Background: Chronic hepatic disease can present with extrapyramidal symptoms. We describe two cases that presented with highly unusual movement disorders: ballism and gait freezing.
Case report: Patient 1 is a 42-year-old man with previous episodes of hepatic encephalopathy (HE) who presented with upper limb dystonia and generalized chorea that progressed to ballism. Patient 2 is a 55-year-old woman who presented with pronounced gait freezing. In both patients, features of HE and acquired hepatocerebral degeneration coexisted. They improved markedly, though transiently, with rifaximin.
Discussion: Ammonia-reducing treatments should be considered in patients presenting with movement disorders due to chronic liver disease.
Compromised liver function as a consequence of either acute hepatic failure or advanced chronic liver disease leads to insufficient detoxification of neurotoxic substances, which accumulate in the blood and are responsible for a broad range of neurological manifestations. These include hepatic encephalopathy (HE) and acquired hepatocerebral degeneration (AHD).1 Clinical manifestations usually involve movement disorders and cognitive features. In HE, improvement may be seen with purgative ammonia-lowering therapies such as lactulose. However, in AHD liver transplant is considered the only effective treatment. We present two cases of advanced liver disease with uncommon movement disorders (freezing of gait and ballismus) that reversed with rifaximin.
A 42-year-old male was first seen in the Neurology clinic due to involuntary movements. His medical history included past alcohol and drug abuse, hepatitis C virus (HCV) infection, and hepatic cirrhosis Child-Pugh C with portal hypertension and hepatopulmonary syndrome. He had been on methadone and olanzapine 2.5 mg id for the last 18 months. There were reports of HE episodes in the previous 8 months with consciousness and attention fluctuation, dysarthria, and motor incoordination that subsided with lactulose treatment. On neurological examination, he presented generalized chorea with orolingual, cervical, and asymmetric limb involvement (worse on the left), exacerbated by movements. There was also a symmetric dystonia of both hands. The gait was slightly ataxic (Video 1, segment 1). Additional neurological examination showed moderate dysarthria and dysphagia. Brain MRI revealed T1-weighted hyperintensities in the globus pallidus (GP), subthalamic nuclei, and midbrain (Figure 1). T2-weighted sequences showed mild bilateral putaminal hyperintensities, particularly on the right side. Ammonia levels were elevated (126 μmol/L, normal range 26–47 μmol/L) and manganese levels were normal (16.5 ng/mL, normal range 4.7–18.3 ng/mL). A basic metabolic screen revealed hypoalbuminemia, thrombocytopenia, and an INR of 1.89 and excluded other metabolic alterations. One month later he was seen at the emergency department, awake and fully oriented, with rapid, involuntary, nonstereotypical, violent ballistic limb movements and choreic movements of the tongue and perioral area (Video 1, segment 2). Ammonia levels were 150 μmol/L. He was treated with haloperidol 5 mg IM with no response and later with lactulose and rifaximin 800 mg id with marked improvement of the ballistic and choreic movements (Video 1, segment 3). Ammonia values 3 days after treatment were 61 μmol/L.
Since then, maintaining treatment with rifaximin, he oscillated between a fully conscious and oriented state with marked hypotonia and dysarthria and episodes of confusion with severe generalized chorea (Video 1, segment 4). Ammonia levels also showed significant variation (between 78 and 327 μmol/L), with higher values usually in the context of a concurrent infection or constipation. There were, however, some doubts on medication adherence.
