Question: Does the level of inferior mesenteric artery (IMA) ligation have an impact on disease-free survival (DFS), disease-specific survival (DSS), and recurrence in patients who underwent low anterior resection (LAR) and total mesorectal excision (TME) for rectal cancer?
Findings: This analysis after 3 years of follow-up of patients enrolled in the HIGHLOW Trial (ClinicalTrials.gov Identifier: NCT02153801) shows that the level of IMA ligation during LAR and TME for rectal cancer does not affect DFS, DSS, and recurrence.
Meaning: The results of the present analysis suggest that when an oncologically adequate intrapelvic dissection is performed, the level of ligation of the IMA is not a main determinant for survival among patients submitted to LAR and TME for rectal cancer.
The standard surgical approach for extraperitoneal rectal cancer is low anterior rectal resection (LAR) with total mesorectal excision (TME).1 The level of inferior mesenteric artery (IMA) ligation, however, remains a highly debated issue and has been reported to affect functional outcomes, anastomotic leak rates, and oncological adequacy.2 Several recent publications have compared genitourinary function, bowel function, and anastomotic leak rate in patients undergoing high ligation (HL) or low ligation (LL) of the IMA during LAR + TME3–6 while data on oncological consequences are sparse.7–9 The most recent meta-analysis suggests that there is no statistically significant difference in lymph nodes harvest or 5-year overall survival (OS) between the 2 techniques.10–12 However, the level of evidence is low because all 8 studies included were retrospective cohort studies, and to date, no randomized studies reporting long-term results are available.
In 2019, the HIGHLOW trial was published in Annals of Surgery with the aim to investigate the genitourinary function of patients undergoing LAR + TME according to high or low tie of the IMA.3 Patients in the LL group reported better genitourinary function, while anastomotic leak and postoperative complication rates were similar between groups. There was no difference in surgical surrogates for oncologic adequacy (number of harvested lymph nodes, Quirke score of mesorectal quality, distal, and circumferential margins) between the 2 techniques. Long-term oncological outcomes from the HIGHLOW trial are awaited to assess whether HL and LL are comparable from an oncological point of view.
The aim of this study is to compare the oncological outcomes of the patients enrolled in the HIGHLOW randomized clinical trial (ClinicalTrials.gov Identifier: NCT02153801)3 after 3 years from the conclusion of the trial.
Between June 2014 and December 2016, patients scheduled to undergo elective laparoscopic LAR + TME in 6 Italian nonacademic hospitals in Northern Italy were randomized into 2 groups in the HIGHLOW Trial according to the level of IMA ligation: high versus low tie of IMA. Patients were randomly assigned to HL or LL of the IMA. The random allocation was generated using a computerized randomization system (www.random.org) and was performed using sealed envelopes. In case of LL, preservation of the left colic artery had to be proven by pathological examination, demonstrating the absence of the left colic artery in the specimen of patients undergoing the LL technique. During LL, apical lymph nodes were dissected from the superior aspect of the IMA with a standardized approach that all participating surgeons were trained for, meant to preserve the hypogastric neural plexus on the inferior portion of the origin of the IMA. The operative techniques for HL and LL were previously described. In the LL technique after the pelvic peritoneum is opened below the sacral promontory, the dissection of the peritoneum from the aortic plane goes upward and then laterally toward the sigmoid colon. The left colic artery is identified and preserved. The superior rectal artery is divided. The superior aspect of the IMA is cleared from lymphatic tissue, while the inferior aspect of the IMA and the surrounding hypogastric plexus are left in place (Fig. 1).13 Oncologic adequacy of HL and LL procedures was evaluated by comparing the difference in harvested lymph nodes and Quirke classification of quality of the mesorectum.14
Details of the study design and short-term outcomes have been previously published.3 The primary outcome of the original HIGHLOW trial was to investigate a difference in genitourinary function using questionnaires and objective evaluations. Postoperative complications were assessed using the Clavien–Dindo classification.15 Conversion to open surgery, anastomotic leak, and other short-term outcomes were collected to investigate differences between groups.
