Centralization of Pediatric Surgical Oncological Care. Why and How

Abstract

In recent years, centralization of care has become popular when treating rare diseases. Little is known of the effect of centralization on the quality of care in pediatric surgical oncology. It is difficult to compare results between “low”-volume and “high”-volume centers, because there is no useful method to risk adjust surgical outcomes due to the broad variation in surgical complexity even among patients with the same tumor histology. There is even controversy about the terms low and high volume. Although common sense would dictate that surgeons become better when they do a procedure more frequently, there is little hard evidence to support this hypothesis.

Most of the evidence in support of centralization comes from the adult literature where it is common practice in highly complex cases with a relatively high mortality. Randomized controlled trials are lacking and it is very unlikely that these will ever happen [1-5].

In non-oncological pediatric surgery also, there is a trend toward centralization; however, the evidence available is mostly comparing pre and post centralization [6, 7]. Also the evidence for continued surveillance of results over many years is important, as results may not remain at the same level. Changes in surgical technique or clinical management, which might be imperceptible, may result in worse outcomes over time [8].

There is evidence of dramatic success of the centralization of the biliary atresia operations in the Great Britain/England. Because of the low mortality of most pediatric surgical operations, other quality measures must be found. In biliary atresia, this was clearance of bile after 30 days, but was not confirmed by all groups [9-11]. Complications, pre- and post-operative, are less clear endpoints, and large groups are needed to detect a signal.

In this article, we are examining different forms of pediatric surgical oncology centralization efforts around the world, and in what ways they are used. In addition, we will examine differences in outcomes before and after the introduction of centralization.

There are different forms of concentration/centralization. From near total centralized care in one hospital to common tumor boards and central reviews of images and operation indications.

We will describe several forms of centralization in different countries, without being comprehensive. Afterwards, we will address the pros and cons of these different ways of centralizing pediatric surgical oncology procedures.

The Netherlands is a country of 18 million inhabitants with 180,000 births per annum and about 650–700 new cases of pediatric malignancy per year.

In 2008, the oncology professionals along with the parents’ organization embarked on an effort to centralize the pediatric oncology (0–18 years) patients in one place instead of the eight pediatric hospitals earlier. After much (political) discussion, it was finally established, and from 2015 solid tumors of chest and abdomen were operated on in one place. From 2018, the rest of all oncology patients were treated in one place, adjacent to the university pediatric hospital, but very importantly, as a different organization. As the doctor and patient organizations were part owners of the hospital, this ensured that the governance of the oncology center was independent from the adjacent university pediatric hospital. The collaboration is intense, and we cannot stress the importance of this part of the process. We use the operating theater and intensive care unit and work with many different pediatric and adult specialists.

This process is described more extensively in the excerpt of the Karl Storz Lecture [11].

In two National cohorts, we have compared the results in terms of complications in neuroblastoma surgery and positive surgical margins in partial nephrectomies in the 15 years before centralization to the 6 years after. Both groups improved significantly after concentration, suggesting its positive effect [12]. In neuroblastoma surgery, major vascular complications diminished (from 26% to 5%, p < 0.001), blood loss decreased (from 450 to 50 mL, p < 0.001), and operating time decreased (from 275 to 168 minutes, p < 0.001). All without an increase in local recurrence.

Although there is no published evidence that training and shortening of the learning curve in a centralized institutions setting is to be expected, but cannot be supported by data in pediatric surgical oncology.

Concentrating to one center, both the patients and staff, has certain limitations. For instance, patients requiring simple care such as antibiotics, blood counts, cultures, simple chemotherapy, and recovery after operations may have to travel longer distances in some cases. However, this is not a major issue in the Netherlands. Our country is small, and our center (Prinses Maxima Center is centrally located) can be reached by anyone within a 2-hour drive, which is often much less. However, in countries with larger distances and less accessibility, this can be a challenge.

In the United Kingdom, the National Health Service (NHS) is split into its devolved nations and encompasses NHS England, NHS Wales, NHS Scotland, and Health and Social Care Northern Ireland. As this population makes up some 67 million people, centralization of any service in the United Kingdom would need to involve all these stakeholders and is likely to require several centers around the country to provide services.

Prior to devolving the NHS, pediatric liver services were confined to three children's hospitals, with Leeds in the North of England, Birmingham in the Midlands, and Kings College London in the South East. This was driven with the knowledge that patients operated on for biliary atresia outside these centers had higher rates of failed Kasai portoenterostomy surgery and subsequent need for liver transplant [13].

This information has characterized the centralization debate for equally rare diseases such as bladder or cloacal exstrophy, which is now managed at two centers in the United Kingdom, Manchester and Great Ormond Street children's hospitals.

