eTable 1. Eligibility Criteria
eTable 2. Clinical Characteristics at Baseline
Data Sharing Statement
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Liu F, Huang C, Xu Z, et al. Morbidity and Mortality of Laparoscopic vs Open Total Gastrectomy for Clinical Stage I Gastric Cancer: The CLASS02 Multicenter Randomized Clinical Trial. JAMA Oncol. 2020;6(10):1590–1597. doi:10.1001/jamaoncol.2020.3152
Is the safety of laparoscopic total gastrectomy for the treatment of clinical stage I gastric cancer noninferior to that of open total gastrectomy?
In this multicenter, open-label, noninferiority, randomized clinical trial of 227 patients with clinical stage I gastric cancer, the overall morbidity and mortality rates were 19.1% in the laparoscopic total gastrectomy group and 20.2% in the open total gastrectomy group, which was not significantly different (rate difference, −1.1%). There were no significant differences in rates of postoperative complications between the 2 groups.
Experienced surgeons can perform laparoscopic total gastrectomy as safely as open total gastrectomy for clinical stage I gastric cancer.
The safety of laparoscopic total gastrectomy (LTG) for the treatment of gastric cancer remains uncertain given the lack of high-level clinical evidence.
To compare the safety of LTG for clinical stage I gastric cancer with that of conventional open total gastrectomy (OTG).
Design, Setting, and Participants
The Chinese Laparoscopic Gastrointestinal Surgery Study (CLASS) Group CLASS02 study was a prospective, multicenter, open-label, noninferiority, randomized clinical trial that compared the safety of LTG vs OTG with lymphadenectomy for patients with clinical stage I gastric cancer. From January 2017 to September 2018, a total of 227 patients were enrolled. Final follow-up was in October 2018.
Eligible patients were randomized to LTG (n = 113) or OTG (n = 114) by an interactive web response system.
Main Outcomes and Measures
The primary outcome was the morbidity and mortality within 30 days following surgeries between LTG and OTG with a noninferiority margin of 10%. The secondary outcomes were recovery courses and postoperative hospital stays.
A total of 214 patients were analyzed for morbidity and mortality (105 patients in the LTG group and 109 patients in the OTG group). The mean (SD) age was 59.8 (9.4) years in the LTG group and 59.4 (9.2) years in the OTG group, and most were male (LTG group, 75 of 105 [71.4%]; OTG group, 80 of 109 [73.4%]). The overall morbidity and mortality rates were not significantly different between the groups (rate difference, −1.1%; 95% CI, −11.8% to 9.6%). Intraoperative complications occurred in 3 patients (2.9%) in the LTG group and 4 patients (3.7%) in the OTG group (rate difference, −0.8%; 95% CI, −6.5% to 4.9%). In addition, there was no significant difference in the overall postoperative complication rate of 18.1% in the LTG group and 17.4% in the OTG group (rate difference, 0.7%; 95% CI, −9.6% to 11.0%). One patient in the LTG group died from intra-abdominal bleeding secondary to splenic artery hemorrhage. However, there was no significant difference in mortality between the LTG group and the OTG group (rate difference, 1.0%; 95% CI, −2.5% to 5.2%), and the distribution of complication severity was similar between the 2 groups.
Conclusions and Relevance
The results of the CLASS02 trial showed that the safety of LTG with lymphadenectomy by experienced surgeons for clinical stage I gastric cancer was comparable to that of OTG.
