Postural tachycardia syndrome (POTS), a disorder that affects an estimated 500 000 people (∼0.18%) in the USA alone, is an important source of disability in young adults.1 It shows a strong predilection for females, typically between the ages of 20 and 50 years.2 POTS is a form of orthostatic intolerance characterised by an excessive increase in heart rate (>30 beats/min) on assuming the upright position associated with orthostatic symptoms but in the absence of orthostatic hypotension.2^–5 Their symptoms of dizziness, nausea, tremor, chronic fatigue and exercise intolerance make even simple activities of daily living such as walking and bathing exhausting prospects.6

Although the role of noradrenergic dysfunction is unclear, there is significant evidence of central dysregulation of the noradrenergic system in POTS and an association between dysregulation of norepinephrine (NE) homeostasis and psychiatric disorders, including depression, panic disorder and attention deficit hyperactivity disorder (ADHD). 7^–9 Centrally acting drugs that inhibit the presynaptic norepinephrine transporter (NET) (eg, tricyclic antidepressants, duloxetine, reboxetine, amphetamines and cocaine) produce the clinical features of POTS.10^–13 Peripheral plasma NE is frequently raised in patients with POTS, particularly when upright, and many clinical features of POTS such as tachycardia, palpitations, shortness of breath, chest discomfort and tremor mimic the hyperadrenergic features of a panic attack. In one family, the POTS phenotype segregated with a heterozygous missense mutation in NET, effecting >98% loss of function and elevated plasma NE levels. The mutation tracked with POTS characteristics and the probands complained of inattention.10 While stimulants used to treat ADHD enhance alertness, attention and working memory, very high levels of catecholamines disrupt these higher cognitive functions. This effect can be reversed by α₂ adrenoreceptor agonists which reduce NE release.9 Together, these findings suggest that the correct balance of central NE activity is required to maintain normal attention and this could be impaired in POTS.

This cross sectional controlled study assessed attention, mood and anxiety symptoms in patients with POTS and contrasted them with ADHD subjects and psychiatrically normal control subjects. We tested the following hypotheses:

1. Patients with POTS will report a higher lifetime prevalence of depression and anxiety disorders, and a higher magnitude of current depression and anxiety symptoms, than controls, and the prevalence will be comparable with subjects with ADHD.

2. Patients with POTS will show significant inattention relative to controls, and similar to levels shown by subjects with ADHD.

METHODS

Patients and study design

The study was approved by the Vanderbilt University Institutional Review Board and all participants provided written informed consent prior to the study. Patients with POTS (n = 23), aged 18–65 years, were admitted as research subjects to the Vanderbilt Clinical Research Centre between January 2003 and December 2004. They were referred to Vanderbilt because of clinical suspicion of POTS or had received a clinical diagnosis. The diagnosis of POTS was confirmed with current criteria, which included (1) symptoms of orthostatic intolerance and an increase in heart rate of ⩾30 beats/min within 10 min of standing from a supine position but in the absence of orthostatic hypotension (fall in blood pressure of ⩾20/10 mm Hg)2^–4; (2) symptoms with standing that improved with recumbence; (3) symptoms had occurred for at least 6 months; and (4) symptoms had occurred in the absence of other major medical comorbidity or another obvious cause of tachycardia (such as acute dehydration). None of the subjects was receiving medications which could affect autonomic function during their admission, including psychotropic drugs. Subjects participated in the seated position and were not acutely symptomatic during the testing.

Patients with POTS were matched by age, gender and race with 20 psychiatrically healthy control subjects who formed a comparison group for the self-report questionnaires used in the study. They were free of current or lifetime history of Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV)14 axis I psychiatric disorders, as assessed by the Structured Clinical Interview for DSM-IV axis I disorders (SCID).15 Age matched adults with ADHD (n = 18) formed a second comparison group with the POTS patients, in particular regarding symptoms of inattention and chronic psychiatric illness. All ADHD patients had been previously diagnosed and treated for ADHD by a psychiatrist or psychologist. To ensure they currently met criteria, we re-evaluated patients using DSM-IV diagnostic criteria, as recommended by Barkley.16 All control subjects and patients with ADHD were free of medical comorbidities and were not taking psychotropic agents. The control subjects and patients with ADHD did not receive formal autonomic function testing, but none complained of autonomic symptoms on screening by a medical doctor (VR), and patients with ADHD did not meet POTS criteria on testing supine and standing heart rate and blood pressure.

