Scrub Typhus-Associated Opsoclonus: Clinical Course and Longitudinal Outcomes in an Indian Cohort.

CONTEXT: Opsoclonus, a rare neurological manifestation in scrub typhus, causes significant distress and disability. There is a paucity of clinical data and outcomes in these patients. AIM: This study aims to describe the clinical and laboratory profile and longitudinal outcomes in a scrub typhus patient cohort with opsoclonus. SETTINGS AND
DESIGN: This retrospective study was conducted in a 2700-bed teaching hospital in South India, in scrub typhus patients with opsoclonus over a 5-year period. PATIENTS AND METHODS: Clinical, laboratory, and radiological data and outcomes at discharge and 6- and 12-weeks postdischarge were documented.
RESULTS: Of 1650 scrub typhus patients, 18 had opsoclonus. 17 had opsoclonus at presentation, while one patient developed opsoclonus on the 5th admission day, 1-day postdefervescence. Opsoclonus was first noted after a median interval of 11 (7-18) days from fever onset. It was associated with myoclonus in 94% (17/18), cerebellar dysfunction in 67% (12/18), extrapyramidal syndrome (EPS) in 33% (6/18), and aseptic meningitis in 17% (3/18) patients. Mean cerebrospinal fluid (CSF) white blood cell (WBC) count was 9 ± 2.7 cells/cumm, with mean CSF protein 118.5 ± 53.9 mg% and mean CSF glucose 97 ± 13 mg% in 1l/15 patients. Brain magnetic resonance imaging was unremarkable in 75% (9/12). Case-fatality rate was 5.5% (1/18). Complete resolution of the index neurological syndrome occurred at 12-week postdischarge.
CONCLUSIONS: Opsoclonus is a rare neurological manifestation in scrub typhus, usually occurring in association with myoclonus, cerebellar dysfunction, or EPS. It appears to occur during the resolving febrile phase, with neurological deficits completely resolving at 12 weeks.

INTRODUCTION

Scrub typhus is a mite-borne zoonosis of epidemiological significance in Southeast Asia and the Indian Subcontinent, with an annual incidence of 1 million and 30% mortality in these endemic areas. Caused by Orientia tsutsugamushi, it classically manifests as an acute multisystem febrile illness with hepatocellular injury, acute renal failure, and acute lung injury, possibly secondary to small-vessel endothelitis.[1] Neurological syndromes are a prominent clinical component with meningitis or meningoencephalitis occurring in 12%–26% of affected patients.[2] In recent years, there have been increasing reports of rarer neurological syndromes in scrub typhus patients. These include cerebellitis, extrapyramidal syndrome (EPS), myoclonus, and opsoclonus.[34567]

Although infrequently reported, opsoclonus causes significant distress and disability in those affected, by preventing retinal image fixation and contributing to restricted activities of daily living and recurrent falls. Best recognized as part of the opsoclonus-myoclonus-ataxia syndrome (OMS), associated with neuroblastoma in children, its occurrence in adults is rare. While the incidence of adult-onset isolated opsoclonus remains unclear, 66 cases have been reported since 1986.[8] Adult-onset opsomyoclonus occurs in 1–2 people/10 million/year, with a 2012 review spanning 65 years at Mayo Clinic, Minnesota, identifying 137 patients.[910]

Broadly, adult opsoclonus is a part of paraneoplastic (adenocarcinoma breast or small-cell carcinoma lung), autoimmune (celiac disease), parainfectious (Epstein-Barr virus, human Immunodeficiency virus, or Streptococcal infections), or idiopathic neurological phenomena.[11121314151617] Scrub typhus-associated opsoclonus is exceedingly rare with six reported cases in literature.[718192021] Given the paucity of clinical data, we set out to describe the clinical profile, laboratory features, and longitudinal outcomes, in a historical cohort of scrub typhus patients with opsoclonus.

