 34 children with recurrent syncope and chronic fatigue had a pronounced decrease in BP with a delayed recovery during the initial phase of standing (0–1 min) when BP was measured with a noninvasive beat-to-beat BP monitoring system (Finapres) (Fig. 1) (8). The initial decrease of BP was found to be closely related to orthostatic symptoms, thus causing impairment of quality of life in these patients. The origin of this disorder seems to be an impaired sympathetic activation of resistance vessels because L-threo-3,4-dihydroxyphenylserine, a newly developed precursor of natural noradrenaline, is effective in increasing BP in these patients (9). We named this disorder “instantaneous orthostatic hypotension” (INOH) (10).

Figure 1
figure 1

Original recordings of HR and finger arterial BP in (a) healthy children, (b) children with INOH of group I, and (c) children with INOH of group II. Arrow S indicates the onset of active standing;arrow ID, initial decrease;arrow OS, overshoot;arrow rp, recovery peak;HR max, maximal increase in HR;HR min, initial decrease in HR (see Ref.12);bpm, beats per minute;7 min, 7 min of standing.

To establish the criteria for INOH, we determined the normal limit of beat-to-beat BP responses to active standing in 140 healthy children aged 6–18 y (11). During a recent 6-y period, we identified 44 children with INOH according to this criteria.

In the present study, we present the clinical characteristics and cardiovascular responses to standing in these patients. Moreover, we discuss the mechanisms responsible for INOH, including catecholamine responses to orthostatic stress.


Study subjects.

We investigated 228 consecutive children and adolescents referred to the Department of Pediatrics, Osaka Medical College, from January 1991 to December 1996 with a diagnosis of suspected orthostatic intolerance, i.e. patients with three or more of the following symptoms for >1 mo: recurrent dizziness, chronic fatigue, headache, morning tiredness, near syncope, palpitation, nausea, motion sickness, recurrent abdominal pain, and loss of appetite (2).

Evaluation of orthostatic intolerance.

In all patients, the following evaluation was done:1) general physical examination including neurologic examination, chest x-ray, and 12 lead ECG, and blood tests including hematologic analysis, serum electrolyte, serum cortisol, and serum thyroxine;2) child general health questionnaire and structured interview including orthostatic symptoms and quality of life of patients and their parents;3) orthostatic test (active standing test); and 4) plasma catecholamine levels in 45 hospitalized patients.

Orthostatic test.

The test protocol was performed in the morning in a soundproof room with the temperature between 23–25°C. For details of the method, see Yamaguchi et al. (11). The experimental settings for patients in the hospital, including the devices, room temperature, and testers (HT and HY), were almost identical to those previously reported for control subjects at school. The orthostatic maneuvers were always supervised by two of our authors (HT and HY) in the same manner. Food intake was not allowed for at least 2 h before the study, and caffeine-containing products had to be avoided for at least 12 h before the study. Digital arterial BP was continuously recorded with the Finapres device (Ohmeda, model 2300). Patients were quietly seated for 15 min in a waiting room before the actual measurement started. BP and HR were recorded during 7 min in the supine position and during 7 min in the upright posture. ECG was also monitored. Before the actual standing test, patients were asked to stand up quickly by themselves from the supine position. The active standing motion was usually completed within 3–4 s. When patients showed near-fainting symptoms of dizziness, nausea, sweating, and pallor, the test was immediately stopped and the patients were returned to the supine position. One of our testers always helped keep the cuff constantly at heart level during arising and the upright position and assisted patients in lying down whenever they fainted.

The present study was approved by the Ethical Committee of Osaka Medical College Hospital.

Data collection and analysis.

Beat-to-beat digital signals of SAP, DAP, MAP, and HR were directly fed into a personal computer (NEC 98 NOTE SX, Tokyo) in real-time via an RS232C interface using an analysis program (developed by NEC San-ei Co., Tokyo). Data were collected during the whole procedure.

Data analysis focused on BP and HR changes in the supine position, the initial phase response (0–30 s after standing), and the later stage (3–7 min) as described in previous studies (11, 12). Briefly, we determined the basal level of SAP, DAP, MAP, and HR, averaged for 30 s before standing, and their changes at the following points (Fig. 1): at the initial BP decrease (usually after approximately 10 s), at the overshoot, and at every minute by averaging consecutive beats over a 10-s period. When the overshoot was not observed in patient groups, circulatory variables at 30 s after standing were used instead because a recovery peak of BP (Fig. 1) was observed at 28 s on average in patient groups.

