The Quality of Life of Long Term Survivors of Patients with Biliary Atresia

Purpose: Advances in surgical techniques and perioperative care have improved patients' short-and mid-term postoperative outcomes with Biliary Atresia (BA). However, the long-term results of these patients have not been thoroughly investigated. This systematic review aims to determine the long-term outcomes and the patients' health-related Quality of life (HrQoL) with their native livers or liver transplantation. Methods: A systematic literature-based search for relevant cohorts was performed using Pubmed/Medline, Cochrane Library from its inception to August 2021. Original studies reporting on BA, Hepatoportoenterostomie, portoenterostomy, Kasai, Liver transplantation, Quality of life, or HrQoL were included. Pooled prevalence has been calculated for cholangitis, secondary liver transplantation, or associated malformations using MetaXL (version 5.3). Subgroup analysis on HrQoL followed surgical treatment after BA was calculated by using RevMan (version 5.4). Results: 12 articles were considered for data synthesis. Nine studies compared biliary atresia patients to an age-matched healthy reference group. 4/9 (n = 338) of these studies indicated lower scores for biliary atresia patients; 5/9 (n = 127) stated similar health status. A Forest plot analysis including all studies with total HrQoL showed a tendency of higher scores towards healthy controls (MD -0.79, 95% CI: -6.00-4.41). Comparing patients after Kasai Hepatoportoenterostomy with healthy controls demonstrated favorable outcomes for the control group (MD -3.22, 95% CI: -7.20-0.75) with no statistical signicance (p = 0.11). The pooled estimation of the prevalence of cholangitis, secondary liver transplantation and associated malformations are 0.33 (95% CI: 0.06–0.66), 0.59 (95% CI: 0,42–0.75) and 0.13 (95% CI: 0,01–0.33). Conclusion: Biliary atresia patients have an overall high prevalence of progressive liver-related complications and risk of lower HrQoL compared to their healthy peers. Furthermore, those patients who received liver transplantation appear to have the same Quality of life as those living with their native livers. Targeted and evidence-based follow-up procedures and transitional care are essential to meet these patients' long-term care needs. Prospective and multicenter research das focuses on the attributes and predictors of the long-term prognosis of patients with biliary atresia are necessary.


Introduction
Biliary atresia (BA) is a neonatal cholangiopathy of unknown cause. It leads if left untreated, to postin ammatory, brotic obliteration of the extrahepatic bile duct with consecutive liver cirrhosis and death [1]. BA occurs in 8000 -18000 live births and is more common in Asians and Europeans in females than males [2]. Hepatoportoenterostomy (HPE; the Kasai procedure) permits bile ow after resection of the brotic tissue via a Roux-y-loop, therefore relieving extrahepatic biliary atresia obstruction [3]. However, it is not a curative procedure. Despite HPE being performed quickly, liver transplantation (LT) is ultimately required for most patients during childhood [4]. The disease is the leading indication for LT in the pediatric population with considerable morbidity and mortality, despite early detection [5].
Approximately one-half of affected infants will require LT in the rst two years of life due to complications of cirrhosis and cholestasis, including severe malnutrition, ascites, portal hypertension, and coagulopathy [6]. The remainder of children with BA may live many years with their native livers, despite chronic, progressive cirrhosis. However, the long-term outcomes of children with BA surviving with or without LT has not been well documented. As with every chronic disease, assessing Health-related Quality of Life (HrQoL) is essential in providing comprehensive long-term care to children. One study revealed that children with BA and surviving with their native livers have a signi cantly impaired healthrelated quality of life, similar to that in patients with BA post-LT [7].
On the other hand, another study revealed that young adult BA survivors are experiencing a similar HrQoL compared with the reference population [8]. Because of the variety of the reported Quality of life in patients with BA, this systematic review aimed to clarify the con icting results in the reported literature.
Further, the present study's information will help facilitate evidence-based management and follow-up design and better care for patients with BA.

Methods
We conducted a meta-analysis according to the review protocol and Meta-analysis using PubMed/Medline, Cochrane Library as databases. For PubMed, the search string contained a combination of the following Medical Subject Heading terms, title/abstract, and topic eld tags: "Biliary atresia," "Kasai," "portoenterostomy," "liver transplantation, "quality of life," "health-related quality of life." The reference lists of the included studies and existing systematic reviews were reviewed for additional relevant studies.