For data on DSM IV axis I diagnoses, subjects with POTS were compared with subjects with ADHD and data from a large population of respondents (n = 9282) to the National Comorbidity Survey Replication (NCS population norms).17 Similarly, data on Anxiety Sensitivity Index score were also compared with published general population normative data.18

Measures

Subjects were assessed by a trained research assistant for current and lifetime history of axis I psychiatric disorders using the SCID. Psychiatric diagnoses were made only if not accounted for by the presence of a medical disorder. Any diagnostic questions were resolved based on the consensus opinion of the project clinicians (KLH, RCS). Subjects completed the following self-report questionnaires (further details are in a supplementary file available online):

1. The Beck Depression Inventory Second Edition (BDI-II), a well validated screening questionnaire for depression.19 20

2. The Beck Anxiety Inventory (BAI), a commonly used measure of anxiety severity.21

3. The Anxiety Sensitivity Index (ASI), which assesses fear of anxiety related symptoms.22

4. The Connors Adult ADHD Rating Scale long form (CAARS-L), which rates items in four factor domains including inattention/memory problems and hyperactivity/restlessness.23

5. The ADHD Behaviour Checklist for Adults (ABC-A), which measures the specific inattentive and hyperactive–impulsive symptoms that comprise the DSM-IV criteria for ADHD.24

Statistical analysis

Differences among the groups were assessed using Fisher’s exact test for categorical variables and one way analysis of variance (ANOVA) for continuous variables, with a post-hoc Scheffe multiple comparisons procedure used for pairwise comparisons. Spearman’s test was performed to analyse for bivariate correlations. A general linear model with inattention as the dependent variable and gender, anxiety and depression as fixed factors was used to test for the influence of potential confounding variables. All tests were two sided and results were considered significant if p<0.05. Data were analysed with SPSS for Windows software, V.14.0 and GraphPad Prism software, V.4 (GraphPad Software, San Diego, California, USA). Results are presented as means (SDs) unless otherwise specified.

RESULTS

Patient characteristics

One patient with POTS declined to participate, reporting that she felt too symptomatic. Data collected on two of the 23 patients with POTS were not analysed because of comorbid ADHD that pre-dated the diagnosis of POTS. One control subject was excluded after meeting criteria for a past diagnosis of major depressive disorder. No ADHD volunteers were excluded. However, three of 23 initial subjects failed to participate and two subjects failed to complete the study so their data could not be included. ADHD subjects could not be closely gender matched with patients with POTS because of the male bias of ADHD compared with the strong female predominance of POTS.

Patients with POTS reported mean length of illness from self-reported onset of symptoms as 6.9 (9.2) years. Characterising data for patients with POTS is presented in table 1 and baseline demographic data for each patient group are compared in table 2.

Groups did not differ significantly in age or education level and there was no difference in gender distribution between the POTS and control groups. There were more men in the ADHD group compared with the control group (p = 0.027), reflecting the male predominance of ADHD. All subjects were Caucasian, with the exception of one African–American patient with POTS who was matched with an African–American control subject.

Prevalence of mental disorders

Lifetime prevalence of DSM-IV axis I psychiatric disorders in patients with POTS and ADHD subjects is presented in table 3.

The lifetime prevalence of anxiety disorders in patients with POTS was not significantly greater than NCS population norms (p = 0.83) and did not differ significantly from subjects with ADHD (p = 0.34). Subjects with ADHD had a non-significant trend towards a higher prevalence of anxiety disorders than NCS population norms (p = 0.08), including panic disorder (p = 0.066). Among anxiety disorders, the prevalence of panic disorder in patients with POTS did not differ from NCS population norms (p = 0.99) or subjects with ADHD (p = 0.32).

There was no difference in the lifetime prevalence of major depressive disorder in patients with POTS compared with NCS population norms (p = 0.99). However, there was a higher prevalence of major depressive disorder in subjects with ADHD compared with patients with POTS (p = 0.024) or NCS population norms (p<0.001).