Forty-five patients needed hospitalization for 1 wk to 2 mo because of severe impairment affecting daily life, such as school absenteeism. Informed consent was obtained from all children and their parents.


Inclusion criteria for children with INOH were determined in our previous study (11). Determination of the normal limits of BP and HR (95% confidence intervals, shaded areas in Fig. 2) was made from the data of healthy adolescents, which had higher specificity than that of preadolescents. Our criteria for INOH are the following:1) three or more of the following symptoms for >1 mo: recurrent orthostatic dizziness, chronic fatigue, headache, morning tiredness, near syncope, palpitation, nausea, sleep disturbance, low-grade fever, loss of appetite;2) a large and prolonged arterial pressure decrease appearing immediately after active standing and evaluated using a noninvasive finger arterial pressure monitoring system (Finapres);3) absence of underlying diseases that potentially could cause orthostatic hypotension, such as dehydration, epileptic seizures, neuropathy, endocrinologic disorders including adrenal insufficiency and hypo- or hyperthyroidism, heart diseases such as hypertrophic cardiomyopathy, congenital heart block, etc. We divided the subjects into two groups (Fig. 1). Group I fulfills either of the two following subcriteria:a) a 60% or greater decrease in MAP with recovery time for MAP of >20 s;b) <60% decrease in MAP with recovery time for MAP of >25 s. Group II shows reduction in SAP of >15% during the later stage of standing (3–7 min after standing) in addition to the abnormality of group I. (According to the data from our controls, the normal limit of SAP reduction during the later stage of standing was determined to be 15%.) The false-positive rate of the standing test with this criteria was found to be 4.3% in our healthy controls (six of 140 subjects aged 6 to 18 y) (11).

Figure 2
figure 2

Changes of HR (C-HR), SAP (C-SAP), and DAP (C-DAP) upon standing in controls (shaded area), group I (open circles), and group II (closed circles) at the ID, OS, and at 1, 3, 5, 7 min of standing. C-HR at ID and OS corresponds to HR max and HR min, respectively (see Fig. 1). Four patients who were near fainting were excluded from the data of 5 and 7 min. Circles with error bars indicate mean ± SE. Difference between group I and II (p< 0.001, p< 0.05). Shaded areas indicate the normal limits (95% confidence interval) as previously reported. An arrow indicates SAP level reduced as much as 25 mm Hg from the supine level (see text).

In patients who needed hospitalization, the first orthostatic test for making a diagnosis of INOH was performed before hospitalization to avoid the influence of prolonged bed rest on the results of the orthostatic test.

Plasma catecholamine.

We investigated the plasma catecholamine level in patients with INOH on another day because intravascular catheterization possibly reduces orthostatic tolerance. Plasma noradrenaline and adrenaline levels were measured in nine patients in group I, in five in group II, and in 23 symptomatic patients with normal BP response to the orthostatic test (group NBP) who were hospitalized. In these patients, group categorization determined by the first orthostatic test was not changed by this test . A catheter was inserted into the right cubital vein 30 min before the orthostatic test. Blood sampling was performed in the supine position 1 and 5 min after standing. Plasma catecholamine levels were determined by HPLC with electrochemical detection (13).


Data are given as mean ± SD unless otherwise noted. The hemodynamic variables were analyzed by repeated-measures ANOVA. When significant p values were obtained, differences between specific means were determined by using a post hoc test (Fisher's protected least significant difference). The difference in plasma catecholamine levels was analyzed by a nonparametric method (Mann-Whitney test). A p value <0.05 was considered to indicate a significant difference.


Of 228 children with suspected orthostatic intolerance, 44 fulfilled the above diagnostic criteria for INOH: 30 in group I and 14 in group II. The demographic details of children of both groups are summarized in Table 1.

Table 1 Demography of children with INOH SAP, DAP, and HR at rest, before the orthostatic test. Criteria for group I: either (a) a 60% or greater decrease in MAP with a recovery time for MAP of >20 s, or (b) <60% decrease in MAP with a recovery time for MAP of >25 s. Group II shows reduction in SAP of >15% during the later stage of standing (3–7 min after standing) in addition to the abnormality of group I.

Clinical characteristics of INOH.