All patients were followed every 3 months for the first year and every 6 months thereafter for the next 2 years according to the National Comprehensive Cancer Network Guidelines.11 Patients were followed-up within a multidisciplinary setting involving surgeons, oncologists, radiologists, and pathologists. Patients were given a physical examination every 6 months for a total of 5 years. A chest–abdominal–pelvic computed tomographic (CT) scan was performed every 6 months for 2 years, then once a year for a total of 5 years from surgery. A colonoscopy was performed for each patient within 1 year after surgery. If an advanced adenoma was found, a colonoscopy was repeated within 1 year. If no advanced adenoma was found, a colonoscopy was repeated after 3 years. Carcinoembryonic antigen analysis was performed every 6 months for a total of 5 years. An increment in carcinoembryonic antigen levels was an indication to perform a CT scan outside the ordinary schedule. If metachronous metastases were found at radiological evaluation, a PET/CT scan was performed to confirm resectability and to evaluate the extension of the disease. Data on OS, disease-free survival (DFS), disease-specific survival (DSS), local recurrence (LR), and distant metastasis (DM) were prospectively collected for each patient.
An LR was defined as tumor recurrence within the lower pelvis, a regional recurrence was defined as tumor recurrence within pelvic lymph nodes, and DM was defined as tumor recurrence elsewhere. A recurrence was diagnosed through radiologic imaging. A biopsy was performed when the radiological imaging was doubtful and when it was deemed to be necessary and feasible by the multidisciplinary team. OS was calculated from the date of surgery to the last date of follow-up or death from any cause. DSS was calculated from the date of surgery to the last date of follow-up or death from cancer. DFS was defined as the time elapsed from the date of surgery until a LR or DM occurred.
Analysis was performed according to the intent-to-treat basis. Time-to-event Kaplan–Meier curves were determined for DFS and DSS and compared with the log-rank test. Cox proportional-hazard models were adjusted by tumor location (low, middle, and high rectum), free circumferential margin of at least 1 mm, Clavien–Dindo complication rate15 equal or higher than 3, completeness of the pathological specimen according to Quirke,14 and disease stage. Rates of local, distant, and regional recurrence were presented as frequencies and percentages. In the original HIGHLOW Trial, a sample size of 212 patients enabled a 2-tailed Fisher exact test applied to 2 cohorts of 100 patients each to have 84.45 power in estimating a 20% difference in the incidence of genitourinary dysfunction (α = 0.05, β = 0.1555).3 The level of significance was set at 5% with 2-sided tests. All statistical analyses were performed using the statistical software Statistical Package for the Social Sciences (SPSS) software (version 22, SPSS, Chicago, IL).
One hundred ninety-six patients from the HIGHLOW trial were analyzed. Patients’ characteristics are described in Table 1.
TABLE 1. -
Tumor Characteristics and Outcomes
||Total Patients, n = 196 (100%)
||High Ligation, n = 101 (51.5%)
||Low Ligation, n = 95 (48.5%)
|Neo adjuvant CRT
|Compliance to neo adjuvant CRT
|Compliance to adjuvant chemotherapy
| Site of tumor in the rectum
|Conversion from lap to open
|Clavien–Dindo, CD ≥ 3
|Quirke completeness of TME
| Nearly complete
|Circumferential margin ≤1 mm
|Stage of rectal cancer
| Stage 1
| Stage 2
| Stage 3
CD, Clavien–Dindo; TME, total mesorectal excision.
From the original HIGHLOW study, there were no significant difference in the mean number of harvested lymph nodes between the groups. Apical lymph nodes were collected separately from the specimen in the LL group and within the specimen in the HL group. There were no patients with apical lymph node metastasis in either group. No left colic artery stumps were found in the specimens of all LL patients, demonstrating that the LL technique had been performed correctly. The completeness of the mesorectum according to Quirke classification did not differ and was higher than 90% in both groups. The incidence of postoperative stage III tumors was greater in the HL group.3
Median follow-up for OS was 41.2 (interquartile range [IQR], 35.2–55.6) and 43 (IQR, 33.7–55.2) months while for DFS was 40 (IQR, 28.1–54.7) and 40 (IQR, 30.7–51.1) months in the HL and LL groups, respectively.
The number of patients with more than 3 years of follow up was 134 (68.4%), 31 (15.8%) patients died, and 31 (15.8%) did not complete the surveillance protocol. The 3-year OS rate of HL and LL patients was 83.1% and 85.2% (P = 0.735) and the DFS was 83.8% and 84.2% (P = 0.567), respectively, with no statistical difference (Figs. 1, 2). The 3-year DSS for HL and LL patients was 90.2% and 92% (P = 0.895), respectively, with no statistical difference found between the 2 groups (Figs. 3, 4).
Laparotomic conversion happened in 10 patients (9.5%) in the LL group and in 10 patients (10.5%) in the HL group (P > 0.05). Reasons for conversion were uncontrolled bleeding in 10 patients (50%), tenacious adhesions in 7 patients (35%), and unclear dissecting plane in 3 patients (15%).
