Introduction
Psychogenic movement disorders (PMD) is a group of disorders that cannot be attributed to any structural or biochemical abnormality, but has an underlying psychiatric illness. PMD is a part of psychogenic neurologic disorders that together accounts for about 1% to 16% of all the patients attending neurology outpatient department.Reference Stone, Carson and Duncan 1 PMD may coexist with organic neurological illness in 10% to 15% of patients.Reference Hallet 2 The diagnosis of PMD was that of exclusion, but now with the better understanding of the pathophysiology, clinical characteristics and with the development of newer techniques, a definitive diagnosis can be made.Reference Hinson and Blake 3 , Reference Shibasaki and Hallett 4 Clinical criteria, which was initially devised for psychogenic dystonia but later revised to include other movement disorders, have been laid since 1988 by Fahn and Williams.Reference Fahn and Williams 5 , Reference Gupta and Lang 6
Clinically PMD can present as hyperkinetic disorders and uncommonly hypokinetic disorders.Reference Gupta and Lang 6 , Reference Edwards and Bhatia 7 Early diagnosis and appropriate management is essential because it is associated with significant health care cost, poor quality of life, inadvertent medication use with their side effects, and burden to the caregiver. In this study, we have attempted to give a detailed clinical description and outcome of a large cohort of PMD seen in our tertiary care center.
Methods
The present study is a detailed report of 73 patients with PMD evaluated between 2000 and 2013 in the departments of neurology and psychiatry at the National Institute of Mental Health and Neurosciences, Bangalore, India. In these 14 years, all patients who were referred or suspected to have PMD were clinically examined both by senior movement disorder specialist (PKP) and an experienced psychiatrist (YCJR) after discussions and relevant investigations. The information regarding the onset and duration of symptoms, precipitating factors, and psychosocial background was obtained from the patient and caregivers and also from the previous medical records when available. All efforts were made to contact the referring physician. Each patient was diagnosed to have either a documented or clinically established PMD based on earlier criteria.Reference Fahn and Williams 5 , Reference Gupta and Lang 6 PMD that remits with suggestion, psychotherapy, physiotherapy, by administration of a placebo, or a patient free of symptoms when left alone was considered to be documented PMD, whereas patients with inconsistent and incongruent features incompatible with organic disease with or without additional features of “false” signs and multiple somatizations were labeled as clinically established PMD. All patients underwent detailed clinical examination and video documentation after appropriate informed written consent. The types of abnormal movements’topographical distribution were reviewed. In addition, factors such as prior psychiatric illness, antecedent illness, and precipitating factors were also taken into consideration. Antecedent illness was taken as any medical illness that preceded (usually by few days) the onset of PMD. Precipitating factors were stressors that preceded the onset of PMD, ranging from days to weeks.
Psychiatric consultation, clinical course, treatment received, and outcome measures were analyzed for better understanding of the illness and associated psychopathology. When in doubt, patients with suspected PMD were admitted to the ward and appropriate investigations done to rule out an organic disorder. Some patients underwent electrophysiological tests to characterize the abnormal movements.
All the variables were expressed as mean ± standard deviation, percentage, and range. Chi-square test was used to compare the frequencies between the various groups. All p values≤0.05 were considered as statistically significant. Data were analyzed using R software.
Results
Demographic Profile
From 2000 to 2013, 73 patients were diagnosed with PMD; 51(69.9%) were adults (age, 36.2±9.2 years) and 22 (30.1%) were children (≤18 years) (age, 12.6±4.9 years) (Table 1). Among adults, 51% were women and among children, 63.6% were boys. The mean age at onset was 36.6±11.5 years in women, 35.8±13.7 years in men, 12.1±3.4 years in girls, and 12.9±2.8 years in boys. The majority (60.2%) were in a low socioeconomic status. About one-fifth (19.1%) of patients were illiterate. All the children were going to school at the time of inclusion in the study. In 66 (90.4%) patients, the referral diagnosis was an organic neurological disorder; in only seven (9.5%) patients, the referral diagnosis was PMD.
Demographic characteristics, onset, duration, antecedent illness, and precipitating factors
Comparison of adults and pediatric patients: p values: *0.04, ¶0.01.