All patients had various somatic complaints such as dizziness, light headedness, blurred vision, syncope, headache, chronic fatigue, etc.; these complaints were evaluated by the general health questionnaire used in our previous study of a health survey at school, as shown in Table 2. The complaints, except abdominal pain, that are listed in Table 2 were more frequently seen in patients with INOH than in controls (803 Japanese school children, aged 10–15 y). Chronic fatigue was noted in the majority of patients with INOH (40 patients, 91%) and was a most disabling complaint in 26% of the patients. Chronic fatigue for >1 mo was also observed in 107 of 228 patients. Consequently, INOH accounted for 37% (40 of 107) of the patients with chronic fatigue. Most patients (73%) also had sleep disturbance, such as impaired sleep induction and insomnia; 37% had low-grade fever (<38°C). No significant difference in these symptoms was observed between group I and group II. Two patients had epilepsy that was well controlled by medication, three had occasional mild asthma attacks, and three were anorexic in the recovery phase. No possible cause of orthostatic hypotension, including dehydration, was found in these patients with normal hematocrit and BUN. No significant difference was found in gender, body weight, and height between group I and group II.

Table 2 Positive rate (%) of complaints in 44 children with INOH Control data are cited from our previous study in 803 healthy Japanese children aged 11–15 y (cited from Ref.26).

The onset of INOH was not acute in most cases: 39 patients (89%) had main physical symptoms that gradually appeared without preceding episodes of other illness. In only one patient did physical symptoms acutely appear 1 wk after the onset of upper respiratory infection, and four patients had preceding febrile illness without apparent symptoms of the common cold.

The combination of the above symptoms associated with orthostatic intolerance was considerably disabling in the patients and, thus, caused impaired quality of life. Twenty-three patients (52%) were school absentees for >1 mo because of their symptoms. Ten patients were completely absent from school, and the remainder were often absent. School absentees were more frequently found in group II (nine of 14) than in group I (14 of 30), but the difference did not reach statistical significance by χ2 analysis.

Circulatory profile.

All patients had supine BP and HR within the normal limits; there was no difference between group I and group II (Table 1). In normal subjects, SAP and DAP showed a decrease followed by immediate recovery at the onset of standing, forming a transient overshoot to a level slightly higher than in the supine position (os in Fig. 1a). Group I and II, on the other hand, showed a marked pressure decrease (−55 ± 14/−28 ± 9 and −55 ± 10/−26 ±7 mm Hg for SAP/DAP, respectively) (Fig. 2). BP recovery after an initial decrease was apparently delayed or absent in the patient groups. Delayed recovery time of >60 s was observed in 10 of 30 patients in group I and more frequently in group II (11 of 14, p< 0.05 by χ2 analysis). Consequently, in these patients without recovery, the BP trace showed a small peak below the basal level instead of forming the overshoot (Fig. 1, b and c).

All patients had an immediate increase in HR of 22 to 67 beats per minute, and its mean was over the normal limits at the onset of standing. HR in group II remained at significantly higher levels at the overshoot (30 s) and at 1 min than in group I, resulting in the disappearance of the dip (HR min, Fig. 1a) in group II.

The averaged response of BP during the later stage of standing was stable in normal subjects, approximately 5 and 10 mm Hg higher than the basal level for SAP and DAP, respectively (11). BP in group I showed a gradual increase during 1–3 min of standing, whereas that in group II still remained low, as much as approximately 25 mm Hg, for SAP 3 min after standing (Fig. 2, arrow), respectively. In controls, the level of DAP was higher than that in the supine position, whereas the patients in group II had persisting decreased levels below the basal level. Four patients (two in group I and two in group II) showed impending syncope due to vasodepressor attack during the later stage.

Increases in HR during the later stage in patients were higher compared with the reference value. Group II showed a much higher increase in HR than group I. Orthostatic tachycardia (>35 beats per minute, the upper limit of the 95% confidence interval at 3 min of standing) was observed in 11 of 30 in group I and more frequently in group II (9 of 14).

Plasma catecholamines.

There was no significant difference in plasma noradrenaline (146 ± 87, 197 ± 122, and 148 ± 92 pg/mL), adrenaline (45 ± 33, 34 ± 23, and 38 ± 23 pg/mL), and dopamine (12 ± 12, 7 ± 13, and 13 ± 14 pg/mL) levels in the supine position among group I, group II, and group NBP, respectively. Responses of plasma noradrenaline at 1 min of standing were significantly lower in group I and II than in group NBP (p< 0.01) (Fig. 3). At 5 min of standing, responses of noradrenaline in group II were still lower than those in group NBP (p< 0.05), although those in group I did not significantly differ from those in group NBP. No significant differences of orthostatic changes in the plasma adrenaline and dopamine level were observed.

