Ghosh, Partha S.*; Hupertz, Vera†; Ghosh, Debabrata*
Liver transplantation (LT) is the only curative treatment in patients with end-stage liver disease (1). Neurological complications (NCs) can affect up to one-third of liver-transplanted patients (2). Many studies have investigated NCs after LT in adults (3–13). Patients with NCs following LT have been associated with a higher mortality rate (5,14), require longer hospitalization (15,16), experience more infections (4), and have lower self-sufficiency and social reintegration than patients without NCs (17). Various factors play a role in the pathogenesis of NCs following LT including poorly functioning graft, electrotype and metabolic derangements, intracranial hemorrhage, cerebral infarction, infection, or immunosuppressant toxicity (18). Seizures are the most common NC after LT (6,19). Various other NCs listed after LT were encephalopathy, posterior reversible leukoencephalopathy syndrome (PRES), stroke, meningitis, central pontine myelinolysis, cerebellar syndromes, headache, neuropsychiatric manifestations, cognitive decline, sleep disturbances, tremors, and peripheral neuropathy (3–13). Routine preoperative neurological evaluation and close follow-up after transplantation are needed to detect signs of complications early and initiate prompt and appropriate treatment or intervention (3,6).
The range of indications for LT in pediatric patients differs from that in adults, and therefore a different spectrum of complications would be expected (20); however, there are only a few studies that have looked into the spectrum of NCs in pediatric LT recipients (3,18,20–22). In the present study, we examined the various NCs noted post-LT in pediatric patients from a single tertiary care center in the United States.
The study was approved by the Cleveland Clinic institutional review board. All of the pediatric LT recipients (age ≤21 years at the time of LT) who received transplants at the Department of Pediatric Gastroenterology and Hepatology, and Transplant Surgery, Cleveland Clinic, and maintained in the pediatric transplant database during a 30-year period (1980–2010) were included in the analysis. The charts of the patients satisfying the following inclusion criteria were analyzed in greater detail.
The inclusion criteria were age ≤21 years at the time of LT and neurological complications after LT. The charts were reviewed retrospectively and the following data were recorded: demographics, indication for LT, spectrum of NCs after LT, the time of onset of each complication, diagnostic work-up, treatment, and follow-up. For each patient in whom NCs developed, the following investigations were done: serum chemistry panel (including ammonia level), serum trough level of the immunosuppressive agent, blood cell counts, and coagulation profile. Electroencephalography (EEG), computerized tomography, or magnetic resonance imaging (MRI) of the brain was performed when clinically indicated. NCs were defined as NCs based on history, neurological examination, and/or investigations in the form of EEG and neuroimaging that developed after LT. We defined encephalopathy when patients presented with altered mental status (lethargy, stupor, coma) with nonfocal neurological examination and without obvious focal structural lesions on neuroimaging. In the postoperative period, a diagnosis of encephalopathy was made only if altered mental status persisted even after the patients were extubated and the effects of sedatives had worn off. Clinical examinations were performed by clinicians from the pediatric intensive care unit and the gastroenterology and transplantation departments, and NCs were confirmed by pediatric neurologists. Routine preoperative neurological examination and postoperative neuropsychological assessment were not performed routinely.
We divided NCs into 2 groups: early (within 3 months of LT) and late (>3 months post-LT). Descriptive statistics regarding age, sex, NCs, and mortality were calculated for each group.
We reviewed the records of 65 patients in our pediatric LT database. We found 20 patients (30.7%) with NCs; 16 of them were girls with a mean age of 11.8 ± 5.9 years (9 months–21 years). Indications for LT in patients who had NCs included a wide spectrum of disorders (Table 1). Two of them had fulminant hepatic failure secondary to autoimmune hepatitis, whereas the rest had end-stage chronic liver disease. Eight patients underwent LT between 1980 and 1990, 2 of which had NCs, whereas the rest had LT between 1990 and 2010, 18 of which had NCs.
