Emergency surgery for splenic flexure cancer: results of the SFC Study Group database

Background The effectiveness of surgical treatment for splenic flexure carcinomas (SFCs) in emergency settings remains unexplored. This study aims to compare the perioperative and long-term outcomes of different alternatives for emergency SFC resection. Method This multicenter retrospective study was based on the SFC Study Group database. For the present analysis, SFC patients were selected if they had received emergency surgical resection with curative intent between 2000 and 2018. Extended right colectomy (ERC), left colectomy (LC), and segmental left colectomy (SLC) were evaluated and compared. Results The study sample was composed of 90 SFC patients who underwent emergency ERC (n = 55, 61.1%), LC (n = 18, 20%), or SLC (n = 17, 18.9%). Bowel obstruction was the most frequent indication for surgery (n = 75, 83.3%), and an open approach was chosen in 81.1% of the patients. A higher incidence of postoperative complications was observed in the ERC group (70.9%) than in the LC (44.4%) and SLC groups (47.1%), with a significant procedure-related difference for severe postoperative complications (Dindo-Clavien ≥ III; adjusted odds ratio for ERC vs. LC:7.23; 95% CI 1.51-34.66; p = 0.013). Anastomotic leakage occurred in 8 (11.2%) patients, with no differences between the groups (p = 0.902). R0 resection was achieved in 98.9% of the procedures, and ≥ 12 lymph nodes were retrieved in 92.2% of patients. Overall and disease-free survival rates at 5 years were similar between the groups and were significantly associated with stage pT4 and the presence of synchronous metastases. Conclusion In the emergency setting, ERC and open surgery are the most frequently performed procedures. ERC is associated with increased odds of severe postoperative complications when compared to more conservative SFC resections. Nonetheless, all the alternatives seem to provide similar pathologic and long-term outcomes, supporting the oncological safety of more conservative resections for emergency SFCs.


Introduction
The surgical treatment of splenic flexure carcinoma (SFC) has been traditionally neglected in the literature, mainly because of its relatively low incidence, as it represents only 3 to 5% of all colonic cancers [1][2][3][4]. In recent years, however, several studies have questioned which type of resection would provide the best surgical and oncological outcomes in patients with SFCs [1,[5][6][7][8][9][10][11][12][13]. Despite a substantial lack of standardization regarding the nomenclature of the different surgical alternatives to resect SFCs, three main surgical procedures are performed, namely, extended right colectomy (ERC), left colectomy (LC), and segmental left colectomy (SLC) [5,14,15]. All these procedures are considered alternatives for curative resections of these tumors, which are located on the border between the right and left colon and have dual lymphatic drainage toward the superior and inferior mesenteric vessels [16]. Indeed, two recent meta-analyses concluded that no procedure-related difference exists in terms of postoperative morbidity, mortality, lymph node yield, and patient survival [14,16]. Consequently, a more conservative SLC, for a long time considered oncologically inadequate [17], is considered a safe and effective option for the treatment of SFCs [5,18].
In contrast, the outcomes of these different resections when performed in acute clinical situations remain substantially unexplored. Although SFCs have been associated with a poorer prognosis than cancers from other colonic subsites due to the high risk of bowel obstruction [3,19,20], the number of emergency cases reported in the literature is limited. Indeed, no prior study has considered emergency SFC resections exclusively. Previous works investigated either samples of elective SFC cases only [5,6,9,11] or mixed populations of patients, including both elective and emergency operations, none of which provided subgroup analyses on emergency cases [7,8,10,18,21]. In these studies, emergency patients represented 5.7 to 50% of the samples, with an absolute number of cases ranging from 34 to 75 patients [2,7,8,10,18,21]. The authors, however, consistently reported that ERC was the preferred approach to treat SFCs presenting as intestinal obstruction [14].
The present study aims to compare the perioperative and long-term outcomes of the different surgical alternatives for SFC resection in emergency settings using the SFC Study Group database [5].

Study design
The present study was designed as an ancillary analysis by the SFC Study Group. As previously described [5] Anonymous patient data were retrieved from each local database and merged into a common database that was centralized by the leading center and further set for the present statistical analyses. Due to the retrospective design of the study, which was conducted exclusively using patient records, no institutional review board approval was required. All personal data were managed in conformity to the principles declared to the National Commission for Data Protection and Liberties. The study was reported following the recommendations listed in the STROBE checklist for cohort and case-control studies [22].

