Training in pediatric hepatology across Europe: a survey of the National Societies Network (2016-2019) of the European Society for Paediatric Gastroenterology, Hepatology and Nutrition

Introduction

Hepatobiliary diseases have a significant impact on pediatric patients. Therefore, trainees in pediatric gastroenterology, hepatology and nutrition (PGHN) should be familiar with their pathophysiological mechanisms and be fully engaged in their diagnosis and management [1,2] so that future pediatric gastroenterologists become independent, qualified and competent specialists in these diseases [3-5].

Trainees should be exposed to a sufficient number of patients and a sufficient number of procedures and be able to develop and implement a plan for the diagnosis, management and treatment of patients with such diseases using a multidisciplinary approach [3-5]. The progressive maturation of pediatric hepatology (PH) has led to the determination of a hepatology-focused curriculum and educational content for PGHN training programs, as well as training requirements for those who wish to pursue further training in this specific area. The aim of our survey was to assess the PH-related infrastructure, staff and patient volumes of the various PGHN training centers across Europe.

Materials and methods

This study constitutes a subgroup analysis of a major project to examine the infrastructure, staffing, patient and procedure volume and organization of PGHN training in PGHN training centers across Europe. Questions related to PH and liver transplantation (LT) were included in the standardized questionnaires (Supplementary Appendix 1) created by the members (AP, AB and CRC) of the Executive Committee of the National Societies Group 2015-2017 of the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) and sent to the presidents/representatives of the ESPGHAN National Societies, who distributed them to the heads of PGHN training centers in their countries and collected the responses. In countries where no official National Society representative participated in the survey (Italy, Portugal and Switzerland), a volunteer among ESPGHAN members from that country was asked to distribute and collect the questionnaires (Italy and Switzerland), or individual centers provided their data after direct communication with the study coordinators (Portugal). The project was approved by the ESPGHAN Council in 2016 and supported by ESPGHAN. It was implemented from 1 June 2016 to 31 December 2019. The manuscript was sent for review and constructive comments to the former ESPGHAN President, former Chair of the ESPGHAN Hepatology Committee and Public Affairs Committee, Professor of Pediatric Hepatology, DK.

Results

A total of 100 training centers from 17 European countries, Turkey and Israel (Supplementary Appendix 2), participated in the survey. Seventy of these were national referral centers—43 in PGHN (Group 1), 14 in PH and/or LT (Group 2), and 13 in pediatric gastroenterology (PG; Group 3)—and 30 were regional referral centers. Thirty training centers were in the capital cities of the 19 countries: 29 were national referral centers (20 in PGHN, 4 in PH and/or LT and 5 in PG), while one was a regional referral center.

The age limit for patient care by most of these centers is 18 years. However, the center in Innsbruck cares for children up to the age of 16 years, while 2 centers (in Rome and Zagreb) care for patients up to the age of 21 years. Dedicated clinics for PH and/or LT were present in 68/91 (75%) centers: 49/64 (77%) of the national referral centers that answered the relevant question and 19/27 (70%) of the regional centers. Among the groups of national referral centers, dedicated clinics for PH and/or LT were present in 25/37 (68%) centers in Group 1, in 14/14 (100%) in Group 2 and in 10/13 (77%) in Group 3. However, dedicated clinics for LT were available in only 9/84 (11%) centers, 8 of which were national referral centers: 0/31 (0%) in Group 1, 7/12 (58.3%) in Group 2 and 1/12 (8%) in Group 3 (jointly with a national PH/LT referral center). Furthermore, dedicated clinics for PH and/or LT were present in 20/28 (71%) of the training centers established in European capitals that answered the relevant question.