Characteristics of Group A (Early NCs)
A total of 9 patients (9/65, 13.8%) had NCs within 3 months of LT (Table 2). Eight were girls with a mean age of 11.9 ± 6.3 years (9 months–20 years). The most common NC in this group was seizures, noted in 7 patients (10.7%). All had generalized tonic-clonic seizures. The etiologies of seizures were PRES (Fig. 1) secondary to toxic levels of tacrolimus (23.4, normal 5–20 ng/mL) in 1, intracerebral hemorrhage (ICH) (Fig. 2) caused by thrombocytopenia (platelet count of 20,000 cells/μL) in 1, electrolyte disturbances (hyponatremia [lowest 125 mmol/L] in 1, hypocalcemia [7.5 mg/dL] and hypomagnesemia in another [1 mg/dL]) in 2, uremia (blood urea nitrogen of 45 with normal creatinine) in 1, and indeterminate in 2. Two patients developed encephalopathy (3%). Neurological examination revealed left-sided hemiplegia, left homonymous hemianopia, and signs of raised intracranial pressure in the patient with ICH. She underwent emergent hemicraniectomy because of worsening of intracranial pressure. The rest of the patients had nonfocal neurological examination.
Neuroimaging (computerized tomography/MRI brain) was abnormal in 4 patients in this group, which included PRES (1), ICH (1), mild cerebral edema (1), and bilateral basal ganglia T1W hyperintensities on MRI (1). EEG was available in 7 patients; findings were diffuse generalized slowing suggestive of encephalopathy in 3, right parieto-occipital EEG seizures in the patient with ICH, and normal in 3. Among the patients with neuroimaging abnormalities, EEG showed encephalopathy in the patient with PRES and seizures in the patient with ICH.
The mean follow-up period was 2.1 ± 2.1 years (range 6 months–3 years). On follow-up, 3 patients died (2 within 3 months posttransplant and 1 after 2 years); none was directly attributable to NCs. Neurological examination remained normal on follow-up in all except 1 with ICH who had residual left hemiparesis and left hemianopia. Four patients were treated with an antiepileptic drug (AED) at the onset of seizures. AEDs used were levetiracetam and lamotrigine. AEDs were stopped in all of the patients within 6 months to 2 years except the patient who had ICH.
Characteristics of Group B (Late NCs)
We found 11 patients (11/65, 16.9%) in this group (Table 3). Eight were girls with a mean age of 10.7 ± 6.5 years (9 months–21 years). The following NCs were noted: seizures (4, 6.1%), headache (4, 6.1%), encephalopathy (3, 4.6%), and paresthesias (1, 1.5%). One patient had both seizures and encephalopathy. Neurological examination was abnormal in 2 patients: papilledema in 1 and right-sided pyramidal tract signs in another. Etiologies of seizures were nodular heterotopia in 1 (coincidental finding), hypoxic-ischemic encephalopathy in 1, and indeterminate in 2. Seizures were generalized in 2 patients and partial in other 2. Headache was noted in 4 patients: 1 had aseptic meningitis (acute onset severe headache for 5 days), 1 had migraine without aura (4-month duration of episodic severe headache), and 2 had chronic daily headache (>3-month duration of moderately severe persistent headache). None of them had headache before LT or had positive family history of migraine. Possible small-fiber neuropathy produced paresthesias in 1 patient.
Neuroimaging abnormalities were noted in 4 patients: coincidental bilateral subependymal heterotopias (1), hypoxic-ischemic encephalopathy (1) (Fig. 3), encephalomalacia secondary to old hemorrhage (1), and MRI showing fluid-attenuated inversion recovery hyperintensity of posterior periventricular white matter (1). EEG was available in 4 patients: 3 had diffuse encephalopathy, whereas the other had normal EEG. Among the patient with neuroimaging abnormalities, EEG showed encephalopathy in 1 patient with hypoxic-ischemic encephalopathy and another with encephalomalacia. The patient with aseptic meningitis had elevated opening pressure on lumbar puncture with lymphocytic pleocytois but negative cultures.
On follow-up, during a mean of 4.5 ± 4.5 years (6 months–17 years), all survived. Neurological examination was abnormal in 2 patients: persistent papilledema with secondary optic atrophy (1) and right-sided pyramidal tract signs (1). Two patients with seizures started taking AED (levetiracetum in both). One is off AED, whereas the other is still taking AED at 1-year follow-up. The patient with aseptic meningitis had worsening visual acuity caused by secondary optic atrophy from persistent papilledema and underwent optic nerve sheath fenestration operation. The patient with migraine had <4 attacks per month. She did not require prophylaxis and headache was controlled with frovatriptan as needed. One patient with chronic daily headache had significant headache relief with amitryptyline prophylaxis, whereas the headache got better in the other patient with nonpharmacological measures.
Characteristics of Both Groups Combined
Overall, combining the 2 groups, we noted that the most common NC after LT was seizures (16.9%) followed by encephalopathy (7.6%) and headache (6.1%). Death occurred in 3 patients in the early NC group compared with none in the late NC group.