Selection criteria
Two reviewers independently scanned the titles and abstracts of the acquired articles for the initial screening. Original articles that reported the outcome or Quality of life of patients with BA surgical history were included. Only studies quantitatively measuring the Quality of life, i.e., using a scoring system, were included in this context. The exclusion criteria were as follows: 1) non-English language papers; 2) reviews, conference proceedings, and case reports or case series; 3) studies conducted on animal models or focused on analyzing the molecular biological or pathological mechanisms of biliary atresia; 4) studies focusing on parental stress and anxiety and 4) publication date prior to 2000. Studies were excluded if they did not meet eligibility criteria. After the initial screening, the full text of 87 included articles was retrieved and read by two reviewers to determine their eligibility for inclusion in the analysis.
The structure of this systematic review followed the PRISMA guidelines ( Figure 1).

Data extraction
The searches were performed in August 2021. A standardized Excel spreadsheet that included the study key characteristics, such as year of publication, geographical region, the study design, time of data collection, patients' age range, the age range of patients, gender, associated malformations, primary surgical BA treatment (Kasai HPE or LT), age at surgery, cholangitis, portal hypertension, reoperations, and deaths was developed for data collection (Table 1). Another separate Excel spreadsheet was created for data concerning HrQoL of BA patients with a surgical history. Table 2 presents the applied HrQoL  assessment instruments of the respective papers, the number of patients enrolled for HrQoL assessment,  HrQoL of patients with Kasai HPE/native liver and LT compared to healthy controls, HrQoL of patients  who underwent Kasai HPE compared with LT, identi ed factors associated with HrQoL and the studies main ndings. Two reviewers independently extracted information from the included articles. Discrepancies in the screening and data extraction process were discussed and resolved by the consensus of the two reviewers.

Statistical analysis
Pooled prevalence has been calculated using MetaXL (version 5.3) software add-in for Microsoft Excel. For the estimation of pooled prevalence, the double arcsine transformation was used. Results were calculated with a 95% CI. Expecting a high heterogeneity, a random-effects model was selected, and heterogeneity was assessed using the I 2 statistic, which describes the percentage of variation across studies not only resulting from sampling error. An I 2 value above 75% was de ned as an indicator for high heterogeneity. Since this analysis aimed to estimate the pooled effect size without testing a hypothesis, there was no p-value calculation for the pooled prevalence analysis. Subgroup analysis on HrQoL after surgical treatment of BA was calculated by using RevMan (version 5.4). A p-value < 0.05 was de ned as statistically signi cant for this analysis, and results were presented with a 95% CI.

Study characteristics
The systematic literature review resulted in the identi cation of 1647 publications ( Figure 1). Of these, 156 were duplicates, 1404 records were excluded following title and abstract screening, and 87 articles were selected for full-text review. Finally, 12 articles were considered for data synthesis, among which eight were multicentred studies [9][10][11][12][13][14][15][16], and the remaining four were single-centered studies [17][18][19][20]. Table 1 presents an overview of the study characteristics of the included paper. Estimates of HrQoL of patients with biliary atresia were found for samples from a total of 6 countries (China, Canada, Netherlands, Finland, Japan, Italy) and three continents (Europe, North America, and Asia). The studies included patient data ranging from the year 1951 to the year 2016. Most of the included studies had HPE as the primary surgical approach (10 of 12) [9,10,[12][13][14][15][16][17][18][19]. Only two studies presented liver transplantation as the primary surgical treatment of BA [11,20]. A total of 1183 patients were enrolled when adding up the total number of patients of all included studies.

