SEARCH

SEARCH BY CITATION

Keywords:

  • Children;
  • cognitive function;
  • intelligence;
  • liver transplantation;
  • WISC-IV

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

To date, the course of cognitive development in children after liver transplantation (Ltx) is poorly understood. Cognitive performance, however, is crucial in all developmental stages and for educational achievement. This cross-sectional single-center study examined the prevalence of long-term cognitive impairment in a cohort of 64 pediatric patients after Ltx. Median age at Ltx was 12 months. The revised Wechsler Intelligence Scale IV was administered to assess cognitive performance. Patients were compared with an age- and gender-matched group of children without a chronic health condition. Liver transplanted children performed significantly worse in three of four cognitive domains as well as in the Total Intelligence Quotient (Total IQ) (p = 0.017 to p = 0.005). Liver transplant recipients showed substantially more “serious delays” (IQ < 70) compared to the reference group (9.4% vs. 4.7%). Children with a genetic–metabolic disease performed worse than the other groups in three of the four WISC Indices and in the Total IQ (p = 0.05 to p = 0.01). The strongest association was revealed between height at Ltx and Verbal Comprehension (R2 = 0.21), Perceptual Reasoning (R2 = 0.30), Working Memory (R2 = 0.23) and Total IQ (R2 = 0.25). Our results indicate a high impact of primary diagnosis and height percentile at Ltx even on children's long-term cognitive performance.


Abbreviations
Ltx

liver transplantation

WISC-IV

Wechsler Intelligence Scale for Children—fourth edition

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

During the last 25 years, survival rates after pediatric liver transplantation (Ltx) have increased steadily, with survival rates currently ranging from 85% to 97% among liver transplanted children [1, 2]. Today, pediatric Ltx is a well-established treatment for a variety of liver diseases in their final stage. Even in relative indications (i.e. certain metabolic diseases), liver transplantation is a promising treatment option that actually is becoming increasingly important in clinical practice [3, 4].

Despite these major advances, (neuro)psychological research on the cognitive and emotional development of liver transplanted children as well as the process of adapting to a life with a chronic health condition has lagged far behind. Only recently has the focus shifted from mere survival to long-term health outcomes such as chronic kidney defects, chronic implant failure or psychosocial functioning [4, 5]. Cognitive performance constitutes an important developmental prerequisite. Importantly, childhood cognitive ability is highly predictive for educational achievement and later occupational outcomes [6] as well as health behavior [7]. Furthermore, average to high intelligence is considered to be one of the most important resilience factors for mental health [8, 9].

Current research indicates that liver transplanted children are at higher risk for developing cognitive deficits compared to the age-matched normal population [10, 11]. However, our understanding of the association between cognitive deficits and certain disease-related variables is still limited. Infantile onset of liver disease [12-14], longer pretransplant duration of disease, postmortem donation [15, 16], elevated serum calcineurin inhibitor level, lower pretransplant growth percentile and elevated serum ammonia [11] have all been suggested as potential risk factors for cognitive deficits in liver transplanted children. However, these findings are based on only a few studies with small samples or date back to the 1980s. Moreover, comparisons between different diagnoses have not yet been performed.

Given the immediate clinical relevance of delayed or impaired cognitive development in children undergoing Ltx, a better understanding of disease-related, therapy-associated and environmental risk factors could contribute to optimizing clinical care of these patients. In particular, more information about the long-term impact on postoperative cognitive abilities is urgently needed.

The current prospective cross-sectional study assessed 64 school-aged, liver transplanted children and compared four areas of cognitive performance as well as Total IQ with that of age- and gender-matched children without a chronic health condition. To this end, the re-edited, newly published German version of the Wechsler Intelligence Scale for Children (WISC-IV) [17] was applied. We expected that liver transplanted children late postoperatively would perform worse compared to the reference group. Furthermore, we hypothesized that the profile of cognitive impairment would differ depending on the primary diagnosis with more severe dysfunction in genetic–metabolic diseases. Moreover, we expected that the cognitive performance would be associated with certain disease-related variables such as age at Ltx, height at Ltx, pretransplant duration of disease or type of donation.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

Participants

The clinical sample consisted of 64 children who underwent pediatric Ltx at the University Hospital Hamburg-Eppendorf, Germany, between 1995 and 2010. Inclusion criteria were as follows: (1) age between 6 and 18 years, (2) currently attending school, (3) residency in Germany, (4) sufficient German language skills, (5) at least 1 year post-Ltx, (6) no acute diseases and (7) no mental handicap. In total, 70 children were eligible for the study and were recruited between June 2010 and May 2012 during their routine annual medical check-up. Four children did not participate because their parents declined participation, and two appointments were cancelled due to logistic reasons.