To date, only a few studies have investigated NCs in pediatric patients after LT. Incidence of NCs following pediatric LT varies from 8% to 46% (3,18–22). In our series of 65 patients, 20 (30.7%) had NCs, which is comparable with the data in the published literature. Seizures were the most common NC in our series (16.9%), which is in keeping with the previously published series. Garg et al (22) documented seizures in 9 of 11 pediatric patients after LT. Seizures can be partial or generalized, most frequently of the tonic-clonic type (23). The majority of patients in our series had generalized tonic-clonic seizures. The prevalence of early seizures may reach 10% to 40% (7,24–28) and decrease later to approximately 10% (29); a trend observed in our series (7 in early NC group compared with 4 in delayed NC group).
The common causes of seizures include metabolic derangements (electrolyte, hepatic, renal derangements), immunosuppressive agents (cyclosporine, tacrolimus), hypoxic-ischemic brain injury, cerebral structural lesions (ischemic or more commonly hemorrhagic strokes), and infections (30). They can occur with or without MRI-detectable structural brain lesions and with or without EEG alterations (23). In our series within the early NC group, 1 had PRES secondary to toxic levels of tacrolimus, 1 had hemorrhagic stroke, 3 had acute metabolic derangements, and 2 had indeterminate causes. In the delayed NC group, seizures were attributable to nodular heterotopia in 1 (coincidental finding), hypoxic-ischemic encephalopathy in 1, and indeterminate in 2. Seizures after LT usually are self-limiting, but some patients may develop severe treatment-resistant epilepsy (31). In our series, 2 patients were taking long-term AED, and in others, AEDs were stopped within 6 months to 2 years.
PRES is a clinicoradiological entity characterized clinically by acute onset of neurological dysfunction in the form of seizures, headache, and visual disturbances. The characteristic radiological findings are bilateral gray and white matter edema predominantly in the posterior regions of the cerebral hemispheres (32). Immunosuppressive agents are often associated with PRES; the reported incidence ranges from 1% to 6% in transplant recipients (33). In our series, the one had PRES associated with toxic levels of tacrolimus and demonstrated characteristic appearance on brain MRI.
We defined encephalopathy when the patient developed altered mental status in the form of lethargy, stupor, or coma, and had nonfocal neurological examination and neuroimaging studies that ruled out focal structural lesions; however, hypoxic-ischemic encephalopathy can lead to diffuse cortical/watershed ischemic changes or basal ganglia/thalamic/cerebellar lesions. In our series, 5 patients (7.6%) had encephalopathy: metabolic derangements were attributed in 4 and hypoxic-ischemic injury in 1. Two of them presented within 3 months post-LT, whereas the other 3 developed later (2 of them had chronic rejection after prior LT and experienced worsening of the hepatocellular function). Neuroimaging did not reveal focal lesions in our patients, with the exception of the patient with hypoxic-ischemic encephalopathy, who had diffuse cortical/watershed ischemic changes. Encephalopathy has been reported in 15% to 33% of pediatric LT recipients (20,22) compared with 11% in adults (4,5).
Headache is a well-recognized complication following LT (23). In our series, we found that 4 patients (6.1%) had headache, which is comparable with that reported by Erol et al (6.2%) (20). Headache manifested only in the delayed group. These patients did not have headache before LT. The cause of headache was attributed to aseptic meningitis in 1, migraine in 1, and chronic daily headache in 2. On follow-up, the patient with papilledema developed worsening of visual acuity and secondary optic atrophy requiring optic nerve sheath fenestration to prevent further worsening of the vision. This complication has not been reported following pediatric LT and underscores the need for regular follow-up to prevent significant morbidity. The patient with migraine had infrequent attacks not requiring prophylaxis and was controlled with frovatriptan as needed. One patient with chronic daily headache had significant improvement with amitryptyline prophylaxis, whereas in the other patient, nonpharmacological measures helped in controlling her headaches.
Peripheral neuropathy may cause paresthesia, dysesthesia, weakness, and hyporeflexia. It may result from pressure or traction of nerves or plexuses more frequent in patients with a prolonged operative course (23). In our series, 1 patient complained of paresthesias after a long interval following LT, thus ruling out compression, traction neuropathy, or plexopathy. The patient had normal sensory and reflex examination; however, a small-fiber neuropathy could not be ruled out.