HrQoL Measurement
In total, ten HrQoL assessment instruments were used. In 42% (5 of 12) of the identi ed publications, HrQoL was measured using the Pediatric Quality of Life Inventory 4.0 Generic Core Scales (PedsQL™ 4.0 GCS), making it the most commonly utilized Quality life instrument. The PedsQL™ 4.0 GCS are child selfreport, and parent proxy-report scales developed to measure health-related Quality of life in children and adolescents ages 2-18, including physical functioning (8 items), emotional functioning (5 items), social functioning (5 items), and school functioning (5 items). These scales produce a Total, Psychosocial, and Physical Summary score. The Short Form-36 Health Survey (SF-36), RAND-36 for Health Status (RAND-36), and WHOQOL-100 for QoL (WHOQOL-100) were each utilized twice in the identi ed publications. The SF-36 questionnaire consists of eight scaled sections: vitality, physical functioning, bodily pain, general health perceptions, physical role functioning, emotional role functioning, social role functioning, and mental health. The score of each section results in a scale from zero to 100, with 100 being the highest score and highest function attainable [21]. The WHOQOL-100 is a generic, cross-cultural multidimensional questionnaire. It consists of 100 items assessing 24 aspects of QoL within six domains (physical health, psychological health, level of independence, social relationships, environment, and spirituality) and a general evaluative aspect (overall QOL and general health). This 100-item questionnaire has a 5-point response scale from 1 to 5. Scores are calculated for each domain, with a maximum possible score of 20 per domain [22]. RAND-36 comprises 36 items that assess eight health concepts: physical functioning, role limitations caused by physical health problems, role limitations caused by emotional problems, social functioning, emotional well-being, energy/fatigue, pain, and general health perceptions. Physical and mental health summary scores are also derived from the eight RAND-36 scales. The scores were converted to a 1 to 100 scale, with higher scores indicating better levels of functioning or well-being [23]. The child self-reported Child Health Questionnaire™ (CHQ-CF87), applied by one study, is a full-length 87-item self-report. Per scale, the items are summed up and transformed into a 0 (worst possible score) to 100 (best possible score) scale [24]. Both studies that included listed liver transplantation as the primary procedure applied an additional disease-speci c questionnaire, respectively: Pediatric Liver Transplant Quality of Life (PeLTQL) and PedsQL™ Transplant Module.

HrQoL Outcome
Complete results of quantitative analysis are provided in Table 2. In total, 811 BA patients were evaluated for HrQoL.

HrQoL compared to the healthy population
Of 12 included studies, nine studies compared BA patients to an age-matched healthy reference group. 4/9 of these studies (n = 338) indicated lower scores for BA patients after surgical treatment when compared to healthy peers [9,11,15,17]; 5/9 studies (n = 127) stated similar health status in both groups [12][13][14]16]. Figure 2 displays the HrQoL of all BA patients compared to healthy peers, including all studies which provided total scores of HrQoL (mean and standard deviation) with a scale from zero to 100, with 100 being the highest score. Therefore, this forest plot analysis only includes seven studies (n = 502) since Lind et al., and Parolini et al. failed to include HrQoL total scores. Comparing both groups for HrQoL, there was a tendency towards healthy controls (MD -0.79, 95% CI: -6.00-4.41, p = < 0.000, I 2 = 93%). Although there was a tendency for higher scores in healthy peers, no signi cant difference was found for both groups (Z = 0.30, p = 0,76).

Kasai HPE/Liver transplantation vs. healthy population
Eight studies compared HrQoL of patients surviving with their native liver after Kasai HPE with healthy controls. Of these studies, 3/8 studies (n = 268) reported lower scores, and 5/8 studies (n = 127) demonstrated comparable HrQoL for BA patients compared to healthy peers, with the only signi cant difference found in "missing school or daycare for hospital visits" and "general health perception" for patients with BA [12,13]. One study even revealed signi cantly higher scores for the social domain in the BA group compared with healthy peers [12]. Figure 3 compares non-transplanted BA patients against healthy peers. Again, only studies with su cient HrQoL data were included for the forest plot analysis. The forest plot demonstrates favourable outcome for the control group (MD -3.22, 95% CI: -7.20-0.75, p = 0.000, I 2 = 77%). However, no statistical signi cance was found for the tendency of higher HrQoL towards disease-free controls (Z = 1.59, p = 0.11). Only three studies included data comparing transplanted patients with healthy controls [11,17]  atresia after a living donor liver transplant in Japan but did not compare HrQoL data with the healthy or general Japanese population [20]. With only three studies providing data regarding HrQoL of LT vs. healthy controls, no reliable statement was possible.