Dedicated beds for pediatric gastroenterology and hepatology inpatients were available in 37/69 (54%) of the national referral centers that answered the relevant question in the total cohort, and in 10/14 (71%) of the national referral centers for PH. Full-time specialists for PH care were available in 31/45 (69%) of centers and part-time specialists in 11/36 (31%) of training centers (Table 1). Liver biopsies (LB) were performed in 85 training centers (Table 2). The performers of LBs in different training centers were the following: PGHN specialists in 35% of training centers, interventional radiologists in 26%, surgeons in 4%, and a combination of them in 35% (Table 1). Transient elastography was available in 60/92 (65%) training centers (Table 1).

Table 1 Availability of specialists for PH care, availability of transient elastography and liver biopsy operators in PGHN training centers

Table 2 Annual numbers of liver biopsies performed at the PGHN training centers across Europe

The annual number of procedures in the training centers of the total cohort is shown in Table 1. Annual numbers of LBs >100 were reported by 6 national referral centers for PH and LT (King’s College Hospital, London, UK; Birmingham Children’s Hospital, UK; Saint-Luc University Hospital, Catholic University de Louvain, Brussels, Belgium; Papa Giovanni XXIII Hospital, Bergamo, Italy; Woman Mother Child Hospital, Lyon, France; Schneider Children’s Medical Center of Israel, Petah Tiqva, Israel). Ten LBs or fewer per year were reported by 61% of regional centers compared to 30% of national referral centers (P=0.022; Table 2). A multinomial logistic regression analysis showed that national PH referral centers were more likely to perform large numbers of LBs (51-100 or >100 liver biopsies per year) than any other centers (P<0.001; Table 2). No significant differences in procedure volume were found between training centers located in European capitals and other cities (P=0.377; Table 2). The annual procedural volume in the training centers of the total cohort, according to the number of outpatients, is shown in Table 3. Thirty-two of 85 (38%) training centers reported a low number (≤10) of LBs per year (Table 3). Low (≤10) annual numbers of LBs were reported by 60% of centers with less than 1500 outpatients per year, but also by 19% of those centers with the highest (>5000) annual numbers of outpatients (Table 3).

Table 3 Annual procedural volume of liver biopsies performed by training centers according to the annual numbers of outpatients

Dividing the total number of LBs performed at the training center per year by the number of trainees, the median (interquartile range, range) annual number of LB per trainee in the cohort of centers performing LBs, is 10 (5-28, 1-125). Significant differences were found between national referral centers and regional centers (P=0.06) and between different categories of training centers (P=0.005), with trainees in PH national referral centers having the greatest exposure to LBs. No significant differences were found between centers located in capital cities compared to other cities (P=0.659; Table 4).

Table 4 Median (IQR; range) number of trainees in training centers performing liver biopsies and median (IQR; range) annual number of liver biopsies per trainee

We were interested in finding out if there are differences between countries in the frequency with which LBs are performed in the population aged 0-19 years. The number of LBs per 100,000 population aged 0-19 years performed annually by all training centers in countries where training centers are fully represented in the survey is shown in Fig. 1. The population aged 0-19 years is from the international database of the United States Census Bureau [6]. Although countries with a larger number of children aged 0-19 years reported a higher number of LBs (r_s=0.714; P=0.047), the number of LBs per 100,000 population aged 0-19 years did not correlate with the number of children per country (r_s=-0.071; P=0.867). We were also interested in knowing the procedural volumes of the major centers in each country among those participating in the survey, to identify potential training centers in each country and at a European level. The annual number of LBs performed by the PGHN centers with the largest procedural volumes among the participating centers in each country is shown in Fig. 2. Twelve of the 18 centers were pediatric liver transplant centers. The total number of PGHN specialists in the top 6 centers (i.e., those with the highest number of LBs) was 52 (37 fulltime), while it was 22 (13 fulltime) in the 6 centers with the lowest number of LBs and 38 (29 fulltime) in the 6 centers with a medium number of LBs. The centers with the highest numbers of procedures had a larger outpatient volume, as 5/6 (83%) of the top 6 centers reported >3000 outpatients per year (3/6, >5000), which was not the case for the centers with a medium or the lowest number of LBs, as only 2/6 (33%) of the centers with a medium number and 2/6 (33%) of centers with the lowest number of LBs have reported >3000 outpatients per year.