ClinicalTrials.gov Identifier: NCT03007550
Gastric cancer is a significant global health problem, especially in East Asia, including Japan, Korea, and China.1-3 Among patients diagnosed preoperatively with early or locally advanced distal gastric cancer, laparoscopic distal gastrectomy has been demonstrated to be safe and effective compared with open distal gastrectomy.4-8 With an increase in incidences of proximal gastric cancer over the last decades, total gastrectomy has been adopted by surgeons, and laparoscopic total gastrectomy (LTG) has become the preferred option.9,10 However, the safety and feasibility of LTG have yet to be proved owing to it being an operation that even well-trained surgeons find technically challenging.11,12
Recently, 2 prospective single-arm trials (the Japan Clinical Oncology Group JCOG1401 trial13 and the Korean Laparoendoscopic Gastrointestinal Surgery Study Group KLASS03 trial14) reported their results, which showed the safety of laparoscopy-assisted total gastrectomy (LATG) for patients with clinical stage I proximal gastric cancer. However, single-arm research does not always reflect the truth in the real world, and the results from prospective randomized clinical trial are more convincing, such as the minimally invasive radical hysterectomy research in early-stage cervical cancer and the laparoscopic pancreatoduodenectomy research for pancreatic tumors.15,16 At present, to our knowledge, no prospective randomized clinical trial has been completed to determine the safety of LTG.17,18
In January 2017, the Chinese Laparoscopic Gastrointestinal Surgery Study (CLASS) Group launched a prospective multicenter randomized clinical trial (CLASS02) with a noninferiority design to compare the safety of LTG for clinical stage I gastric cancer with the conventional open total gastrectomy (OTG).19 In the present study, we report the morbidity and mortality results of this trial, which demonstrate that LTG can be safely performed by experienced surgeons for clinical stage I gastric cancer.
The CLASS02 trial was a prospective, multicenter, open-label, noninferiority, randomized clinical trial that compared the safety of LTG vs OTG with D1+/D2 lymphadenectomy for patients with clinical stage I (T1N0M0, T1N1M0, T2N0M0) gastric cancer in the upper or middle third of the stomach. The primary outcome was the morbidity and mortality within 30 days following surgery. The secondary outcomes were the recovery courses and the postoperative hospital stays. Eligibility criteria are listed in eTable 1 in Supplement 1. The study protocol was approved by the institutional review boards of all participating hospitals and is available in Supplement 2. Signed informed consent was obtained from all patients. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.
In this trial, eligible patients were randomized by an interactive web response system using the minimization method, and the randomization process was controlled for the following factors: age (<60 or ≥60 years), sex (female or male), body mass index (<25 or ≥25; calculated as weight in kilograms divided by height in meters squared), and centers.
For the safety of enrolled patients and the quality of this study, and to minimize the differences in surgical skills among the surgeons, well-trained surgeons from gastric cancer centers in China were encouraged to participate in this clinical trial. Furthermore, we set selection criteria for the participants. First, they had to complete at least 50 cases of LTG and OTG with D2 lymphadenectomy respectively, which is the basic requirement for completing the learning curve reported by many studies.4,6,14 Second, they had to submit unedited videos to the CLASS Research Council. These videos were assessed through blind reviews before the surgeons could be accredited to participate in this trial. In brief, these 2 criteria were set to make sure that the participants had completed the learning curves and could perform LTG and OTG according to the treatment guidelines. Finally, 13 surgeons from 13 gastric cancer centers were permitted to participate in this trial as researchers.
To ensure the consistency of the surgical procedures, standard operation procedures of LTG and OTG were set up. And a series of photographs from the surgeries were required for assessment, including the resection margin of specimen, the dissection areas of lymphadenectomy, and the abdominal incision. All photographs were submitted to the CLASS data center within 1 week after each operation. The CLASS Research Committee monitored and reviewed the photographs regularly to ensure the quality of the operations.
The total gastrectomy with D1+/D2 lymphadenectomy was performed according to the Japanese gastric cancer treatment guidelines (fourth edition).20 The proximal resection margin was routinely sent for frozen section analysis intraoperatively to confirm a negative resection margin. In our study, Roux-en-Y digestive tract reconstruction was recommended, and linear or circular staplers were both acceptable, which should be selected based on the surgeons’ experiences and the specific intraoperative circumstances. Until now, there was still no confident evidence to confirm superiority of the linear stapler and circular stapler over each other; both circular and linear staplers for esophagojejunal anastomoses are feasible and safe.21,22 In the LTG group, laparoscopic operations include perigastric devascularization, lymph node dissection, and blood vessel ligation. Gastrectomy and digestive tract reconstruction could be completed via a mini-laparotomy through an abdominal incision not greater than 10 cm. In the OTG group, the procedure was similar to that of laparoscopic surgery except it was performed under direct visualization. In this study, “conversion to open” in the LTG group was defined as cases requiring laparotomy or auxiliary incision lengths greater than 10 cm.
