Abstract
Background. Regarding low rectal cancer (RC) low anterior resection (LAR), a specific consensus regarding the optimal level of inferior mesenteric artery (IMA) ligation does not exist.
Objectives. To systematically evaluate the effects of different IMA ligation methods on the prognosis of patients with low RC, so as to further guide clinical treatment.
Materials and methods. Between January 2013 and December 2018, 158 patients with low RC underwent LAR. According to the IMA ligation method used, the cases were divided into the low-ligation group (LL group; n = 66) and the high-ligation group (HL group; n = 92). The basic information, operation indicators, postoperative data, and long-term survival in the 2 groups were compared.
Results. Sixty cases in the HL group and 60 cases in the LL group were successfully matched using propensity score matching (PSM). There were no statistically significant differences in intraoperative bleeding, intraoperative time, postoperative hospital stay, harvested lymph nodes (LNs), postoperative complications (including urinary retention, urinary incontinence, anastomotic leaks, bowel obstruction, incisional infection, and anal function 3 months after surgery), overall survival (OS), disease-free survival (DFS), local recurrence, and distant metastasis between the 2 groups (p > 0.05). Compared with the HL group, the time to first flatus and the time to fluid intake were shorter in the LL group (p < 0.05).
Conclusions. In general, the different IMA ligation methods have no significantly different effects on the prognosis of patients with low RC, but the LL group showed restored intestinal motility earlier.
Key words: prognosis, propensity score matching, low rectal cancer, ligation of inferior mesenteric artery
Background
Only 200 years have passed since the industrial revolution, when rectal cancer (RC) patients were first treated surgically and the basic principle of total mesorectal excision (TME) was widely accepted by surgeons.1 Although RC surgical techniques flourished during this period, the management of the left colonic artery (LCA) in RC surgery has not been systematically defined. In the early 20th century, Miles2 and Moynihan3 proposed 2 different approaches to the inferior mesenteric artery (IMA): low ligation (LL) and high ligation (HL). Low ligation involves ligation of the IMA immediately distal to the origin of the LCA, and a complete removal of the mesenteric vessels and lymph nodes (LNs) within its limits in order to provide rich blood flow to the anastomosis. High ligation involves dissection of the vessels at the root of the IMA from the abdominal aorta to allow complete clearance of the LNs at the root of the IMA.
The question of whether the LCA needs to be preserved has been widely discussed by clinical scholars.4, 5, 6, 7 However, in previous studies, RC or colorectal cancer (CRC) has always been analyzed as a whole, with little distinction made between the different segments and locations of tumors in the colorectum. Low RC has a unique anatomy compared to mid-RC, high RC or colon cancer. In particular, it is located lower, in the narrow pelvis. Because of its location, the anastomosis is likely to receive less blood flow after low RC surgery, as it is further from the proximal vessels, thus contributing to the development of anastomotic leakage (AL). In recent years, the use of anal-saving operations has become more widespread as the demands for quality of life have increased. Therefore, the proper choice of IMA management for low RC is crucial. This study examines the effects of different IMA treatments on low RC.
Objectives
Although there have been many studies comparing the safety and feasibility of different IMA ligation methods in patients with CRC, few studies have investigated low RC separately. Thus, the purpose of this study is to systematically assess the impact of different ICA management techniques on the prognosis of patients with low RC.
Materials and methods
Study design
Patients diagnosed with low RC and undergoing low anterior resection (LAR) performed by the same surgical team at the Second Affiliated Hospital of Harbin Medical University, China, between January 2013 and December 2018 were included in the study. Depending on the intraoperative LCA treatment method used, patients were divided into 2 groups: high ligation of the IMA (HL group; n = 92) and low ligation of the IMA (LL group; n = 66).