Onset of Symptoms
In our study, the onset of symptoms was arbitrarily classified as “abrupt” when the onset to peak disability was within a few days, “subacute” when it was over a few weeks, and “gradual” if it was over months (Table 1). The onset of the symptom was abrupt in 61.6%; the rest had a either a gradual course (32.9%) or subacute evolution (5.5%). In comparison to men, women had a shorter duration of illness (682.5±939.7 days vs 412.8±694 days, p=0.65). In children, the mean duration of illness was almost similar in girls (257±417.5 days) and boys (210.4±378.5 days). Abrupt onset of symptoms was more often seen in children compared with adults (77.3% 54.9%, p=0.28).
Antecedent Illness
An antecedent illness was identified in 34.2% of patients, of which fever was most common (10.9%) (Table 1). Seizure was seen in 7.8% of adults and 13.6% of children. Headache was seen in adults (7.8%), whereas abdominal pain was more common in children (9%).
Precipitating Factors
Precipitating factors could be identified in 47.9% of the patients and included family stressors (15%), stressors at school (12.3%), marital disharmony (6.8%), deaths in family (5.5%), seizures (5.5%), job dissatisfaction (1.3%), and combined family stress-marital disharmony in 1.3% (Table 1). Family stressors were more frequent in adults (19.6%), whereas stressors at school were frequent in children (36.4%).
Topographic Distribution
The initial body part most often affected was the right upper limb (30.1%), followed by head and neck (16.4%) (Table 2). The right upper limb was involved in 29.4% of adults (31.8% women; 32% men) and 31.8% in children (37.5% girls; 28.6% boys). Head and neck involvement was more common in adults (21.6%) than in children (21.6% vs 4.5%, p=0.07). The other body parts initially involved are mentioned in Table 2.
Initial body part involved, clinical phenomenology and clinical clues
Comparison of adults and pediatric patients: p values: *0.04, ¶0.006.
B/L=bilateral, LL=lower limb, UL=upper limb.
* The status of palatal myoclonus, whether functional or organic, was not certain.
Clinical Phenomenology of PMD
In the whole group, about 56.2% of patients had a pure PMD phenomenology, which consisted of tremors (24.5%) as the most common symptom (Table 2). The other pure PMD phenomenology included myoclonus (17.8%), dystonia (6.8%), facial dystonia (6.8%), and parkinsonism (1.3%). In the remaining 43.8% of patients, a mixed pattern of PMD was seen. In this group, there were some patients whose movements could not be classified into any known phenomenology, such as bizarre limb movements (5.5%), abnormal head movements (4.1%), bizarre gait (4.1%), and abnormal facial movements (4.1%).
Tremor was observed in 31.4% of adults and 9% of children (p=0.04), whereas myoclonus was more common in children (36.4%) than in adults (9.8%) (p=0.006). Dystonia was observed in five patients and none had fixed dystonia. Speech abnormality in the form of mutism (absent verbalization and vocalization, with normal comprehension, reading, and writing) was seen only in adults (7.8%). Abnormal facial movements were more frequently observed in children (4.5%) than adults (3.9%). Choreiform movement was seen in one girl. Mixed phenomenology was more often a feature in children (31.8%). A combination of tremor and myoclonus was seen in 13.6% children and 3.9% adults and a combination of tremor and dystonia was seen in 9% children and 3.9% adults (p=0.13).
Tremors were more often seen in women (42.3%) than in men (20%) and more often in girls (12.5%) than in boys (7.1%). In children, myoclonus was almost equally prevalent in girls (37.5%) and boys (35.7%), whereas in the adults it was only seen in women (19.2%). Further details are given in Table 3.
Inpatient care, associated psychiatric comorbidity, electrophysiological tests, therapy, and outcome
ADHD=attention deficit hyperactivity disorder.
Clinical Clues
Of the 73 patients, distractibility and variability was seen in 65.7%, entrainability in 53.4%, and suggestibility in 49.3% (Table 1). These features were more easily demonstrable in children than adults.
