Abstract
Objective
Hepatic encephalopathy (HE) is a marker of poor prognosis in adults with chronic liver disease (CLD). We prospectively studied the prevalence and precipitants of HE in children with CLD as there is a paucity of literature on the same.
Methods
Children (1–18 years) admitted with CLD were examined daily for the presence and grading of HE (West Haven/Whittington grading). Precipitants were classified as infection, dyselectrolytemia, gastrointestinal bleeding, constipation and dehydration. Changes in grades of HE and outcome were noted.
Results
One hundred and sixty children (age 120 [84–168] months) were enrolled. HE was present in 50 (31.2%) patients with a total of 61 episodes. Maximum grade of HE was grade I (n = 16), II (n = 23), III (n = 11) and IV (n = 11). Forty-two cases had single and 8 had recurrent (2–5) episodes. Median duration of HE episodes was 96 (72–192) hours. Precipitants were identified in 55/61 (90.2%) episodes with infection (45/61, 73.7%) and dyselectrolytemia (33/61, 54%) being the most common. Lower albumin and sodium, higher INR and presence of infection were significantly associated with presence of HE. Overall, HE resolved in 33 (54%) episodes, while it progressed and persisted in 28 (45.9%) episodes. Patients with HE had a poorer outcome (25/50 vs 13/110; P < 0.01) with both higher in-hospital (11/50 vs 9/110; P = 0.02) and 1-month post discharge (14/39 vs 4/101; P < 0.01) mortality than those without HE.
Conclusion
One-third of admitted CLD children have HE, with identifiable precipitants in 90% of cases. Children with HE have poorer liver functions, higher rate of infections and worse outcome than those without HE.
Keywords: encephalopathy, cirrhosis, pediatric, infection, liver disease, outcome
Since the first description of mood disorders in liver disease by Hippocrates our understanding of hepatic encephalopathy (HE) has come a long way.1 HE marks a watershed moment in the natural history of cirrhosis and is one of the defining features of decompensation. In adults with cirrhosis, overt HE can develop in 25–40% cases during their clinical course.2,3 Of all the complications of cirrhosis, HE is the most dreaded and affects the quality of life, requires repeated hospitalizations and increases the risk of mortality.4,5 Children are different from adults with respect to the etiology and prognosis of chronic liver disease (CLD). There is a paucity of literature on the prevalence of HE in children with CLD. The precipitating factors and reversibility of HE in children may be different compared to adults. So extrapolating adult data to children would be inaccurate and there is a need for exclusive pediatric studies on HE. This prospective cohort study was undertaken to determine the prevalence and precipitants of HE in hospitalized children with CLD.
Methods
Children (1–18 years) admitted with CLD from April 2017 to December 2018 were prospectively enrolled. CLD was diagnosed based on liver biopsy if available or on clinical, imaging and endoscopic findings (≥ grade II varix). Details about the etiology of CLD, laboratory parameters, endoscopic findings and treatment given were noted. Esophageal varices were graded by Paquet classification,6 and further classified as small (≤grade II) and large (≥grade III). Gastric varices were graded by Sarin (location) and Hashizume (size) classification.7,8
Child-Pugh score and Pediatric end-stage liver disease (PELD) score were calculated.9 PELD score was calculated in children >12 years of age also for the purpose of homogeneity.10 Extra-hepatic organ failure was defined as per the pediatric CLIF-SOFA score.11For comparison of patients, the worst value of laboratory test during that admission was taken and organ failure was assessed as that present at admission or developing at any time during hospital stay.
HE was diagnosed clinically and graded as per West Haven grading (>3 years old) and Whittington grading (≤3 years) by a single observer.12,13 Precipitants of HE which included infection, gastrointestinal bleeding, diuretic overdose, constipation and electrolyte disorders, were actively looked for and specific treatment was given as per standard practice.14 Standard criteria were used for defining the presence and site of infection.15
Venous ammonia (normal <54 umol/L) was measured in all patients with HE at the time of first appearance. Ammonia was measured using microdiffusion method based on colorimetric assay (PocketChem BA-PA-4140 Arkray Factory, Inc. Shiga, Japan).
Intravenous antibiotics were given for spontaneous bacterial peritonitis (SBP), pneumonia and other sites of infections as per standard protocol and modified based on the antibiotic sensitivity pattern. Gastrointestinal bleeding, renal failure and hypotension were managed as per standard guidelines.16 For treatment of HE, oral lactulose (0.5 ml/kg/dose up to 30 ml/dose) was given to achieve 2–4 soft stools per day.14 If there was no response beyond 48 h or worsening in grade of HE while on lactulose, rifaximin (20–30 mg/kg/day) was added. If there was no response beyond 48 h of treatment or further worsening of HE grade (on lactulose and rifaximin) then oral branched-chain amino acids (BCAA, 0.25 g/kg/day) and intravenous l-Ornithine-l-apartate (LOLA - 13 to 18 years–20 g, 6 to 12 years- 10 g, <6 years – 5 g) were added together. All patients received etiology specific therapy and a diet with adequate protein (1−1.5 g/kg) and calories, fat-soluble vitamins and micronutrient supplements. After recovery from HE, secondary prophylaxis with lactulose was given.