The reference group included 64 age- and gender-matched children without a chronic health condition. It was derived from the normative sample of the German version of the WISC-IV. Data were collected by the group of F. Petermann between 2005 and 2007. Sixty-four children were randomly selected and matched individually to each liver transplanted child by age and gender. The subtests of the WISC-IV are highly standardized and show good to excellent interrater reliability so that no bias was expected due to the fact that Ltx and control children were assessed by different examiners.

Procedures

This study was approved by the ethics committee of the Medical Council of Hamburg, Germany. All participants were informed about the study by letter and phone before their annual medical check-up. Informed consent was obtained from children and parents. During the annual medical exam, each child was individually tested on a day with no scheduled medical procedures. The WISC-IV was administered and scored in a standardized manner by a trained psychologist and lasted between 70 and 100 min. During the children's assessment, their parents were asked to complete a brief questionnaire on the family's current living conditions, including family income per month and the parents' level of education. All medical information was obtained from the patients' charts. The following pre- and posttransplant disease–related variables were included: diagnosis, duration of disease, days on the waiting-list, high urgency listing, age at Ltx, height and weight percentile at Ltx, days in hospital, days in the intensive care unit, number of complications, number of (viral or bacterial) infections, number of acute and chronic organ rejections, number of surgical revisions, number of re-Ltx and time since Ltx. Regarding primary diagnosis, four categories were defined: biliary atresia, cholestatic diseases, genetic–metabolic diseases and others. Laboratory parameters were obtained from the medical records at three time points: (i) immediately pre-Ltx, (ii) 1-year post-Ltx, and (iii) at the time of assessment. The following laboratory parameters were included: serum bilirubin level (mg/dL), serum albumin level (g/L), creatinine level (mg/dL), GOT/AST (U/I), GPT/ALT (U/I), GGT (U/I) and cholinesterase. With regard to children's immunosuppressive medication, we included the following variables at two measurement points (i.e. 1-year post-Ltx and at the time of assessment): type of main immunosuppressant, daily dose of CSA (mg), tacrolimus (mg), steroids (mg) and children's blood levels of CSA or Tacrolimus (µg/L). PELD scores were not included since most children had received their transplant before PELD scores had been in use.

Measures

Wechsler Intelligence Scale IV (WISC-IV)

Cognitive functioning was measured using the newly released and completely revised German version of the Wechsler Intelligence Scale for Children—fourth edition (WISC-IV) [17]. A global intelligence score, the Total IQ, is comprised of four index scores: (i) Verbal Comprehension Index, which includes the subtests Similarities, Vocabulary and Comprehension; (ii) Perceptual Reasoning Index, which consists of Block Design, Picture Concepts and Matrix Reasoning; (iii) Working Memory Index, which encompasses the subtests Digit Span and Letter Number Sequencing; (iv) Processing Speed Index, which is composed of the subtests Coding and Symbol Search. The normative populations mean (M) is 100, and the standard deviation (SD) is 15 for all indices. The WISC-IV was standardized in a population of 1.650 German-speaking children aged between 6.00 and 16.11 years in Germany, Austria and Switzerland. The criterion and factorial validity of the WISC-IV were satisfactory [17]. Retest reliability is excellent for both, the Total IQ (r = 0.97) and the index scores (r = 0.87–0.94). Compared to its predecessor WISC-III, the WISC-IV shows substantial improvements [18]: (a) a robust four-factor structure replaced the difficult-to-interpret and often criticized composite of Verbal- and Performance IQ and yields a broader representation of general intellectual functioning. It represents fluid reasoning and working memory as fundamentals of intellectual functioning and adheres more closely to contemporary intelligence theory. (b) Floor- and ceiling effects were reduced by adding items with low and high levels of difficulty, respectively. (c) Less dependency from educational level and increased developmental appropriateness (e.g. via modified instructions or the addition of specific teaching times). (d) Improved psychometric properties. In particular, the WISC-IV provides contemporary standard values for an appropriate comparison with the current norm population.

