Thoracic and cardiovascular surgeries in Japan during 2017

Committee for Scientific Affairs, The Japanese Association for Thoracic Surgery · Hideyuki Shimizu · Morihito Okada · Akira Tangoku · Yuichiro Doki · Shunsuke Endo · Hirotsugu Fukuda · Yasutaka Hirata · Hisashi Iwata · Junjiro Kobayashi · Hiraku Kumamaru · Hiroaki Miyata · Noboru Motomura · Shoji Natsugoe · Soji Ozawa · Yoshikatsu Saiki · Aya Saito · Hisashi Saji · Yukio Sato · Tsuyoshi Taketani · Kazuo Tanemoto · Wataru Tatsuishi · Yasushi Toh · Hiroyuki Tsukihara · Masayuki Watanabe · Hiroyuki Yamamoto · Kohei Yokoi · Yutaka Okita

The Japanese Association for Thoracic Surgery has conducted annual surveys of thoracic surgery throughout Japan since 1986 to determine statistics pertaining to the number of procedures performed, based on surgical category. Herein, we summarize the results of the association's annual survey of thoracic surgery performed in 2017.
Adhering to the norm to date, thoracic surgery was classified into three categories: cardiovascular, general thoracic, and esophageal surgeries. Patient data were examined and analyzed for each group. Access to computerized data is available to all members of the association. We honor and value all members' continued professional support and contributions (Tables 1, 2).
Incidence of hospital mortality was included in the survey to determine nationwide status, which has contributed to Japanese surgeons' understanding of the present status of thoracic surgery in Japan, while helping to effect progress for improving operative results by enabling them to compare their work with that of others. In this way, the association has been able to gain a better understanding of present problems and future prospects, which is reflected in its activities and the education of its members.
Thirty-day mortality (so-called operative mortality) is defined as death within 30 days of surgery, regardless of the patient's geographic location, including after the patient is discharged from hospital. Hospital mortality is defined as death within any time interval following surgery if the patient has not been discharged from hospital.
Hospital-to-hospital transfer in the category of esophageal surgery is not considered a form of discharge; transfer to a nursing home or a rehabilitation unit is considered hospital discharge, unless the patient subsequently dies because of complications from surgery. Contrastingly, hospital-to-hospital transfer 30 days following surgery in the categories of cardiovascular and general thoracic surgery is considered discharge, as data related to the National Clinical Database (NCD 2017) were employed in this category, and hospital-to-hospital transfer 30 days following surgery is considered discharge according to the NCD.

Survey abstract
All data pertaining to cardiovascular and thoracic surgeries were obtained from the NCD, whereas data regarding esophageal surgery were collected from a survey questionnaire derived from the Japanese Association for Thoracic Surgery documentation. The reason for this was that NCD information regarding esophageal surgery does not include non-surgical cases (i.e., patients with adjuvant chemotherapy or radiation only).
Because of changes in data collection related to cardiovascular surgery (initially self-reported using questionnaire sheets in each participating institution up to 2014, then by downloading an automatic package from the Japanese Cardiovascular Surgery Database (JCVSD), a cardiovascular sub-section of the NCD), the response rate is not available and is, therefore, not indicated in the cardiovascular surgery category (Table 1). Additionally, the number of institutions (based on surgery count) was not calculated in the cardiovascular surgery category (Table 2). Morihito Okada and Akira Tangoku have equally contributed to this work.

Final report: 2017 (A) Cardiovascular surgery
We are extremely pleased with the cooperation of our colleagues (members) in terms of completing the cardiovascular surgery survey, thereby undoubtedly improving the quality of this annual report. We are truly grateful for the significant efforts made by all within each participating institution in completing the JCVSD/NCD. Figure 1 illustrates the development of cardiovascular surgery in Japan over the past 30 years. Aneurysm surgery includes only surgeries for thoracic and thoracoabdominal aortic aneurysms. Extra-anatomic bypass surgery for thoracic aneurysm and pacemaker implantation has been excluded from the survey since 2015. The number of assist device implantation surgeries is not included in the total number of surgical procedures but was nonetheless included in the survey.
A total of 70,078 cardiovascular surgeries including 56 heart transplants were performed in 2017, an increase of 3.3% compared with that in the 2016 survey results (n= 67,867). The number of cardiovascular surgeries is continuously increasing, despite an apparent decrease in 2015, likely due to major changes in data collection and aggregation approaches.
