Quality of life for pediatric liver recipients


  • Estella M. Alonso

    Corresponding author
    1. Siragusa Transplant Center, Children's Memorial Hospital, Chicago, IL; and Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, IL
    • Siragusa Transplant Center, Children's Memorial Hospital, 2300 Children's Plaza, Box 65, Chicago, IL 60614
    Search for more papers by this author
    • Telephone: 773-975-8837; FAX: 773-975-8671


Key Points

1. The measurement of health-related quality of life (HRQOL) in children must be performed with instruments that are sensitive to developmental changes and that include assessments of domains of function that are specific to childhood.

2. Pediatric liver transplant recipients report HRQOL outcomes that are lower than normative samples. Although both physical function and psychosocial function are reported to be lower, the largest differences are observed in the area of school functioning. It is yet unknown whether these differences are related to an increased prevalence of developmental delays or learning disabilities in this population. It is also possible that missing days of school because of illness or visits to doctors may play an important role.

3. The impact of variables such as age and the interval from transplantation on HRQOL in this population has not been fully examined. However, it appears that age at the time of testing may have an impact, with lower function observed in older children. Liver Transpl 15:S57–S62, 2009. © 2009 AASLD.

In considering all aspects of health, transplant physicians must look beyond survival and graft function and consider overall well-being and functional outcomes. Examples of these outcomes include physical function, mental health, cognitive development, and daily living skills. These constructs are included in assessments of health-related quality of life (HRQOL) and are an important aspect of long-term follow-up for survivors of transplantation. HRQOL refers to those aspects of quality of life that are directly related to health and interaction with health care systems. It can be measured directly from the patient's perspective or, in situations in which the patient is unable to adequately respond, from the perspective of a proxy, such as a parent. HRQOL may be measured with generic or disease-specific approaches. Generic HRQOL instruments are multidimensional, consisting at least of the physical, psychological, and social health dimensions delineated by the World Health Organization. HRQOL may also be assessed with a disease-specific methodology. An assessment using a generic approach allows comparison with healthy children and across other disease samples, whereas an assessment using a disease-specific instrument focuses on dimensions that are uniquely relevant to patients with that disease. Disease-specific instruments may also be more sensitive to changes that are related to treatment choices and are thus important tools for assessing outcomes and effectiveness in clinical trials.


The measurement of HRQOL in children is somewhat more difficult than in adults. Instruments that test children must be sensitive to developmental changes and frequently require several age-specific versions. These tools must include the assessment of domains that are specific to childhood such as school function. Although children may possess the most accurate information regarding their own level of function in different settings, younger children may not be reliable respondents. They may answer items that question the same concept differently or overestimate or underestimate aspects of their function. Thus, in children younger than 5 to 8 years, using the parents as proxy respondents may yield a more reliable account. As with the measurement of HRQOL in adults, the measurements can be performed with generic or disease-specific tools, but disease-specific tools for pediatric conditions are less well developed. Table 1 lists the most commonly used pediatric HRQOL instruments.

Table 1. Pediatric Health-Related Quality of Life Instruments
  1. Abbreviation: PedsQL, Pediatric Quality of Life Inventory.

PedsQL 4.0 Generic Core Scales
Child Health Questionnaire Parent Report Form 50
Infant Toddler Quality of Life Instrument
Child Health Questionnaire Child Form 87
Health Utilities Index Mark 3
PedsQL Multidimensional Fatigue Scale
PedsQL 3.0 Transplant Module

CHQ CF-87, Child Health Questionnaire Child Form 87; CHQ PF-50, Child Health Questionnaire Parent Form 50; FOG, Functional Outcomes Group; HRQOL, health-related quality of life; NS, not significant; PedsQL, Pediatric Quality of Life Inventory; SPLIT, Studies of Pediatric Liver Transplantation.


Life after transplantation presents many challenges for both patients and their families. The requirement for daily medications, diagnostic monitoring, and office visits may interfere with some of the regular family routines and may contribute to ongoing anxiety for the parents. The medications that patients take frequently alter their physical appearance and contribute to their sense that they are different from their peers. Furthermore, parents of transplant recipients may be more fearful and protective and limit some of their children's opportunities and activities. It is not uncommon for older, adolescent patients to feel ambivalent and confused about their transplant care.


