Pediatric opsoclonus-myoclonus-ataxia syndrome: Experience from a tertiary care university hospital
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.241404
Source of Support: None, Conflict of Interest: None
Keywords: Methyl prednisolone, neuroblastoma, paraganglioma, paraneoplastic, pediatric opsoclonus-myoclonus-ataxia, steroids
Opsoclonus-myoclonus-ataxia syndrome (OMAS) was first described by Kinsborne in 1962. It is a rare and distinct syndrome of early childhood characterized by truncal and appendicular ataxia, rapid conjugate multidirectional eye movements (opsoclonus), and jerky movements of the body with tremulousness, behavioral, and sleep disturbances. OMAS is often misdiagnosed as acute cerebellar ataxia in children. It is associated with a tumor, viz. neuroblastoma and ganglioneuroma among others, in 40–50% of the cases. OMAS has a good therapeutic response to immunomodulation. Early diagnosis and prompt treatment prevent neuropsychiatric sequelae and morbidity in children. There are relatively few studies which describe the therapeutic response and outcome in children with OMAS and only one case series from India.,,,,,, In this study, we retrospectively analyzed the clinical profile, laboratory investigations, and therapeutic outcome in pediatric OMAS.
This study was carried out at the National Institute of Mental Health and Neurosciences (NIMHANS), Bengaluru, India, which is a tertiary care referral centre for neuropsychiatric cases. Patients were diagnosed with OMAS based on the previously proposed criteria, i.e., the presence of three or more of the following: (1) Opsoclonus; (2) myoclonus and/or ataxia; (3) behavioral abnormalities and/or sleep disturbances; and (4) the presence of a neuroblastoma. Patients aged 5 years or less, diagnosed with OMAS between 2006 and 2015, who were evaluated and followed up in a single neurology unit were included in the study. Their case records were retrieved and reviewed.
Clinicodemographic data were extracted from case files. The data included the patients' age, gender, birth and development history, and preceding events, if any. Clinical profile details including the age of onset, duration, symptomatology, investigations, follow-up duration, relapses, and response to treatment were noted. All patients underwent brain magnetic resonance imaging (MRI), urinary vanillylmandelicacid (VMA), and lumbar cerebrospinal fluid (CSF) analysis. Other laboratory tests such as the complete blood count, hepatic and renal function tests, lactate, and vitamin B12, among others, were performed based on the clinical discretion. Imaging studies performed to identify the underlying tumor included computed tomography (CT) scan of the abdomen (n = 12), thorax (n = 3), and neck (n = 1); metaiodobenzylguanidine (MIBG) scan (n = 3); and fluorodeoxyglucose positron emission tomography [FDG-PET] (n = 2). The protocol used for various imaging studies was as follows. A helical CT scan from the diaphragm to the pelvic diaphragm was performed using 90 kV and 200 mAs (Philips Brilliance Royal Philips Electronics, The Netherlands). An oral (1200 ml of 2% ionic contrast) and an intravenous contrast (50 ml of nonionic contrast) 0.938 pitch was also simultaneously administered. Serial 5-mm axial reconstructions together with coronal and sagittal constructions were obtained. The MIBG scan was performed 24 and 48 hours after intravenous injection of 300 mci of MIBG. Whole body FDG-PET was performed in fasting state 60 minutes after the intravenous injection of 3.5 mci of gallium (Ga)-68 DOTANOC/10 mci of fluoro-18-L-dihydroxyphenylalanine (18F-FDOPA). CT images were retrieved from the Picture Archival and Communication System (PACS) and reviewed (AH, PSB, MN, and RDB). Doubts were resolved by discussion with other authors (ABT and SS).
All patients received intravenous methylprednisolone (20–30 mg/kg/day infusion for 5 days) as induction therapy. Subsequently, the patients received oral prednisolone (0.75–1.0 mg/kg/day) or monthly intravenous pulsed methylprednisolone. The patients were followed up and assessed clinically for symptomatic response to treatment, relapses, persisting behavioral, and neurological abnormalities. The data were entered into a predesigned proforma and incorporated into a Microsoft Excel spreadsheet for analysis.