Ethics and Statistical Analysis
The study was approved by the institute's ethics committee (IEC-2017-35-DM-96). Written informed consent was taken from the parents/guardians of all participants.
Statistical analysis was performed using STATA 12.0 (Stata Corp LP, Texas, USA). Assuming a prevalence of HE as 30% in hospitalized adult CLD patients, a 25% relative error of the given prevalence and level of confidence of 95%, a sample size of 144 cases was calculated. Results are presented as percentage and median with interquartile range (IQR). Continuous variables were compared using the Mann-Whitney U test and categorical variables using the Chi-square test or Fischer exact test as applicable. Predictors with P < 0.05 in univariate analysis were entered into a multivariate analysis to determine independent predictors for HE.
Results
Baseline Characteristics
One hundred and sixty children (93 boys, age 120[IQR: 84–168] months) with CLD were prospectively enrolled. The demographic and laboratory profile of the entire group of 160 cases is shown in Table 1. The etiology of CLD was as follows: Autoimmune hepatitis (AIH) −48 (30%), Wilson disease–39 (24.4%), Budd Chiari syndrome (BCS)–23 (14.4%), Progressive familial intrahepatic cholestasis–10 (6.2%), sclerosing cholangitis–7 (4.4%), biliary atresia–7 (4.4%), Caroli syndrome–6 (3.7%), chronic hepatitis B–5 (3.1%), choledochal cyst–3 (1.9%), autoimmune sclerosing cholangitis– 3 (1.9%) and cryptogenic- 9 (5.6%). Child-Turcott-Pugh (CTP) score at the time of enrollment was 9 (7–12) with 34 (21.5%) in Child A, 57 (36.1%) in Child B and 67 (42.4%) in Child C [not calculated in 2 BCS cases who were on warfarin]. PELD score at study enrollment was 11 (3–23).
Table 1.
Comparison of Clinical and Laboratory Profile of CLD Children With and Without Hepatic Encephalopathy.
| Parameter | Patients(n = 160) | With HE(n = 50) | No HE(n = 110) | P valuea(univariate) | P valuec(multivariate) |
|---|---|---|---|---|---|
| Age (months) | 120 (84–168) | 120 (96–172) | 120 (72–156) | 0.18 | – |
| Gender (M:F) | 93:67 | 27:23 | 66:44 | 0.47 | – |
| Etiology of CLD | – | ||||
| AIH | 48 | 19 | 29 | 0.13 | |
| Wilson | 39 | 16 | 23 | 0.12 | |
| BCS | 23 | 5 | 18 | 0.28 | |
| Cholestatic liver diseases | 30 | 5 | 25 | 0.06 | |
| Others | 20 | 5 | 15 | 0.51 | |
| Total bilirubin (mg/dL) | 1.8 (1.2–11.4) | 11.2 (5.2–29) | 1.7 (1–6.7) | <0.001 | 0.33 |
| Albumin (g/L) | 2.8 (2.4–3.5) | 2.4 (2–2.6) | 3.1 (2.7–3.9) | <0.001 | 0.005 |
| PT-INR | 1.7 (1.2–2.6) | 3.05 (2.2–5.6) | 1.37 (1.14–1.84) | <0.001 | 0.02 |
| Serum sodium (mEq/L) | 137.5 (134–140) | 134 (129–136) | 138 (135–140) | <0.001 | 0.009 |
| Infection (n, %) | 64 (40) | 36 (72) | 28 (25) | <0.001 | 0.03 |
| Ascites (n, %) | 91 (57) | 43 (86) | 48 (43.6) | <0.001 | 0.57 |
| Eso Vx present,n (%) (large: small) (n = 151) | 119 (78.8) 45:74 |
38 (88.3) 13:25 | 81 (75) 32:49 | 0.08 0.57 |
– |
| GI bleed (n, %) | 22 (13.7) | 8 (16) | 14 (12.7) | 0.57 | – |
| Circulatory failure (n, %) | 18 (11.2) | 16 (32) | 2 (1.8) | <0.001 | 0.85 |
| Respiratory failure (n, %) | 24 (15) | 21 (42) | 3 (2.7) | <0.001 | 0.06 |
| Renal failure (n, %) | 11 (6.8) | 9 (18) | 2 (1.8) | 0.001 | 0.98 |
| Specific treatment (n, %) | 49 (30.6) | 18 (36) | 31 (28) | 0.32 | – |
| Deaths (n, %) | 38 (23.7) | 25 (50) | 13 (11.8) | <0.001 | – |
| CTP A/B/C (n)b | 34/57/67 | 0/5/44 | 34/52/23 | <0.001 | |
| PELD (at admission) | 11 (3–23) | 27 (19–38) | 6 (−1 to 13) | <0.001 |
CLD, Chronic Liver disease; AIH, Autoimmune hepatitis; BCS, Budd Chiari syndrome; HE, hepatic encephalopathy; Eso Vx, Esophageal varix; PT-INR, Prothrombin time-International normalized ratio; GI, gastrointestinal; PELD, pediatric end-stage liver disease; CTP, Child Turcott Pugh.