HRQOL in children after liver transplantation has been evaluated in several single-center studies that have been reviewed in a recent publication.1 To this point, HRQOL in pediatric liver transplant recipients has been assessed with only generic instruments and, with few exceptions, only from the parents' perspective.2–9 These studies have suggested that the majority of patients experience a relatively normal level of physical function. However, when they are considered as a group, physical function scores are lower than what is reported for healthy children. Psychosocial function is also diminished, especially in the area of school function. The altered school function that is observed in this group may be secondary to an increased prevalence of developmental delays or learning disabilities. One study of long-term recipients who were 3 to 9 years old revealed that almost 20% had an intelligence quotient of less than 70.10 A recent survey of school outcomes conducted by the Studies of Pediatric Liver Transplantation (SPLIT) network revealed that 40% had required some type of special educational services, with 8% having been classified as learning disabled and 5% having a reported intelligence quotient of less than 70. Missed school days may also contribute to compromised function. Data from the SPLIT network suggests that 30% to 40% of patients in long-term follow-up miss more than 10 days of school per year, with teens having the highest rate of absences.

Bucuvalas et al.6 conducted a cross-sectional study designed to assess the HRQOL of pediatric liver transplant recipients (5–18 years old) and to compare and contrast the findings from 2 well-validated HRQOL instruments, the Child Health Questionnaire Parent Form 50 (CHQ PF-50) and the Pediatric Quality of Life Inventory (PedsQL) 4.0 Generic Core Scale. The mean age at transplantation of the 77 participants was 3.8 years, and the average time since transplantation was 5.8 years (0.6–15 years). Children with biliary atresia accounted for 53% of the study population. Seventy-nine percent of the patients were white, 38% were Medicaid recipients, and the maternal education level was greater than high school for 60% of the population. Forty-two percent of the patients had been hospitalized at least once within the previous 12 months. The mean scores for physical and psychosocial health and for emotional, social, and school functioning were lower in this study population than population norms. Table 2 displays the results of the PedsQL 4.0 Generic Core Scales. With the exception of school functioning and psychosocial health, which were lower in the liver transplant recipients, the scale scores were similar to the scores for children with chronic illness previously reported.11 For the CHQ PF-50, physical and psychosocial summary scores were lower for transplant recipients than for the normal population, and no ceiling or floor effect was detected. When individual subscales were examined, the most significant differences were seen in the scales of physical functioning, general health perceptions, and parental impact on time and emotion. Scores for family cohesion, mental health, and roles were similar between the 2 groups.

Table 2. PedsQL 4.0 Generic Core Scales Scores in Comparison with the General Population and Children with Chronic Illness
SubscaleStudy PopulationGeneral PopulationPChronic IllnessP
  1. NOTE: Data for the general population and children with chronic illness were retrieved from Varni et al.11 Other data were taken from Bucuvalas et al.6

  2. Abbreviations: NS, not significant; PedsQL, Pediatric Quality of Life Inventory.

Physical Health77.8 ± 20.389.3 ± 16.3<0.0173.2 ± 27.0NS
Emotional Functioning73.2 ± 19.082.6 ± 17.5<0.0173.1 ± 23.8NS
Social Functioning75.5 ± 17.991.6 ± 14.2<0.0179.8 ± 21.9NS
School Functioning64.4 ± 20.785.5 ± 17.6<0.0171.1 ± 24.0<0.01
Psychosocial Health71.0 ± 15.686.6 ± 12.8<0.0174.8 ± 18.2<0.05

The SPLIT Functional Outcomes Group (FOG) recently conducted a large cross-sectional analysis of generic HRQOL in 675 (270 self-report) pediatric liver transplant recipients between the ages of 2 and 18 years, which included both parent and self-report responses. The mean age of the patients included in the sample was 8.0 ± 4.4 years, and 55% were female. The median interval from transplant to survey was 4.1 ± 2.5 years. These HRQOL outcomes were compared to a sample of healthy children randomly matched by age group, gender, and race/ethnicity. Findings in this large cross-sectional study were consistent with what is described above. The Total Scale Score and subscales of the PedsQL 4.0 Generic Core Scales were all significantly lower than those of healthy children (P < 0.001), with effect sizes ranging from 0.27 for self-reported Emotional Functioning to 0.68 for self-reported School Functioning. In a subanalysis of the determinants of the school function subscale, we found that questions related to missed school days had the largest impact, but that differences specific to cognitive function were also noted.