Kasai vs. liver transplantation
Five studies compared HrQoL of native liver patients with liver transplant recipients (n = 327). [9,[13][14][15]17]. All ve studies described similar/comparable scores for both groups. However, even though De Vries et al. described comparable scores for both groups, non-transplanted BA patients had signi cantly lower scores on general health perception than transplanted patients. Furthermore, native liver patients tended to have lower physical role functioning than the reference and transplanted patient groups [14].

Factors associated with HrQoL
The included studies identi ed several factors associated with HrQoL. Concerning patient characteristics, children's higher age at survey was signi cantly related to higher generic HrQoL [20]. Furthermore, patients with and without associated anomalies had no signi cant difference in HrQoL scores [13]. Regarding gender, females tend to report lower HrQoL than males [14]. Interestingly, "age at surgery" revealed no signi cant difference between patients with good and impaired Quality of life (p = 0.56) [18]. However, four studies failed to include age at operation in data. The number of drugs for immunosuppression was stated in two studies to be related to a lower transplant-speci c HrQoL [11,20]. Miserachs et al. de ned immunosuppression polytherapy as more than two immunosuppressant drugs at a time. Three studies tested laboratory-based parameters for correlation with HrQoL. De Vries et al. found no signi cant correlations for "serum Aspartate-Aminotransferase," "bilirubin," "albumin levels," "age at LT," or "time elapsed since LT" with HrQoL. Then again, another study described "Aspartate-Aminotransferase level at one year" after Kasai HPE to be a signi cant independent predictor for HrQoL (p=0.

Pooled Prevalence
All in all, as studies with HrQoL as the primary outcome, the included studies only provided very little data regarding long-term clinical outcomes after surgical biliary atresia treatment.