Finally, the prevalence of substance abuse or dependence in patients with POTS did not differ significantly from NCS population norms (p = 0.73) but was significantly less than in subjects with ADHD (p = 0.025).

Prevalence of depression symptoms

The BDI-II score compared the magnitude of current depression symptoms in patients with POTS, subjects with ADHD and control subjects free of DSM-IV axis I disorders. Patients with POTS scored as mildly depressed (4.8 (8.5)). Overall, BDI-II scores differed significantly among the groups (p<0.001). Post hoc analysis showed a significantly higher score in patients with POTS compared with control subjects (14.8 (8.5) vs 1.4 (1.5); p<0.001) and in ADHD subjects compared with controls (10.7 (9.8) vs 1.4 (1.5); p<0.001) but there was no significant difference between the POTS and ADHD groups (p = 0.239). For patients with POTS, there was a significant inverse correlation between length of illness and BDI-II score (r_s = −0.47, p = 0.04). There was no significant correlation of supine, upright or delta NE with BDI-II score in the POTS group.

Prevalence of anxiety symptoms

The BAI compared the magnitude of current anxiety symptoms in the three groups (fig 1A). Patients with POTS scored as moderately anxious (22.5 (8)). Overall, BAI scores differed significantly among the groups (p<0.001). Post hoc analysis showed a significantly higher score in patients with POTS compared with control subjects (1.3 (1.6); p<0.001), and in patients with POTS compared with subjects with ADHD (5.9 (5.7); p<0.001), but there was no significant difference between the ADHD and control groups (p = 0.115). In the POTS group, there was no significant correlation between length of illness and BAI score (r_s = −0.40, p = 0.09). There was also no significant correlation of supine, upright or delta NE with BAI score in the POTS subjects. The five BAI items scoring with highest frequency in the POTS group compared with the other groups were heart pounding and racing (2.3 (0.6)), feeling dizzy and light-headed (1.8 (0.9)), numbness and tingling (1.6 (0.8)), feeling unable to relax (1.5 (0.8)) and feeling unsteady (1.5 (0.9)).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 The Beck Anxiety Inventory (BAI) scores (A) and the Anxiety Sensitivity Index (ASI) scores (B) are shown for healthy control subjects (normals), patients with postural tachycardia syndrome (POTS), patients with attention deficit hyperactivity disorder (ADHD), and published population norms (dark grey). The overall p value was measured using analysis of variance (ANOVA) for the three study groups (not including the published population norms) and paired p values were performed using a post hoc Scheffe test. Data are presented as mean (SEM). Differences between the two anxiety tools may reflect inclusion of somatic anxiety symptoms in the BAI whereas the ASI scores only for cognitive anxiety.

The ASI compared the magnitude of current fear of anxiety related symptoms in the three groups (fig 1B). Overall, ASI scores differed significantly between the groups (p = 0.001). Post hoc analysis showed a significantly higher ASI score for patients with POTS (15.4 (9.8)) than the control group (4.6 (4.5); p = 0.001), and in subjects with ADHD (11.8 (9.7)) compared with controls (p = 0.031) but there was no significant difference between the POTS and ADHD groups (p = 0.504). ASI scores for the three groups were also compared with published general population normative data of 19.1 (9.11) (obtained from 12 non-clinical samples comprising >4500 participants).18 Patients with POTS experienced a trend towards lower anxiety sensitivity than the general population (p = 0.07) while subjects with ADHD experienced higher anxiety sensitivity than the general population (p = 0.001). As expected, our psychiatrically normal control subjects experienced significantly less anxiety sensitivity than the general population (p<0.001).

Among patients with POTS, there was a significant inverse correlation between length of illness and ASI score (r_s = −0.56, p = 0.01). There was no significant correlation between supine, upright or delta NE with ASI score in patients with POTS.

Prevalence of inattention symptoms

Significant group differences were noted on all eight CAARS-L subscales. Post hoc analysis showed that patients with POTS did not differ significantly from controls for hyperactivity (p = 0.170), impulsivity (p = 0.064), self-concept (p = 0.196) and DSM-IV hyperactivity (p = 0.060) subscales but both groups scored significantly lower than the ADHD group.