PATIENTS AND METHODS

Setting

This retrospective study was carried out at Christian Medical College, Vellore, a 2700-bed teaching hospital in South India, with an average of 2500 inpatients and 8000 outpatients/day. The predominant catchment area for the hospital includes the districts of Vellore and Tiruvannamalai in Tamil Nadu state and Chittoor district of Andhra Pradesh. The study was approved by the Institutional Review Board and Ethics Committee, and the study protocol followed the principles of the Declaration of Helsinki. The hospital electronic database was screened for patients with a diagnosis of scrub typhus with opsoclonus, admitted between January 2012 and December 2017. Data were collected from inpatient and outpatient medical records. Patient confidentiality was maintained using unique identifiers. Patient data included age, gender, admission date, discharge date or date of death, and admission diagnosis. Clinical and laboratory profiles, radiological data, outcomes and neurological status at discharge, and 6- and 12-weeks postdischarge were documented.

Inclusion criteria

All patients aged >16 years with a diagnosis of scrub typhus, who presented with or developed opsoclonus during their hospital stay were included. A diagnosis of scrub typhus was considered valid in patients hospitalized in Intensive Care Units (ICU)/wards with a febrile illness >3 days and <14 days, with positive scrub typhus immunoglobulin M (IgM), enzyme-linked immunosorbent assay (ELISA), and other serologies negative and/or an eschar characteristic of scrub typhus demonstrable on physical examination. Scrub typhus was confirmed by IgM, ELISA positivity, and/or the presence of a pathognomonic eschar with polymerase chain reaction (PCR) confirmation where feasible. IgM ELISA was performed on serum samples using Scrub Typhus Detect (InBios International Inc., Seattle, WA, USA) as per the manufacturer's instructions. An optical density >0.5 was considered positive. A standard PCR targeting, the 56-kDa outer membrane protein genes was carried out in a subset of cases for the diagnostic confirmation of scrub typhus. A subset of patients was also screened for serum anti-onconeural (anti-Hu, anti-Ri, anti-Yo, anti-Ma/Ta, anti-CV2, anti-amphiphysin, Zic-4, and Sox-1), antineuronal (antimyelin, antimyelin-associated glycoprotein, and antiglutamic acid decarboxylase), and antiganglioside antibody panels.

Exclusion criteria

Patients with an acute febrile illness with definitive etiological diagnosis other than scrub typhus; with diagnosed malignancies; on immunosuppressant therapy; or diagnosed with immunodeficiency diseases and with autoimmune disorders, irrespective of treatment status, were excluded from the study.

RESULTS

Demography

A total of 18 of 1650 scrub typhus patients were diagnosed with opsoclonus during the study period. Fifteen patients were from Tamil Nadu, 2 from Andhra Pradesh, and 1 from Karnataka. There were nine males and nine females. The mean age of the patients was 38 ± 16.8 years. The patient characteristics are summarized in Table 1.

Table 1

Demography, clinical profile, and laboratory investigations (n=18)

Variable	n (%)
Age (years), mean±SD	38±16.8
Male: female	1:1
Duration of febrile illness at admission median days (range)	10 (4-15)
Onset of opsoclonus relative to onset of illness median days (range)	7 (5-11)
Presenting clinical features
Headache	13 (72)
Nausea/vomiting	12 (66)
Altered sensorium	7 (39)
Seizures	2 (11)
Eschar	8 (44)
Laboratory features
Total WBC counts (cells/cumm), mean±SD	10,890±4670
Platelet counts (cells/cumm), mean±SD	130,564±75,000
Serum SGOT/SGPT** (U/L), mean±SD	158±32/69±28
Serum creatinine (mg/dl), mean±SD	2.86±1.1
Comorbidities
Diabetes mellitus	2 (11)
Hypertension	1 (6)
Alcohol consumption	1 (6)
Current cigarette smoking	4 (22)
Chronic lung disease	1 (6)

SD=Standard deviation, WBC=White blood cell, SGOT=Serum glutamic oxaloacetic transferase, reference range 8–40 U/L, SGPT=Serum glutamic pyruvic transferase, reference range 5–35 U/L