P value: comparison between those with and without HE.
CTP not calculated for 2 children on warfarin. Note: CTP also contains HE as one of the components.
Multivariate analysis was done after excluding PELD. All laboratory parameters are the worst values during the hospital stay and organ failure is any time from admission to discharge or death in hospital.
The predominant reasons for admission to hospital (n-171) in these 160 children were HE (n = 40), GI bleeding (n = 22), ascites (n = 63), fever (n = 26) and work-up for etiology (n = 20). However, the majority of cases [85.3% (146/171)] had ≥1 reason for admission like HE + fever (n = 15), GI bleeding + HE (n = 13), ascites + HE (n = 25), fever + ascites (n = 10).
Of the 160 children, 50 (31.2%) cases had HE and they had a total of 61 episodes of HE during the study period. HE was present at admission in 47/61 (77%) episodes and developed during hospitalization in 14/61 (22.9%) episodes. Forty-two patients had a single episode, 7 experienced 2 episodes and 1 child had 5 episodes of HE. Of the 160 enrolled children, 18 (11.2%) were <3 years (Median PELD: 8, CTP: 7) and 142 (88.7%) were ≥3 years (median PELD: 11, CTP: 9). Only 2/18 children who were <3 years of age had HE (grade II) in comparison to 48/142 children who were >3 years old (P = 0.05). The CTP and PELD scores were not significantly different between the younger and older children. The etiology of CLD was not different in the patients with and without HE (Table 1). The PELD score of the cases with HE was significantly higher than those without HE (Table 1).
Precipitants were identified in 55/61 episodes (90.2%), one precipitant in 21/61, two precipitants in 17/61, three precipitants in 14/61 and four precipitants in 3/61 episodes of HE. Thus, multiple precipitants were seen in 55% (n = 34/61) episodes of HE. Infections were the most common precipitant (45/61, 73.7%) followed by electrolyte disorder (33/61, 54%), diuretic overdose (14/61, 22.9%), gastrointestinal (GI) bleeding (13/61, 21.3%) and constipation (5/61, 8.2%) as shown in Figure 1.
Figure 1.
Presence of various precipitants and the proportion of cases with respective precipitants who developed hepatic encephalopathy. SBP: Spontaneous bacterial peritonitis, UTI: Urinary tract infection, GI: Gastrointestinal.
There were 68 infectious episodes in total in 160 children. Of these, single-site infection was seen in 28/68 (41.2%) and multiple-site infection in 40/68 (58.8%) episodes. SBP was the most common site (n = 38) followed by pneumonia (n = 27), blood stream infection (n = 18) and UTI (n = 14). Cellulitis and cholangitis were documented in 5 episodes each. Of the total 68 infectious episodes among 160 children, 45 (66%) precipitated HE (Figure 1). Multiple-site infection was seen in 32 episodes of HE. SBP was the most common infectious precipitant (n = 25) followed by pneumonia (n = 18), bloodstream infection (n = 13) and urinary tract infection (UTI, n = 10). The other infections encountered among children with HE were cellulitis (n = 5), pyoderma (n = 1), pyomyositis (n = 1), cholangitis (n = 1), swine flu (n = 1), otitis media (n = 1) and endophthalmitis (n = 1). Table 2 shows the presence and site of infection in patients with and without HE.
Table 2.
Comparison of Presence and Site of Infections in Children With and Without HE.
| Infection | HE episodes (n = 61) n (%) | No HE (n = 110) n (%) | P value |
|---|---|---|---|
| Present (n-68) | 45/61 (73) | 23/110 (20.9) | <0.01 |
| Multiple sites (n-40) | 32/61 (52.4) | 8/110 (7.2) | <0.01 |
| SBP (n-38) | 25/61 (40.9) | 13/110 (11.8) | <0.01 |
| Pneumonia (n-27) | 18/61 (29.5) | 9/110 (8.2) | <0.01 |
| UTI (n-14) | 10/61 (16.4) | 4/110 (3.6) | <0.01 |
| Blood stream infection (n-18) | 13/61 (21.3) | 5/110 (4.5) | <0.01 |
| Cellulitis (n-5) | 5/61 (8.2) | 0/110 (0) | <0.01 |
| Cholangitis (n-5) | 1/61 (1.6) | 4/110 (3.6) | 0.45 |
HE, Hepatic encephalopathy; SBP, Spontaneous bacterial peritonitis; UTI, Urinary tract infection
Hyponatremia (Na < 130 mEq/L) was the precipitant of HE in 23/61 episodes (27.7%), hypernatremia (Na > 150 mEq/L) in 1/61 (1.6%), hypokalemia (K < 3.5 mEq/L) in 13/61 (21.3%) and hyperkalemia (K > 5.5 mEq/L) in 6/61 (9.8%). Hyperkalemia was associated with renal failure in 4/6 and diuretic usage in 6/6 cases. Four HE episodes were associated with acute gastroenteritis with dehydration and they had hypernatremia in 1 and hypokalemia in 3 episodes.