Twenty-three patients were diagnosed with OMAS during the study period. Of these, 14 were aged 5 years or less (the mean age: 21.7 ± 7 months, age range: 10–36 months, male:female = 1:2.3) and were included for analysis. The mean duration of symptoms was 3.8 ± 5.0 months (range: 15 days to 16 months). All patients had uneventful birth and development milestones. [Table 1] summarizes and compares the clinical features, treatment, and follow-up details of patients with and without tumor. The most common referral diagnosis was acute cerebellar ataxia (n = 8). Antecedent events were noted in 10 patients, with the most common being upper respiratory tract infection (50%). The clinical features noted at presentation in the order of frequency were ataxia (71%), tremulousness (57%), abnormal eye movements (50%), behavioral abnormalities (28%), and regression of milestones (21%). Behavioral abnormalities included irritability (60%) followed by anger outbursts (40%) and clinging to mother (20%).
Brain MRI, urinary vanillylmandelic acid (VMA), cerebrospinal fluid (CSF) analysis, complete blood count, hepatic and renal function tests, and lactate were normal in all patients. An underlying neuroblastoma/paraganglioma was detected in 7 patients. The sites of the neuroblastoma included the abdomen (n = 4), thorax (n = 2), and neck (n = 1) [Figure 1]. Histopathological confirmation was available in 4 patients.
Treatment and follow-up
The mean duration between the onset of symptoms and the initiation of immunomodulation was 6.1 ± 7.5 months (range: 2–24 months), and the mean duration of treatment was 22.3 ± 20 months (range: 3 months to 5 years). Intravenous immunoglobulin (IVIG) was administered in 1 patient because of the presence of severe disease. Following induction with intravenous methylprednisolone, maintenance therapy was initiated (n = 13), including monthly methylprednisolone alone (n = 4), methylprednisolone followed by oral steroids (n = 4), and oral steroids alone (n = 5). The number of pulses of methylprednisolone was decided based on the clinical response and the patient's compliance to treatment. Among patients who were on oral steroids, 3 received pulsed methylprednisolone for relapse of the symptoms. Symptomatic treatment included sodium valproate (n = 5), clonazepam (n = 6), and clonidine (n = 2).
Out of 7 patients with tumor, 3 patients underwent surgical removal. The mean duration between the onset of symptoms and surgery (n = 3) was 28 ± 17 months (12–46 months). The duration of follow-up following surgery ranged from 2 weeks to 2 years. Two patients had persisting symptoms and required immunomodulation even after surgery.
The mean follow-up (n = 13) was 31.3 ± 19 months (range: 7–62 months). Three patients did not attain remission. In the remaining patients, the mean duration from initiation of treatment to remission was 12 ± 14 months (range: 2 months to 4 years). Twelve relapses were noted in 6 patients. In 5 patients, relapses were precipitated by fever and infection and in 1 patient due to stopping of steroids. Relapses were mainly in the form of increased ataxia or tremulousness, rather than the presence of a full syndrome. At the time of the last follow-up, only 5 patients had achieved complete remission, whereas the rest had persisting mild behavioral and cognitive abnormalities [Table 1]. In 4 patients, an underlying neuroblastoma or a paraganglioma was detected at the time of first evaluation itself, whereas in the remaining 3 patients, it was detected during the follow-up visit.
OMAS is rare with an incidence of 0.27 cases/million population in Japan and 1.8/million in the United Kingdom. A previous study from India showed that this disorder contributes to nearly 7% of acute movement disorders in children. There is only one case series from India describing 11 children with OMAS. The most common age of onset is 1–3 years.,, The common underlying etiologies include paraneoplastic (21%), and parainfectious causes (55%). Less common etiologies such as vaccination and intoxication with phenytoin have been proposed.,,,,,,, Recent case studies from India have reported infectious causes such as scrub typhus,, dengue, malaria, and varicella zoster virus. Due to these associations and symptomatic response to immunomodulatory therapies, the pathophysiology is thought to be autoimmune in nature.,, Autoantibodies against antigens on neuronal and dendritic surface, namely N-methyl-D-aspartate (NMDA), glutamic acid deacetylase (GAD), anti-Hu and gamma amino butyric acid (GABA B), are reported.,,,, Recently, glucose transporter-1 deficiency has also been suggested as an etiological cause. Recent studies have reported a new hypothesis that proposes disinhibition of the fastigial nucleus in the cerebellum. This is evidenced by an increased activity in bilateral cerebellar nuclei, as detectable by functional MRI scans, in response to administration of clonazepam;,, and, an atrophy of vermis and floculonodular lobes in patients with persisting and severe symptoms. Damage to omnipause cells in the nucleus raphe interpositus of pons (which prevents excessive saccades by inhibiting burst neurons in the para-pontine reticular formation) and rostral interstitial nucleus of Cajal has also been reported.