Clinical profile

The median duration of fever at admission was 10 (range 4–15) days. An eschar was documented in eight patients, with common sites being the groin, genitalia, axillae, neck, and inframammary folds. Thirteen patients presented with headache. Nausea and vomiting were present in 12 patients, altered sensorium in 7, and seizures in 2. One of 18 patients required mechanical ventilation and inotropic support in the ICU for acute respiratory distress syndrome (ARDS) and septic shock. 17/18 patients had opsoclonus at admission with a mean duration of 1.8 ± 2 day prior presentation. In these patients, opsoclonus first occurred after a median interval of 7 (range 5–11) days from fever onset. One patient developed the isolated opsoclonus following hospitalization, on the 5th admission day. This patient had presented with acute-onset high-grade intermittent fever for 6 days prior admission. His laboratory investigations at admission were significant for thrombocytopenia and elevated serum transaminase levels. He was initiated on therapy with oral doxycycline 100 mg twice daily. His vital parameters, platelet counts, and serum transaminase levels were monitored daily. He was noted to have developed opsoclonus 1 day following defervescence. At this time point, his thrombocytopenia had resolved and serum transaminase levels displayed a decremental trend [Figure 1].

Figure 1

Temporal profile of fever and serum glutamic oxaloacetic transaminase levels in a patient with scrub typhus with opsoclonus onset on day 5

Nine of the 18 (50%) patients had concomitant positive myoclonus. Of these, four had focal, four multifocal, and one generalized myoclonus. The distal upper limbs were affected in 8 of the 9 patients. Myoclonic jerks were precipitated by voluntary action or by the stimulus of touch. Eight patients had concomitant signs of bihemispheric cerebellar dysfunction and six had associated EPS. Of these, all patients had asymmetrical cogwheel rigidity in their upper limbs, five displayed resting tremors, and five had a Parkinsonian gait. Three of 18 patients had associated clinical meningitis/meningoencephalitis [Table 2].

Table 2

Neurological associations, cerebrospinal fluid analysis, and radiological features (n=18)

Variable	n (%)
Neurological syndromes-associated with opsoclonus
Myoclonus	9 (50)
Cerebellar dysfunction	8 (44)
Extrapyramidal syndrome	6 (33)
Concomitant meningitis/meningoencephalitis	3 (17)
CSF analysis
Meningitis group (n=3)
Mean CSF WBC (cell/cumm)	55±12.2
Mean CSF protein (mg/dl)	69.3±12.2
Mean CSF glucose (mg/dl)	50±8
Nonmeningitis group (n=11)
Mean CSF WBC (cell/cumm)	9±2.7
Mean CSF protein (mg/dl)	118.5±53.9
Mean CSF glucose (mg/dl)	97±13
MRI brain (n=12)
Cerebellitis	3 (25)
Brain stem enhancement	0
Normal	9 (75)
Outcome
Duration of hospital stay (mean±SD days)	11±7.8
Case fatality rate	1 (6)
Resolution of nonneurological illnesses at discharge	16 (94)
Persistence of neurological syndrome at discharge	17 (100)

SD=Standard deviation, WBC=White blood cell, CSF=Cerebrospinal fluid

Laboratory and radiological characteristics

Scrub typhus ELISA was positive in all patients. Thrombocytopenia (platelet count <100,000 cell/cumm) was seen in 7/18 (38.8%) patients; hepatic involvement (threefold elevation of serum glutamic oxaloacetic transaminase/serum glutamic pyruvate transaminase) in 12/18 (66.6%); pulmonary involvement (PF ratio <300) in 1/18 (5.5%); and renal dysfunction (serum creatinine >2 mg/dl or need for dialysis) in 3/18 (16.6%) patients. Cardiovascular involvement (redistributive shock or inotropic support) occurred in one patient (5.5%). Concomitant involvement of ≥3 organ systems (severe scrub typhus) occurred in 3/18 (16.6%) of patients.