When compared with data for 2016 [1] and 2007 [2], the number of surgeries in 2017 for congenital heart disease increased by 7.1% (9368 vs. 8744) and 0.2%, respectively; procedures for valvular heart disease increased by 0.2% (23,312 vs. 23,254) and 53.2%, respectively; surgery for thoracic aortic aneurysm increased by 8.7% (20,746 vs. 19,078) and 114.6%, respectively; ischemic heart procedures decreased by 6.6% (13,898 vs. 14,874) and 23.6%,                              Tables 3, 4, 5, 6, 7, and 8. In 2017, among 9368 procedures for congenital heart disease, 7072 open-heart surgeries were performed with an overall hospital mortality of 1.9%. Compared with data for 2007, the number of surgeries for neonates and infants has not changed significantly; however, hospital mortality decreased significantly from 13.7 to 7.7% for neonates and from 3.9 to 2.0% for infants. In 2017, atrial septal defect was once again the most common disease (1418 cases) for the first time in 3 years. This was primarily due to a doubling in the number of surgeries for patients above the age of 18 years (from 372 to 761 cases). Ventricular septal defect (perimemb/muscular) was the second most common health issue (1092 cases) and had previously been the most common disease in 2015 and 2016. In the past 10 years, hospital mortality for complex congenital heart disease was as follows (2007 [2], 2012 [3], and 2017): complete atrioventricular septal defect (4.3%, 3.2%, and 2.7%, respectively), tetralogy of Fallot (1.4%, 1.1%, and 0.9%, respectively), transposition of the great arteries with intact septum (2.7%, 2.6%, and 4.5%, respectively) and with ventricular septal defect (5.6%, 3.2%, and 1.5%, respectively), and single ventricle (5.4%, 5.5%, and 2.2%, respectively) and hypoplastic left heart syndrome (20.1%, 10.2%, and 8.8%, respectively). Right heart bypass surgery is now commonly performed (350 bidirectional Glenn procedures excluding 28 Damus-Kaye-Stansel procedures and 429 Fontan type procedures including total cavopulmonary connection) at an acceptable hospital mortality rate (1.1% and 1.4%). The Norwood type I procedure was performed in 108 cases, with a relatively low hospital mortality rate of 12.0% (Table 9).
The total number of procedures for valvular heart disease is increasing. The number of isolated aortic valve replacement/repair with/without coronary artery bypass grafting (CABG) (n=10,690) increased by 12.9% from the previous year (n=9472) and by 10.3% from 5 years ago (n =9688), despite the rapid spread of transcatheter aortic valve replacement (n=4632 in 2017). On the other hand, the number of isolated mitral valve replacement/repair with/without CABG (n=4687) remained stable, with a 2.4% increase from the previous year (n=4576) and a 1.5% increase from 5 years ago (n=4617). Aortic and mitral valve replacements with bioprosthesis procedures were performed in 10,871 cases and 2747 cases, respectively. The ratio for employing bioprosthesis increased dramatically from the 30% level in the early 2000s [4,5] and was 83.9% and 68.7% in the aortic and mitral positions, respectively, in 2017. Additionally, CABG was performed as a concomitant procedure in 17.8% for all valvular procedures (15.7% in 2007 [2] and 18.2% in 2012 [3]). Repair of the valve was a popular procedure in mitral and tricuspid     3.1% and 5.3% for the aortic and mitral positions, respectively, whereas for mitral valve repair, this was only 1.2%. The hospital mortality for redo valve surgery was 3.4% and 5.7% in the aortic and mitral positions, respectively. Finally, overall hospital mortality did not show dramatic improvement during the past 10 years (3.8% in 2007 [2], 3.2% in 2012 [3], and 3.6% in 2017). Isolated CABG was performed in 12,629 cases, representing only 73.0% of the number performed 10 years ago (n=17,295) [2]. Among these, off-pump CABG was intended in 7431 cases (58.8%) at a success rate of 97.1%. The percentage of intended off-pump CABG in 2017 was less than 60% for the first time in 13 years, since 2004 [4]. Hospital mortality associated with primary elective CABG procedures in 10,283 cases was 1.6%, unchanged from 2003 (1.5%) [5]. Hospital mortality for primary emergency CABG in 2211 cases was still as high as 7.1%. The result of conversion from off-pump CABG was 2.9%, and hospital mortality in this context was 6.9%. Hospital mortality was higher in patients with end-stage renal failure on dialysis, regardless of surgical procedures (on-pump arrest, on-pump beating, and off-pump). In this report, the number of concomitant CABGs alongside other major procedures was not included in the category of ischemic heart disease but in other categories such as valvular heart disease and thoracic aortic aneurysm. Accordingly, the overall number of CABGs, including concomitant CABG with other major procedures, remained more than 18,000 cases per year (18,327 cases) in 2017.