Discussion
Overall, children with BA are at risk of impaired HrQoL, both physically and psychologically, signi cantly when younger, whether they are native liver survivors or received an LT during disease progression ( Figure  2). Whereas bilirubin levels, the number of cholangitis episodes, variceal bleeding, and nutritional status mainly de ne the pretransplant HrQoL, the HrQoL posttransplant is mainly affected by the number of prescribed immunosuppressive medications. Interestingly, age at surgery and associated malformations seem not to affect the HrQoL (Figure 4). However, children at higher ages score higher in HrQoL than their younger peers. This might be a result of their long-term adaption to their disease. This phenomenon is known as "response shift," a change in individuals' perceptions of their HrQoL based on their individual experiences [25] Moreover, children who already had reached school age are clinically stable, with a similar or even improved health status compared with the years before, most likely score their HrQoL higher. Also, the score can be affected by the reporter. Parents might tend to report a lower HrQoL because of the intense years after their child was diagnosed with BA and their experience coping with the disease [20].
An important nding is the often neglected aspect of gender difference in HrQoL. In our study, we found that females had lower scores in all domains of HrQol questionnaires. Especially in bodily pain, females show increased sensitivity to nociceptive stimuli [26]. Studies reported that the hormones like estrogen or testosterone had a direct impact on perceived pain. For instance, estrogen increases the nervous system activity, which increases the transmission of pain stimuli and pain sensitivity of a woman [27].
On the other hand, testosterone has an analgesic effect and can reduce pain sensitivity in men [26,27]. Also, in another study concerning "fatigue" after LT, Berg-Emons et al. concluded that the female gender perceives "fatigue" with greater severity, suggesting impaired HrQol [28]. However, although females score lower in "physical function" and "physical role functioning," studies are suggesting that females score higher in domains "emotional role functioning" and "mental health," suggesting females have a better ability to manage stress and psychosocial demands in the posttransplant setting [26]. However, these ndings could not be substantiated in the present study.
Nevertheless, gender differences also exist in response to painful situations in teenagers and adolescents. Males would primarily seek out distracting behaviors while females use social support and positive self-statements [26,29]. Hence, it is essential to recognize gender differences in the setting of chronic liver disease to direct attention to being active, thinking positively, and expressing emotions positively, which will be correlated with higher scores in HrQol questionnaires. However, among all pretransplant clinical covariates, serum bilirubin following HPE appears to be the most predictive biomarker of the outcome [30]. With increasing levels of bilirubin, the likelihood of the need for LT increases. For example, children with Bilirubin levels > 50 µl/ml at 2-and ve years had a transplant-free survival of 4.8 % and 0, respetively [31]. Cholangitis is one factor that has a direct impact on HrQol in patients with BA. In the present study, the pooled prevalence of cholangitis was 33% which correlates with the incidence of cholangitis reported in the literature (40 % and 90 %) (Figure ). In general, most patients experience at least one episode prior to 2 years of age. Recurrent cholangitis may predict the need for LT as it can lead to progressive cirrhosis [30,32]. On the other hand, one episode of cholangitis does not predict early LT [30]. Also, other liver disease-associated symptoms like therapy refractory ascites with accompanying risk of spontaneous bacterial peritonitis or increased portal hypertension with the risk of gastrointestinal hemorrhage, which is present throughout the lives of children with BA, are often correlated with a poor outcome and early liver transplantation needs. However, HrQol is mainly dependent on whether adequate bile drainage is achieved after HPE [33] [30,34].
A signi cant common clinical problem in BA is poor nutrition and, along nondrainage of the HPE, the most common indication for LT. Unfortunately, nutritional problems in BA are challenging to overcome.
Children with BA have a constant caloric de cit and need to be supplemented with extra calories, fat-soluble vitamins, and medium-chain triglycerides (MCT) to prevent weight loss and failure to growth [35]. The perceived dependency from external nutritional support is also re ected by the correlation of a low HrQoL and the number of prescribed drugs. We found that immunosuppression polytherapy and the number of prescribed medications were associated with a signi cantly lower HrQoL [11,20]. Immunosuppressive medications are necessary post-LT because of the risk of organ rejection and graft loss, and the need for re-transplantation is a lifelong danger. However, those drugs produce side effects that need to be monitored. In one study, the ideal survivor after liver transplantation was on single-agent immunosuppression, not receiving prednisone, antihypertensive or antiseizure medications with no impact on linear growth or immunosuppression-related comorbid conditions [37]. A condition that is barely achieved in those patients who received an LT. So, it is not a surprise that Miserachs et al. reported signi cantly lower HrQoL scores in liver transplant recipients with biliary atresia (p < 0.001) than healthy peers in physical, psychosocial, social, and family functioning domains [11]. Despite the constant improvement of the survival rates following LT, future research on immunosuppression withdrawal will be necessary to improve the HrQoL of children with BA.
Although we found that the HrQoL of those who lived with their native livers and BA patients who received an LT is comparable to healthy peers, the survival of BA patients with their native liver is poor and reported in the literature between 30% -55%, 30% -40%, and 20% -40% at 5, 10, and 20 years respectively [38]. Most patients will develop chronic liver disease and require an LT. In the present study, the pooled prevalence of having an LT after a "failed HPE" was 59 %,which is in line with the reported literature [39]. It has been suggested that LT as a primary approach for managing patients with BA is possible. However, HPE remains to be the most reasonable initial approach. If the patient is stable, it is better to delay the LT to increase the chances of a successful liver transplant [38].
One often-overlooked factor is that patients with BA often show a history of neurodevelopmental delay, especially in early childhood [37]. Also, motor impairments or behavioral problems are common. Additionally, up to 25 % f the patients are suffering from anxiety and depression. While addressing the numerous liver disease complications, it is essential to identify patients with neurodevelopmental and psychological issues as early as possible for timely referral to a psychosocial health care provider [12].

Conclusion
The development of preventive and therapeutic modalities in treating patients with BA depends on understanding the disease. Although considerable progress has been made in basic research and the clinical setting, the disease remains a dark chapter in pediatric hepatology [38]. The present study underscores the urgent need to improve the various factors in uencing the HrQoL in patients with BA. This includes individualized patient-speci c immunosuppression, including withdrawal from polydrug immunosuppressive regimes if possible, modifying psychoemotional problems, and recognizing the differences in disease burden and treatment between males and females.

Availability of data and materials
All data generated or analyzed during this study are included in this article.

Competing interests
The authors declare that they have no competing interests.

Funding
The authors declare that they did not receive funding for this research from any source.

Authors' contributions
CT designed the study and drafted the manuscript. CT and ML screened and evaluated the Quality of studies and extracted data. CT and ML interpret the results. KR reviewed the manuscript. All authors read and approved the manuscript.     Health-related Quality of Life: Kasai hepatoportoenterostomy vs. healthy controls