Cerebrospinal fluid (CSF) analysis in the three patients with meningitis/meningoencephalitis, revealed a mean CSF white blood cell (WBC) count of 55 ± 12.2 cells/cumm with lymphocytic pleocytosis. The mean CSF protein was 69.3 ± 12.2 mg% and mean CSF 50 ± 8 glucose mg%. Among patients without concomitant meningitis/meningoencephalitis (15/18), CSF analysis was not performed in two patients. Two of the remaining 13 patients had normal parameters on CSF analysis. These were not included in the calculation of mean to prevent an outlier effect. In the other 11 patients, the mean CSF WBC count was 9 ± 2.7 cells/cumm with lymphocytic pleocytosis in all samples. The mean CSF protein was 118.5 ± 53.9 mg/dl and mean CSF glucose 97 ± 13 mg/dl [Table 2]. Serum screens for anti-onconeural, antineuronal, and antiganglioside antibody panels, performed in two patients from this subgroup were negative.

Magnetic resonance imaging (MRI) scans of the brain were performed in 12 patients. In three patients, imaging demonstrated the effacement of bilateral cerebellar folia with mild increase in T2 fluid-attenuated inversion recovery intensity suggestive of diffuse cerebellar edema [Figure 2]. There was no postcontrast enhancement or diffusion restriction to suggest cytotoxic edema or infarct. No focal lesion was seen. One of the three patients was also noted to have diffuse leptomeningeal enhancement suggesting meningitis. About two-third patients with radiological cerebellar abnormalities had clinical features of cerebellar dysfunction. No significant radiological abnormalities were noted in 9/12 patients. None of the patients imaged had hyperintense or contrast-enhancing pontine, midbrain, or medullary lesions [Table 2].

Figure 2

Magnetic resonance imaging scans of a patient with scrub typhus opsoclonus with cerebellar dysfunction. Images at the level of cerebellum showing effacement of cerebellar folia on T2W (arrow, image 2a) with mild increase in T2-fluid attenuated inversion recovery intensity (arrow, image 2b) and no postcontrast enhancement (image 2c)

Therapy and outcomes

All patients were treated with a 1-week course of doxycycline with or without intravenous azithromycin. The patient requiring ICU admission was also treated with intravenous systemic glucocorticoids for ARDS. Six patients were initiated on clonazepam due to significant myoclonus. The mean dose of clonazepam on the day preceding discharge was 1.86 (±0.5) mg. The mean duration of hospital stay was 11.3 (±6.5) days. The case fatality rate was 6% (1/18) with the patient with severe scrub typhus succumbing to ventilator acquired pneumonia and refractory shock on the 15th hospital day [Table 2].

Before discharge, convalescing patients were initiated into relevant rehabilitation programs focused on passive physiotherapy for muscle strength improvement, exercises to improve supported gait, coordination and balance, and a daily session of occupational therapy to improve autonomy in activities of everyday life and prevent falls. They were counseled to continue physical and occupational therapy following discharge.

A total of 9/17 (52.9%) patients had a complete resolution of myoclonus at discharge. Improvement in the clinical severity of other neurological syndromes was noted in all patients at discharge. All patients except one had a complete resolution of other clinical and laboratory abnormalities. This patient had resolving renal failure at discharge and was subsequently noted to have normal serum creatinine levels as an outpatient, 1 week later.

Four patients were lost to follow-up at 6-week postdischarge. Of the 13/17 patients, myoclonus was noted to have completely resolved in all patients, while opsoclonus persisted in nine. One patient with cerebellar dysfunction and opsoclonus reported two occasions of fall. In both instances, the patient fell, while attempting unaided ambulation. He was evaluated in a medical facility in his hometown and was not noted to have suffered significant injuries. Other persisting neurological features have been mentioned in Table 3. A total of 12/17 patients presented for clinical review at 3 months. A complete resolution of opsoclonus and cerebellar dysfunction was noted in all 12 patients with persisting EPS in one patient. This patient was noted to have persisting EPS, characterized by cogwheel rigidity and resting tremors in the left upper limb, activated rigidity in the right upper limb, and Parkinsonian gait. Downtitration of clonazepam in myoclonus patients and treatment with oral trihexyphenidyl 2 mg thrice daily, in the patient with persisting EPS were initiated.