Inclusion and exclusion criteria
Patients were considered eligible for the study if: 1) they were diagnosed with RC during pathological examination and the distance from the lower edge of the tumor lesion to the dentate line was ≤5 cm; 2) had stage I–III disease; and 3) had complete clinical records. The exclusion criteria were: 1) patients who received neoadjuvant radiotherapy or chemotherapy; 2) RC patients who underwent emergency surgery due to obstruction, massive bleeding, perforation, or other emergencies; 3) patients who underwent abdominoperineal resection or diverting ileostomy; and 4) patients with American Society of Anesthesiologists (ASA) grade IV who were considered by anesthesiologists to be unfit to undergo surgery.
Surgery
Surgical methods included laparotomy and laparoscopy. For all of the enrolled patients, the TME principle was followed during radical resection, and complete resection of the diseased bowel segment and the corresponding mesentery was performed, including complete dissection of the vessels, nerves, lymphatic vessels, and LNs, followed by anastomosis of the bowel. The HL was performed when preoperative computed tomography (CT) showed suspicious metastatic LNs at the root of the IMA or when metastatic LNs were diagnosed using frozen biopsy at the root of IMA. In the other cases, LL was performed. For the LL group, the vessels were severed below the LCA emanating from IMA, and lymph nodes were dissected. For the HL group, the vessels were severed about 1 cm below the IMA root, and LN dissection was performed at the IMA root. All arteries and veins were skeletonized, and LNs larger than 0.5 cm were dissected.
Observation indicators
The intraoperative variables included intraoperative blood loss and operation time. Postoperative data included pathological information, the postoperative recovery index, postoperative complications, postoperative LN detection, lymphatic invasion, vascular invasion, perineural invasion, postoperative tumor differentiation, pathological type, time to first flatus, time to fluid intake, postoperative hospitalization days, postoperative intestinal obstruction, urinary retention, urinary incontinence, AL, and anal function 3 months after surgery. The postoperative long-term survival variables included overall survival (OS), disease-free survival (DFS), local recurrence, and distant metastasis. All patients with stage II (including high-risk factors) and stage III according to the tumor−nodule−metastasis (TNM) staging system wereroutinely treated with the standard XELOX protocol (8 cycles of capecitabine and 8 cycles of oxaliplatin) for chemotherapy. The patients were followed up to December 2021 for at least 36 months or until death.
Statistical analyses
The IBM Statistical Package for Social Sciences (SPSS) v. 26.0 software (IBM Corp., Armonk, USA) was used for data analysis. The baseline information in the 2 groups were matched using propensity score matching (PSM), and variables including gender, age, tumor size, body mass index (BMI), pathological tumor (pT) stage, pathological lymph node (pN) stage, carcinoembryonic antigen (CEA) level, and ASA grade were considered covariates in the performance of 1:1 PSM (caliper value = 0.1). All calculated data are expressed in terms of frequency and percentage. The Kolmogorov–Smirnov test was used to test the normality of the subgroups and the Levene’s test was used to evaluate the homogeneity of variance. Mean and standard deviation (M ±SD) were used to present the measurement data conforming to a normal distribution, and Student’s t-test or Mann–Whitney U test were used for further comparisons. The χ2 test or Fisher’s exact test were used for comparisons of counted data. Survival curves were drawn using the Kaplan–Meier method and analyzed with the log-rank test using GraphPad Prism v. 8.0 software (GraphPad Software, San Diego, USA). The value of p ≤ 0.05 was considered statistically significant.
Results
Baseline data in the LL and HL groups
According to the inclusion and exclusion criteria, 158 patients with low RC were included in this retrospective study, and were divided into the LL (n = 66) and HL (n = 92) groups, according to whether LCA was retained during operation. The basic patient characteristics in the LL and HL groups were analyzed. Before matching, BMI (p = 0.004) and tumor size (p = 0.025) were significantly different between the 2 groups. After PSM, the characteristics of the 2 groups tended to be balanced, and 60 pairs of cases were successfully matched. There were no statistical differences in the baseline data of these 2 groups (p > 0.05; Table 1). Normality and homogeneity of variance tests were performed before the analysis of data.
Postoperative pathological information
In terms of the postoperative pathological data, there were no significant differences in differentiation degree (p = 0.114), number of LNs harvested (p = 0.973), vascular invasion (p = 0.699), perineural invasion (p = 0.350), lymphatic invasion (p = 0.079), or pathological type (p = 0.810) between the 2 groups (Table 2).