Data analyses

Statistical analyses were conducted using the Predictive Analysis SoftWare, version 18.0 (PASW, IBM, Ehningen, Germany). Comparisons between liver transplant recipients and the reference group and between different subgroups of the cohort were conducted using two-tailed independent t-tests or the Mann–Whitney U-test. For comparison with normative data from the general population, one-sample t-tests were performed. Analysis of variance (ANOVA) was performed to detect intergroup differences with respect to the defined diagnostic categories. For post hoc analysis, the Scheffé method was used. Effect sizes (d) for differences in means are designated as small (0.20), medium (0.50) and large (0.80) in magnitude [19]. For correlation analyses, Pearson's r was used. Simultaneous multiple regression analyses were performed with the four index scores and the Total IQ as dependent variables and disease–related variables such as type of donation, age at Ltx, duration of disease, height percentile at Ltx and time since Ltx as independent variables. Independent variables were selected based on our previous findings [15, 16]. Furthermore, explorative regression analyses (“stepwise” method) were performed with all described disease-related variables and laboratory parameters as independent variables in order to detect further predictors. If interrelations were based on outliers, the variable was excluded from stepwise regression analysis (Re-Tx, days in intensive care unit and days in hospital pretransplant). For regression analyses, laboratory parameters were log-transformed. For stepwise regression analysis we used z-scores of a combined daily dose of Calcineurin inhibitors (CNI; mg) and combined blood levels of CNI (µg/L).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

Patient characteristics are presented in Tables 1 and 2. Table 1 shows socio-demographic characteristics of participants and controls. At the time of assessment, children ranged from 6.3 to 16.11 years of age with a mean age of 11 years and 10 months. The mean elapsed time between Ltx and assessment was 8 years and 11 months (SD = 49 months; range: 12–191 months).

Table 1. Socio-demographic and clinical characteristics of 64 liver transplanted children and reference group
 Liver transplanted childrenChildren without chronic health conditionGeneral German populationt; p
  • 1

    Data on German income from n = 11 058; median income in Germany is about 1400 Euro in employed people; mean income per household is about 2700 Euro; relative poverty is defined as max. 50% of the median income, for example, less than 1640 Euro in a family with two children. Within the reference group, data of family income was not allowed to be collected due to restrictions of the ethical review committee.

  • 2

    Including special education school (father N = 1, mother N = 1).

  • 3

    Data from primary caregiver (80% mother).

  • 4

    “No graduation” includes special education school and also 3.5%, who are still attending school; all data about income and education are derived from the Federal Office of Statistics, Germany; —, not applicable.

Age at assessment (month)
M (SD)142.8 (34.0)142.2 (34.5)t = 0.10; p = 0.92
Range75–20375–203 
Gender (% female)45.3%44.8%t = 0.18; p = 0.86
Family income per month (n (%))
<1000 Euro5 (7.8%) 13.3%1 
1000–2000 Euro26 (40.6%) 35.4% 
2000–3000 Euro21 (32.8%) 23.4% 
3000–4000 Euro4 (6.3%) 11.8% 
>4000 Euro4 (6.3%) 8.4% 
Not specified4 (6.3%) 7.6% 
Education (father/mother) (n (%))
No graduation2 (3.2%)/4 (6.3%)23 (4.7%)37.6%4 
Graduation after 8 or 9 years of education22 (34.4%)/15 (23.4%)17 (26.6%)36.3% 
Graduation after 10 years of education25 (39.1%)/27 (42.2%)24 (37.5%)28.9% 
Graduation after 12 years of education9 (14.1%)/15 (23.4%)20 (31.3%)26.6% 
Not specified6 (9.4%)/3 (4.7%)00.6% 
Employment of primary caregiver
Full time9 (14.0%)15 (23.4%)31.3% 
At least part time22 (34.3%)19 (29.7%)30.5% 
Less than part time7 (10.9%)14 (21.9%)7.3% 
Not employed22 (34.4%)14 (21.9%)30.4% 
Not specified4 (6.4%)2 (3.1%)0.5% 
Main spoken language of the child
German48 (75.0%)51 (79.7%) 
German and other15 (23.4%)11 (17.2%) 
Other1 (1.6%)2 (3.1%) 
Table 2. Disease-related variables for liver transplanted children (n = 64)
 Liver transplanted children
  • 1

    Including PFIC (n = 4), neonatal hepatitis (n = 2), congenital hepatic fibrosis (n = 1) and primary sclerosing cholangitis (n = 1).