Measures for arrhythmia were performed primarily as concomitant procedures in 5066 cases, with hospital mortality of 2.8%. Implantation of pacemaker and implantable cardioverter-defibrillator was not included in this category.
In 2017, 20,746 procedures were performed for thoracic and thoracoabdominal aortic diseases; 10,086 and 10,660 were for aortic dissection and non-dissection, respectively. The number of surgeries for aortic dissection increased by 6.9% this year, compared with that in the preceding year (n =9441). The hospital mortality of procedures for 5995 Stanford type A acute aortic dissections remained as high as 11.2%. The number of procedures for non-dissected aneurysm increased by 10.6%, with overall hospital mortality of 5.5%, and 4.3% and 22.5% for unruptured and ruptured aneurysms, respectively. The rate of thoracic endovascular aortic repair (TEVAR) among all operative

(B) General thoracic surgery
The 2017 survey of general thoracic surgery comprised 678 surgical units, with the bulk of data submitted via a webbased collection system of the NCD [1]. In total, 85,307 procedures were reported by general thoracic surgery departments in 2017, twice the number of operations in 2000 and 12,560 more than in 2012 (Fig. 2). In 2017, 44,140 procedures for primary lung cancer were performed, a number that has increased annually. The 2017 value was 2.4 times that of 2000. Procedures for lung cancer represented 52% of all general thoracic surgery instances.
The number of video-assisted thoracoscopic surgery (VATS) instances, defined by a surgical procedure utilizing a skin incision longer than 8 cm and/or a minithoracotomy (hybrid) approach, has been noted since the 2015 annual report. The number of VATS procedures for benign pulmonary tumors and primary lung cancer, and the total number of VATS procedures in 2016 are shown in Tables 10, 11, 13, 16, 17, 18, 19, 20, 21, 22, 23, 25, 26, and 27, respectively.
In 2017, 2197 procedures were conducted for benign pulmonary tumors (Table 10), a similar number to that recorded in 2016. Hamartoma was the most frequent diagnosis in procedures for benign pulmonary tumors. VATS was performed for 2071 patients (94%).
Additional information on primary malignant pulmonary tumors is shown in Tables 11 and 12. With regard to lung cancer sub-type, adenocarcinoma was by far the most frequent diagnosis (71% of all lung cancer procedures), followed by squamous cell carcinoma (18%). Sublobar resection was performed in 11,784 lung cancer cases (27% of all cases) and lobectomy in 31,584 cases (72% of There were 46 procedures for malignant tracheal tumor in 2017; however, 21 patients were treated with sleeve resection and reconstruction (Table 14).
Pleural tumors numbered 698 in 2017 (Table 15). Diffuse malignant pleural mesothelioma was the most frequent histologic diagnosis. Total pleurectomy was performed in 104 cases and extrapleural pneumonectomy in 65 cases. The 30-day mortality rate was 0% following total pleurectomy and 2% after extrapleural pneumonectomy, both representing better outcomes than before.
Thymectomy for myasthenia gravis was performed in 508 patients (Table 18); 319 procedures were associated with thymoma, and the remaining was not associated with thymoma.
Procedures for non-neoplastic disease were performed for 22,634 patients. There were 2423 cases of lung resection for inflammatory lung diseases (Table 19), 20% of which were associated with atypical mycobacterium infections and 15% with fungal infections. Procedures for inflammatory nodules were performed in cases where lung cancer was suspected prior to surgery (in 918 cases, 38%).
The 2962 procedures for empyema (Table 20) comprised 2226 cases (75%) of acute empyema and 736 cases of chronic empyema. Bronchopleural fistula was reported in 453 patients with acute empyema and in 355 patients with chronic empyema. The hospital mortality rate was 15% among patients with acute empyema with fistula.
In 2017, 101 operations were performed for descending necrotizing mediastinitis (Table 21). The hospital mortality rate was 7%. Furthermore, 400 procedures were conducted for bullous diseases (Table 22); lung volume reduction surgery was performed in only 28 patients.
A total of 14,499 procedures were performed for spontaneous pneumothorax (Table 23). The 11,113 procedures for primary pneumothorax comprised 2838 patients (26%) who underwent bullectomy only and 7488 patients (67%) who underwent an additional procedure. There were 3386     procedures for secondary pneumothorax, where COPD was by far the most prevalent associated disease (69%). The hospital mortality rate for secondary pneumothorax associated with COPD was 2.6%. The 2017 survey reported 193 procedures for chest wall deformity (Table 24). However, this may have been underestimated, because the Nuss procedure for pectus excavatum was more likely to have been performed in pediatric surgery centers not associated with the Japanese Association for Thoracic Surgery.