In addition, hypoglycemia (<70 mg/dL) was associated with HE in 18/61 (29.5%) and hyperglycemia (>200 mg/dL) in 2/61 (3.2%) episodes. Drugs affecting mentation were received prior to the development of HE in 5 cases [ketamine −3, midazolam-4 (administered for procedural sedation) and alprazolam-1]. All these cases also had other precipitants.
Dominant portosystemic shunt was diagnosed by ultrasound Doppler in 3/50 patients (prominent omental collateral- 1 and splenorenal- 2). Ultrasound Doppler was performed as a routine in all patients with HE. One of these 3 cases had 5 episodes of HE and 2 cases had a single HE episode. All three patients had other precipitants of HE. The PELD scores of children with single [PELD: 27.5 (19.5–42)] and multiple episodes of HE [PELD: 26 (17.5–31.5)] were not significantly different (P: 0.49).
Children with HE were compared with those without HE (Table 1). On univariate analysis, patients with HE had higher PELD score, higher total bilirubin, higher PT-INR, lower albumin, lower sodium, higher rates of infection, ascites, organ failure (circulatory, respiratory and renal) and death. Multivariate analysis (after excluding PELD score) showed that lower albumin, lower sodium, higher PT-INR and presence of infection were significantly associated with HE.
The maximum severity of HE was Grade IV (n = 11), grade III (n = 11), grade II (n = 23) and grade I (n = 16). The median duration of HE in the 61 episodes was 96 h (72–192). The episodes with a higher grade of HE (grade III-IV), more often had multiple precipitants, electrolyte disorders, diuretic overdose, higher ammonia, higher bilirubin and higher PELD scores in comparison to episodes with lower HE grades (I-II) as shown in Table 3. On multivariate analysis, serum bilirubin was significantly different between lower and higher grades of HE.
Table 3.
Comparison of Precipitants in 61 Episodes of HE With Mild (Grade I– II; n-39) and Severe HE (Grade III- IV, n-22).
| Parameter | Grade I to II (N = 39) | Grade III to IV (N = 22) | P value (univariate) | P value (multi-variate) |
|---|---|---|---|---|
| Agea | 120 (78–164) | 132 (96–192) | 0.21 | |
| Age ≤3 y | 2/28 | 0/22 | 0.57 | |
| Gender (M:F)∗ | 16:12 | 11:11 | 0.82 | |
| Precipitant present, n (%) | 34 (87) | 21 (95.4) | 0.28 | |
| Infection, n (%) | 27 (69.2) | 18 (81.8) | 0.28 | |
| Constipation, n (%) | 4 (10.3) | 1 (4.5) | 0.40 | |
| GI bleed, n (%) | 6 (15.4) | 7 (31.8) | 0.13 | |
| Electrolyte disorder, n (%) | 17 (43.6) | 16 (72.7) | 0.03 | 0.71 |
| Diuretic overdose, n (%) | 6 (15.4) | 8 (36.4) | 0.05 | 0.38 |
| Single vs. multiple precipitants (n) | 17:17′ | 4:17 | 0.02 | 0.47 |
| Total Bilirubin (mg/dL) | 5.7 | 20.8 | 0.006 | 0.01 |
| PT-INR | 3.0 | 5.7 | 0.23 | |
| Albumin (g/L) | 2.6 | 2.2 | 0.32 | |
| Sodium (mEq/L) | 136 | 129.5 | 0.57 | |
| Spontaneous shunta | 2/28 | 1/22 | 0.82 | |
| PELDb | 23 (17–31) | 39 (30–53) | <0.01 | |
| Blood Ammonia (umol/L) | 99 (80–126) | 180.5 (129–255) | <0.01 | - |
HE: hepatic encephalopathy; M:F – Male:Female, GI: Gastrointestinal; PELD: Pediatric end-stage liver disease, PT-INR: Prothrombin time International normalized ratio.
These parameters are expressed for the total cases (n-50).
PELD excluded from multivariate analysis.