Other conditions are hotly debated with data for and against moving some services from the current 24 centers down to three or more, which would cover the whole of the United Kingdom. As usual, there is concern over deskilling and devaluing workforces in smaller centers that will lose these services, but this must be countered with the possibility of not just improving outcomes but driving excellence and innovation, which only comes with a high-volume caseload.

In 2021, the Getting It Right First Time report confirmed that 13 Operational Delivery Networks (ODNs) would be established with a broad aim of ensuring that “expertise and resources for children with the rarest surgical problems are concentrated into, indicatively, 10 centers” including solid tumor surgery.

Additionally, special interest groups (SIGs) and national advisory panels (NAPs) under the umbrella of the UK Childrens Cancer and Leukemia Group (CCLG) provide a core of experts who are increasingly receiving more complex patients such as those with Stage 5 Wilms tumor. This movement will continue to be driven by patient and parent groups, along with colleagues in oncology.

There is a widespread reluctance to discontinue certain medical services, and a recent attempt to reduce the number of trusts providing congenital cardiac surgery from 11 to seven was met with legal challenges from hospitals that would have been affected. This demonstrates how difficult it can be to make changes in this area. However, it is important to note that this opposition was not based on clinical outcomes, as pointed out in an article from the Lancet [13, 14].

People are willing to travel to specialized medical centers if they believe that the outcomes there are superior. In addition, these centers should have the ability to provide comprehensive care and support to both patients and their families. This should also include appropriate accommodation for parents and families who often have to travel long distances to reach these centers.

Germany is divided in so called “Bundeslander,” comparable to the different states in the United States. This makes a national policy difficult to arrange. Additionally, this has resulted in very few efforts to centralize pediatric surgical oncology care. It is not uncommon for a hospital to treat one or two pediatric oncology cases like neuroblastoma per year. This has led to a doctor-driven initiative that allows difficult neuroblastoma cases to be discussed in an expert panel that can advise on the type of operation and the best place for the operation. Results are expected soon.

There are various forms of centralization based on factors such as geography, healthcare organization (nationalized or otherwise), and involvement of patient organizations.

The main question here is whether concentration of care has a significant impact on improving the outcome of pediatric surgical procedures. The available data are mostly based on adult patients, with scant data from pediatric centers; however, these reports generally show improvements in complication rates, including mortality, local residual tumor, or local recurrence. As pediatric surgical care is a specialty with low patient numbers, proving a difference with centralization is challenging as randomization is not feasible, and historical cohorts have their limitations. However, more literature is becoming available that suggests that concentration of care does improve the outcome of pediatric surgeries. It is important to note that relatively few high-risk operations can benefit from this approach, for instance, routine Wilms tumors do not seem to have any advantage, but partial nephrectomies do [15, 16]. So, if the answer to the previous question is “yes,” then how to do this? How to tackle the resistance from doctors and administrators? How to reduce collateral damage that also is inevitable when centers lose these patient groups and knowledge/experience that comes with that?

First, get together as doctors and join up with parents’ organizations, insurance companies, and politicians. In this era, it is important to have and keep the initiative, or you will be overruled by politics and financial arguments.

In some remote areas, it may not be feasible to restrict emergency operations. However, losing knowledge to deal with such cases would not improve individual patient care. Distances require either patients or doctors to travel, these two models have their specific pros and cons. Multidisciplinary teams can meet online, study the images and clinical notes, and give advice regarding the operative strategy. This can mean an operation elsewhere, a surgeon traveling or an operation by the local staff. Within the setting of German pediatric oncology surgery, this seems to work out and leads to better stratification of risks. In Great Britain, the service is nationalized, and this has led to some hospitals doing some operations, while their cardiac colleagues are centralized in another hospital. This practice could result in inadequate cancer care units without centralized facilities such as pediatric cardiac services. Centralization of one specialty always has consequences for other specialties and this causes collateral damage. For example, without cardiac anesthesia there can be no full lymphoma service possible. Without plastic surgery service, a full sarcoma service is difficult, and so forth.

A frequently heard argument against centralization is that there should be competition within a country. When there is no competition, benchmarking is crucial to ensure quality of care in all centers. This means that competition is global, not national. Benchmarking (while ensuring anonymity) is essential and ensures that everyone can compare their results. Openness on what is best practice can give insight regarding true numbers and outcomes. This can lead to adapting protocols or techniques to obtain the best possible outcomes.

There can no longer be any doubt that centralization of (pediatric) surgical oncology care can benefit outcome. However, a benefit for a relatively small number of patients can also lead to collateral damage for a large group of non-oncological patients. Therefore, every region has its own dilemmas and local consequences that influence its policy. It is however dangerous to use this reasoning to not do anything. Centralizing difficult pediatric oncology and non-oncology care for the most complication-prone operations should be considered in any region, regardless of politics and professional stakeholder desires.

The authors declare no conflicts of interest.