Standardizing the Treatment of Esophagogastric Junction Tumors: Centralization, Registries and Surgical Training

Dear Editor

We have read with interest the article published by Osorio et al.¹ We congratulate the authors for their review, but we wanted to share our comments.

We agree that centralization of certain complex procedures could improve results. However, as the authors point out, when choosing hospitals, other structure and result requirements should be met in addition to volume.² We believe that the basic factors for centralization should be the 90-day morbidity and mortality results and 3- or 5-year survival results, as externally audited by impartial authorities with no conflicts of interests, such as the National Healthcare Administration. However, these results are not known for any service and, therefore, the basic principles of quality are overlooked.³ These hospitals should offer a structure that provides permanent access to interventional radiology and endoscopy teams, critical care units and, of course, trained surgeons. Only then will failure-to-rescue rates decrease. If we do not know what the actual results are, there is no ‘textbook outcome’.⁴

No one questions the need for multidisciplinary cancer committees. But these should be based on scientific evidence and demonstrated protocols, and not be a ‘committee of experts’. Practices with no evidence to support them should be integrated into randomized clinical trials. Without results and structure, volume is of no interest.

The auditing process should be permanent so that results improve or at least remain the same. You cannot talk about benchmarks in results —as frequently published— or ‘service benchmarking’ without impartial auditing.

Filling the registries with patient complications and communicating them does not imply that the results are objective.⁵ National or multiregional cancer registries are objective only if their reliability is validated.⁶ Therefore, it does not seem coherent that the audit to verify the veracity of the registry data is conducted by professionals who are connected to it. The registries should have 2 sources, the surgical service itself and the auditors, which should be compared and any discrepancies resolved. A postoperative complication is any negative event.⁷ All morbidities should be noted on specific forms in the medical record. Simultaneously in the registry, morbidity forms should be used, as well as the medical and nursing progress notes, particularly the latter. Surgical societies insist on publishing endless lists of complications that are not valid for comparison.⁸ In addition, these listings do not include minor complications, which also affect patient quality of life, clinical results and economic costs.^9,10 We believe that all complications should be compiled, classified according to the Clavien Dindo Classification⁷ and with calculated scores like the Comprehensive Complication Index,¹¹ in order to compare the results.^9,10 Obviously, important and specific results from each intervention should also be used, such as the percentage of fistulae, their location and reoperations.³ In addition, if we compare results, the complexity of the patient should be considered, using scales like the Charlson index.¹² Otherwise, it could lead us to reject patients with severe comorbidities who are at risk for worse results.

Long-term cancer outcomes are more complex to follow, but necessary. They are the only results that summarize the global and multidisciplinary care received by the patient throughout their process. In this case, it is possible that the infrastructure necessary for its analysis is somewhat more expensive.

We are surprised by the high number of minimally invasive gastrectomies (MIG) of the Spanish EURECCA group: 37%. We believe that the use of MIG in western countries is well above what randomized clinical trials (RCT) conducted by experienced groups currently allow: early distal gastric cancer (EDGC),^13,14 although in advanced gastric cancer (AGC) it has shown some benefit¹⁵ in very experienced groups. It should be borne in mind that, to participate in these trials, for instance, surgeons must have an experience of more than 50 MIG and 50 laparotomic procedures with D2 lymphadenectomy, the hospitals should perform at least 300 gastrectomies/year, and the qualification of surgeons is determined by videos of their surgeries.¹⁶ These figures are far from the caseload and experience in Spain and Europe.

RCT are underway in long-term EDGC¹⁷ and in short-term^18–20 and long-term^21,22 AGC. There are short-term results in EGC of total MIG,²³ but no long-term results.

Currently, MIG should be reserved for EDGC, or RCT. Total MIG should not be practiced outside the RCT setting. Patients admitted for MIG should be reserved for experienced surgeons at hospitals with high volumes.²⁴

Training should be led by those who have the best 90-day results and long-term oncological results, as demonstrated by audit.

Unfortunately, there is no real evidence from audited results of national or international services. Aside from that, everything else is conjecture. In a society that defends and requires transparency and quality, we cannot defend the accreditation of units and the centralization of complex processes without previously auditing short- and long-term results in a prospective, impartial manner. Only then will we be able to improve and determine benchmarking services. The cost-efficiency analysis should come afterwards. The cost of audits is insignificant compared to the health and economic benefits. The biggest problem will be to mitigate the fear of surgery services. The Health Administration should take the lead and audit these results, since, together with patients, they will be the true beneficiaries of this healthcare policy.

Funding

This manuscript has not been funded.