A 55-year-old woman was referred to our movement disorder clinic due to gait imbalance and falls with a fluctuating pattern. She had had a liver transplant due to alcoholic cirrhosis and developed graft dysfunction in the first year after transplant, coinciding with imbalance onset. Six years later, she had a self-limited period of confusion associated with hyperammonemia. During the following year, she maintained episodic confusion and developed difficulty initiating gait, turning, and walking through narrow spaces. In her first neurological observation, she presented slight mental slowness, hypomimia, a wide-based gait with small shuffling steps with evident start and turn hesitation, freezing, and impaired postural reflexes (Video 2, segment 1). The remaining neurological evaluation was normal. Brain MRI revealed bilateral and symmetric T1-weighted hyperintensities involving the GP and substantia nigra (SN) (Figure 1). T2-FLAIR sequences showed small subcortical and deep white matter hyperintensities in both hemispheres alongside slight increase in the striatum signal. Ammonia levels were elevated, fluctuating between 100 and 200 μmol/L. Metabolic screen showed mild stable hyperbilirrubinemia without coagulopathy. At last, she was admitted to the emergency unit confused and unable to walk. Hepatic ultrasound and liver biopsy showed regenerative nodular hyperplasia causing portal hypertension, without cirrhosis or rejection. She was initiated on rifaximin with impressive improvement in cognition and gait, which became practically normal in 1 month (Video 2, segment 2). Manganese levels at this point were within the normal range (10.2 ng/mL). Ammonia levels also showed significant reduction from pretreatment (249 μmol/L) to posttreatment levels (55 μmol/L, 1 month later). Neuropsychological evaluation at this point revealed a mild multidomain cognitive impairment; brain MRI showed reduction of the SN T1-weighted hyperintensity.
Despite initial clinical improvement under rifaximin, during the following months she oscillated between days with a normal gait and others in which forward gait was impossible due to freezing. Freezing partially improved with backward and side-to-side gait, but was not affected by visual or auditory cues (Video 2, segment 3). The patient was proposed for liver retransplant.
Herein, we report two patients with chronic hepatic disease presenting with unusual movement disorders, in which features of both HE and AHD are present (Table 1).
|Patient 1||Patient 2|
|Suggest hepatic encephalopathy||Fluctuating course||+||+|
|Transient consciousness and attention impairment||+||+|
|T2 hyperintensities of basal ganglia||+||+|
|Improvement with rifaximin||+||+|
|Suggest AHD||Persistent gait alteration||+ (ataxia)||+|
|Cerebellar dysarthria and ataxia||+||−|
|T1 hyperintensity of globus pallidus||+||+|
The presence of portosystemic shunt in cirrhotic or noncirrhotic portal hypertension allows neurotoxic substances such as ammonia, glutamine, and manganese to bypass the liver and gain access to the cerebral circulation. A proinflammatory state leads to brain blood barrier disruption and accumulation of these substances intraparenchymally.1 Intracellularly, ammonia is converted into glutamine that produces oxidative stress, metabolic disruption, and astrocyte structural changes. Ammonia also interferes in the communication between astrocytes and neurons, leading to widespread neurological manifestations.1
Hepatic encephalopathy presents with consciousness fluctuation ranging from somnolence to coma, neuropsychiatric symptoms including attention deficit, psychomotor slowing, and sleep disturbances. Movement disorders associated with HE are less prominent than mental status features and include asterixis, tremor, hypokinesia, and rigidity.2 HE usually has a fluctuating course and can be episodic, recurrent, or persistent.1 It is generally considered to be caused by hyperammonemia and, hence, potentially reversible. Treatment involves reduction of ammonia production through laxatives and nonabsorbable antibiotics such as rifaximin.2 On the contrary, AHD is a chronic disorder thought to be related primarily to manganese deposition in basal ganglia, where it causes presynaptic dopaminergic dysfunction and loss of postsynaptic dopaminergic receptors.3 The manganese theory is supported by MRI findings, including T1-high signal intensities in GP and adjacent areas. This preferential deposition is linked to a different clinical phenotype, mainly with cognitive deterioration and movement disorders, including tremor, parkinsonism, dystonia, chorea, and ataxia.2 These symptoms are usually irreversible, progressive, and resistant to pharmacological treatment.1 The main therapeutic option is hepatic transplantation.1
In our cases, the fluctuating course with attention and consciousness impairment, hyperammonemia, and the impressive improvement with rifaximin are suggestive of HE. On the contrary, the combination of dystonia, oromandibular dyskinesias, and ballistic movements (case 1) and a persistent gait disorder (case 2) with typical T1 hyperintensities in both patients support an AHD diagnosis. Patient 1 had previously been on olanzapine, which could contribute to the oromandibular dyskinesias; it cannot, however, justify the whole clinical presentation, giving its fluctuating pattern. This patient also presented a HCV infection, which has also been associated with movement disorders.4 Hyperkinetic disorders have been described in patients with HCV while under inferferon-alpha treatment, but these have disappeared with treatment interruption.5 Case 1 was not under any HCV treatment and therefore we believe the virus had no role in the clinical presentation.