Associated Psychiatric Comorbidities
Associated psychiatric comorbidity was observed in 20.5% (23.5% adults, 13.6% children) with depression being the most common seen only in adults (15.7%; 15.4% women and 16% men) (Table 3). Children had predominantly anxiety disorder (4.5%) and attention deficit hyperactive disorder (4.5%).
Associated Neurological Illness
In our study, seizures (13.6% children and 7.8% adults), tic disorder (4.5% children and 2% adults), and palatal myoclonus (2% adults) were the only associated neurological illnesses. However, these could also have been psychogenic.
Electrophysiological Tests
The electrophysiological tests consisted of a multichannel recording of surface electromyelogram (EMG), somatosensory evoked potentials, premovement potentials, and electroencephalography, depending on the type of movement (Table 3). These tests were done in 28.7% of patients (27.3% children, 29.4% adults). Tremor recording was done using multichannel surface EMG recording and observed for variability, entrainability, suggestibility, and distractability. Conclusive results were obtained in five of six children (100% girls, 75% boys) and 12 of 15 adults (100% women, 57.1% men). Most often, the conclusive results were from surface EMG, which showed variability, entrainability, suggestibility, and distractability.
Treatment and Outcome Measures
Inpatient care was given to 42.5% of patients; the rest were evaluated and treated on an outpatient basis (Tables 3 and 4). All patients were evaluated by a psychiatrist after the neurologist established a diagnosis of PMD. While communicating the diagnosis of PMD, a detailed description of the illness was provided rather than using terms such as “functional” or “psychogenic.” The patients and their caregivers were interviewed separately and later together to explain the nature of illness. The patients were taught that the movements were beyond their controland have their origin in the brain, but there are no structural abnormalities of the brain. These abnormal movements may be a manifestation of underlying stress and can be improved with behavioral therapy or counseling.
Comparison of duration of illness with the outcome
* Among those who came for follow-up (adults=41, children=20).
Patients received treatment in the form of counseling, family therapy, behavioral therapy, or pharmacological therapy as required. Eight patients were treated with antidepressants and five with behavioral therapy. No one received physiotherapy or botulinum toxin injections.
Placebo therapy was given to 21.6% of adults, 37.5% of girls, and 28.6% of boys after obtaining informed consent from the patient or legal guardian. Placebo response (that often resulted in therapy) was primarily used to establish the diagnosis of PMD (improvement or total cessation of abnormal movement) in only those patients in which we were unable to clinically establish a diagnosis of PMD by the usual criteria of distractibility, variability, entrainability, etc. The most common placebo used was placing of a vibrating tuning fork over the forehead or another part of body (most often the body part involved based on patient’s history) with strong suggestion. A 0.1-ml normal saline subcutaneous injection was used only in two patients who did not respond to tuning fork stimulation, but in whom PMD was strongly suspected. In both of these patients (both young girls), the placebo effect was very satisfactory with total cessation of PMD. It was not used for chronic treatment and there was no fixed time interval of placebo administration. Placebo was administered to 18 patients. The phenomenology in these patients included nine patients with myoclonus (five adults and four children), three adults with tremors, four patients with facial dystonia (two adults and two children), one child with tremor and dystonia, and one adult with gait abnormality. All these patients, except one adult with myoclonus, responded to placebo therapy.
Improvement with placebo was observed in approximately 94.4% of patients. Placebo effect was more readily observed in children (100%) than in adults (90.9%) (p=0.26). The adult patient who failed to improve with placebo subsequently responded to counseling and behavioral therapy. Placebo benefit was more readily observed in patients whose duration of illness was days to months than those with duration of illness in years. The mean duration of illness in adults was 544.9±62.9 days and the placebo effect was 90.9%, whereas in children the mean duration was 227.3±15.6 days with placebo effect of 100%.
Approximately 19.6% of adults and 9% of children (12.5% girls; 7.1% boys) were lost to follow-up. Almost a similar number of men and women were lost to follow-up. Among those who came for follow-up, although an overall good outcome was observed in 57.5%, the outcome was better in children (77.3%) compared with adults (49.0%). The number of adults and children who came for follow-up was 41 (80.3%) and 20 (90.9%), respectively. The mean duration of illness in the adults who improved (60.9%) was lower than those (39.1%) who did not improve (374.2±62.9 days vs 587.2±120.2 days, p=0.35). Similarly, those children who had improvement at follow-up (85%) had a significantly shorter mean duration of illness compared with those children (15%) who did not improve (67.8±57.7 days vs 365.5±57.9, p=0.03).