Study population
For the present analyses, SFC patients were selected if they met the following criteria: (1) age > 18 years; (2) colon cancer located at the splenic flexure (i.e., 10 cm proximal toward the transverse colon or 10 cm distal to the descending colon [2,5,17]) as assessed by preoperative computed tomography (CT) and confirmed during surgery and at the pathologic report [2,6,17]; (3) any AJCC TNM stage [23]; (4) obstructive or perforated neoplasm requiring emergency surgery; (5) curative-intent surgical resection; and (6) one-or two-stage surgery (via temporary stoma). Patients with synchronous colonic carcinomas, untreatable metastatic disease, and polyposis coli were excluded. Emergency surgery was defined as an unplanned procedure performed within 48 h of hospital admission [10,18]. As previously described [5], three types of surgical procedures were performed for SFC resection: ERC, LC, and SLC [2,5,14,15]. Both laparoscopic and open procedures were considered. Interventions were carried out by general and colorectal surgeons. Patients were treated and followed-up according to national protocols [5]. Adjuvant chemotherapy was generally indicated in patients with TNM III/IV tumors or in those presenting cancers with unfavorable histopathological features.

Study outcomes
The study outcomes were the same as those reported previously on the elective management of SFCs [5] and included intraoperative variables (e.g., operative time, blood loss), postoperative variables (e.g., postoperative morbidity and mortality, length of hospital stay), quality of the surgical resection (e.g., resection margin status, number of retrieved lymph nodes), and overall survival (OS) and disease-free survival (DFS) up to 5 years. Conversion was defined as a premature interruption of the laparoscopic approach before the resection phase was concluded [24,25]. Postoperative morbidity and mortality were defined as events that occurred during the hospital stay or within 90 days after surgery. Postoperative complications were graded according to the Dindo-Clavien classification [26], with grades of III or more identifying severe complications. Postoperative ileus was defined as the absence of bowel movements or flatus associated with intolerance of oral intake lasting more than 3 days postoperatively [27,28]. Resections were classified as R0 when a complete removal of the tumor with free resection margins and no peritoneal spread were objectivized macro-and microscopically.

Statistical analysis
Demographics, clinical characteristics, and study outcomes were compared between the 3 procedures applied to resect SFCs using the chi-squared test for categorical variables and Kruskal-Wallis tests for continuous variables. A multivariate analysis was performed by including demographic, preoperative, and oncological variables that reached a p value < 0.1 in the univariate analysis. Adjusted p values were reported for the overall comparisons, and whenever significant, two-by-two group comparisons were also explored. Bonferroni correction was applied. Adjusted odds ratios (AORs) were calculated and presented with the 95% confidence interval (CI).
The Kaplan-Meier method was used for the survival analyses, and the log-rank (Mantel-Cox) test was applied for group comparisons. OS was defined as the time from surgery to disease-related death and was censored at the last follow-up date if no events occurred. DFS was defined as the time from surgery to disease recurrence and was censored at the last follow-up date if no events occurred. Univariate and multivariate Cox proportional hazards regression models were used to identify independent predictors of survival.
Statistics were carried out with SPSS (Statistical Package for the Social Sciences, IBM SPSS Statistics, Version 23 for Macintosh; IBM Corp., Armonk, NY, USA). A p value < 0.05 was considered statistically significant.

Results
The original SFC Study Group database included 494 SFC patients. Of these, 95 patients underwent emergency surgery. Five patients were excluded because they did not meet all the inclusion criteria or due to missing data. Finally, the present study sample was composed of 90 patients who underwent emergency ERC (n = 55, 61.1%), LC (n = 18, 20%), or SLC (n = 17, 18.9%) (Fig. 1). SFCs were located at the splenic flexure (n = 56, 62.2%), up to 10 cm proximal toward the transverse colon (n = 17, 18.9%), or up to 10 cm distal toward the descending colon (n = 17, 18.9%), without differences between the 3 surgical procedure groups (p = 0.389). For the majority of patients, surgery was indicated due to bowel obstruction (n = 75, 83.3%), whereas the remaining presented with a perforation (n = 15, 16.7%). No procedure-related differences were observed (p = 0.451). No endoscopic colonic stent was used as a bridge to surgery. Overall, 73 patients (81.1%) underwent open surgery; a higher frequency of laparoscopy was observed for LC and SLC procedures than ERC, but the difference was not statistically significant (p = 0.051). The majority of interventions (57.8%) were performed by colorectal surgeons without procedure-related differences (p = 0.379). Demographic, clinical, preoperative, and histopathological characteristics for the total study sample and by surgical procedure are displayed in Table 1.

