National Clinical Database (NCD) shows the trend for centralization of major surgery: Should it depend on hospital or surgeon volume ?

As a previous paper reported,1 Japan's National Clinical Database (NCD) is a large‐scale nationwide web‐based data entry system for surgical procedures that was established in 2010, and data entry began in 2011 and has grown considerably in annual volume. One hundred and fifteen surgical procedures in gastroenterological fields are registered. Among them, the detailed information including preoperative morbidity, postoperative complications, and 30‐ and 90‐day mortality is required in the eight major gastroenterological surgeries: esophagectomy, distal gastrectomy, total gastrectomy, right hemicolectomy, low anterior resection, hepatectomy, pancreaticoduodenectomy, and surgery to treat acute diffuse peritonitis. Table 1 shows year‐by‐year trends in those eight procedures from 2011 to 2017. The registrations from the participating hospitals to the NCD office via internet are actually owed by the data managers’ voluntary works of each site. However, the estimated coverages of NCD were reported to be 90%‐95% by comparison with regional government report data and medical charts,2 and the audit works verified the NCD's data and found high accuracy of data entry.3 These show the quite high reliability of NCD as big data. Hasegawa et al. reported, in this issue of the Annals of Gastroenterological Surgery, 4 that in total, 3 818 414 cases of gastroenterological surgery have been registered from 2011 to 2017 and the postoperative mortality rate has remained low in 2017, although the rates of preoperative comorbidities and postoperative complications have been increasing. These results indicate the advancement of perioperative management and the high quality of gastroenterological surgery in Japan. Furthermore, in this issue of the Annals of Gastroenterological Surgery, 5 Marubashi et al. reported the collaborative clinical projects between Japan and the USA with the goal of achieving further improvement of surgical quality in both countries by using NCD. These papers show the potential of contribution of NCD.

Table 1.

Frequency of eight major gastroenterological surgery procedures (2011‐2017)

Procedure	2011	2012	2013	2014	2015	2016	2017
Esophagectomy	4914	5947	5694	6092	6058	6041	6100
Distal gastrectomy	32 250	36 689	39 094	37 719	37 083	36 197	34 682
Total gastrectomy	18 048	20 462	18 775	17 963	17 387	16 191	14 840
Right hemicolectomy	17 885	21 022	21 816	22 444	22 851	22 829	22 543
Low anterior resection	16 981	20 306	21 097	21 854	22 496	21 387	20 879
Hepatectomy	7439	8242	7937	7663	7439	7610	7698
Pancreaticoduodenectomy	8306	9331	10 069	10 395	10 577	11 028	11 580
Acute diffuse peritonitis surgery based on Hasegawa's paper in this issue	7751	9179	10 452	12 085	13 030	13 981	14 423

Table 2 shows the year‐by‐year numbers of institutions performing the eight major procedures from 2011 to 2017. NCD data clearly show the trends for centralization of esophagectomy. Worldwide, centralization has been a common strategy.6 However, the importance of hospital volume versus surgeon volume in obtaining superior outcomes has been debated.7 Also, there has been no consensus on the minimal requirement for the number of procedures per hospital or per surgeon.6 Some recent papers showed the beneficial effect of the hospital volume in major surgeries, esophagectomy,8 pancreaticoduodenectomy,9 resection of cholangiocarcinoma,10 while the surgeon volume was reported to be attributable largely in liver resection.11 Several papers focusing on this issue by using NCD have been already published. One of the advantages of NCD is to evaluate the participation of board‐certificated surgeons. Konno et al. demonstrated that the board‐certificated surgeons contribute to favorable outcomes of gastroenterological surgery in Japan and the number of board‐certificated surgeons per hospital is a surrogate marker of operative mortality,12 while some papers based on NCD have recently reported the importance of the hospital volume to reduce mortality in esophagectomy,13 distal gastrectomy,14 hepatectomy,15 and pancreaticoduodenectomy.16 Nimptsch et al. described that the volume effect seems to be determined by the ability to rescue patients who experience complications.17 The integrated surgical team is undoubtedly essential to improve outcomes. The aim of centralization is to offer its benefit to patients. But, the evaluation of the integrated surgical team is too hard; how do we analyze the quality of the surgeon and hospital field effect? In the near future, NCD is expected to answer this quite important issue.

Table 2.

Number of institutions performing the procedure (2011‐2017)

Procedure	2011	2012	2013	2014	2015	2016	2017
Esophagectomy	683	741	731	723	695	675	656
Distal gastrectomy	1713	1907	1921	1899	1922	1879	1877
Total gastrectomy	1563	1750	1717	1717	1734	1699	1668
Right hemicolectomy	1644	1839	1848	1841	1885	1888	1855
Low anterior resection	1588	1786	1775	1784	1810	1800	1770
Hepatectomy	963	1055	1053	1032	1040	1030	1017
Pancreaticoduodenectomy	1126	1184	1209	1207	1199	1181	1162
Acute diffuse peritonitis surgery	1223	1385	1385	1432	1473	1448	1449

DISCLOSURE

Conflicts of Interest: The author declares no conflicts of interest for this article.