Standard and multivisceral colectomy in locally advanced colon cancer

Colon is the second most common site for cancer both in men and women.^1,2 Significant advances in its treatment have been achieved over the last decades due to the improvements in surgical technique, chemotherapy, targeted therapy and examination of tumor biomarkers.^3,4 Unfortunately, many of these patients are diagnosed relatively late presenting with locally advanced colon cancer (LACC), which is challenging to manage.₅ The definition of LACC is controversial as no uniform approach exists in the literature to date. Some authors qualify stage T3 cancer with extramural invasion ≥ 5 mm and T4 as locally advanced, while others define LACC as stage T4 cancer solely.^{6, 7}

Multimodal treatment is the standard approach for LACC, although surgery remains its cornerstone. Microscopically complete resection provides satisfactory oncologic outcomes, however,

complex multiorgan resection may be required in these patients.₈ The latter is associated with postoperative complications, prolonged hospital stay, increased treatment-associated costs and later start of adjuvant chemotherapy, which may ultimately increase the risk of tumor recurrence.₉ Thus, potential improvement in survival after surgery for LACC should outweigh the aforementioned risks.

This study explores the outcomes of colectomy in patients with LACC. The impact of surgical approaches on postoperative results as well as potential risk factors for complications were evaluated.

A total number of 474 patients with colon cancer underwent surgery throughout the study period. Of those, 203 (42.8%) were operated for LACC.

Patient characteristics, perioperative data and pathology findings are presented in Table 1. The most common location of LACC was the left colon (58.1%), while 82 (40.4%) patients had colon obstruction. More than one-fourth of the patients (28.6%) had distant metastases at surgery. In patients with non-metastatic LACC (n = 145), the rate of curative resections (R0) was 97%. Morbidity and mortality rates were 12.3% and 2.5%, respectively. Median length of stay was 11 (5–44) days.

SC was performed in 112 (55.2%) patients and MVC in 91 (44.8%). The latter included resections of one (n = 70) or ≥ 2 organs/structures (n = 21) (Table 2). In patients with ≥ 2 organs/structures resected the resection of the small bowel, stomach and pancreas was most often carried out.

A comparative analysis between the outcomes of MVC and SC was performed (Table 3). Preoperative parameters were similar except the significantly higher total protein levels in the MVC group. The latter was almost always performed in patients with T4b adenocarcinoma (94.5% vs. 33.9%, p < 0.01). Operative time was significantly longer for MVC (180 minutes vs. 140 minutes, p<0.01) and so was the median blood loss (200 ml vs. 100 ml, p = 0.01). The red blood cell transfusion rate was significantly higher for MVC - 22% vs. 8.9%, p = 0.01. Microscopically complete resection was achieved in a similar number of patients (98.5% vs. 95.3%, p = 0.56). Proportions of postoperative complications and their types were comparable between the groups. The length of stay after surgery was longer following MVC (11 days vs. 10 days, p = 0.04).

Univariable analysis of factors associated with postoperative complications was performed (Table 4). Male gender, presence of ≥ 3 comorbidities, tumor location in the left colon and red blood cell transfusion were associated with grade ≥ II complications. These factors were analyzed together in the multivariable model (Table 5). The latter demonstrated that only presence of ≥ 3 comorbidities was associated with grade ≥ II morbidity. Specifically, their risk increased more than three times in these patients - OR 3.1 (1.1–9.2), p = 0.038.

Our findings indicate that colectomy in patients with LACC is a safe procedure providing satisfactory surgical outcomes when performed in a specialized surgical unit. This is applicable to both SC and MVC. In the literature, postoperative morbidity and mortality rates following surgery for LACC are 25–38% and 3.3–6.9%, respectively.^{5,14, 15, 16, 17} Hoffman et al., demonstrated that morbidity rate may increase with the use of multiorgan resections.₁₈ In this series, postoperative morbidity rate was 12.8% including 7.4% severe complications. Although MVC was associated with an increased blood loss, need for blood transfusion, as well as with longer operative time and hospital stay, postoperative morbidity-associated parameters and mortality were comparable to those of SC.

Severe complications were mostly caused by the anastomotic leakage (4.9%), which is consistent with the data in the literature.^{14, 15, 16, 17, 18, 19} Multiple parameters are found to be associated with the risk of leakage including patient-specific variables, intraoperative complications, surgeon- and technique-related factors. In this report, such analysis was not possible due to the small number of cases. However, when analyzing risk factors for complications technical parameters such as single-layer anastomosis suture or its end-to-end type did not increase the rate of complications. We believe that single-layer suture is a simple technique that significantly expedites the procedure, while the end-to-end anastomosis avoids the need for additional closure of the intestinal stumps on the proximal and/or distal loops. The effectiveness of this technique was reported also by Liu et al.₂₀ Our results do not confirm the association between the postoperative results and blood transfusion suggested by Marinello et al.₁₄ Despite being a significant predictor in the univariable analysis, blood transfusion was not an independent predictor of complications in the multivariable model.

According to the literature, MVC is performed in only 1.2–12% of patients with colon cancer.^14,15,21,22 In our study, MVC was performed in 44.8% of patients with LACC, which accounts for 19.2% of a total number of colon resections for cancer. Higher incidence of MVC in our series can be attributed to strict selection criteria in the aforementioned studies, as well as to a significantly higher proportion of late diagnosed patients in our population. This is confirmed also by the fact that nearly 40% of our patients presented with partial or total colon obstruction prior to surgery.

Intraoperatively, it is not always possible to assess whether or not colectomy will be curative, thus the main goal is to achieve a complete resection of the primary tumor and suspicious adjacent tissues if these are found at surgery.₉ According to the literature, the most common invasion sites for T4b colon cancer are the small bowel, urinary bladder and abdominal wall.^{15, 16, 17} In this series, most often these patients had tumor ingrowth into ≥ 2 organs, predominantly to the small bowel, distal pancreas and stomach. Tumor invasion into adjacent structure(s) was verified by the pathology examination in about 95% of patients who had undergone MVC. Given that this parameter ranges from 44% to 72.5% in the literature^5,9,15,16, the choice of surgical approach was adequate in this series.

This report has several limitations, including retrospective design with its inherent biases. Furthermore, we did not register grade I complications (according to Clavien-Dindo)₁₁, which somewhat limits the information on postoperative results of colectomy for LACC. It is also worth mentioning that our data are based on an experience of a specialized center of colorectal surgery, thus the reproducibility of our results is limited and surgical outcomes should be interpreted with caution.

In conclusion, colectomy including MVC is a safe procedure in the setting of LACC. In experienced hands, the postoperative outcomes are acceptable showing no differences between the SC and MVC. However, their oncologic benefits require further investigation. Given the complexity of MVC, these procedures should be reserved to qualified expert centers that are familiar with colorectal procedures as well as with the surgery of other organ systems.