HE resolved in 25 (40.9%) episodes, progressed to a higher grade and then resolved in 8 (13.1%) and progressed to higher grades without resolution in 28 (45.9%) episodes. Overall, HE resolved in 33 (54%) episodes (Figure 2). Among the 50 cases with HE, 25 (50%) died; 11 (22%) in-hospital, 14 (28%) within 1 month of discharge and 3 (6%) had persistent HE over a follow-up period of 6 months. In comparison, 13/110 (11.8%) children without HE died; 9 (8.2%) patients died in-hospital and 4 (3.6%) within 1-month post-discharge. Out of 39 episodes of grade 1–2 HE there were 8 deaths (4 in hospital) compared to 17 deaths (16 in hospital) in 22 episodes of grade 3–4 HE (P < 0.01) (Figure 3).
Figure 2.
Evolution of hepatic encephalopathy from admission to end of hospital stay. HE: Hepatic encephalopathy.
Figure 3.
Outcome of children with no HE, grade I-II HE and grade III-IV HE. HE: Hepatic encephalopathy.
Discussion
Prevalence and Predictors of HE
We found the prevalence of HE in children admitted with CLD to be 31.2%. In two studies of 63 and 301 adults with cirrhosis followed up in the outpatient department, overt HE developed in 30.1% and 41% cases respectively.2,17 In our study, we included patients with grade I HE also whereas the adult studies looked at overt HE (>grade I). Prevalence of overt HE in our study was 25% which closely resembles the adult experience. In decompensated cirrhosis in adults defined by history of ascites, HE and GI bleeding, HE prevalence was 45%.18 Among children with decompensated CLD, HE was the second common mode of decompensation after ascites and was seen in 43%.11
There is a lack of pediatric studies which have specifically evaluated the prevalence and profile of HE in CLD. The patients who developed HE in our study had more severe liver disease as shown by lower albumin, lower sodium and higher PT-INR. The predictive factors of overt HE in adults were subclinical HE in the past, lower glutamine levels and higher Child score in one study17 and hyponatremia, higher bilirubin and creatinine and diuretic treatment for ascites in another.18 This supports our observation that development of HE is a manifestation of advanced CLD.
Precipitants of Hepatic Encephalopathy
Identifying a precipitating factor is of vital importance as correction of the factor offers the probability of reversibility of HE. Ninety percent of HE episodes in our study had identifiable precipitants and more than half had multiple precipitants. We found infection to be the most common precipitant (73.7%), followed by electrolyte abnormality, diuretic overdose, GI bleeding and constipation. Similarly, precipitants were identified in 68% and 90% in children and adults respectively.19,20 In a pediatric study of 22 HE cases, the most common precipitant was infection (27%) followed by gastrointestinal bleeding (23%), constipation (14%) and protein overload (4%), which is similar to our observation.19
In an adult study comprising 194 adults with cirrhosis, dehydration (53%) and acute renal failure (42%) were more common overall followed by constipation (39%) and infections (25%). In this study, 74% episodes had >1 precipitant compared to 56% episodes with multiple precipitants in our study. Hence, identification of one precipitant should not preclude the investigations for other precipitants as all of them need to be addressed for a good outcome. Benzodiazepines and opioids precipitated HE in 22% episodes in adults20 in comparison to 8% in our study. The medications were used for pre-procedure sedation in most of our children; however, the reason for prescribing these medications is not available in the adult study. This highlights the need for extra caution and monitoring while using these medications in children with advanced CLD.
Infections are common in CLD patients including those with acute or chronic liver failure and these patients have a worse prognosis.21 The overall prevalence of infections in decompensated CLD in children was 57% and those with infections had more advanced liver disease and poorer outcome.15 SBP and pneumonia were the two most common sites of infection in our study and >1 site of infection was seen in 52% of HE episodes. This highlights the importance of actively investigating CLD patients for infection and treating them aggressively to improve outcome. This assumes further importance as many of the infections especially SBP may be asymptomatic.22 Also, some of the CLD cases like autoimmune liver disease may be on immunosuppressants which further increases their risk of infections. It is well known that cirrhosis-associated immune dysfunction progresses with the severity of liver disease predisposing these patients to infections.23 Thus, poorer liver function is the background for development of infections and subsequently triggering HE.