Several other NCs have been reported by Amodio et al (23), which included tremors, sleep disorders, psychiatric manifestations, cerebellar syndromes, central pontine myelinolysis, and vegetative state. Tremors frequently appear after LT and resemble tremors caused by sleep deprivation, stress, and immunosuppressive agents. We did not find any patient with tremors. The reason may be either younger age (yet to manifest tremors because of threshold effect) or tremors too mild to be appreciated clinically. Central pontine myelinolysis was reported in 10% to 30% of patients who died after LT at the beginning of the liver transplant era (34,35). With the advances in post-LT management, the incidence dropped to about 2% to 3.5% (4,36). We have not encountered any patient who presented with or had radiological signs of central pontine myelinolysis. The fact that the majority of our patients underwent LT after the 1990s may explain the absence of this dreaded complication caused by improved peri- and postoperative care. Cerebellar syndrome following LT is characterized by headache, nausea, dizziness, emesis, encephalopathy, nystagmus, and ataxia. Neuroimaging studies may show ventricular widening or cerebral edema. Cerebellar syndrome after LT has been associated with cyclosporine (37). We did not encounter any patient with this complication.
In our study, mortality rates were similar in the LT recipients with NC (3/20, 15%) compared with those without NC (6/45, 13.3%). Pujol et al (25) found that adult LT recipients with NC had a higher mortality rate than patients without NC (55% vs 17%, respectively). Menegaux et al (3) noted higher mortality in pediatric patients (50%) than in adults (14%), and concluded that LT recipients with NC had a higher mortality rate; however, other studies suggested that mortality after LT is not affected by NC (5,8,20), a finding also observed in our study.
Previous studies have found that the majority of NCs develop during the first month after LT (3,5,20,25). It is presumed that metabolic derangements caused by transplantation and the alteration of the blood-brain barrier may render LT recipients more vulnerable to NC during the early posttransplant period; however, various studies reported NC happening between 1 and 6 months following LT in both adults and children (7,20,22). Because early NCs are associated with increased mortality, we arbitrarily made a 3-month cutoff period for early versus delayed complications to better stratify the risk; however, in our study, there was an increased prevalence of delayed complications following LT (16.9% vs 13.8%). It may be that improved immediate postoperative management did play a key role in reducing the early NCs. The small number of patients precludes us from drawing a firm conclusion.
NCs are common in children with LT. It is of utmost importance that the transplantation team should collaborate with pediatric neurologists to ensure rapid and accurate diagnosis of NCs and to institute appropriate management. The spectrum of NCs after LT is varied, although seizures seem to predominate as the most common complication. Although NCs are common in the immediate posttransplant period with improved care and longer lifespan of LT recipients, NCs can develop much later in the course.
1. Campagna F, Biancardi A, Cillo U, et al. Neurocognitive-neurological complications of liver transplantation: a review. Metab Brain Dis 2010; 25:115–124.
2. Wang WL, Yang ZF, Lo CM, et al. Intracerebral hemorrhage after liver transplantation. Liver Transpl 2000; 6:345–348.
3. Menegaux F, Keeffe EB, Andrews BT, et al. Neurological complications of liver transplantation in adult versus pediatric patients. Transplantation 1994; 58:447–450.
4. Lewis MB, Howdle PD. Neurologic complications of liver transplantation in adults. Neurology 2003; 61:1174–1178.
5. Bronster DJ, Emre S, Boccagni P, et al. Central nervous system complications in liver transplant recipients-incidence, timing, and long-term follow-up. Clin Transplant 2000; 14:1–7.
6. Ghaus N, Bohlega S, Rezeig M. Neurological complications in liver transplantation. J Neurol 2001; 248:1042–1048.
7. Vogt DP, Lederman RJ, Carey WD, et al. Neurologic complications of liver transplantation. Transplantation 1988; 45:1057–1061.
8. Stein DP, Lederman RJ, Vogt DP, et al. Neurological complications following liver transplantation. Ann Neurol 1992; 31:644–649.
9. Moreno E, Gomez SR, Gonzalez I, et al. Neurologic complications in liver transplantation. Acta Neurol Scand 1993; 87:25–31.
10. Boin IF, Falcao AE, Luzo AC, et al. 1; Analysis of neurologic complications within the first 30 days after orthotopic liver transplantation. Transplant Proc 2001; 33:3695–3696.