The patients were evaluated twice daily. The times to first ambulation and flatus, as well as other considerations, were recorded until discharged. Laboratory results were recorded on the first, third, and fifth postoperative days. Postoperative fluid infusion or nutritional support (enteral/parenteral) was performed according to the clinical pathways. Patients could be arranged for discharge when the following conditions were met: postoperative complications had resolved, body temperature was less than 37 °C, pain could be well tolerated, and more than one-third normal diet could be taken orally.
Morbidity and mortality within 30 days following surgeries were examined. Intraoperative complications were defined as intraoperative massive hemorrhage (≥400 mL), injuries to visceral organs, pneumoperitoneum-related complications (eg, hypercapnia, emphysema, aeroembolism), and anesthesia-related or other complications that occurred during the operation. Postoperative complications were categorized into system-related or surgery-related complications. Briefly, system-related complications included pulmonary, cerebrovascular, renal, pleural effusion, deep vein thrombosis, or miscellaneous. Surgery-related complications included wound complications, intra-abdominal bleeding or abscess, anastomotic complications (eg, leaking, bleeding, stenosis), ileus, lymphatic leakage, pancreatic fistula, and pancreatitis. Complications were diagnosed based on the clinical evidence and image-based physical evaluations. For example, with anastomotic leakage and stenosis, the clinical features, such as fever, leukocytosis or leukopenia, tachycardia, and local abdominal pain, were the bases of the anastomotic leakage diagnosis. Dysphagia, nausea, and vomiting that impaired food intake sometimes suggested the existence of anastomotic stenosis. To confirm the diagnosis, endoscopy and imaging examinations were used, including computed tomography (CT) scan and fluoroscopy with water-soluble contrast. The Clavien-Dindo classification system was used to stratify for the severity of surgical complications.23
The sample size was calculated by 2 independent statisticians according to the hypothesis of noninferiority design. As stated in the previous reports,24-26 the morbidity and mortality rates were assumed to be 20% in the OTG group and 15% in the LTG group. Setting 1-sided α of .025, β of .2, and the noninferiority margin δ of 10%, the sample size required for each group was 100 cases. Considering the maximum dropout rate for this clinical study was about 10%, the sample size was determined to be 112 cases in each group. The patients in this study were from 13 gastric cancer centers in different regions of China.
Continuous variables were described as means with SDs or medians with interquartile ranges, and categorical variables were described as frequencies and percentages. Besides the complication rates, the differences between groups were examined by t test, Mann-Whitney test, Pearson χ2 test, Fisher exact test, or Cochran-Mantel-Haenszel χ2 test as appropriate. This study was a noninferiority trial; modified intention-to-treat analysis was conducted. If the upper limit of the 95% CI for between-group difference calculated by the Newcombe method was less than 10%, it could be inferred that LTG was not inferior to OTG in the incidence of early complications. Unless otherwise specified, the 2-sided P values less than .05 indicated statistical significance. All statistical analyses were performed using SAS 9.4 software (SAS Institute).
From January 2017 to September 2018, a total of 227 patients were enrolled in this clinical trial and randomized to the LTG group (n = 113) or the OTG group (n = 114). In the LTG group, 8 patients were excluded. This included 1 patient who withdrew informed consent, 4 patients who were noted to have distant metastases, 2 patients who underwent distal gastrectomy, and 1 patient who underwent proximal gastrectomy. In the OTG group, 5 patients were excluded. This included 1 patient who withdrew informed consent, 1 patient who was noted to have distant metastases, 1 patient who underwent distal gastrectomy, 1 patient who underwent proximal gastrectomy, and 1 patient who underwent total gastrectomy via the thoracoabdominal approach. After exclusion, 105 patients in the LTG group and 109 patients in the OTG group were analyzed for morbidity and mortality (Figure).