Hyponatremia, hypokalemia and hypoglycemia were the common metabolic derangements encountered in our study. Diuretic use and advanced liver disease contributed to dyselectrolytemia and hypoglycemia. The relationship between hyponatremia and HE was elegantly demonstrated in a study where 75% patients with cirrhosis developed HE with a drop in serum sodium by 5 mEq/L on follow-up, and magnetic resonance spectrometry showed osmolyte imbalance (lower myoinositol, creatine and choline) in HE patients.18 Hypokalemia causes intracellular acidosis due to increased H+ ion uptake in exchange for K+ ion in proximal tubular cells of the kidneys. To maintain the physiological acid-base milieu inside these cells there is increased uptake and metabolism of glutamine to ammonia.24
Spontaneous portosystemic shunts were seen in 3 (6%) children with HE, one of whom had recurrent episodes. In a large study of 982 cirrhotics, 34% had spontaneous portosystemic shunts, umbilical vein shunt was the most common followed by splenorenal shunt. Subjects with shunts had advanced liver disease with higher Child-Pugh score, incidence of HE and esophageal varices.25 We have not actively looked for a shunt in all cases and a possibility of a spontaneous shunt being missed on Doppler ultrasound cannot be negated. In those with recurrent episodes of HE with relatively preserved liver functions it is prudent to look for portosystemic shunts. The European Association for the Study of the Liver guidelines suggest obliteration of accessible shunts in adults if the MELD (Model for End-Stage Liver Disease) is <11.26
Diagnosing HE in younger children has always been a challenge. In our study, HE was detected in 11% of children <3 years of age and in 34% of those >3 years old, although they had comparable PELD scores. The possibility of missing lower grades of HE in younger children cannot be excluded.
In another pediatric study that included children with CLD and high ammonia (>150 mg/dL in infants and 100 mg/dL in older Children) only 53.4% episodes of high ammonia had overt HE. This cohort had younger children with a mean age of 5.6 years, with nearly half (47.7%) being <3 years of age.27 In the Whittington scale for HE in young children, grades I-II are clubbed together as detecting grade I HE is difficult. This raises the issue that lower grades of HE may be missed in small children as the criteria used to diagnose HE are subjective. Since a reliable biomarker for HE is lacking, a thorough history with inputs from parents about the recent changes in behavioral states of the child will add value to diagnosis of early HE.
Low-grade HE (grade I-II) was seen in 64% and high grade (grade III-IV) in 36% episodes in our study. In a pediatric study (mean age 13.5 ± 2.5 years) with 22 cases, 45% had grade II HE, 50% had grade III and 4.5% had grade IV but none were in grade I HE.19 In another pediatric study the prevalence of HE was 43% among decompensated CLD and among them, 61% had high-grade HE.11
Among adults 70–84% were in low-grade HE and 16–47% had high-grade HE.20,28 The wide variation can be explained by the difference in the severity of the liver disease. As the cohort with lower MELD score (16 ± 6.8) had lower prevalence of high-grade HE whereas the group with higher MELD score (22.7 + 7.2) had higher prevalence of high-grade HE. The higher prevalence of advanced HE in our study could be due to the fact that we have taken only hospitalized patients and milder grades of HE may be missed in young children. Those with high-grade HE had multiple precipitants, higher incidence of dyselectrolytemia, diuretic overdose, higher ammonia and higher PELD score in our study. This suggests that the grade of HE is determined both by the severity of liver disease and number and nature of precipitants.20
Mortality in children with HE was 50% compared to 11% in those without HE. This is likely to be due to the overall worse liver functions in children with HE and similar to native liver survival of 52% at day 90 in children with CLD and hyperammonemia.27 The reported 30-day mortality in adults with HE grade III-IV is 38% and grade I-II HE is 8%.28 Although those with grade III-IV HE had higher MELD score compared to those with grade I-II, on regression analysis, 30-day mortality was associated with severity of HE and it was independent of MELD score and extra-hepatic organ failure.28 After an episode of overt HE one year survival was as low as 36%.29 These observations highlight the poor prognosis of overt HE both in the short and long term and across ages.
The limitations of our study include the wide age range of children studied and that ammonia levels were not measured in all cases. There is a possibility that we may have missed some cases with Grade I HE, especially in the younger children. This could have affected our results of response to therapy and outcome also. We actively looked for spontaneous portosystemic shunts only in children with HE and not in those without HE and we used ultrasound Doppler rather than computed tomography (better diagnostic accuracy) and thus could have missed some cases with spontaneous portosystemic shunts. Also, we have not been able to offer liver transplant to our cases due to various logistic constraints. However, this is the first prospective study with a good number of cases to study HE in children with CLD.
In conclusion, One-third of all admitted children with CLD have HE and precipitants are seen in 90% episodes of HE, with multiple precipitants in 55% episodes. Infection is the most common precipitant followed by electrolyte abnormalities and gastrointestinal bleeding. Patients with CLD and HE are characterized by poorer liver functions and higher infections than those without HE. The in-hospital outcome of CLD children with HE is poor with 22% mortality.
Credit authorship contribution statement
Aathira Ravindranath – Conceptualised the study, collected and analyzed data and co-drafted the manuscript.
Anshu Srivastava – Conceptualised the study, analysis and interpretation of data, co-drafted the manuscript.
Surender Kumar Yachha –contribution to study design and critical revision for important intellectual content.
Ujjal Poddar – interpretation of data and critical revision for important intellectual content.
Moinak Sen Sarma – analysis of data and critical revision for important intellectual content.