  • 2

    Including α-1-antitrypsin deficiency (n = 1), Wilson's disease (n = 1), hyperoxaluria (n = 1), Allagille syndrome (n = 2), Crigler Najjar syndrome (n = 2), OTC deficiency (n = 1), citrullinemia (n = 1) and argininosuccinic aciduria (n = 1).

  • 3

    Including liver tumor (n = 2), acute liver failure (n = 3) and hepatic failure of unknown origin (n = 7).

Age at transplantation (months)
M (SD); range35 (46); 0–142
Median12
Duration of disease (months)
M (SD); range29 (39); 0–149
Days on waitinglist
M (SD); range105 (115); 0–546
Basic immunosuppression at test (n (%))
CSA36 (56.3%)
Tacrolimus28 (43.8%)
Diagnosis (n (%))
Biliary atresia34 (53.1%)
Cholestatic diseases18 (15.6%)
Genetic–metabolic diseases210 (12.5%)
Others312 (18.8%)
Type of donation (n (%))
Postmortem donation49 (76.6%)
Living-related donation15 (23.4%)
Mean time elapsed since transplantation (months)107 (48.9)
Height percentile at Ltx (median)18
Weight percentile at Ltx (median)10
Number of hospitalizations pre-Ltx, M (SD)3.0 (1.8)
Days in intensive care (1-year post-Ltx)9.0 (7.4)
Number of surgical revisions (1-year post-Ltx)0.3 (0.7)
Number of rejections (1-year post-Ltx)0.5 (0.6)
Number of virus infections (1-year post-Ltx)0.6 (0.7)
Number of bacterial infections (1-year post-Ltx)0.9 (1.1)
Number of complications (1-year post-Ltx)0.9 (0.80)

Biliary atresia was the most common diagnosis (n = 34). There were significant group differences between the diagnostic categories concerning duration of disease and age at Ltx (for all p < 0.05). Children with the primary diagnosis of a genetic–metabolic disease experienced the longest duration of disease (F(37,26) = 5.48, p = 0.002) and were the oldest at Ltx (F(37,26) = 6.6, p = 0.001). Children who received a living-related donation were significantly younger at the time of Ltx (t(63) = −2.78, p = 0.007) (Table 2).

Comparison between children after Ltx and the reference group

Liver transplanted children scored within the lower normal range in all indices and in the Total IQ of the WISC-IV, ranging between M = 93.08, SD = 16.25 and M = 94.94, SD = 15.49. Compared to the reference group, the scores of transplant recipients were lower in all indices and in the Total IQ. These differences were statistically significant for three of the four index scales (Table 3).

Table 3. Comparison of WISC-IV indices and Total IQ between 64 liver transplanted children and children without a chronic health condition
WISC-IV indicesLtx children, M (SD)Children without a chronic health condition, M (SD)tpd
Verbal Comprehension95.16 (14.54)102.00 (17.29)−2.40.0170.44
Perceptual Reasoning94.94 (15.49)102.13 (15.37)−2.60.0090.52
Working Memory94.52 (18.17)99.53 (14.23)−1.70.0850.31
Processing Speed94.13 (15.56)101.08 (16.63)−2.40.0160.45
Total IQ93.08 (16.25)101.59 (17.41)−2.90.0050.48

Compared to normative data of the WISC-IV, patients scored below the population mean in all assessed indices and in Total IQ, t(63) = −3.41, p = 0.001 to t(63) = −2.42; p = 0.019. Children who were transplanted in their first year of life did not differ significantly from the population mean. Differences between WISC-IV indices were not significant (Supplementary Material, Tables S1 and S2).