The baseline clinical characteristics of patients enrolled in this clinical trial are summarized in eTable 2 in Supplement 1. The mean (SD) age of the patients was 59.8 (9.4) years in the LTG group and 59.4 (9.2) years in the OTG group. The majority of patients were male (LTG group, 75 of 105, 71.4%; OTG group, 80 of 109, 73.4%). About half of all patients had 1 or more comorbidities (LTG group, 43 of 105, 41.0%; OTG group, 55 of 109, 50.5%). Ten patients (9.5%) in the LTG group and 8 patients (7.3%) in the OTG group underwent endoscopic mucosal resection/endoscopic submucosal dissection first. In addition, there were no significant differences in the distributions of body mass index, Eastern Cooperative Oncology Group performance status, American Society of Anesthesiologists class, tumor size, histologic type, or pathologic TNM (tumor, node, metastasis) stage.
The mean (SD) operation time of the LTG group (230.5 [67.3] minutes) was significantly longer than the OTG group (190.4 [62.6] minutes) (P < .001). The mean estimated blood loss was significantly lower in the LTG group than in the OTG group (92.0 mL vs 121.3 mL; P = .048). The majority of patients had undergone D1+/D2 lymphadenectomy in each group (LTG group: 100%; OTG group: 96.3%; P = .11), and the mean (SD) number of retrieved lymph nodes were 35.0 (12.7) in the LTG group and 37.2 (13.6) in the OTG group (P = .23). Combined gallbladder resections were performed for 3 patients (2.9%) in the LTG group and 1 patient (0.9%) in the OTG group (P = .36). Among patients in the LTG group, open conversion occurred in 2 patients (1.9%) because of massive hemorrhage (1 patient) and severe abdominal adhesion (1 patient). The recovery courses, including time to ambulation, time to first flatus, time to first liquid intake, and postoperative hospital stays, were not significantly different between the 2 groups (Table 1).
The overall morbidity and mortality rates were not significantly different between the 2 groups (LTG group: 19.1%; 95% CI, 11.5%-26.6%; OTG group: 20.2%; 95% CI, 12.7%-27.7%; rate difference [RD], −1.1%; 95% CI, −11.8% to 9.6%). Intraoperative complications occurred in 3 patients (2.9%) in the LTG group (massive hemorrhage) and 4 patients (3.7%) in the OTG group (massive hemorrhage in 3 patients and spleen injury in 1 patient) (RD, −0.8%; 95% CI, −6.5% to 4.9%). In addition, there was no significant difference in the overall postoperative complication rate of 18.1% (95% CI, 10.7%-25.5%) in the LTG group and 17.4% (95% CI, 10.3%-24.6%) in the OTG group (RD, 0.7%; 95% CI, −9.6% to 11.0%). In terms of each subtype of postoperative complications, including surgery-related and system-related complications, no statistical difference was noted between the 2 groups. Among the patients in the LTG group with postoperative complications, 1 patient died from intra-abdominal bleeding secondary to splenic artery hemorrhage. However, there was no significant difference in mortality between the LTG group and the OTG group (RD, 1.0%; 95% CI, −2.5% to 5.2%) (Table 2).
According to the Clavien-Dindo classification of surgical complications,23 in the LTG group, 11 patients were classified as grade II or lower, 7 patients as grade III, 1 patient as grade V in the LTG group; in the OTG group, 15 patients were classified as grade II or lower, 3 patients as grade III, 1 patient as grade IV. The distribution of severity was similar between the 2 groups (Table 3). Only 1 case of intra-abdominal bleeding resulted in death after operation in the LTG group, and patients with other complications in both groups were discharged successfully after conservative treatment or surgical interventions.
The trends of esophagogastric junctional adenocarcinoma and proximal gastric cancer have continued to increase over the last decades, and LTG is becoming an accepted option of minimally invasive gastrectomy.9,10 However, the safety and feasibility of LTG have not been verified. In 2019, two prospective single-arm trials (JCOG1401 and KLASS03) reported their results about the safety of LATG/LTG for early gastric cancer.13,14 In the JCOG1401 study,13 rates of grade 2 to 4 esophagojejunal anastomotic leakage were 2.1%, and the overall proportion of in-hospital grade 3 to 4 adverse events was 30.8%. In the KLASS03 study,14 postoperative morbidity and mortality rates were 20.6% and 0.6%, respectively. The results of the 2 studies showed the safety of LATG/LTG for early gastric cancer to a certain extent.