Amrita Mathias – collection of data, co-drafting of manuscript.
All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
Conflicts of interest
The authors have none to declare.
Funding
None.
Article guarantor
Anshu Srivastava.
Footnotes
It has been presented as a poster in the 70th Annual meeting of the American Association for the Study of Liver Diseases: The Liver Meeting in November 2019 at Boston and published as an abstract in Hepatology. 2019; 70 S1: 188–1382. https://doi.org/10.1002/hep.30941.
References
- 1.Amodio P. Hepatic encephalopathy: historical remarks. J Clin Exp Hepatol. 2014;5:S4–S6. doi: 10.1016/j.jceh.2014.12.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Amodio P., Del Piccolo F., Pettenò E., Mapelli D., Angeli P., Iemmolo R. Prevalence and prognostic value of quantified electroencephalogram (EEG) alterations in cirrhotic patients. J Hepatol. 2001;35:37–45. doi: 10.1016/s0168-8278(01)00129-5. [DOI] [PubMed] [Google Scholar]
- 3.Tapper E.B., Parikh N.D., Sengupta N., et al. A risk score to predict the development of hepatic encephalopathy in a population-based cohort of patients with cirrhosis. Hepatology. 2018;68:1498–1507. doi: 10.1002/hep.29628. [DOI] [PubMed] [Google Scholar]
- 4.Tapper E.B., Halbert B., Mellinger J. Rates of and reasons for hospital readmissions in patients with cirrhosis: a multistate population-based cohort study. Clin Gastroenterol Hepatol. 2016;14:1181–1188. doi: 10.1016/j.cgh.2016.04.009. [DOI] [PubMed] [Google Scholar]
- 5.Jepsen P., Ott P., Andersen P.K., Sørensen H.T., Vilstrup H. Clinical course of alcoholic liver cirrhosis: a Danish population-based cohort study. Hepatology. 2010;51:1675–1682. doi: 10.1002/hep.23500. [DOI] [PubMed] [Google Scholar]
- 6.Paquet K.J. Prophylactic endoscopic sclerosing treatment of the esophageal wall in varices—a prospective controlled randomized trial. Endoscopy. 1982;14:4–5. doi: 10.1055/s-2007-1021560. [DOI] [PubMed] [Google Scholar]
- 7.Sarin S.K., Lahoti D., Saxena S.P., Murthy N.S., Makwana U.K. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients. Hepatology. 1992;16:1343–1349. doi: 10.1002/hep.1840160607. [DOI] [PubMed] [Google Scholar]
- 8.Hashizume M., Kitano S., Yamaga H., Koyanagi N., Sugimachi K. Endoscopic classification of gastric varices. Gastrointest Endosc. 1990;36:276–280. doi: 10.1016/s0016-5107(90)71023-1. [DOI] [PubMed] [Google Scholar]
- 9.McDiarmid S.V., Anand R., Lindblad A.S. Principal Investigators and Institutions of the Studies of Pediatric Liver Transplantation (SPLIT)Research Group. Development of a pediatric end-stage liver disease score to predict poor outcome in children awaiting liver transplantation. Transplantation. 2002;74:173–181. doi: 10.1097/00007890-200207270-00006. [DOI] [PubMed] [Google Scholar]
- 10.Dehghani S.M., Gholami S., Bahador A., et al. Comparison of Child-Turcotte-Pugh and pediatric end-stage liver disease scoring systems to predict morbidity and mortality of children awaiting liver transplantation. Transplant Proc. 2007;39:3175–3177. doi: 10.1016/j.transproceed.2007.07.080. [DOI] [PubMed] [Google Scholar]
- 11.Bolia R., Srivastava A., Yachha S.K., Poddar U. Pediatric CLIF-SOFA score is the best predictor of 28-day mortality in children with decompensated chronic liver disease. J Hepatol. 2018;68:449–455. doi: 10.1016/j.jhep.2017.10.001. [DOI] [PubMed] [Google Scholar]
- 12.Conn H.O. In: Diseases of the Liver. 7th ed. Schiff L., Schiff E.R., editors. Lippicott; Philadelphia: 1993. Hepaticencephalopathy; pp. 1036–1060. [Google Scholar]
- 13.Whitington P.F., Alonso A.E. In: Paediatric Liver Disease. DA D., editor. Blackwell; Oxford: 2003. Fulminant hepatitis and acute liver failure; pp. 107–126. [Google Scholar]
- 14.Vilstrup H., Amodio P., Bajaj J., et al. Hepatic encephalopathy in chronic liver disease: 2014 practice guideline by the American association for the study of liver diseases and the European association for the study of the liver. Hepatology. 2014;60:715–735. doi: 10.1002/hep.27210. [DOI] [PubMed] [Google Scholar]