Ballismus and isolated gait abnormalities constitute exceptional manifestations of chronic hepatic disease. Choreic and ballistic movements are usually attributed to decreased output from basal ganglia, caused by lesion or dysregulation of subthalamic afferents.6 The T1-hyperintense lesions involving the subthalamic nuclei in the brain MRI of case 1 seem to be the cause of his movement disorder. We found only two previous reports of two patients with bilateral ballism associated with hepatic disease, preceded by episodes of HE. They responded to treatment with haloperidol and lactulose.7,8 A previous paper reported on three patients with hepatic cirrhosis and akinetic-rigid parkinsonian syndromes that also improved after rifaximin. This was accompanied by a T1 signal reduction in GP.9
Regarding the phenotype of the second patient, there are several reports of gait abnormalities and postural instability in patients with hepatic disease.1,2 Usually, they are noticed as part of a parkinsonian syndrome and are more often related to AHD than HE. We have not found cases reporting freezing in the setting of hepatic disease.10 Our patient presented with freezing of gait without other parkinsonian signs apart from hypomimia, evoking a frontal gait. The frontal lobe has a crucial role in gait initiation, resulting in freezing and magnetism when damaged. Furthermore, disruptions in the networks connecting the cortical areas (frontal and parietal), basal ganglia, and the mesencephalic locomotor region involved in gait control are considered the main pathogenic mechanisms of freezing and parkinsonian motor features.1,11 This patient presented GP damage/T1 hyperintensities and although these are usually associated with AHD, the improvement with rifaximin suggests that ammonia (and, hence, HE) has an essential role. In fact, HE can also display GP T1 hyperintensities as well as white matter/periventricular T2 hyperintensities, as our patient did.1,2 The presence of abnormalities in both T1W1 and T2-FLAIR sequences also argues in favor of the coexistence of ammonia and manganese as pathogenic elements (HE and ADH) in the same patient. The link between freezing and hyperammonemia has also been described in epileptic patients under valproate treatment.12 It is conceivable that ammonia is toxic to basal ganglia and its gait networks, particularly in susceptible individuals as ours.
Lastly, it is interesting to notice that our patient’s freezing improved with backward or side-to-side gait. Freezing associated with parkinsonism is usually exacerbated in nonforward gait. Isolated backward freezing has even been described in patients with neurodegeneration with iron brain accumulation.13 A mechanism of external cueing (such as touching the wall) in a parkinsonian gait may be responsible for this phenomenon. Alternatively, we hypothesize that backward or side-to-side gait improvement could be a form of geste antagoniste in a dystonic gait, despite the fact that no leg dystonia is evident. A “crab-like” gait that improves in side-to-side walking has been described in dystonia.14
A clear distinction between HE and AHD is probably artificial, as both processes may happen simultaneously and share pathogenic pathways. Other cases with overlapping features between HE and AHD have been reported.8 In a small case series of AHD, previous episodes of HE could be identified in up to 60% of patients.10 Whether HE constitutes a risk factor for AHD remains to be clarified.
In conclusion, ballism and freezing of gait should be included in the neurological manifestations of portal hypertension, with or without cirrhosis. Treatments that reduce ammonia production, such as rifaximin, must be considered even in patients with atypical manifestations. However, these cases illustrate that improvement may be transient and that the disease will still progress, favoring the use of rifaximin as a bridge to hepatic transplantation in patients with persistent movement disorders.
1 Citation: Sousa AL, Salgado P, Alves JE, Silva S, Ferreira S, Magalhães M. Uncommon movement disorders in chronic hepatic disease with response to rifaximin. Tremor Other Hyperkinet Mov. 2019: 9. doi: 10.7916/tohm.v0.649.