Laparoscopic total gastrectomy for gastric cancer is a relatively complicated procedure that requires higher skills, and the long-term oncologic efficacy has not yet been clarified.13,14 So, at present, for the safety of patients, LTG should be performed by well-trained laparoscopic surgeons in a cautious manner. A well-trained laparoscopic surgeon should be proficient in gastric cancer treatment and have gone through the learning curves of LTG.11
Although the distribution of severity was not significantly different between the 2 groups (P = .52), there were more severe complications (grade 3 or greater) in the LTG group vs the OTG group according to the Clavien-Dindo classification system (7.6% vs 3.8%).23 The relatively high incidence rate of severe complications of LTG may have been caused by the surgeons’ relative inadequate experience. Besides, the mean surgical time was longer in the LTG group compared with the OTG group (230.5 minutes vs 190.4 minutes); this also reflected the above situation. However, this did not mean that there was a significant difference in the safety of LTG compared with OTG. In addition, the incidence rate of severe complications of LTG in our study decreased slightly compared with that of the KLASS03 study (7.6% vs 9.4%).14 This further indicated that with the accumulation of experience, the incidence rate of severe complications would gradually decrease, and the differences in severe complication rates between LTG and OTG might be gradually narrowed.
Compared with laparoscopic gastrectomy, robotic gastrectomy is also an option for minimally invasive surgery and offers surgeons an advanced system for viewing and manipulation.27 In addition, the robotic console reduces ergonomic discomfort, enabling surgeons to maintain comfortable positions when operating.28 However, the laparoscopic approach is still the most common of the minimally invasive gastrectomies, for it is more easily accessed and has a more reasonable performance to price ratio than the robotic approach.27 Robotic gastrectomy is possibly promising in the future, as the latest evidence suggests that both short-term and long-term outcomes are comparable with those of the laparoscopic approach after the surgeons go through the associated learning curves.29
This study has several limitations. Of enrolled patients, 19.0% (20 of 105) in the LTG group and 16.5% (18 of 109) in the OTG group were misclassified to early gastric cancer preoperatively. The accuracy of preoperative tumor staging assessment was similar to that of the JCOG1401 study.13 There are 5 patients who were found to have distant metastasis and 5 patients who had an operation other than total gastrectomy; these were mostly owing to the limitations of preoperative examinations and the biological nature of some types of gastric cancers. Computed tomography (CT) is not sufficiently sensitive to detect or exclude peritoneal metastasis in some patients with gastric cancer.30 Additionally, the biological nature of gastric cancer potentially reduced the accuracy of preoperative assessments, because some types of gastric cancers, especially the poorly differentiated and undifferentiated carcinomas (signet ring cell carcinoma and other variants, 37.1% in the LTG group and 35.8% in the OTG group), can spread extensively in the mucosa and/or submucosa of the stomach (T1a/T1b) and result in distant metastases (peritoneal seeding, M1) but cannot always be detected through endoscopy and CT scan.31,32 So, according to the withdrawal criteria, patients with distant metastases enrolled in this study were excluded in the final analysis set. Some well-differentiated cases were small and confined in the upper or the middle third of the stomach. Therefore, proximal or distal gastrectomy would be more proper for the radical resection according to the treatment guidelines.20 Based on these considerations, we performed proximal or distal gastrectomy for the benefit of these patients.
In conclusion, the results of the CLASS02 trial showed that the safety of LTG with lymphadenectomy by experienced surgeons for clinical stage I gastric cancer is comparable with that of OTG. This study provides baseline evidence for future oncological safety studies of LTG for early gastric cancer, and even more so for advanced gastric cancer.
Accepted for Publication: May 28, 2020.
Corresponding Author: Yihong Sun, MD, PhD, Zhongshan Hospital, Department of General Surgery, Fudan University, 180 Fenglin Rd, Shanghai 200032, China (firstname.lastname@example.org).
Published Online: August 20, 2020. doi:10.1001/jamaoncol.2020.3152
Author Contributions: Dr Sun had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Liu, C. Huang, Xu, and Su contributed equally to this work.