Surgical operation indicators, postoperative recovery indicators and postoperative complications
In the analysis of intraoperative indicators, there were no significant differences in operative duration (p = 0.720), intraoperative blood loss (p = 0.143) or the time of first postoperative feeding (p = 0.151) between the 2 groups. In the postoperative information analysis, the first postoperative flatus time in the LL group (48 h, range: 10–96 h) was earlier than that in the HL group (50 h, range: 14–110 h; p = 0.022). There were no statistical differences between the 2 groups in the length of postoperative hospital stay (p = 0.261), AL (p = 0.186), postoperative intestinal obstruction (p = 1.000), postoperative urinary retention (p = 1.000), urinary incontinence (p = 0.752), or incision infection (p = 1.000). Anal function was evaluated using the Wexner scale 3 months after surgery, and no significant difference was found between the HL (9.58 ±1.52) and LL (9.77 ±1.52) groups (Table 3, Figure 1).
Survival analysis
The median postoperative follow-up time was 75.5 months in the HL group and 80 months in the LL group. At the end of follow-up, there were 10 deaths in the LL group and 12 deaths in the HL group due to recurrence or metastasis of RC. Survival curves showed that the HL and LL groups had no obvious differences in OS (p = 0.573), DFS (p = 0.925), local recurrence (p = 0.952) or distant metastasis (p = 0.852; Figure 2).
Discussion
For the same surgical team, the length of the operation and the amount of intraoperative blood loss can reflect the difficulty of surgery.8 In the HL and LL comparisons, the increased difficulty and time of the LL operation was mainly due to the dissection of the LNs. In HL, the IMA is severed within 2 cm from the abdominal aorta, and then the LNs at the root of the IMA are dissected together. However, in LL, the IMA is dissected from below where the LCA emanates, and then cleared along the area of the IMA root where LNs are enclosed by the IMA, LCA, inferior mesenteric vein (IMV), etc. The technical difficulties in identifying anatomical variants of the IMA and LCA, protection of nerves, performing vascular skeletonization, and dissection of peripheral and dorsal LNs are significantly higher in LL. In addition, LL involves preservation of the LCA. According to previous reports, preserving the LCA has the potential to increase anastomotic tension, and the surgeon tends to free more of the bowel and even the splenic flexure of the colon to provide a longer, movable bowel for anastomosis in order to reduce anastomotic tension. In this study, there were no significant differences in intraoperative blood loss or operation time between the HL and LL groups (p > 0.05). This is likely due to the fact that, in the actual operation process, in order to ensure adequate blood supply of anastomotic stoma, the surgeon often excises more intestinal loops during HL surgery, thus increasing the possibility of having to mobilize the splenic flexure. Furthermore, in our view, in a mature surgical team that completed the surgeon’s learning curve and developed minimally invasive laparoscopic techniques, the different ligation methods will not have an evident impact on the course of the procedure, even if they do have some impact on its difficulty.
The influence on postoperative complications, among which AL is a complication that clinicians pay more attention to, has been analyzed. Once AL occurs, it will not only aggravate the pain and stress of patients, but also increase hospitalization costs and length of stay, and can endanger patients’ lives. There are 2 main factors that affect AL: anastomotic tension and anastomotic blood supply. Low ligation is more likely to increase intraoperative anastomotic tension compared to HL, due to the retention of the LCA. However, surgeons often use various methods during surgery (freeing more intestinal loops or the splenic flexure, diverting ileostomy, minimally invasive liver surgery (MILS), etc.) to ensure patient safety. Other reports9 have indicated that IMA LL can make the anastomosis tension-free in 80% of patients; thus, anastomotic tension is not unavoidable in LL. With regard to the anastomotic blood supply, in HL of the IMA, the supply mainly comes from the left branch of MCA, reaches the left colon through the Riolan’s arch, and finally reaches the anastomotic site. In LL of the IMA, apart from MCA, the retained LCA also provides part of the blood supply. It has been reported that Riolan’s arch is present in less than 50% of the population, and it is more likely to cause vascular stenosis, occlusion or blood flow through longer intestinal canals in elderly men, patients with cardiovascular and cerebrovascular diseases, diabetics, and patients with a lower tumor location.10 For these patients, the risk of insufficient blood supply to the intestinal canal and anastomotic stoma may increase.11, 12 In a study using a Doppler flow analyzer to measure anastomotic blood supply at different IMA ligation levels, it was found that the anastomotic blood supply in the LL group was significantly higher than in the HL group.13 In another study, Park et al. reported similar results.14 In the current paper, there was no statistical difference in AL between the 2 groups (p = 0.186). There were 7 AL patients in the HL group (11.7%) and 3 in the LL group (5.0%). The reason for this result may be that, although IMA HL can cause a decreased anastomotic blood supply, the decreased blood flow is not enough to cause AL. Therefore, the different ligation methods produce no significant differences in low RC AL.