Diaphragmatic hernia was treated surgically in 28 patients (Table 25). This figure may also have been underestimated, as procedures may have been classified as gastrointestinal surgery.
The survey reported 443 procedures for chest trauma excluding iatrogenic injuries (Table 26). In this context, hospital mortality rate was 5%. Table 27 shows procedures for other diseases, including 89 cases of arteriovenous malformation and 92 cases of pulmonary sequestration.
A total of 68 lung transplantations were performed in 2017 (Table 28): 57 patients received lung transplants from brain-dead donors, and 11 patients received transplants from living, related donors.
The number of VATS procedures has increased annually to reach 68,458 (80% of all general thoracic surgeries) in 2017 (Table 29).
The details of tracheobronchoplasty, pediatric surgery, and combined resection of neighboring organs are shown in Tables 30, 31, 32, and 33.                     (Tables 36 and 37). Annual trends among registered inpatients with esophageal diseases have not changed for the past last decades (Fig. 3). Among benign esophageal diseases (Table 35), hiatal hernia, achalasia, esophageal varices, and esophagitis (including reflux esophagitis) were the most common conditions in Japan. On the other hand, benign esophageal tumors, spontaneous rupture of the esophagus, and congenital esophageal atresia were common diseases that were surgically treated in addition to the aforementioned diseases. Open surgery was performed in 1009 (57.7%) patients with a benign esophageal disease, with 30-day mortality in 3 (0.3%) patients, whereas thoracoscopic and/ or laparoscopic surgery was performed for 605 (37.5%) patients, with no instances of 30-day mortality. The difference in these death rates between open and scopic surgery appears to be related to conditions requiring open surgery.
The majority of malignant diseases were carcinomas (Table 36). Among esophageal carcinomas, the incidence of squamous cell carcinoma was 88.8%, whereas that of adenocarcinomas, including Barrett's cancer, was 4.3%. The resection rate for patients with a squamous cell carcinoma was 79.9%, whereas that for patients with adenocarcinoma was 93.8%.
On the basis of location, cancer in the thoracic esophagus was the most common (Table 37). Among 4303 patients (40.9% of total esophageal malignancies) with superficial esophageal cancers within mucosal and submucosal layers, 6319 (60.1%) patients underwent esophagectomy, whereas 2170 (20.6%) patients underwent EMR or ESD. The 30-day mortality rate and hospital mortality rate after esophagectomy for patients with a superficial cancer were 0.5% and 1.0%, respectively.
Multiple primary cancers were observed in 2077 (19.8%) of all 10,514 patients with esophageal cancer. Synchronous cancer was found in 1008 (9.3%) patients, whereas metachronous cancer was observed in 1063 (10.1%) patients. The stomach and head and neck were common sites for both synchronous and metachronous malignancies (Table 37).
Among esophagectomy procedures, transthoracic esophagectomy via right thoracotomy was most commonly adopted for patients with a superficial cancer, as well as for those with advanced cancer (Table 38). Transhiatal esophagectomy, which is commonly performed in Western countries, was adopted in only 9.1% of patients with a superficial cancer or advanced cancer who underwent esophagectomy in Japan. Thoracoscopic and/or laparoscopic esophagectomy was adopted for 1434 patients (74.0%) with a superficial cancer and for 2422 patients (55.3%) with an advanced cancer. The number of cases of thoracoscopic and/or laparoscopic surgery for superficial or advanced cancer has been increasing for a number of years (Fig. 4).
Combined resection of the neighboring organs during the resection of an esophageal cancer was performed in 296 patients (Tables 38 and 39). Resection of the aorta, together with esophagectomy, was performed in eight cases. Tracheal and/or bronchial resection combined with esophagectomy was performed for 20 patients, with both 30-day mortality rate and hospital mortality rate at 0%. Lung resection combined with esophagectomy was performed for 60 patients, with both 30-day mortality rate and hospital mortality rate at 0%. Salvage surgery following definitive (chemo)radiotherapy was performed for 230 patients, with 30-day mortality rate at 1.7% and hospital mortality rate at 2.6% (Table 38).         We aim to continue our efforts to gather all-encompassing survey data via more active collaboration with the Japan Esophageal Society and other related institutions.
Acknowledgements On behalf of the Japanese Association for Thoracic Surgery, the authors thank the Heads of the Affiliate and Satellite Institutes of Thoracic Surgery for their cooperation, and the Councilors of the Japan Esophageal Society.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.