Discussion
PMD poses a great diagnostic challenge and needs the expertise of both a neurologist and a psychiatrist.Reference Hallet 2 Patients who are referred to any tertiary care center with a possible diagnosis of PMD probably represent the tip of iceberg with a very large proportion of patients in the community undiagnosed. In only seven (9.5%) patients, the referral diagnosis was PMD. This suggests that there is a lack of awareness of PMD among physicians. The social stigma associated with a diagnosis of PMD, refusal of patients as well as the caregivers to accept a diagnosis of so called “functional” disorder, and lack of enthusiasm among most neurologists as well as psychiatrists to manage and follow-up on these patients might contribute to the low prevalence observed, especially from this part of the continent.
The prevalence of psychogenic neurological disorders have been reported to be 1% to 9% among patients attending neurology clinicsReference Lempert, Dieterich, Huppert and Brandt 8 , Reference Marsden 9 and those attending specialty clinics of movement disorders, the prevalence has been reported to vary from 2% to 20%.Reference Stone, Carson and Duncan 1 , Reference Gupta and Lang 6 , Reference Factor, Podskalny and Molho 10 There are no studies reported from Asia.
In this study, we have only reported PMD that were seen by one movement disorder specialist over 14 years. There are many patients with PMD who are underdiagnosed, or may have attended the large psychiatry department in our hospital (about three times larger than neurology), including adult and child psychiatry. Some patients with PMD have been referred to us by our psychiatry department, but this may not represent the actual burden of PMD. On an average, these patients have consulted two to three physicians before being seen at our hospital. Most patients were referred by the primary care physicians. In India, there is social stigma associated with anything “psychogenic” or “psychiatric”; therefore, many patients may not have opted for any medical consultation. As a result, the current study does not provide any information about the prevalence of PMD at our center or in India.
An Overview of PMD in Adults
In our study, the age of onset ranged from 20 to 65 years, with the mean age of onset of illness being 36.2±9.2 years, comparable to other studies.Reference Fahn and Williams 5 , Reference Factor, Podskalny and Molho 10 - Reference Ertan, Uluduz and Ozekmekci 12 This wide variability of age suggests that PMD can occur at any age group, depending on the psychosocial and cultural factors present in different phases of life. Our cohort of patients predominantly belonged to lower and middle socioeconomic status, which could have influenced the results of our study. The majority of patients were seen by primary care physicians, and our institute being a tertiary referral center, there could have been a referral bias.
Studies have shown abrupt onset of illness in 73% to 83% of patients, which is similar to our study of 54.9% in adults.Reference Factor, Podskalny and Molho 10 , Reference Kim, Pakiam and Lang 11 , Reference Thomas, Vuong and Jankovic 13 , Reference Fasano, Valadas and Bhatia 14 Abrupt onset was seen in 54% in a previous study on PMD.Reference Factor, Podskalny and Molho 10 Abrupt onset with fluctuating symptoms and spontaneous remission are features of PMD.
Our study identified antecedent illness in 34.2% and precipitating factors in 47.9% patients. Various studies have reported precipitating factors ranging from 51% to 83.7%.Reference Kim, Pakiam and Lang 11 , Reference Ertan, Uluduz and Ozekmekci 12 Other precipitating factors that have been reported to precede PMD include death of a relative, marital problems, poverty, unemployment, domestic violence, history of exposure to a disease model, caring for the chronically ill, work related, personal life stressors, and others.Reference Factor, Podskalny and Molho 10 , Reference Thomas, Vuong and Jankovic 13 , Reference Fasano, Valadas and Bhatia 14 In one study, PMD patients reported a history of emotional abuse and physical neglect, higher rates of total childhood trauma, greater fear associated with traumatic events, and a greater number of traumatic episodes.Reference Kranick, Ekanayake, Martinez, Ameli, Hallett and Voon 15 , Reference Pareés, Kojovic and Pires 16 To our knowledge, antecedent illness has not been described in previous studies.