According to the authors of the WISC-IV [17], scores <90 indicate below-average performance. We found that 28.6% (Working Memory) to 42.2% (Processing Speed) of the liver transplant recipients scored below average (expected 25%). With respect to the Total IQ, we found that 9.4% of the patients were in the clinical range of mental retardation (<70) compared to 4.7% in the reference group. 15.6% of the patients scored within borderline range (IQ 70–84) compared to 6.4% in the reference group. The frequency distribution of Total IQ scores differed significantly between patients and reference group (p = 0.032), with liver transplant recipients showing scores below the average more frequently (Figure 1).

image

Figure 1. Frequency distributions of Total IQ-Scores in liver transplanted children and the reference group. Mann–Whitney U-test = 1618.5; p = 0.032.

Download figure to PowerPoint

Differences in cognitive performance between diagnostic categories

Regarding liver transplant recipients, significant differences in cognitive performance were detected between the four diagnostic categories (Table 4). Children with the diagnosis of a genetic–metabolic disease performed worse in all indices and in the Total IQ.

Table 4. Differences between diagnostic categories regarding WISC-IV indices and Total IQ
WISC-IV indicesDiagnosisMSDnFp
Note
  • Levene statistics show that homogeneity of variance is given within all subgroups (p = 0.851 to p = 0.183).

  • 1

    A Scheffé post hoc test revealed that this group differed significantly from the other diseases group on a p < 0.05 level.

  • 2

    A Scheffé post hoc test revealed that this group differed significantly from the biliary atresia and the cholestatic disease group on a p < 0.05 level.

  • 3

    A Scheffé post hoc test revealed that this group differed significantly from the biliary atresia and the others group on a p < 0.05 level and on a p < 0.01 level, respectively.

  • 4

    A Scheffé post hoc test revealed that this group differed significantly from the biliary atresia, the cholestatic disease and the others group on a p < 0.05 level.

Verbal ComprehensionBiliary atresia96.1513.66342.760.050
 Cholestatic disease94.5013.428  
 Genetic–metabolic disease184.7016.3610  
 Others101.5013.0812  
Perceptual ReasoningBiliary atresia96.4113.64343.530.020
 Cholestatic disease102.5013.868  
 Genetic–metabolic disease281.9019.6810  
 Others96.5813.0612  
Working MemoryBiliary atresia94.9118.35341.750.167
 Cholestatic disease101.6310.458  
 Genetic–metabolic disease83.8020.4010  
 Others97.6817.8112  
Processing SpeedBiliary atresia95.9715.04343.310.026
 Cholestatic disease97.3810.088  
 Genetic–metabolic disease380.7010.7010  
 Others97.9218.9312  
Total IQBiliary atresia94.3814.69343.940.012
 Cholestatic disease98.5012.688  
 Genetic–metabolic disease478.4017.8510  
 Others98.0015.8712  

Considering the significant differences between the various diagnostic categories, we compared the cognitive performance in each category to the population norm (M = 100, SD = 15). Figure 2 displays means and standard deviations of the WISC-IV indices and the Total IQ for each of the four diagnostic categories. Children with a genetic–metabolic disease scored significantly (i.e. more than 1 SD) below the population mean in all indices and in the Total IQ, t(9) = −2.5, p = 0.004 to t(9) = −5.70, p < 0.001. There were no significant differences regarding height percentile at Ltx between the diagnostic categories (F(3,60) = 0.63, p = 0.599). Children with biliary atresia scored significantly below the population mean only in the Total IQ, t(33) = −2.23, p = 0.029.

image

Figure 2. Means of the WISC-IV indices and the Total IQ grouped by diagnostic categories. M = 100, SD = 15; scaling of the y-axis is M ± 3 SD.

Download figure to PowerPoint

Prediction

In a simultaneous multiple regression analysis, five disease-related variables (i.e. duration of disease, age at Ltx, type of donation, height percentile at Ltx and time since Ltx) were included. Height at Ltx appeared as the only significant predictor (Table 5). It was inversely associated with Verbal Comprehension, Perceptual Reasoning, Working Memory and Total IQ, and explained between 21% and 30% of variance of children's cognitive performance.

Table 5. Predictors for WISC-IV indices and Total IQ (simultaneous regression analysis)
CriterionPredictorsβSEpR2
Note
  1. SE, standard error; R2, the % of explained variance.