There were statistically significant differences between HL and LL groups in terms of first postoperative flatus (p < 0.05), which is consistent with the fact that a retained LCA can provide more blood supply to the early postoperative bowel and promote the recovery of peristalsis. No significant difference could be found in anal function scores between the 2 groups 3 months after surgery (p > 0.05), indicating that there was no significant difference in intraoperative pelvic floor and anal muscle injury and related nerve injury between the 2 ligation methods. In the 2 groups, we achieved similar results in terms of postoperative urinary retention and urinary incontinence (p > 0.05), which is consistent with an earlier retrospective cohort study.15 Thus, the different ligation levels have no significant effects on urinary function injuries and postoperative recovery.
In RC, LN metastasis notably affects the long-term survival of patients. Complete LN dissection can more accurately determine the stage and prognosis of RC, and can aid in guiding treatment after surgery.16, 17 Therefore, the number of dissected LNs can be used to evaluate the outcome of RC surgery. In this study, there were no statistically significant differences in the number of postoperative LN examinations, OS, DFS, local recurrence, or distant metastasis between the HL and LL groups (p > 0.05). Thus, IMA ligation level has no evident impact on LN dissection and long-term oncology prognosis, and both methods can ensure a good therapeutic effect. These results are consistent with previous studies.18, 19 Although the 2 different ligation methods entail different dissection methods for IMA root LNs, as long as the TME principle is strictly followed and the mesorectum is completely removed, the radical cure in oncology can be achieved.
Although neoadjuvant therapy is considered a vital treatment method for locally advanced RC, patients receiving neoadjuvant therapy were excluded from the current study. According to previous work, neoadjuvant therapy can significantly influence the extent of LN examination, and treatment impact cannot be analyzed reliably. As the number of harvested LNs in the HL and LL groups was a crucial point that needed to be taken into account in this study in order to objectively compare the oncology outcomes of the 2 different approaches, patients receiving neoadjuvant therapy were excluded from the study. In our paper, a notable number of patients who were characterized by a high risk of recurrence did not receive neoadjuvant therapy. This was due to the fact that the patients’ conditions, such as the presence of serious comorbidities, significant bleeding or obstruction, did not allow them to receive standard therapy.
Limitations
The current study has some important limitations. First, as it is a retrospective study, there is inevitably a selection bias, which may reduce the persuasiveness of the findings. To minimize selection bias, we performed PSM on patient baseline information and discussed the results objectively and adequately. Second, the sample size is relatively small and a larger sample size is needed to confirm our results. Third, since most of the patients in this study were elderly, there was no systematic study of the effects on sexual life and reproductive functions. Moreover, as the center was not able to routinely send the third station LN (station 253 LN) separately for examination, a deeper understanding of the root LNs of the IMA was not possible. Therefore, further research on the issue of IMA root LN metastasis is needed.
Conclusions
In conclusion, in low RC LAR, IMA LL not only achieves the same oncological prognosis as HL, but also shows no significant differences in intraoperative and postoperative complications, and is superior to HL in terms of the recovery of early bowel function.