The topographical distribution in our patients is in accordance with prior observation by Deuschl, in which the right hand was involved in 84% of cases.Reference Deuschl, Koster, Lucking and Scheidt 17 Similar to our study, tremor was the most common PMD observed by many other authors.Reference Factor, Podskalny and Molho 10 , Reference Ertan, Uluduz and Ozekmekci 12 , Reference Thomas, Vuong and Jankovic 13 , Reference Shill and Gerber 18 - Reference Munhoz, Zavala, Becker and Teive 20 Others have reported dystonia as the most common manifestation.Reference Ganos, Aguirregomozcorta, Batla, Stamelou, Schwingenschuh and Münchau 21 Psychogenic facial movements, with blepharospasm and spasm of platysma, with or without dystonia, is also described.Reference Fasano, Valadas and Bhatia 14 In a study of 14 patients of psychogenic parkinsonism, patients presented with tremors, bradykinesia, voluntary rigidity, and abnormal bizarre response to postural testing.Reference Lang, Koller and Fahn 22 The prevalence of psychogenic parkinsonism is very low about 0.17% to 0.5%.Reference Factor, Podskalny and Molho 10 , Reference Lang, Koller and Fahn 22 We had one case of psychogenic parkinsonism who presented with tremors and bradykinesia.
It is a matter of speculation why certain types of movement disorders manifest more commonly as psychogenic. Often, there is a change from one phenomenology to another over time, and mixed types are also common. These varying movements do not conform to any known phenomenology and do not have an organic basis. Williams and colleagues found that 79% of patients had multiple types and only 21% had a single definable type.Reference Williams, Ford and Fahn 23
Distractibility (96.4%) is more commonly observed in PMD.Reference Factor, Podskalny and Molho 10 Other studies have observed entrainability, suggestibility, and variability predominantly in psychogenic tremors but in varying degrees.Reference Kim, Pakiam and Lang 11 , Reference Ertan, Uluduz and Ozekmekci 12 These specific tests aid in differentiating organic from PMD and should always be performed in suspected cases.
Depression was the most common psychiatric illness observed in our study with predominance in women. However the prevalence was far less when compared to other studies.Reference Factor, Podskalny and Molho 10 - Reference Fasano, Valadas and Bhatia 14 Others have found somatoform disorders as the predominant psychiatric illness.Reference Munhoz, Zavala, Becker and Teive 20 , Reference Ganos, Aguirregomozcorta, Batla, Stamelou, Schwingenschuh and Münchau 21 Point and life time prevalence rates for psychiatric illness have also been determined.Reference Feinstein, Stergiopoulus, Fine and Lang 24 The lower prevalence of psychiatric comorbidities in our series could be due to social stigma and poor follow-up. It is possible that patients with PMD have significant psychosocial stressors at home or workplace that leads to depression as the most common associated psychiatric comorbidity.
It is not uncommon to find PMD in patients with neurological disorders. Repeated observations and unresponsiveness to medications may help differentiate between the two. In our study, seizures, tic disorders, and palatal myoclonus was the associated neurological illness. Associated neurological illness has been identified in other studies.Reference Fahn and Williams 5 , Reference Factor, Podskalny and Molho 10 , Reference Thomas, Vuong and Jankovic 13 , Reference Ranawaya, Riley and Lang 25 , Reference Monday and Jankovic 26 Research in cognitive neuroscience has tried to explain the relation between organic movement disorder and PMD. There exist various circuits in the brain that give a sense of intention to movement that are disrupted in organic neurological disorder. Functional imaging studies have shown a strong association between the amygdala, supplementary motor area, and prefrontal cortex, suggesting that emotional stimuli influences these areas and culminates in the generation of movement.Reference Voon, Brezing, Gallea and Hallett 27
Patients with PMD are less likely to accept a psychological basis of their symptoms and are prone to consult multiple doctors. Therefore, follow-up of these patients, even in the best hospitals, is dismal and difficult to treat. Response rates in other studies were found to be 4.5% to 57%.Reference Factor, Podskalny and Molho 10 - Reference Thomas, Vuong and Jankovic 13 , Reference Munhoz, Zavala, Becker and Teive 20 Patients with a short duration of illness had better outcomes, which was also observed in our study.Reference Lempert, Dieterich, Huppert and Brandt 8 , Reference Feinstein, Stergiopoulus, Fine and Lang 24 PMD may become chronic because of delayed medical attention, being treated as a neurological disorder for a long duration, coexisting neurological disorders, or social and cultural community factors. The outcome depends on age, onset, and duration; underlying psychopathology; level of literacy; interventional methods; and family support.