Verbal ComprehensionHeight percentile at Ltx0.3850.0720.0070.214
 Duration of disease−0.0700.0710.679 
 Age at Ltx−0.0480.0710.822 
 Type of donation0.1084.6100.433 
 Time since Ltx−0.1860.0550.281 
Perceptual ReasoningHeight percentile at Ltx0.5050.067<0.0010.296
 Duration of disease−0.1280.0660.425 
 Age at Ltx−0.0320.0660.875 
 Type of donation−0.0354.2900.787 
 Time since Ltx<0.0010.0510.998 
Working MemoryHeight percentile at Ltx0.4400.0890.0020.228
 Duration of disease−0.1320.0870.433 
 Age at Ltx0.0390.0870.854 
 Type of donation−0.1885.6400.173 
 Time since Ltx0.0580.0670.731 
Processing SpeedHeight percentile at Ltx0.1220.0860.4150.084
 Duration of disease−0.0160.0840.930 
 Age at Ltx−0.0620.0850.787 
 Type of donation0.1325.4600.376 
 Time since Ltx0.1240.0650.504 
Total IQHeight percentile at Ltx0.4660.0770.0010.249
 Duration of disease−0.0880.0750.593 
 Age at Ltx−0.0350.0760.865 
 Type of donation0.0144.8900.916 
 Time since Ltx−0.0080.0580.960 

Exploratory analysis of interrelations between disease-related variables, laboratory parameters and cognitive performance

Exploratory stepwise regression analyses were performed to detect further possible predictors for the cognitive performance. All eligible disease-related variables and laboratory parameters were included. Table 6 shows disease-related variables that served as predictors. Height at Ltx was confirmed as the most important predictor. Additionally, the number of days spent in intensive care within the first year posttransplant was found to significantly predict Verbal Comprehension, Perceptual Reasoning and Total IQ. Processing Speed was associated with age at Ltx, with younger age predicting better performance. Neither the daily dose of CNI nor blood levels of CNI nor any laboratory parameter contributed to the regression model.

Table 6. Predictors for WISC-IV indices and Total IQ in exploratory stepwise regression analysis
CriterionPredictorsβSEpR2
Note
  1. SE, standard error; R2, the % of explained variance.

Verbal ComprehensionDays in the intensive care unit within first year posttransplant−0.4300.3000.0200.185
Perceptual ReasoningHeight percentile at Ltx0.5040.0790.0020.445
 Days in the intensive care unit within first year posttransplant−0.4150.2550.009 
Working MemoryHeight percentile at Ltx0.4550.1280.0130.207
Processing SpeedAge at Ltx−0.4240.0610.0220.180
Total IQHeight percentile at Ltx0.4800.0940.0040.385
 Days in intensive care unit within first year posttransplant−0.3730.3020.023 

Furthermore, dosage or blood level of immunosuppressive medication at 1-year posttransplant or at the time of assessment did not significantly correlate with cognitive performance (Supplementary Material, Table S3).

Relations to gender and socioeconomic status

There were no significant gender-related differences (p = 0.104 to p = 0.841) or differences due to family income (p = 0.130 to p = 0.958) in cognitive performance. The mother's level of education was significantly correlated with Verbal Comprehension (r = 0.345, p = 0.007), Perceptual Reasoning (r = 0.370, p = 0.004) and Total IQ (r = 0.310, p = 0.017) of the child. Level of education of the father, however, was only associated with Verbal Comprehension (r = 0.336, p = 0.011).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

The present findings demonstrate long-term cognitive dysfunction in liver transplanted children compared to an age- and gender-matched reference group of children without a chronic health condition as well as compared to the normal population. In comparison to the reference group, twice as many patients displayed serious cognitive delay. Children with a primary diagnosis of genetic–metabolic disease exhibited the lowest scores. Height percentile at Ltx proved to be the most relevant predictor of posttransplant cognitive performance. These findings support our hypothesis and confirm and expand previous research in this area [10, 11, 20, 21].