A total of 3 LR occurred. There was no statistically significant difference in LR rate between groups (2% HL vs 2.1% LL; P = 0.958). Two cases of LR after HL occurred in patients with an incomplete mesorectal fascia at the pathological examination. In the LL group, LR happened in one patient with an incomplete mesorectum fascia at the pathological examination and in one patient with a nearly complete mesorectum fascia at the pathological examination.
There was no statistically significant difference found in the regional recurrence rate between the 2 groups (3.25% HL vs 2.1% LL; P = 0.719). All regional lymph node recurrences occurred in patients who were stage IIIb, irrespective of the allocation.
No statistically significant difference was found in the distant recurrence rate (13 HL vs 13 LL; P = 0.970) nor in the sites of the recurrences (mainly liver and lung in both groups). In the HL group, distant recurrence occurred in 2 patients with a stage I, in 1 patient with stage II, and in 10 patients with stage III rectal cancer. In the LL group, distant recurrence occurred in 2 patients with stage I, 2 patients with stage II, and 9 patients with stage III rectal cancer. There was no difference in stage-specific recurrence between groups.
Upon univariate analysis, conversion to open surgery (hazard ratio [HR], 2.68; CI 95%, 1.1–6.2; P = 0.037) and stage III disease (HR, 2.5; CI 95%, 1.1–5.7; P = 0.024) were significantly associated with OS. Quirke score incomplete (HR, 4.66; CI 95%, 1.1–19.6; P = 0.0360), adjuvant (HR, 5.54; CI 95%, 2.3–13.3; P < 0.001) and neoadjuvant therapy (HR, 3.27; CI 95%, 1.7–6.4; P < 0.001), stage III disease (HR, 14.3; CI 95%, 5.0–40.9; P < 0.001), middle (HR, 2.21; CI 95%, 1.1–4.3; P = 0.021) and low (HR, 7.66; CI 95%, 1.0–57.6; P = 0.048) rectal cancer, conversion to open (HR, 4.94; CI 95%, 2.4–10.1; P < 0.001), and CD ≥3 (HR, 3.57; CI 95%, 1.67–7.63; P = 0.003) were all significantly associated with DFS upon univariate analysis. The level of ligation was not significantly associated with either OS or DFS.
After multivariate analysis, conversion rate was found to be a significant determinant for DFS (HR, 3.9; CI 95%, 1.85–8.22; P = 0.003), but not for OS (HR, 2.4; CI 95%, 0.9–5.7; P = 0.067)]. At multivariate analysis, stage III rectal cancer retained significance for both OS (HR, 2.32; CI 95%, 1.0–5.38; P = 0.047) and DFS (HR, 12.97; CI 95%, 4.44–37.86; P < 0.001). The level of ligation did not affect OS and DFS significantly after multivariate analysis (Table 2).
TABLE 2. -
Factors Affecting OS and DFS: Multivariate Analysis
||OS HR (CI 95%)
||DFS (CI 95%)
| High vs low
|Conversion from lap to open
| Yes vs no
|Stage of rectal cancer (Ref. Stage 1)
| Stage 3
| Stage 2
DFS, disease-free survival; HR, hazard ratio; IMA, inferior mesenteric artery; OS, overall survival.
The analysis of oncological outcomes after 3 years from the conclusion of the HIGHLOW trial found that the DFS, DSS, and local and distant recurrence rates were not found to be different between patients who had HL or LL of IMA during laparoscopic LAR for extraperitoneal rectal cancer.
Based on our findings, it seems that the main determinant of survival was not the level of vascular ligation, but the quality of the intrapelvic dissection. Indeed, all cases of local relapse occurred in patients who had an incomplete or nearly complete mesorectum excision, according to the Quirke criteria.14 This is in line with the randomized trials comparing open to minimally invasive LAR reporting a 1.8% and 2.1%, and 3.1% and 5.4% for LR rates at 2 years for stage II or III rectal cancer, respectively, with no difference in circumferential margin involvement rate between the 2 techniques.16,17 These results are in concordance with an earlier randomized clinical trial that went unnoticed because the words “high tie” and “low tie” were not mentioned specifically in the title, abstract, or even in the text.18 Indeed, this trial compared the 10-year survival and early complication rates after curative resection for carcinoma of the left colon by “left hemicolectomy” or “left segmental colectomy”, the terms that were in vogue at that time. Of note, the 2 groups were specifically described as undergoing high or low tie of the IMA, respectively. At 12 years of follow up, there was no statistically significant difference in “survival” between the patients in either group irrespective of Dukes stage (the terms “disease-free survival” or “disease-specific survival” had not yet been clearly defined in 1994). The median survival rate was 10 years, with no statistically significant difference between the 2 groups. The only difference noted was increased frequency of bowel movements during the first postoperative year for LCG (or HL), but this difference disappeared after 12 months.