In this study, children constituted 60% of the entire cohort. Similar to previous studies, we observed that respiratory illness was the most common antecedent event and ataxia was the most common presenting feature. Behavioral and sleep abnormalities were noted in 70–80% of the patients, which is also similar to previous studies.,, However, girls outnumbered boys in contrast to earlier studies.,,,, Klein et al., also reported a female preponderance. There was no significant difference in the clinical profile or sequelae between the patients with or without tumor, a finding that was similar to that obtained in the previous studies., A comparison between this study and previous case series is presented in [Table 2].
Our study highlights that, even when urinary VMA and iodine-123 meta-iodobenzylguanidine (MIBG) are normal, the CT scan aids in detection of an underlying tumor. This has been noted in previous studies and has been attributed to the low grade of the tumor. A neuroblastoma/paraganglioma was detected in 50% of the patients, whereas in the previous studies, this incidence ranged between 7–65%.,,,,, Thus, a diligent search with a CT scan of the abdomen, thorax, and neck not only has a high yield but is also cost-effective compared to the FDG-PET. This observation is particularly relevant in countries such as India where cost and availability of investigations is a major constraint. In recent studies, the presence of infection has been found to influence the treatment and outcome., In the current study, investigation for infectious causes was not done as most of cases were referred in the subacute or chronic stage.
A good therapeutic response to immunomodulation was noted, similar to the findings of previous studies.,,,,,, Steroids were used as the mainstay of treatment in the current study, either in the form of intravenous pulse methylprednisolone and/or oral prednisolone. Clinical remission was noted in a majority of the patients (>80%), thus eliminating the need for other agents such as cyclophosphamide and rituximab, which are more toxic and expensive. Clinical remission, relapses, and sequelae noted in the present study were similar to the findings of previously reported studies, where intravenous immunoglobulin (IVIG), adrenocorticotrophic hormone (ACTH), and immunomodulation were used., Rituximab and cyclophosphamide were not used in the present study in view of a good therapeutic response with steroids as well as the cost constraints. However, most of the patients required a prolonged immunomodulation therapy. The usefulness of IVIG  and plasmapheresis  in cases with partial recovery to prednisolone has been reported. In this study, only 1 patient received IVIG due to partial recovery with prednisolone. Recent studies have suggested a multimodal approach and the use of rituximab based on the role of B cell and B cell activating factor (BAFF) in the pathogenesis of OMAS.,, Monotherapy with clonazepam ,, or doxycycline in the case of an infectious etiology, particularly due to scrub typhus,, leading to complete recovery, has also been reported. Based on the current literature, we propose an algorithm for the evaluation and management of OMAS [Figure 2].
There was no difference in the duration of immunomodulation and outcome in patients with or without a tumor. This observation is similar to the findings of previous studies.[4-7,9] In contrast, Singhi et al., reported that the presence of a tumor was an indicator of a worse outcome. Larger randomized controlled studies are required for confirming these results. It was noted that most patients required prolonged treatment for maintenance of their improved outcome.
Approximately half of the patients had relapses that were temporally related to infection or tapering of steroids. Persisting neuropsychiatric features, including hyperactivity and a poor scholastic performance, were noted in less than half of the cohort, in contrast to the speech and motor disturbances noted in the previous studies [Table 2]. A follow-up is required in these patients not only for monitoring treatment, relapses and persisting abnormalities but also for the detection of tumor, as an underlying tumor was detected in 3 patients during the follow-up visit in the present study.
This study has limitations of being a retrospective, smaller cohort study and a lack of a uniform treatment protocol. Nevertheless, it highlights the therapeutic response to pulsed or oral steroids and the requirement for a prolonged maintenance therapy. We already know that OMAS is a rare and distinct pediatric syndrome associated with an underlying tumor in 40–50% of cases. Behavioral and sleep disturbances are commonly seen. It is responsive to immunomodulation. What the current study adds is that there is a female preponderance, especially in the subset with an underlying tumor. A diligent search with CT scan of the abdomen, thorax, and neck is the most sensitive, cost-effective, and yielding investigation to detect an underlying tumor. Monotherapy with pulse methylprednisolone, both for acute and maintenance phase, appears to be a good therapeutic option. What remains to be known is the optimal duration of immunomodulation required to balance the relapse rates versus the side effects of treatment.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2]