- 15.Bolia R., Srivastava A., Marak R., Yachha S.K., Poddar U. Prevalence and impact of bacterial infections in children with liver disease-A prospective study. J Clin Exp Hepatol. 2018;8:35–41. doi: 10.1016/j.jceh.2017.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Angeli P., Bernardi M., Villanueva C., et al. EASL Clinical Practice Guidelines for the management of patients with decompensated cirrhosis. J Hepatol. 2018;69:406–460. doi: 10.1016/j.jhep.2018.03.024. [DOI] [PubMed] [Google Scholar]
- 17.Romero-Gomez M., Boza F., GarciaValdecasas M.S., García E., Aguilar-Reina J. Subclinical hepatic encephalopathy predicts the development of overt hepatic encephalopathy. Am J Gastroenterol. 2001;96:2718–2723. doi: 10.1111/j.1572-0241.2001.04130.x. [DOI] [PubMed] [Google Scholar]
- 18.Guevara M., Baccaro M.E., Torre A., et al. Hyponatremia is a risk factor of hepatic encephalopathy in patients with cirrhosis: a prospective study with time-dependent analysis. Am J Gastroenterol. 2009;104:1382–1389. doi: 10.1038/ajg.2009.293. [DOI] [PubMed] [Google Scholar]
- 19.Sharma P., Sharma B.C. Profile of hepatic encephalopathy in children with cirrhosis and response to lactulose. Saudi J Gastroenterol. 2011;17:138–141. doi: 10.4103/1319-3767.77246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pantham G., Post A., Venkat D., Einstadter D., Mullen K.D. A new look at precipitants of overt hepatic encephalopathy in cirrhosis. Dig Dis Sci. 2017;62:2166–2173. doi: 10.1007/s10620-017-4630-y. [DOI] [PubMed] [Google Scholar]
- 21.Fernández J., Acevedo J., Wiest R., et al. Bacterial and fungal infections in acute-on-chronic liver failure: prevalence, characteristics and impact on prognosis. Gut. 2018;67:1870–1880. doi: 10.1136/gutjnl-2017-314240. [DOI] [PubMed] [Google Scholar]
- 22.Srivastava A., Malik R., Bolia R., Yachha S.K., Poddar U. Prevalence, clinical profile, and outcome of ascitic fluid infection in children with liver disease. J Pediatr Gastroenterol Nutr. 2017;64:194–199. doi: 10.1097/MPG.0000000000001348. [DOI] [PubMed] [Google Scholar]
- 23.Albillos A., Martin-Mateos R., Van der Merwe S., Wiest R., Jalan R., Álvarez-Mon M. Cirrhosis-associated immune dysfunction. Nat Rev Gastroenterol Hepatol. 2022;19:112–134. doi: 10.1038/s41575-021-00520-7. [DOI] [PubMed] [Google Scholar]
- 24.Jiménez J.V., Carrillo-Pérez D.L., Rosado-Canto R., et al. Electrolyte and acid-base disturbances in end-stage liver disease: a physiopathological approach. Dig Dis Sci. 2017;62:1855–1871. doi: 10.1007/s10620-017-4597-8. [DOI] [PubMed] [Google Scholar]
- 25.Lipinski M., Saborowski M., Heidrich B., et al. Clinical characteristics of patients with liver cirrhosis and spontaneous portosystemic shunts detected by ultrasound in a tertiary care and transplantation centre. Scand J Gastroenterol. 2018;53:1107–1113. doi: 10.1080/00365521.2018.1498913. [DOI] [PubMed] [Google Scholar]
- 26.European Association for the Study of the Liver Electronic address:easloffice@easloffice.eu; European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of hepatic encephalopathy. J Hepatol. 2022;77:807–824. doi: 10.1016/j.jhep.2022.06.001. [DOI] [PubMed] [Google Scholar]
- 27.Snehavardhan P., Lal B.B., Sood V., Khanna R., Alam S. Efficacy and safety of sodium benzoate in the management of hyperammonemia in decompensated chronic liver disease of the childhood-A double-blind randomized controlled trial. J Pediatr Gastroenterol Nutr. 2020;70:165–170. doi: 10.1097/MPG.0000000000002521. [DOI] [PubMed] [Google Scholar]
- 28.Bajaj J.S., O'Leary J.G., Tandon P., et al. Hepatic encephalopathy is associated with mortality in patients with cirrhosis independent of other extrahepatic organ failures. Clin Gastroenterol Hepatol. 2017;15:565–574.e4. doi: 10.1016/j.cgh.2016.09.157. [DOI] [PubMed] [Google Scholar]
- 29.Jepsen P., Ott P., Andersen P.K., Sørensen H.T., Vilstrup H. Clinical course of alcoholic liver cirrhosis: a Danish population-based cohort study. Hepatology. 2010;51:1675–1682. doi: 10.1002/hep.23500. [DOI] [PubMed] [Google Scholar]