Figure 3
figure 3

Increases in plasma noradrenaline in group NBP, group I, and group II at 1 and 5 min during standing. Bars with error bars indicate mean ± SE. Difference between controls and group I and II, p< 0.05.


The majority of the patients (38 of 44) were treated with oral medication, either of the noradrenaline precursor (L-threo-3,4-dihydroxyphenylserine), midodrine hydrochloride (α1-agonist), dihydroergotamine, or amezinium methylsulfate (inhibitor of noradrenaline reuptake). As a result, 29 of 44 patients improved, although the details of outcome will be further described elsewhere.


In the present study, we showed that 44 (19%) of 228 children with orthostatic symptoms had apparently abnormal circulatory responses to standing: a marked instantaneous pressure decrease at the onset of standing with delayed recovery as shown in group I (mild form) and its combination with prolonged reduction of BP during the later stage of standing as in group II (severe form).

To our knowledge, this is the first article to describe the existence of INOH in as many as 19% of children with complaints of orthostatic intolerance. As shown in Table 2, compared with normal controls, patients with INOH have various somatic complaints associated with hypotension. It is noteworthy that 91% of INOH patients had complaints of chronic fatigue and that 40 (37%) of 107 children with chronic fatigue were diagnosed as having INOH. This finding suggests that INOH can be an unrecognized cause of chronic fatigue syndrome in addition to neurally mediated syncope (14). Quality of life in these patients was impaired to various degrees, and 52% of them could not attend school because of orthostatic dizziness and strong fatigue.

Headache, one of the complaints of adults with orthostatic intolerance, was also more frequently seen in the patients but may not be a good distinguishing characteristic compared with chronic fatigue, orthostatic dizziness, etc. Further analysis will be needed for distinguishing INOH and other illnesses with symptoms of orthostatic intolerance .

It was difficult to diagnose INOH when using conventional BP measurement by the auscultatory method with a sphygmomanometer because of its slow measurement. The drama of an instantaneous decrease in BP at the onset of standing was usually finished within the first 1 min in patients of group I. Such a conventional method, even though recorded in 1-min intervals during the orthostatic test, will miss the crucial evidence of INOH. This may partly explain the lack of knowledge in the literature describing BP changes in relation to orthostatic dizziness in children. In fact, before the first visit to our clinics, patients with INOH were often misdiagnosed by general practitioners as having psychiatric diseases such as anxiety disorder, somatization disorder, and feigned illness.

It is likely that INOH is more common in children and adolescents than expected not only in Japan but also in Western countries. Recently, pediatricians in the United States have paid attention to children and adolescents presenting symptoms similar to INOH, such as chronic fatigue (3) and postural tachycardia (15) of unknown origin. Some of their patients may be diagnosed with INOH if our method is used. We, therefore, emphasize that an active standing test (a more practical maneuver than a passive head-up tilt test) in which a beat-to-beat BP monitoring system is used should be added to clinical practice.

To reliably establish the criteria for orthostatic intolerance, including INOH and orthostatic hypotension in children, we have determined the cutoff level of abnormal circulatory responses to active standing in healthy Japanese school children and adolescents: either a recovery time for MAP of >25 s or a recovery time of >20 s with a 60% or greater decrease in MAP at the initial decrease (11). Specificity of these criteria was as high as 95.7%. This original cutoff level at the initial decrease (which corresponds to 45 mm Hg for mean BP reduction) is a stricter limit than that proposed by Dambrink et al. (16) who reported 40/25 mm Hg for systolic/diastolic as a normal finding. Moreover, a reduction of >15% in systolic BP during the later stage of standing (3–7 min) was considered a hypotensive response. This is almost consistent with the adult criteria for orthostatic hypotension proposed by the American Autonomic Society (17). These cutoff levels of BP in the orthostatic test should be acceptable for INOH and orthostatic hypotension in children.