While several studies to date have demonstrated a lower IQ in pediatric Ltx patients [20, 22, 23], only one of these specifically addressed verbal deficits [24]. Moreover, all studies were conducted in very small patient groups with a sample size ranging from N = 15 to N = 28. In our study, we found substantial impairments across several domains of cognitive function such as verbal comprehension, perceptual reasoning and processing speed. This should be emphasized as the revised version of the Wechsler battery (WISC-IV) is better suited to differentiate between these subdomains of cognition compared to the WISC-III [25] used in previous reports. In addition, our data may provide a more accurate estimate of cognitive impairment in pediatric Ltx patients by limiting the Flynn effect (i.e. norms become outdated with an increase of approximately 3 IQ-points per 10 years) [26]: several studies have demonstrated that the WISC-III indeed appears to overestimate a child's intelligence because of this shift in norm values [27]. Therefore, the results of previous studies in liver transplanted children utilizing the WISC-III [11, 20, 24] likely underestimated the degree of cognitive impairment in this population. Against this background, the present results gained with the WISC-IV are even more meaningful. In addition, the new 4-factor structure enables clinicians to evaluate a specific intellectual profile of every child [28]. Combined with guidelines for interpretation, this may serve as a basis for individually tailored interventions [18]. Moreover, the newly added Index Working Memory allows for comparisons with our previous studies on attention and executive functioning [16].

Although differences in Working Memory between patients and the selected reference group fell short of statistical significance, the difference to the population mean was significant. Therefore, this finding confirms our earlier reports. We have previously demonstrated substantial impairment in a Working Memory task of a computer-based assessment tool in 59 liver transplanted children [16]. Yssaad-Fesselier et al [21] also found Working Memory scores significantly below the population mean. Furthermore, Sorensen et al [10] detected impaired Working Memory performance based on a rating of executive functioning completed by parents and teachers. Thus, Working Memory may be particularly affected in pediatric liver transplant recipients. Neurobiologically, this is plausible as corticosteroids are toxic to the hippocampus, a limbic structure relevant for learning and memory [29].

Our results support the assumption that liver transplanted children with different diagnoses show significant differences of cognitive performance. Children who received Ltx following a primary genetic–metabolic disease scored much worse compared to the others. Wilson's disease, hyperoxaluria, Crigler Najjar syndrome and especially urea cycle defects are known to affect the brain. Our results are in line with recently published work by Stevenson et al [30] who demonstrated significant mental deficits in a group of 33 pediatric liver transplant recipients with metabolic disorders. As children with a genetic–metabolic disease also suffered from the longest duration of disease preoperatively, our results may argue for early transplantation in these children. Our results might be of special relevance, as “after biliary atresia, metabolic conditions […] are the second most common disease category leading to liver transplantation in children” ([31], p. 59). However, at present, these findings have to be interpreted with caution due to the heterogeneity of diseases included in our study and the small sample size in the genetic–metabolic group. Therefore, larger, multicenter studies are warranted to further explore these associations.

We expected duration of disease, age and height at Ltx, and type of donation to predict cognitive performance. This hypothesis was partially confirmed since only height at Ltx significantly contributed to all cognitive indices except for Processing Speed. Available previous studies have rarely included laboratory parameters, usually examined small samples [11, 24, 32], or were published more than 20 years ago [12, 13, 22]. Therefore, we also performed exploratory stepwise regression analysis by including disease-related variables and laboratory parameters in order to detect combinations of variables that might contribute to cognitive deficits after Ltx. This procedure confirmed the impact of height at Ltx as a significant predictor. As this is in line with earlier findings [11], children with lower height percentile at Ltx should be carefully monitored posttransplant regarding cognitive performance. Growth might be interpreted as a marker for nutritional status. Several studies in the recent literature have shown that malnutrition, especially in early childhood, is highly interrelated to cognitive and behavioral dysfunction and adult mental disorders [33, 34].