Table 3

Ambulatory review

Variable	n/N (%)
6-week postdischarge
Lost to follow-up persisting neurological syndrome	4/17 (22)
Opsoclonus	9/13 (64)
Myoclonus	0 (0)
Cerebellar dysfunction	4/13 (31)
Extrapyramidal syndrome	2/13 (15)
12-week postdischarge
Lost to follow-up	5/17 (29)
Persisting neurological syndrome
Myoclonus	0/12 (0)
Cerebellar dysfunction	0/12 (0)
Extrapyramidal syndrome	1/12 (8)

DISCUSSION

To the best of our knowledge, this is the largest descriptive study on the clinical course and longitudinal outcomes in scrub typhus patients with opsoclonus.

Opsoclonus results from a disruption of the pontine reticular omnipause neuronal inhibitory tone on saccadic burst neurons and cerebellar fastigial nuclear disinhibition due to dysfunction of Purkinje cells in the dorsal vermis.[2223] In children, it usually occurs as the opsoclonus–myoclonus syndrome (OMS) (opsoclonus, myoclonus, ataxia, and sleep or behavioral disturbances) affecting those <3 years of age and associated with neuroblastomas in 50% of all cases.[24] A adult-onset opsoclonus may either be isolated or occur as opsomyoclonus. While opsoclonus is only one of several eye movement disturbances-associated with myoclonus, myoclonus is the dyskinesia most often associated with opsoclonus.[25]

Fifty percent (9/18) of patients in our study presented with opsomyoclonus. Clinical characteristics in this subgroup were indicative of cortical myoclonus, possibly localizable to the sensorimotor cortex in those with focal manifestations. A review of existing literature on opsoclonus in scrub typhus, available as five case reports, detailing clinical features, and outcomes in six patients, revealed that four patients presented with isolated opsoclonus.[71820] The other two patients presented with opsomyoclonus.[1921] Although reported individually and concomitantly in scrub typhus patients, cerebellar dysfunction and EPS have not been described previously in association with opsoclonus in this illness.[3456] We demonstrate a clinical pattern, wherein scrub typhus-associated opsoclonus is frequently but not always accompanied by varying combinations of the above-mentioned neurological manifestations.

CSF analysis in opsoclonus patients without meningitis was significant for albuminocytological dissociation. In a study on clinical outcomes in adult-onset idiopathic or paraneoplastic opsoclonus-myoclonus involving 24 patients, CSF analysis was noted to be normal or show mild elevations of protein and mild lymphocytic pleocytosis.[26] Of the six patients with scrub typhus opsoclonus reported in literature, CSF analysis was performed in five.[7192021] The analysis was normal in two patients.[721] CSF albuminocytological dissociation was seen in two patients opsomyoclonus opsomyoclonus. Details of CSF analysis were unclear in the fifth patient.[19]

Radiological abnormalities were present only in 25% (3/12) patients, with poor clinicoradiological correlation. Cerebellar abnormalities were detected in only 25% (2/8) patients, with clinical features of cerebellar dysfunction. Pontine and midbrain hyperintensity or postcontrast enhancement was not seen in any of the study patients. Brain MRI is usually normal in patients with idiopathic OMS but may occasionally show hyperintensity in the dorsal pons or midbrain in T2-weighted images.[27] Among the six cases of scrub typhus opsoclonus reported in literature, MR imaging of the brain was performed in five patients and was normal in all.[7192021]

The significant time interval between the onset of fever and opsoclonus (11 days [range 7–18 days]) and CSF albuminocytological dissociation in a majority of patients (69%) when combined with observations in the study patient who developed opsoclonus postadmission, at a timepoint when other clinical parameters (fever, thrombocytopenia, and serum transaminase elevation) were resolving [Figure 1], suggest an underlying immune-mediated pathogenic mechanism.