Study concept and design: Liu, Xu, Su, G. Zhao, H. Huang, Hu, G. Li, Yu, Y. Li, Suo, N. Zhao, Sun.
Acquisition, analysis, or interpretation of data: Liu, C. Huang, G. Zhao, Ye, Du, Hu, N. Zhao, Zhang, H. Li, He, Sun.
Drafting of the manuscript: Liu, G. Zhao, H. Huang, G. Li, Yu, Suo, He, Sun.
Critical revision of the manuscript for important intellectual content: Liu, C. Huang, Xu, Su, G. Zhao, Ye, Du, Hu, Y. Li, N. Zhao, Zhang, H. Li, Sun.
Statistical analysis: Liu, G. Zhao, N. Zhao, Zhang, Sun.
Obtained funding: Liu, Sun.
Administrative, technical, or material support: Liu, Xu, Su, G. Zhao, Ye, Du, H. Huang, Hu, G. Li, Yu, Y. Li, Suo, H. Li, Sun.
Study supervision: C. Huang, G. Zhao, Sun
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported by grants from Clinical Trial Fund of Zhongshan Hospital (2016ZSLC13) and Johnson & Johnson Medical Ltd (IIS MIP-2017-11).
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
The Chinese Laparoscopic Gastrointestinal Surgery Study (CLASS) Group: Hui Cao, MD, and Gang Zhao, MD: Renji Hospital, Shanghai Jiao Tong University, Shanghai, China; Xiaohui Du, MD: Chinese PLA General Hospital, Beijing, China; Xuedong Fang, MD: China-Japan Union Hospital of Jilin University, Changchun, China; Changming Huang, MD: Fujian Medical University Union Hospital, Fuzhou, China; Hua Huang, MD: Fudan University Shanghai Cancer Center, Shanghai, China; Jiankun Hu, MD: West China Hospital, Sichuan University, Chengdu, China; Weiguo Hu, MD: Ruijin Hospital, Shanghai Jiao Tong University, Shanghai, China; Xianli He, MD: Tangdu Hospital, Air Force Medical University, Xi’an, China; Jiafu Ji, MD, Xiangqian Su, MD, and Ziyu Li, MD: Beijing Cancer Hospital, Peking University, Beijing, China; Zhiwei Jiang, MD: Jiangsu Province Hospital of Chinese Medicine, Nanjing, China; Guoxin Li, MD: Nanfang Hospital, Southern Medical University, Guangzhou, China; Yong Li, MD: Guangdong General Hospital, Guangzhou, China; Zhongchen Liu, MD: Shanghai Tenth People’s Hospital Affiliated to Tongji University, Shanghai, China; Xiang Peng, MD: The First People’s Hospital of Foshan, Foshan, China; Jian Suo, MD: The First Hospital of Jilin University, Changchun, China; Yihong Sun, MD, PhD: Zhongshan Hospital, Fudan University, Shanghai, China; Kaixiong Tao, MD: Union Hospital, Huazhong University of Science and Technology, Wuhan, China; Baolin Wang, MD: The Second Affiliated Hospital of Nanjing Medical University, Nanjing, China; Hongbo Wei, MD: The Third Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China; Kuan Wang, MD: The Cancer Hospital of Harbin Medical University, Harbin, China; Jian Xu, MD: Jiaozhou Central Hospital of Qingdao, Qingdao, China; Zekuan Xu, MD: The First Affiliated Hospital with Nanjing Medical University, Nanjing, China; Jianxin Ye, MD: The First Affiliated Hospital of Fujian Medical University, Fuzhou, China; Mingang Ying, MD: Fujian Provincial Cancer Hospital, Fuzhou, China; Peiwu Yu, MD: The First Hospital Affiliated to AMU, Chongqing, China.
Meeting Presentation: This work was presented at the 2020 Gastrointestinal Cancers Symposium of the American Society of Clinical Oncology; January 23, 2020; San Francisco, California.
Data Sharing Statement: See Supplement 3.
Additional Contributions: We thank Lau Wen Liang, Joel, MMed (National University Hospital, Singapore) for his critical revision of the manuscript with no compensation.