Patients with POTS experienced significantly more symptoms than controls on three CAARS-L subscales: inattention/memory problems (p = 0.033), the CAARS-L “DSM-IV inattention” score (p<0.001) and the “DSM-IV ADHD” score (p = 0.003) (table 4). They also scored significantly higher than controls on the ADHD Index (p = 0.027). The ADHD group scored significantly higher than either of the other groups on all CAARS-L subscales.

A general linear model analysis using gender, BDI score and ASI score was used to test for a possible confounding influence by gender, current depression symptoms and current anxiety sensitivity on inattention scores. Firstly, using Connors DSM-IV inattention as the dependent variable, the model showed no statistically significant effect of gender (p = 0.307). There was a statistically significant effect of BDI score (p = 0.021) and ASI score (p = 0.019) but group remained a statistically significant driver of the CAARS-L DSM-IV inattention score (p<0.001), even when BDI and ASI scores were taken into account. Secondly, using CAARS-L ADHD score as the dependent variable, the model showed no statistically significant effect of gender (p = 0.444) or ASI score (p = 0.155). There was a statistically significant effect of BDI score (p = 0.005) but group remained a statistically significant driver of the CAARS-L ADHD score (p<0.001), even when BDI score was taken into account.

In the POTS group, there were statistically significant inverse correlations between length of illness and the following CAARS-L subscales: inattention/memory score (r_s = -0.46, p = 0.05), self-concept score (r_s = −0.47, p = 0.04), Connors DSM-IV hyperactivity score (r_s = −0.53, p = 0.02), DSM-IV ADHD score (r_s = −0.49, p = 0.04) and ADHD Index (r_s = −0.52, p = 0.02). There was no significant correlation of supine, upright or delta NE with any of the CAARS-L subscale scores in subjects with POTS.

Prevalence of ADHD symptoms in childhood and adulthood

Scores on the four ABC-A subscales were compared between the three groups, and data on inattention are presented graphically in fig 2. There were significant overall group differences on all subscales (attention problems in childhood: p<0.001; hyperactivity in childhood: p<0.001; inattention over the past 6 months: p<0.001; hyperactivity over the past 6 months: p<0.001).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 The Attention Deficit Hyperactivity Disorder (ADHD) Behaviour Checklist for Adults Inattention scores (ABC-A score) are presented for the three groups (as outlined earlier) during their childhood (A) and during the last 6 months (B). The overall p value is measured using analysis of variance (ANOVA) for the three groups, and paired p values were performed using a post hoc Scheffe test. Data are presented as mean (SEM). POTS, postural tachycardia syndrome.

For inattention in childhood, post hoc analysis showed that patients with POTS (4.8 (5.4)) had significantly less childhood symptoms than ADHD subjects (20.2 (7.0); p<0.001) and did not differ significantly from control subjects (2.9 (2.6); p = 0.499). However, for inattention over the past 6 months, patients with POTS (10.7 (7.8)) had significantly more symptoms than control subjects (2.2 (2.1); p<0.001) but less than ADHD subjects (16.7 (7.7); p = 0.021).

For hyperactivity in childhood, post hoc analysis showed that patients with POTS (5.7 (6.2)) had significantly less childhood symptoms than ADHD subjects (18.8 (9.3); p<0.001) and did not differ significantly from control subjects (2.5 (2.2); p = 0.281). For hyperactivity over the past 6 months, patients with POTS (7.9 (6.6)) had significantly more symptoms than control subjects (1.8 (2.0); p = 0.007) but less than ADHD subjects (13.7 (7.7); p = 0.015).

DISCUSSION

Several studies have reported that patients with POTS have comorbid psychiatric symptoms, particularly anxiety and panic disorder.8 25 Using the Structured Clinical Interview for DSM-IV axis I disorders, we found that there was no significant difference in lifetime prevalence of major depressive disorder in patients with POTS compared with a general population sample.17 Patients with POTS do, however, score as mildly depressed on quantitative measures of current depression symptoms (BDI). This may reflect living with a chronic medical illness rather than a phenomenon specific to POTS, as medical illness is commonly associated with an increased frequency of depressive disorders.26^–28 Certainly, patients with POTS show limitations across multiple domains of quality of life in common with other chronic medical illnesses, including chronic obstructive pulmonary disease and heart failure.6 Interestingly, there is a significant reduction in depressive symptoms (BDI score) as POTS progresses. This could reflect a reduction in symptom burden over time, or indicate the psychological adjustment of patients to living with a chronic illness.