Seven–ten days after exposure to O. tsutsugamushi and just before the germinal-center formation, rapid class-switch recombination from IgM to IgG would result in rising anti-O.tsutsugamushi IgG titers. Anti-O. tsutsugamushi IgG antibodies would then gain access across the blood–brain barrier, triggering a Type-2 hypersensitivity reaction, possibly directed against self-antigens in the pons, midbrain, and/or cerebellum. In contrast, other manifestations such as meningitis/meningoencephalitis, ARDS, myocarditis, and hepatitis in scrub typhus, have been shown to be consequent to disseminated vasculitis following rickettsial vascular endothelial invasion rather than immune-mediated injury.[1] In addition, studies also suggest the direct CSF invasion by O. tsutsugamushi in patients with meningitis/meningoencephalitis.[28]

Pediatric paraneoplastic and nonparaneoplastic opsomyoclonus are believed to be immune mediated, with antibodies directed against the cerebellar granular neurons. The identity of the autoantigen remains unknown.[29] Well-characterized autoantibodies are rare in cases of adult opsomyoclonus with only a small subset of patients demonstrating anti-Ri antibodies against Nova proteins breast cancer-associated opsoclonus (rarely gynecologic, lung, and bladder) and anti-Hu antibodies in small-cell lung cancer.[1030] Anti-neuroleukin antibodies have been recently demonstrated in patients with Streptococcal infection-associated opsomyoclonus.[16] Two patients in our cohort screened negative for serum anti-onconeural, antineural, and antiganglioside antibody panels. Antibody screening was performed in an attempt to identify putative opsoclonus-associated autoantigens.

All patients had a significant improvement in the extent of their neurological syndromes at discharge. The earliest neurological syndrome to completely resolve was myoclonus, possibly attributable to the initiation of clonazepam. While opsoclonus and cerebellar dysfunction resolved completely at 12 weeks, EPS persisted in one patient. Of the two reported cases of scrub typhus-associated EPS, the syndrome resolved spontaneously within 2-week postdischarge in one patient and required amantadine therapy due to persistence in the second patient.[56] The complete resolution of scrub typhus-associated opsoclonus has also been documented in all cases in literature. The time to resolution in these patients varied between 2 days and 2 weeks.[718192021] Opsoclonus due to other diseases appears to be more persistent, but usually resolves with immunosuppression in adults.[10] Pediatric OMS poses a greater challenge, with nonparaneoplastic pediatric OMS requiring aggressive immunotherapy. Paraneoplastic opsoclonus in children with neuroblastoma usually persists despite tumor removal.[10]

While opsoclonus in the study patients resolved without immunomodulatory intervention, the role of immunosuppression in shortening the duration of opsoclonus and its associated neurological syndromes remains unclear. In 5 of 6 cases in literature, opsoclonus and its associated syndromes resolved without glucocorticoid therapy.[7192021] The reported resolution of opsoclonus within 48-h, in a 60-year-old Indian male with scrub typhus, multiorgan dysfunction, and isolated opsoclonus, who received intravenous systemic glucocorticoids for ARDS, may suggest a role of systemic glucocorticoids in shortening opsoclonus duration.[18] This fact cannot however be corroborated by our study since the study patient receiving intravenous glucocorticoids succumbed to his illness before opsoclonus resolution.

Our study was not without limitations. The use of video-oculography in opsoclonus assessment could have contributed to a greater accuracy in diagnosis and in objectivity in documenting disease resolution. A comprehensive antibody screening of CSF and serum samples were also not performed in all study patients. The small sample size due to the rarity of this condition limits the generalizability of our findings.

CONCLUSIONS

Opsoclonus is a rare neurological manifestation of scrub typhus. It may either have an isolated presentation or occur in combination with myoclonus, cerebellar dysfunction, and/or EPS. The occurrence of opsoclonus during the resolving phase of illness associated with albuminocytological dissociation on CSF analysis suggests an immune-mediated pathogenic mechanism. Brain imaging does not reveal localizing lesions in a majority of cases. Unlike other causes of opsoclonus, neurological syndromes usually resolve without immunosuppression by 12-week postdischarge.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.