Mechanisms responsible for a pronounced BP decrease associated with delayed recovery in patients with INOH may relate to altered arterial-resistance properties such as reduced vasoconstriction, as indicated by prolonged decreases in DAP during standing (Fig. 2). Two possible mechanisms may explain reduced vasoconstriction: decreased noradrenaline secretion from the sympathetic nerve endings or hyposensitivity of α-adrenoceptors in the resistance vessels. In the present study, the increase in plasma noradrenaline level at 1 min after standing was significantly lower in group I and II than in the NBP group. The reduced noradrenaline level was also found in group II at 5 min of standing but not in group I. Although we did not measure catecholamine clearance in the patients, these findings could be interpreted as partially reduced noradrenaline secretion in response to standing in group I and further reduction in group II.

Hyposensitivity of α-adrenoceptors in the resistance vessels may be less important as a cause of INOH. In our previous study on autonomic function tests in which a pharmacologic method was used, we found higher BP responses to α-adrenergic stimulation (supersensitivity) in children with clinical manifestation similar to INOH than in controls. Moreover, we reported a patient in group II with supersensitivity of α-adrenoceptor (18). Further investigation will be needed for the evaluation of α-adrenoceptor function in patients with INOH.

Reduced venous return from the lower part of the body caused by postural change and, hence, reduction in cardiac output may not take part in forming an initial decrease of BP at the onset of standing. Using the pulse contour method, Smit et al. (19) reported that the initial decrease reflected a sudden reduction in total peripheral resistance but not in cardiac output in patients with orthostatic intolerance. We, also, found the similar hemodynamic pattern in patients with INOH in a preliminary study in which impedance cardiography was used (our unpublished data).

These findings suggest that group I had a delayed reflex activation of the sympathetic nervous system but with a normal gain by postural change, whereas group II had further disturbed sympathetic activation during the whole period of standing. The role of venous tone in the later stage of standing, however, should be further studied.

A concomitant increase in HR in response to reduction in BP at the onset of standing was higher in both patient groups than the normal limits. This may reflect intact autonomic innervation of the heart in patients with INOH. Moreover, the patients had no evidence of autonomic damage in the intestinal organs, the sweat glands, and the eyes. INOH, therefore, seems to be different from pure autonomic failure (Bradbury-Eggleston type idiopathic orthostatic hypotension) in terms of pathophysiology.

Although orthostatic tachycardia was found in 20 of 44 patients, INOH is also considered to differ from orthostatic intolerance with hypovolemia (20), the postural tachycardia syndrome (21), and hyperadrenergic orthostatic hypotension (22, 23) for the following reasons. These disorders show a) normal or enhanced noradrenaline secretion during the orthostatic maneuver and b) hyposensitivity of α-adrenoceptor of the resistance vessels. Our patients had a decreased noradrenaline response to postural change. Further, treatment with the noradrenaline precursor L-threo-3,4-dihydroxyphenylserine was found to be effective in these patients, as previously reported in a smaller sample of patients with orthostatic intolerance (9). In keeping with this, children with orthostatic intolerance had a higher response of BP to intravenous infusion of an α-agonist (phenylephrine) compared with normal children in another study (24). Taken together with these findings, we conclude that primary disorders of INOH are associated with insufficient activation of muscle sympathetic nerves. Further study than the orthostatic test should be done to investigate a generalized reduction in noradrenaline output in response to other stresses.

Although the cause of reduced sympathetic activation in our patients is unclear, centrally mediated inhibition of the sympathetic nervous system is most likely. There was no evidence of peripheral impairment of the autonomic nervous system in our patients, including sweat disturbance, diarrhea, etc., as is seen in peripheral neuropathy and idiopathic orthostatic hypotension (pure autonomic failure). Moreover, basal plasma catecholamine levels were similar between the patients and controls, suggesting that peripheral autonomic nerves were not impaired. Fritsch Yelle et al. (25) reported that postspace-flight astronauts who were prone to develop orthostatic hypotension and presyncope showed similar circulatory and pathophysiologic changes to those of our group II, that is, orthostatic hypotension and tachycardia associated with reduced noradrenaline responses to standing. They speculated that mechanisms responsible for reversible hypoadrenergic responsiveness were possibly centrally mediated.

In conclusion, INOH is an important pathologic condition in children with symptoms of orthostatic intolerance and chronic fatigue. This can be identified by using a noninvasive beat-to-beat continuous BP monitoring system. The mechanism responsible for INOH may depend on reduced noradrenaline secretion during orthostatic stress associated with centrally mediated sympathetic inhibition. The patients in group II, who can be diagnosed with child-type orthostatic hypotension, seemed to have a more pronounced decrease of sympathetic activation and thus a more impaired quality of life than those in group I.