In our study, there was no significant difference in the prevalence of PMD among men and women. Female predominance of 61% to 90% has been observed in many studies.Reference Factor, Podskalny and Molho 10 - Reference Ertan, Uluduz and Ozekmekci 12 , Reference Shill and Gerber 18 The low prevalence of women in our study is probably from the usual practice of low referral of women in our country owing to low educational status, social stigma and misconceptions of the disease.
An Overview of PMD in Children
There is a paucity of literature on PMD in children. The prevalence of PMD in children is about 3.1%.Reference Ferrara and Jankovic 28 The mean age of onset of symptoms have been reported to be 12.3±2.6–14.2±2.1 (7.0–17.7) years.Reference Ferrara and Jankovic 28 , Reference Schwingenschuh, Pont-Sunyer, Surtees, Edwards and Bhatia 29 The mean age of onset of PMD in pediatric population in our study was 12.1±3.4 (8–18) years, which is consistent with previous studies. In our series, boys outnumbered girls. This is perhaps an important observation, which suggests that there are probably various psychosocial factors starting from adolescence.Reference Ferrara and Jankovic 28
In a study of 15 cases, the most common type of PMD in children was dystonia (47%) with eight patients having precipitating factor.Reference Schwingenschuh, Pont-Sunyer, Surtees, Edwards and Bhatia 29 Six patients showed complete improvement, and the majority had a short duration of illness. In a study of 54 children, 75% were girls, and sexual assault was observed in 6% and 50% had depression or anxiety. Tremor was observed in 65% followed by dystonia in 43%.Reference Ferrara and Jankovic 28 Follow-up studies of pediatric conversion disorder reports remission rates were between 85% and 97%.Reference Leary 30 Acute onset and short duration symptoms predict good response.Reference Lempert, Dieterich, Huppert and Brandt 8 A multidisciplinary approach with family support is essential in the management of PMD in children.Reference Faust and Soman 31 Tremor (36%) and dystonia (29%) were also commonly observed in another study.Reference Canavese, Ciano, Zibordi, Zorzi, Cavallera and Nardocci 32
Comparison of PMD in Children and Adults
Among adults, the gender ratio was almost same, whereas in children boys predominated. In both the groups, the right upper limb was the most common initial body part involved. However, the clinical phenomenology differed among the two, with tremors being more common in adults and myoclonus being more common in children. Because tremors are rhythmic and require being produced continuously, it may be difficult for the children to perform. Myoclonus that are irregularand arrhythmic, occurring at varying periods, are easy to produce. Children showed a better response to placebo when compared with adults, which is difficult to explain, but it may be related to the different perception of the illness, treatment, and underlying psychology.
A comparison of other studies with the present study is given in Table 5.
Comparison of various studies with present study
Conclusions
The series presented here is the first performed in India including both an adult and pediatric population with data collected systematically and using established and well-accepted diagnostic criteria. In our study, the majority of patients were referred from different states of the country to our tertiary care center. Hence this information cannot be generalized to the general population. Because there is a paucity of studies in PMD, more research needs to be done to better understand the psychopathology and create awareness among physicians and neurologists. We firmly believe that if PMDs are diagnosed early and accurately, appropriate therapy can be instituted in a timely manner and symptoms ameliorate before the development of chronicity.
Disclosures
YCJR served as a principal investigator for and received a research grant from the Indian Council of Medical Research, Department of Science and Technology, and the Department of Biotechnology and an advisor for and honoraria from GSK Pharmaceuticals Ltd. NK, DKP, MJ, MN, and PKP do not have anything to disclose.