Several smaller studies of pediatric liver transplant patients have suggested that HRQOL may improve over time. Cole et al.8 performed a longitudinal assessment of 38 children under the age of 5 years at the time of transplant. Scores for all physical and psychosocial domains increased from transplant up to 12 months post-transplant. The largest change was observed in the first 6 months. Thus far, this single-center study is the only longitudinal assessment of HRQOL in this population. To further explore this relationship between time from transplant and functional outcomes, the parent-reported HRQOL scores from the SPLIT/FOG cross-sectional study detailed previously were compared at different intervals from transplant (Table 3).

Table 3. PedsQL 4.0 Generic Scale Scores by Interval from Liver Transplantation
PedsQL 4.0 Scale ScoreInterval from LT [Median (Interquartile Range)]
<2 Years (n = 262)2–4 Years (n = 355)5–7 Years (n = 217)≥8 Years (n = 92)
  1. Abbreviations: NS, not significant; PedsQL, Pediatric Quality of Life Inventory.

Total Score (P = 0.0067)78.3 (62.5–90.5)80.6 (65.2–91.7)78.3 (63.0–90.2)73.4 (60.3–84.8)
Physical Health (P = NS)85.7 (62.5–96.9)87.5 (71.9–96.9)84.4 (65.6–96.9)84.4 (65.6–93.8)
Psychosocial Health (P = 0.0008)77.2 (62.5–90.0)76.9 (63.3–90.0)75.0 (60.0–86.7)67.5 (57.5–80.8)

Contrary to our expectations, generic HRQOL did not seem to improve with time from transplant. Although physical function appeared to be similar across groups, psychosocial function was markedly lower at intervals of 8 years or more. We considered whether this observation might also be dependent on the age at testing because such a high percentage of patients received liver transplants as infants and would be reaching adolescence if they were tested 8 or more years later. The contribution of age at testing and interval from transplant will be tested in multivariate models predicting HRQOL in this data set, but large-scale longitudinal studies will be needed to fully examine this question.


Age may also be a factor that influences HRQOL. The SPLIT group conducted an earlier study of parent proxy-reported generic HRQOL at a mid-term follow-up time point: 2 years following liver transplantation. This included 102 children of 2 age groups: younger (n = 67) and older (n = 35) than 5 years.2 Results in the 2 age groups were found to be different. Older children had HRQOL similar to that in previously published reports, with physical function that was significantly lower than that of a normative group. In contrast, the group of younger children (2-5 years old) did not differ from the control group in subscales measuring physical or psychosocial function (Table 4). Univariate analysis among the subscales identified demographic variables but not clinical variables as significant predictors of HRQOL in these younger children.

Table 4. Infant Toddler Quality of Life Instrument Comparison
Item/ScaleMedian (Interquartile Range)P Value
LT Patients (n = 67)Control (n = 84)
  • The data were taken from Alonso et al.2

  • Abbreviation: NS, not significant.

  • *

    Controls scored lower than cases on this subscale.

Global Health Item85 (85–100)100 (85–100)0.004
Physical Abilities Scale100 (93–100)100 (97–100)NS
Growth and Development Scale93 (83–100)95 (88–100)NS
Discomfort and Pain Scale*92 (75–100)83 (71–100)0.009
Temperament and Mood Scale83 (74–90)80 (75–86)NS
General Behavior Scale76 (60–85)75 (66–83)NS
Global Behavior Item85 (60–85)85 (60–100)NS
Getting Along with Others Scale73 (64–80)75 (67–82)NS
General Health Perceptions Scale54 (45–63)78 (67–85)<0.001
Change in Health Item100 (75–100)50 (50–75)<0.001
Parental Impact-Emotional Scale82 (68–93)89 (75–96)NS
Parental Impact-Time Scale95 (81–100)95 (81–100)NS
Family Cohesion Item85 (60–100)85 (60–100)NS

The SPLIT/FOG cross-sectional data set was also analyzed to examine the impact of age at testing on parent-reported HRQOL. Results suggest that age at testing may indeed be important, with younger children having the highest scores (Table 5). In fact, the impact of age at testing appears to be more significant than the interval from transplant. Both of these variables will be included in the development of models to predict both self-reported and parent-reported HRQOL in the FOG data set.