Apart from height at Ltx, other previously suggested disease-related factors that may contribute to cognitive impairment were only partially confirmed. For example, in line with Krull et al [24], we found that more days spent in the intensive care unit after the Ltx was correlated with verbal delays. However, regarding the impact of age at Ltx, our results are contradictory to those obtained by other research groups. Young age at onset of disease and consecutively at Ltx was associated with an increased risk for cognitive delay [5]. In our study, half of the liver transplanted children received their transplant within the first year of life and their cognitive performance did not differ significantly from the population mean. Additionally, age at Ltx was only related to Processing Speed, with younger age predicting better performance. These findings are in line to our previously published reports [16, 35]. However, it has clearly been demonstrated that adverse early biological and environmental experiences exert dramatic influences on the developing brain [36]. It is therefore conceivable that, similarly, there might be a critical window for liver disease to affect the brain during early childhood. On the other hand, if Ltx is performed very early in life, neuronal plasticity might alleviate potential toxic influences. Therefore, the contradictory results might be due to a vague definition of “younger” and “older.” Usually this is only defined within the examined respective samples. Instead, further studies should compare groups of liver transplanted children with the same primary disease, but with clearly different ages at Ltx.

We did not find any laboratory parameter to serve as a significant predictor of cognitive performance. Moreover, there was no relation between daily dosage of immunosuppressive medication or CNI levels (measured at 1-year posttransplant, or time of assessment) and cognitive performance. However, these findings are limited as blood levels were only determined once, either 1-year posttransplant and at the time of assessment.

The degree of cognitive impairment detected in our cohort is similar to the results obtained in pediatric kidney and heart transplant recipients. A recent study found cognitive impairment in 49 children after kidney transplantation with a reported Total IQ of M = 83.9, SD = 20.0 [37]. Older studies mainly showed deficits regarding nonverbal abilities [38, 39]. IQs of children after heart transplantation are relative consistently reported to be 10–15 IQ points below nonclinical comparison groups [40, 41]. In children with biliary atresia, cognitive deficits can already be detected preoperatively [42, 43]. Given the starkly different etiologies and pathogenetic processes leading to organ failure in these diseases, the underlying neuropsychological and -biological mechanisms remain to be elucidated. Further research should also take into account that stress [44] and anesthesia [45, 46] in early childhood may negatively affect brain development and consequently cognitive function.

The obtained interrelations between parental education level and some IQ Indices are within the expectable scope in the population [47].

In summary, our current study provides clear evidence for substantial cognitive impairment in a relatively large cohort of pediatric liver transplant recipients. These results were obtained in a cohort representative of the clinical heterogeneity of children undergoing Ltx as different types of primary diagnoses were considered. In addition, our results are further strengthened by the fact that all patients were school aged and were assessed with the same instrument. Furthermore, children with known mental retardation disorders were excluded in order to minimize confounding effects of outliers. We also included an appropriate reference group and used a newly revised assessment method.

Cognitive deficits have tremendous consequences for future development and well-being, especially in children, as cognitive performance impacts daily life as well as educational success. Ltx in children is meant to result not only in survival but in the best achievable health status in the long term. Therefore, cognitive performance should be regarded as equally important measure of therapeutic success in addition to medical outcomes.

The high prevalence of cognitive deficits demonstrated in the present study emphasizes the urgent need to implement (neuro)psycholocial screening procedures using appropriate and standardized assessment methods into routine clinical care. Counseling for the parents should also be offered on a regular basis. Importantly, further research should explore (neuro)psychological treatment options in order to develop evidence-based intervention strategies.

Authors' Contributions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

Tanja Kaller participated in research design, collected and analyzed the data, and wrote the article. Nadine Langguth participated in performance of the research and revised the manuscript. Franz Petermann contributed to the acquisition of the data and revised the article. Rainer Ganschow and Björn Nashan participated in medical and health care and contributed to the acquisition of the data. Karl-Heinz Schulz participated in research design, data analysis, and planning, discussing and revising the article.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

This work was supported by Homann Foundation, Jovita Foundation, Friede Springer Foundation, and Paul und Helmi Nitsch-Foundation and the Registered Association Hamburg Heals Children. We are grateful to Sebastian Kohlmann, Anja Fischer and Stephan Gold for language support.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Authors' Contributions
  8. Acknowledgments
  9. Disclosure
  10. References
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
ajt12408-sup-0001-SupTab-S1.docx19K

Table S1: Confidence intervals of the WISC-IV indices and Total IQ in liver transplanted children (n = 64)

Table S2: Results of children transplanted in their first year of life (n = 33) in comparison to the population mean

Table S3: Correlations between immunosuppressive medication and WISC-IV indices and Total IQ in liver transplanted children (r, p)

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.