Figure 1 Annual number of liver biopsies per 100,000 population aged 0-19 years [7] in countries with full representation of the pediatric gastroenterology, hepatology and nutrition training centers

Figure 2 Annual number of liver biopsies performed by the pediatric gastroenterology, hepatology and nutrition center with the largest procedural volume amongst participating centers in each country. The centers shown in the figure are the following: London (King’s College Hospital); Brussels (Saint-Luc University Hospital); Bergamo (Papa Giovanni XXIII Hospital); Lyon (Woman Mother Child Hospital); Petah Tiqva (Schneider Children’s Medical Center of Israel); Malatya (Inönü University Faculty of Medicine); Groningen (University Medical Center Groningen); Tübingen (University Children´s Hospital, Tubingen); Zagreb (University Hospital Center Zagreb); Barcelona (Vall d’Hebron University Hospital); Athens (Agia Sofia Children’s hospital); Innsbruck (Innsbruck Medical University); Sofia (Ivan Mitev Specialized Hospital for Active Treatment of Children’s Diseases); Olomouc (Olomouc University Hospital), Ljubljana (Ljubljana University Medical Center); Porto (São João University Hospital); Vilnius (Vilnius University Hospital Santaros Klinikos); Pécs (Pécs University Medical School). (*) Centers with pediatric liver transplant programs

Discussion

The survey shows that the infrastructure, staff, training program, patients and procedural volumes related to PH vary widely across PGHN training centers in Europe. Dedicated clinics for PH were present in 3/4 of the training centers, while dedicated clinics for LT were present in only one tenth. Full-time specialists to care for PH patients were reported by 2/3 of the training centers and part-time specialists by almost 1/3. Dedicated beds for pediatric gastroenterology and hepatology inpatients were reported by half of the training centers. The number of LBs varied considerably among the different training centers, as 38% of them performed 10 LBs or fewer per year, while 6 national reference centers for PH and LT performed more than 100 LBs per year. Trainees on fellowship at the latter centers were exposed to double to quadruple the number of LBs. The latter training centers also had the largest outpatient volumes, with 2/5 of them exceeding 5000 outpatients per year. In about 1/3 of the training centers LBs were performed only by PGHN specialists, in 1/4 only by interventional radiologists, while in about 1/3 LBs involved a combination of different specialists (PGHN, interventional radiologists or pediatric surgeons).

Williams et al highlighted the increasing burden of liver diseases and specific problems in hepatology training in the UK [7]. In 2013, the Hepatology Advanced Training Program was launched to provide clinicians with the opportunity to obtain a subspecialty certificate in hepatology, with only 20 posts available each year [8]. The PGHN training curriculum proposed by both ESPGHAN [3,4] and the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) [5] provides a broad knowledge of hepatobiliary diseases. In addition, NASPGHAN envisages a special PH fellowship program to improve expertise in pediatric PH and LT [5].

There is reportedly a variety of training opportunities in hepatology in the UK, even for adult specialists [9], and the Lancet Commission proposed the establishment of dedicated liver centers in district general hospitals to improve training in this important area [7]. Early identification of training fellows interested in hepatology can provide future specialists with the necessary knowledge, experience and skills to provide the best possible care to patients with hepatobiliary diseases [10,11].

Our survey showed that the age range of the children cared for by the different centers varied: it ranged from 16 years in Innsbruck to 21 years in Rome and Zagreb PH and LT centers. However, it is common knowledge that patients with chronic diseases need medical care into adulthood. Health care transition should be understood as a targeted, planned transition of adolescents with chronic diseases from a child-centered to an adult-centered healthcare system [12], taking into account that this particular patient age group is at high risk for developmental and psychosocial problems. A successful transition program can limit some of the psychosocial effects of chronic illness by improving self-efficacy and autonomy, as well as treatment adherence [13-15].