Table 5. Pediatric Quality of Life Inventory 4.0 Generic Core Scale Scores by Age at Testing
Scale ScoreAge at Testing [Median (Interquartile Range)]
<2 Years (n = 259)2–4 Years (n = 254)5–7 Years (n = 244)≥8 Years (n = 169)
Total Score (P < 0.0001)85.7 (73.8–94.4)79.4 (63.0–90.2)73.9 (59.8–84.2)76.1 (59.8–88.0)
Physical Health (P < 0.0001)93.8 (78.1–100.0)87.5 (68.8–96.9)83.3 (62.5–93.8)81.3 (62.5–93.8)
Psychosocial Health (P < 0.0001)82.7 (70.0–92.3)76.7 (61.7–88.3)69.2 (56.7–81.7)73.3 (56.7–90.0)


Functional outcomes have also been specifically examined in adolescent liver transplant recipients. Our group conducted a survey of adolescent liver and kidney transplant recipients (11-18 years old) with the Child Health Questionnaire Child Form 87 and CHQ PF-50.12 These surveys included questions tapping constructs such as general health, bodily pain, self-esteem, and family impact that are not addressed in detail by the PedsQL 4.0 Generic Core Scale Score. Respondents included 36 pediatric liver transplant recipient/parent pairs and 18 kidney recipient pairs. Parents of liver transplant recipients reported physical function and general health perceptions that were lower than those of the normative group and similar to those of the kidney recipients (Figs. 1 and 2). Reported outcomes in psychosocial domains were closer to those of the normative group, but dimensions related to the emotional impact on parents and family activities were lower than those of norms. The self-reported outcomes of these patients were closer to those of a healthy population, with the only significantly lower score being in the area of general health perceptions. This contrast in viewpoint supports the concept that ratings of both parents and patients should be considered to gain a fuller picture of these outcomes.

Figure 1.

Child Health Questionnaire Parent Form 50 (CHQ PF-50) scores as reported by the parents/caregivers of 36 liver transplant recipients and 18 kidney transplant recipients versus the scores of a published normative population. *P ≤ 0.001; **P = 0.0001.

Figure 2.

Child Health Questionnaire Child Form 87 (CHQ CF 87) scores as reported by 36 liver transplant and 18 kidney transplant recipients. *Adolescent kidney transplant recipients perceived their general health (54.1 ± 10.5) to be significantly worse than the normative population (66.4 ± 14.6). P = 0.0006.


Adherence to the medication regimen required for the average liver transplant recipient depends on the cooperation of the entire family. The nonadherence rate for pediatric liver transplant recipients may be as high as 30% and may contribute to 15% of grafts lost in long-term follow-up. It is generally accepted that clinicians are poor judges of adherence and that noncompliance is rarely admitted unless queried in a nonjudgmental fashion. Several factors that may have a common impact on patients affecting both adherence and HRQOL are listed in Table 6.

Table 6. Factors Influencing Adherence Behaviors
Anger related to experience of chronic illness
Limited abstract reasoning and lack of vision of consequences
Concerns regarding physical appearance
Poor family communication and unclear roles and responsibilities
Ethnic and cultural influences

Fredericks et al.13 recently reported that measures of nonadherent behavior, including the standard deviation of immunosuppressive medication levels, were associated with lower generic HRQOL in adolescent patients, with strong associations noted for physical and school functioning. These data raise the question of whether variations in blood levels of immunosuppressive medications are related to neurocognitive function. Alternatively, do children with lower school functioning have cognitive delays that affect adherence behaviors? Nonadherence may also contribute to chronic graft dysfunction or episodes of rejection, which could influence physical function.


There are factors that contribute to and determine HRQOL after pediatric liver transplantation. A large-scale data collection gathered through a SPLIT/FOG HRQOL study suggests that both physical and psychosocial domains are lower than those reported for healthy children. We suspect that age at survey may influence reports of HRQOL, with parents reporting better outcomes for infants and younger children than school-aged children. It is unclear if time since transplant also influences HRQOL outcomes, but investigations are underway to better determine these associations. Health maintenance behaviors may be related to HRQOL, and adolescents who struggle with issues related to medication adherence may be at higher risk for compromised functional outcomes.