2 Editor: Elan D. Louis, Yale University, USA
3 Funding: None.
4 Financial Disclosures: None.
5 Conflicts of Interest: The authors report no conflicts of interest.
6 Ethics Statement: All patients that appear on video have provided written informed consent; authorization for the videotaping and for publication of the videotape was provided.
The authors would like to thank the patients and families for their kind collaboration.
Feltracco, P, , Cagnin, A, , Carollo, C , Barbieri, S, , Ori, C, Neurological disorders in liver transplant candidates: pathophysiology and clinical assessment. Transplant Rev 2017;31(3):193–206. doi: https://doi.org/10.1016/j.trre.2017.02.006
Ferro, JM and , Oliveira, S, Neurologic manifestations of gastrointestinal and liver diseases. Curr Neurol Neurosci Rep 2014;14(10):487. doi: https://doi.org/10.1007/s11910-014-0487-z25171900
Mileti, V, , Ozreti, D , Relja, M, Parkinsonian syndrome and ataxia as a presenting finding of acquired hepatocerebral degeneration. Metab Brain Dis 2014;29(1):207–209. doi: https://doi.org/10.1007/s11011-013-9478-z24390157
Wijarnpreecha, K, , Chesdachai, S and , Jaruvongvanich, V, , Ungprasert, P, Hepatitis C virus infection and risk of Parkinson’s disease: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol 2018;30(1):9–13. doi: https://doi.org/10.1097/MEG.000000000000099129049127
Brito, MO and , Doyle, T, Movement and extrapyramidal disorders associated with interferon use in HIV/hepatitis C coinfection. AIDS 2007;21(14):1987–1989. doi: https://doi.org/10.1097/QAD.0b013e32829fb36917721114
Albin, RL The pathophysiology of chorea/ballism and Parkinsonism. Park Disord 1995;1(1353–8020):3–11. doi: https://doi.org/10.1016/1353-8020(95)00011-T
Zaman, Q, , Ahmad, A and , Khokar, N, , Khan, MF, Bilateral ballismus as a presenting feature of acquired hepatocerebral degeneration. Parkinsonism Relat Disord 2016;25:104–105. doi: https://doi.org/10.1016/j.parkreldis.2016.02.01426923522
Romeiro, FG , Américo, MF , Yamashiro, FS , Caramori, CA , Schelp, AO , Santos, AC , et al. Acquired hepatocerebral degeneration and hepatic encephalopathy: correlations and variety of clinical presentations in overt and subclinical liver disease. Arq Neuropsiquiatr 2011;69(3):496–501. doi: https://doi.org/10.1590/S0004-282X201100040001721755129
Kok, B, , Foxton, MR and , Clough, C, , Shawcross, DL, Rifaximin is an efficacious treatment for the parkinsonian phenotype of hepatic encephalopathy. Hepatology 2013;58(4):1516–1517. doi: https://doi.org/10.1002/hep.26364
Fernández-Rodriguez, R, , Contreras, A and , De Villoria, JG, , Grandas, F, Acquired hepatocerebral degeneration: clinical characteristics and MRI findings. Eur J Neurol 2010;17(12):1463–1470. doi: https://doi.org/10.1111/j.1468-1331.2010.03076.x20491897
Takakusaki, K Neurophysiology of gait: from the spinal cord to the frontal lobe. Mov Disord 2013;28(11):1483–1491. doi: https://doi.org/10.1002/mds.2566924132836
Kipervasser, S, , Elger, CE, , Korczyn, AD and , Nass, RD, , Quesada, CM, , Neufeld, MY, Gait instability in valproate-treated patients: call to measure ammonia levels. Acta Neurol Scand 2017;136(5):401–406. doi: https://doi.org/10.1111/ane.1276528436001
Martinez-Fernandez, R, , Caballol, N and , Castrillo, L, , Krack, P, Freezing of Backward Gait. Mov Disord Clin Pract 2014;1(3):255–257. doi: https://doi.org/10.1002/mdc3.1204630713861