11. Vecino MC, Cantisani G, Zanotelli ML, et al. Neurological complications in liver transplantation. Transplant Proc 1999; 31:3048–3049.
12. Shibolet O, Galun E, Bishara A, et al. Major neurological complications following liver transplantation. Transplant Proc 2959; 33:
13. Beltran J, Taura P, Garcia-valdecasas JC, et al. Neurological disorders and intraoperative factors in liver transplantation. Transplant Proc 1993; 25:1749–1750.
14. Guarino M, Stracciari A, Pazzaglia P, et al. Neurological complications of liver transplantation. J Neurol 1996; 243:137–142.
15. Dhar R, Young GB, Marotta P. Perioperative neurological complications after liver transplantation are best predicted by pretransplant hepatic encephalopathy. Neurocrit Care 2008; 8:253–258.
16. Kim BS, Lee SG, Hwang S, et al. Neurologic complications in adult living donor liver transplant recipients. Clin Transplant 2007; 2:544–547.
17. Kaplan PE, Clinchot DM, Arnett JA. Cognitive deficits after hepatic transplantation: relevance to the rehabilitation potential. Brain Inj 1996; 10:599–607.
18. Donmez FY, Guvenc Z, Emiroglu FK, et al. Evaluation of neurological complications in pediatric liver transplant recipients: MRI versus CT. J Child Neurol 2009; 24:656–663.
19. Choi EJ, Kang JK, Lee SA, et al. New-onset seizures after liver transplantation: clinical implications and prognosis in survivors. Eur Neurol 2004; 52:230–236.
20. Erol I, Alehan F, Ozcay F, et al. Neurological complications of liver transplantation in pediatric patients: a single center experience. Pediatr Transplant 2007; 11:152–159.
21. Gridelli B, Lucianetti A, Rodriguez G, et al. Neurologic complications following pediatric orthotopic liver transplantation. Transplant Proc 1994; 26:193–194.
22. Garg BP, Walsh LE, Pescovitz MD, et al. Neurologic complications of pediatric liver transplantation. Pediatr Neurol 1993; 9:444–448.
23. Amodio P, Biancardi A, Montagnese S, et al. Neurological complications after orthotopic liver transplantation. Dig Liver Dis 2007; 39:740–747.
24. Adams DH, Ponsford S, Gunson B, et al. Neurological complications following liver transplantation. Lancet 1987; 1:949–951.
25. Pujol A, Graus F, Rimola A, et al. Predictive factors of in-hospital CNS complications following liver transplantation. Neurology 1994; 44:1226–1230.
26. Ardizzone G, Arrigo A, Schellino MM, et al. Neurological complications of liver cirrhosis and orthotopic liver transplant. Transplant Proc 2006; 38:789–792.
27. Estol CJ, Lopez O, Brenner RP, et al. Seizures after liver transplantation: a clinicopathologic study. Neurology 1989; 39:1297–1301.
28. Saner F, Gu Y, Minouchehr S, et al. Neurological complications after cadaveric and living donor liver transplantation. J Neurol 2006; 253:612–617.
29. Wszolek ZK, Steg RE. Seizures after orthotopic liver transplantation. Seizure 1997; 6:31–39.
30. Bronster DJ, Emre S, Mor E, et al. Neurologic complications of orthotopic liver transplantation. Mt Sinai J Med 1994; 61:63–69.
31. Bechstein WO. Neurotoxicity of calcineurin inhibitors: impact and clinical management. Transpl Int 2000; 13:313–326.
32. Hinchey J, Chaves C, Appignani B, et al. A reversible posterior leukoencephalopathy syndrome. N Engl J Med 1996; 334:494–500.
33. Singh N, Bonham A, Fukui M. Immunosuppressive-associated leukoencephalopathy in organ transplant recipients. Transplantation 2000; 69:467–472.
34. Boon AP, Adams DH, Buckels JA, et al. Neuropathological findings in autopsies after liver transplantation. Transplant Proc 1991; 23:1471–1472.
35. Wszolek ZK, McComb RD, Pfeiffer RF, et al. Pontine and extrapontine myelinolysis following liver transplantation. Relationship to serum sodium. Transplantation 1989; 48:1006–1012.
36. Yu J, Zheng SS, Liang TB, et al. Possible causes of central pontine myelinolysis after liver transplantation. World J Gastroenterol 2004; 10:2540–2543.
37. Eidelman BH. Neurological complications of immunosuppressive agents. Gastroenterol Int Congress Proc 1999; 12:128–134.