Risk factors for early postoperative complications after D3 dissection for stage II or III colon cancer: Supplementary analysis of a multicenter randomized controlled trial in Japan (JCOG0404)

Abstract Objective To determine risk factors for early postoperative complications after D3 dissection for stage II/III colon cancer. Background Identification of risk factors for postoperative complications is essential in patients surgically treated for colon cancer. The Japan Clinical Oncology Group (JCOG) conducted a randomized controlled trial, JCOG0404, to confirm the non‐inferiority of laparoscopic surgery (LAP) to open surgery (OP) with D3 dissection for stage II/III colon cancer. This supplementary analysis was made to assess risk factors for surgery requiring D3 dissection using data from JCOG0404. Methods Proportion of postoperative complications of any grade (CTCAE ver. 3.0) until first discharge and risk factors for the most frequent complications were analyzed by univariable and multivariable analysis. Results Among 1057 randomized patients treated between October 2004 and March 2009, 520 patients with OP and 525 patients with LAP were analyzed. Overall postoperative complications of all grades occurred in 190 patients (18.2%). Multivariable analysis showed that the risk factors for overall early postoperative complications were OP itself (odds ratio [OR] 2.01, 95% confidence interval [CI]: 1.38‐2.91, P = 0.0003) and operation time of >240 minutes (OR 1.94, 95% CI: 1.24‐3.02, P = 0.0036). The most frequent adverse event was wound complication (50/1045, 4.8%). In the univariable analysis, reconstruction, greater blood loss, OP, and higher body mass index were significantly associated with wound complication. Conclusion Open surgery and longer operation time of >240 minutes were significant risk factors for postoperative complications. LAP surgery and shorter operation time could contribute to fewer postoperative complications in patients undergoing colectomy with D3 dissection. (Japan Clinical Oncology Group study JCOG 0404: NCT00147134/UMIN‐CTR: C000000105.)


| INTRODUC TI ON
Additionally, JCOG0404 enrolled more than 1000 patients, making it one of the largest RCT for patients with colon cancer requiring D3 dissection in Japan.
We aimed to identify the risk factors for postoperative early complications in patients with stage II and III colon cancer by exploratory analyses using the data from JCOG0404. To the best of our knowledge, the present study is the first to evaluate risk factors for early postoperative complications after surgery requiring D3 dissection from prospectively collected data.

| Operative methods of D3 dissection
For right-sided tumors, the vascular pedicles were divided at their origin, and the draining lymph nodes along the superior mesenteric vein were removed. For left-sided tumors, removal of lymph nodes at the root of the inferior mesenteric artery was carried out along with high ligation or with preservation of the left colic artery and ligation of the root of the superior rectal artery.

| Endpoints and statistical considerations
Adverse events were evaluated according to CTCAE 3.0.
Postoperative mortality and morbidity were respectively defined as death from any cause and any adverse event of grade 1 or higher including anastomotic leakage, paralytic ileus, bowel obstruction, and (viii) body mass index (BMI) (≤20 vs 20 < BMI ≤ 25 vs >25 kg/ m 2 ). Univariable analysis was carried out by Fisher's exact test to compare treatment arms in terms of operative morbidity and mortality. To investigate risk factors for postoperative morbidity, univariable and multivariable logistic regression analysis was done and the odds ratio (OR) and its 95% confidence interval (CI) for postoperative morbidity were estimated. For wound complications and leakage, only univariable logistic regression analysis was carried out as a result of insufficient events to perform multivariable analysis. The population included in this analysis was defined as those patients who received the assigned surgery; hence, patients who did not receive the assigned surgery were excluded from this analysis. A twosided P-value of <0.05 was considered statistically significant. All statistical analyses were carried out using SAS ver. 9.2 or higher.  Figure 1).  triangle anastomosis in 25 patients and stoma creation in 1 patient), greater blood loss, OP itself, and higher BMI were significantly associated with wound-related complications (Table 3).

| RE SULTS
Anastomotic leakage occurred in 37 patients (3.5%). There was no difference in its occurrence between the OP group (18; 3.5%) and the LAP group (19; 3.6%). Univariable analysis showed that gender, operation time, and type of procedure were significantly associated with anastomotic leakage (Table 4). So far, several reports have shown the risk factors for overall complications to be male gender, 10 cancer stage, 10 BMI, 10 visceral fat area, 11 age, 12 OP 12,13 and blood loss. 13 The OP procedure was also shown to be a risk factor in our analysis in comparison with LAP.

| D ISCUSS I ON
However, male gender, BMI, and blood loss were not risk factors for overall postoperative complications in our analysis. The present study is the first, to our knowledge, to show that in addition to the OP procedure, operation time of over 240 minutes was a risk factor for overall postoperative complications. We speculated that This is why the present study did not need to take into consideration the surgical learning curve as a potential risk factor because the JCOG0404 study chairman certifies surgeons at each participating institution according to the following criteria. For OP, surgeons must There are limitations in the present study. The inclusion criteria of JCOG0404 did not include disease in the transverse or descending colon, which would have required high-level laparoscopic skills.
The incidences of wound complications and anastomotic leakage were insufficient for multivariable analysis. However, 37 patients (3.5%) developed anastomotic leakage in our study. With anastomotic leakage being one of the most critical complications, these are not satisfactory data. Thus, further investigation is necessary to identify the risk factors for anastomotic leakage.
In conclusion, our analysis from prospectively collected data of D3 dissection for stage II/III colon cancer indicated that the risk factors for overall postoperative complications were the OP procedure and operation time of over 240 minutes. LAP and shorter operation time may contribute to a decrease in postoperative complications for patients undergoing colectomy with D3 dissection.

ACK N OWLED G EM ENTS
This study was supported in part by the National Cancer Center