Patients with POTS did not experience a significantly increased lifetime prevalence of anxiety disorders (including panic disorder) compared with a general population sample17 or subjects with ADHD. For current anxiety symptoms, patients with POTS scored significantly higher than controls and ADHD subjects on the BAI, and within the moderate anxiety range. However, when using the cognitively focused ASI, patients with POTS did not score higher than the general population (they actually had a trend towards a lower score) although they still scored higher than control subjects free of axis I psychiatric disorders. This is supported by the findings of Masuki et al who reported that ASI scores in 14 patients with POTS were significantly higher than in control subjects but within the limits of published general population data.29

The contrast between the BAI score (moderate anxiety) and the ASI score (comparable with the general population) in the POTS group is striking. A major difference between the BAI and ASI is that the former measures both somatic symptoms and subjective anxiety and panic symptoms on factor analysis while the latter measures sensitivity to anxiety provoking stimuli but not somatic symptoms. Since the orthostatic symptoms experienced by patients with POTS are similar to the somatic criteria required for the diagnosis of panic attacks or other anxiety disorders (eg, palpitations), it is likely that these symptoms, rather than psychological factors, elevate the BAI scores. Unlike subjects with panic disorder, patients with POTS show several important clinical differences that can be used to distinguish between the two disorders. These include predictable precipitants for the onset of somatic symptoms in subjects with POTS (eg, dehydration and standing) while patients with panic disorder experience repeated episodes without a clear precipitant. Furthermore, patients with POTS experience a significant worsening of somatic symptoms on adoption of the upright position while subjects with panic disorder typically do not.30 It is therefore likely that the somatic symptoms in POTS patients do not represent panic disorder, which is appropriately reflected in the lower ASI score. This finding is supported by two recent studies. Masuki et al showed that excessive increase in heart rate in patients with POTS in response to orthostatic stress is not caused by anxiety.29 Khurana collected data on anxiety and somatic symptoms in response to anxiety provoking stimuli and found that symptoms of POTS were distinguishable from symptoms of panic disorder.31 Taken together, the evidence suggests that clinically observed anxiety in patients with POTS is caused by biological rather than psychological factors.

As POTS progresses, there is a significant decrease in ASI score but no significant change in BAI score. This suggests that there is a diminution in cognitive anxiety symptoms with time, but not in somatic symptoms.

Our data suggest that patients with POTS experience inattention of a magnitude that is greater than in controls but less severe than in adult ADHD subjects. Unlike ADHD subjects, patients with POTS did not report more inattention during childhood. This could indicate a causal role for POTS in attention difficulties, or suggest that the two problems share common antecedents. Moreover, this might suggest common genetic underpinnings as noradrenergic gene products regulate sympathetic tone, cardiac output and attention. Interestingly, there is a significant perceived reduction in the magnitude of inattention and ADHD symptoms as POTS progresses. This could reflect patient adaptations or an improvement in POTS symptoms over time.

Study limitations

Firstly, the sample size in this study is relatively small although it does compare favourably with other clinical research studies in POTS. It is possible that with a much larger cohort, some of our non-significant trends might have become significant. Secondly, whereas psychiatric diagnoses were evaluated using the gold standard of a structured interview, data on attention were collected by subjective self-rated questionnaires. While these tools have been well validated, future studies might benefit from using more objective outcome measures.

CONCLUSIONS

Patients with POTS do not have a higher than normal prevalence of anxiety disorders. While they register with moderate anxiety using somatic assessment tools such as the BAI, they have no more anxiety than the general population using cognitive tools such as the ASI. Similarly, patients with POTS do not have a significantly increased prevalence of major depression but do exhibit mild depressive symptoms. The main abnormality noted in patients with POTS was difficulty with maintaining attention. These attention difficulties develop later in life, in contrast with patients with ADHD. Further investigation is required in order to identify the aetiology of these problems, and therefore guide clinicians in the selection of targeted treatments.