Furthermore, differences in the number of procedures per 100,000 population aged 0-19 years were evident between countries, with Croatia reporting the highest number of LBs, while the Czech Republic and Bulgaria report the lowest, although the reasons for these differences are not known. The ESPGHAN Position Paper on LB [16] describes the clinical indications for performing LBs to diagnose liver disease. However, implementation of the guidelines in clinical practice is not always possible. Strategies are needed to improve the adoption, implementation and sustainability of guidelines. At the same time, the most important criterion by which aspects of medical care should be measured is disease outcome, which was outside the scope of this survey, with improvement in outcomes following a particular intervention being the real benefit.

Regarding training, there are currently no universally accepted recommendations for the training and experience required to qualify a person to perform LBs. The American Association for the Study of Liver Disease recommended a minimum of 40 LBs under supervision [17], while NASPGHAN recommended a minimum of 20 LBs performed independently, of which half should be in infants and children aged <3 years [18]. A similar number of LBs was also recommended by the American Gastroenterology Association in “Training the Gastroenterologist for the Future” [19]. A minimum number of 50 procedures was recommended by the British and Irish Committee on Higher Medical Training [20]. The effect of the experience of the person performing LBs on the adequacy of the obtained liver tissue and on complication rate is poorly defined [21]. An audit carried out by the British Society of Gastroenterology in 1991 reported a slightly higher incidence of complications if the operator had performed <20 LBs compared with >100 (3.2% vs. 1.1%) [22]. Regarding the ability to obtain adequate tissue, one report claimed that physician assistants who observed 10 LBs, practiced on inanimate objects, and then performed 30 LBs under supervision were able to obtain adequate tissue in 99.1% of attempts after a single pass and in 99.8% after 2 passes, with a mean tissue length of 3.2 cm [23]. Another report claimed that, in a series of 101 LBs performed independently by a hepatologist with no prior experience in ultrasound diagnostics, after observing 64 LBs performed by a certified ultrasound technician, no differences in complication rate or in adequacy of the obtained liver tissue were found [24]. The authors of the Position Paper on LB of the ESPGHAN Hepatology Committee suggested that a minimum of 20 LBs should be performed under supervision [16]. Based on the above, it is reasonable for low procedure volume training centers to work with larger centers within a clinical network. Clinical networks provide opportunities for multidisciplinary team meetings between doctors, surgeons and other professionals, audits and other quality improvement activities, data management, shared information technology protocols and research [25]. They provide access to specialized care, regardless of where patients live, and ensure that access to specialist advice is available by telephone 24/7. Specialist hepatology services and training in hepatology are provided in the UK by 3 leading specialist centers in Leeds, Birmingham and London [25]. Access to these services is by direct referral from primary or secondary services or by referral from the specialist gastroenterology service. Shared care is provided through network arrangements, including outreach clinics organized in close collaboration with local and specialist pediatric consultants, offering joint consultations and agreed local investigations and treatment.

This study had a number of limitations, such as the cross-sectional study design and the variability of the response rate, as there were countries where PGHN training centers were fully represented (Austria, Bulgaria, Czech Republic, Croatia, Greece, Israel, Lithuania and Slovenia), while other countries such as Germany and Turkey had limited representation, as only a few centers participated in the survey. However, the greatest value of this survey is that this first collaborative work of the ESPGHAN National Societies provides the largest set of data on the infrastructure, staffing, number of procedures and training programs in PH across Europe, identifying discrepancies that can help ESPGHAN, together with the National Societies, to develop actions to fill the identified gaps to achieve excellence in training and medical care across Europe.

In conclusion, this survey showed clear differences and deficits in terms of PH infrastructure, staff, patient and procedure volumes, and thus training opportunities, between PGHN training centers across Europe. ESPGHAN, together with the National Societies, should take initiatives for a well-structured training program according to the ESPGHAN curriculum, supporting the rotation of trainees to referral centers with large patient and procedure volumes and their participation in training courses and schools, as well as for developing competence assessment and accreditation measures where these are lacking.