Comparisons between the three surgical procedures for SFCs
No procedure-related differences were observed except for age, ASA score, and suspected extracolic organ involvement upon CT scan (Table 1). In the histopathological analysis, patients in the LC group had the largest tumors and presented a higher prevalence of perineural invasion and moderately or poorly differentiated cancers. Regarding the intra-and postoperative outcomes, a higher incidence of postoperative complications was observed in the ERC group (70.9%) than in the LC (44.4%) and SLC groups (47.1%); however, a significant procedure-related difference was noted only when considering severe postoperative complications (Dindo-Clavien ≥ III) (p = 0.011). A similar trend was also observed for the time to return to regular diet, which was longer for patients operated on by ERC (p = 0.054) ( Table 2). In the adjusted model, the only significant betweengroup difference was observed for the rate of severe postoperative complications; ERC was associated with an increased risk of Dindo-Clavien ≥ III, particularly when compared to LC (AOR 7.23; 95% CI 1.51-34.66; p = 0.013).
Stage pT4 and the presence of synchronous metastases were found to be independent predictors for OS and DFS (Table 3).

Discussion
The present multicenter retrospective study is the first, to our knowledge, to describe and compare the outcomes of the different surgical resections performed for SFCs presenting as a surgical emergency. Our results confirm that ERC and open surgery are the most frequently performed procedures in this clinical situation. Moreover, ERC appears to be associated with higher rates of severe postoperative complications. Comparable short-and long-term oncologic outcomes are observed between ERC, LC, and SLC, supporting the feasibility and safety of more conservative SFC resections in emergency settings.
The choice of which procedure to perform for cancers located at the splenic flexure remains essentially based  on the surgeon's experience and preference. In the setting of elective surgery, accumulating evidence consistently shows no difference in terms of oncological outcomes and survival between ERC, LC, and SLC [14,16]; however, some previous comparative studies reported that extensive SFC resections are associated with higher rates of postoperative complications, particularly postoperative ileus [5,6], and longer recovery times [5], suggesting a short-term benefit in performing more limited SFC resections (LC and SLC) [5,6,11,18]. For emergency surgery, however, a few studies did not provide any subgroup analysis separately assessing the outcomes for the different types of SFC resection performed as emergency and elective surgery. The main reason for this lack of evidence could be related to the very low incidence of emergency SFCs. To date, all the   [5] in which 74.4% were laparoscopic procedures. This difference can be easily linked to the clinical scenario in which the surgeon has to choose the surgical approach facing an SFC presenting as bowel obstruction (83% of cases) or perforation. ERC and open surgery are usually preferred because they may be technically easier in the presence of proximal colonic distension or may be mandatory due to concerns about the viability of the cecum [8,14]. Indeed, the experience of the surgeon on call, who may not always be a specialized colorectal and/or laparoscopic surgeon, could have an impact on this choice [10,21]. Binda et al. [21] observed some differences in terms of the choice of the surgical procedure and approach when comparing general surgeons vs. colorectal surgeons, but only for elective surgery. In the present study, almost 60% of the procedures were performed by specialized colorectal surgeons, and no difference was observed between the ERC, LC, and SLC groups. The type of resection and the surgeon's experience have also been related to some surgical outcomes, such as postoperative complications, R0 resection, and lymph node yield. The present findings suggest that ERC may be associated with increased odds of developing severe postoperative complications when compared with LC and SLC; however, similar and favorable pathologic outcomes were observed for all SFC resection alternatives, with an R0 obtained in almost 99% of cases and more than 12 lymph nodes harvested in 92% of patients. Importantly, these data confirm the findings already reported for elective ERC, LC, and SLC [5,10,14,16], supporting the oncological safety of more conservative SFC resections in emergency settings. Patient survival rates were comparable among the ERC, LC, and SLC groups, with a global sample 5-year OS of 58.4%. These rates are inferior to those reported in studies concerning elective SFC resections or mixed populations [14,16], but they are comparable to the study by Odermatt et al. [10] that considered the highest percentage of emergency surgery in their sample (50% of patients). Poorer OS and DFS may be expected in emergency patients who present with significant comorbidities, unstable conditions, or advanced tumor stage at diagnosis [10]. The present results, taking into account potential confounders, showed that for patients operated on in emergency settings, predictors for OS and DFS are essentially associated with tumor characteristics, particularly stage pT4 and the presence of synchronous metastasis. The present retrospective study has some limitations, and these findings should be interpreted and generalized with caution because they are based on a sample of patients operated on in several European referral centers over a long period of time. The potential impact of selection and reporting bias cannot be excluded.

Conclusion
The present study, based on a relatively large sample of SFC patients undergoing emergency surgery, confirms that ERC and open surgery are the most frequently performed procedures. ERC appears to be associated with a higher incidence of severe postoperative complications, whereas R0 resection, lymph node yield, tumor recurrence, and survival rates are comparable between ERC, LC, and SLC. These findings also support the feasibility and safety of more conservative SFC resections in emergency settings.