With regard to distant recurrence, previous studies were mainly retrospective. One recent cohort study did not report any statistically significant differences in surgical, histological, short-term, or long-term oncological outcomes between patients treated with either HL or LL.19 Similarly, in a case–control study comparing HL and LL approach, where LL was associated with IMA root lymph node retrieval, there was no statistically significant intergroup difference in recurrence or OS rates. Conversely, the DFS in the LL group was lower than that in the HL group.20 The French randomized study did not individualize LR.18
Matsuda et al5 conducted a randomized clinical trial comparing defecatory function according to whether patients underwent HL or LL during anterior resection for rectal cancer. However, even if this study tended to show that the level of ligation of the IMA did not affect defecatory function or the rate of postoperative complications, this study was underpowered as at 3 months, only 88 of the 100 patients necessary for the power calculation were assessed (high [n = 47], low [n = 41]). These same authors published the oncological follow-up of the same group of patients 2 years later.7 Again, even though the study concluded that the ligation level of the IMA in rectal cancer did not influence the oncological outcomes, these conclusions were not justified according to the methodology used. They advocated large-scale RCTs to conclude this issue7.
Fujii et al6 conducted a randomized trial comparing the anastomotic leak rate between HL and LL. The sample size was tailored on that primary endpoint. The study was stopped prematurely because of slow accrual. However, it did not detect any significant differences in terms of anastomotic leak rate between the 2 groups. Five-year overall survival rate was listed as a secondary endpoint. The trial did not report a statistically significant difference in the 5-year relapse-free survival rate and in the 5 years overall survival rate.6 The authors did not power the study on oncological outcomes nor discuss a possible influence of underpowered analysis in the limitations.
Our protocol specified that all patients should undergo lymph node dissection at the root of the IMA irrespective of the level of ligation. The appropriate extent of lymphadenectomy at the origin of the IMA for lower rectal carcinoma remains an unsolved issue. The rationale in favor or against it has been debated for years.18 Several authors argue that HL alone could adequately control and improve survival for left colonic cancer because the root lymph nodes are involved in more than 10% of cases, the possibility of skip lesions, or control of retrograde lymph flow, improved survival of patients with involvement of more than 5 lymph nodes as well as providing extra length for a tension-free anastomosis.21–23 Those who say root lymph node dissection is not necessary proclaim less postoperative mortality and morbidity with LT and the absence of definitive convincing arguments in favor of longer survival with HL.24 No positive apical lymph nodes were found in either group in our study.3
Conversion to laparotomy was found to be associated with a greater risk of recurrence. This contrasts with the data from a meta-analysis performed in 2017.25 Worse DFS survival in patients where conversion was necessary might be related to the stage of disease more than to conversion itself. Indeed, stage III rectal cancer is associated with worse DFS in itself,17 and 40% of the conversions occurred among patients with stage III disease. The size of the tumor may also be a plausible explanation, all the more that the definition of conversion for specimen extraction has never been formally closed.26
The sample size of the HIGHLOW trial was calculated to investigate functional outcomes rather than oncological outcomes, and this represents the main limitation of the study. The calculation of the sample size necessary to perform a powered study aimed to compare OS, DSS, and DFS among patients undergoing laparoscopic LAR + TME according to HL or LL should be performed with a noninferiority criterion of the LL technique compared to the HL one. The number of patients would therefore grow significantly, making even a multicenter study likely to be excessively long. Despite this, planning an ad hoc randomized study tailored on the oncological outcomes remains the only way to achieve a sufficient statistical probability to adequately compare LL and HL in terms of cancer survival. Despite this, the HIGHLOW trial remains one of the largest studies on HL versus LL of the IMA reporting on oncological outcomes.
Performing the LL technique could become a risk factor for understaging stage III rectal cancer, thus producing a “high-risk LL-related stage II rectal cancer.” Although possible, this eventuality seems unlikely given the similar node counts in the 2 groups and the lymph nodes sampling of the superior aspect of the root of the IMA performed in the LL group.
In conclusion, LL of the IMA does not seem to impair oncological outcomes compared to HL after 3 years from the conclusion of the HIGHLOW trial. The purported advantages of HL have to be weighed against increased morbidity. Data are awaited to confirm these findings when a longer follow-up will be available for analysis.
Because of the sensitive nature of the data collected for this study, requests to access the dataset from qualified researchers trained in human subject confidentiality protocols may be sent to Dr. Giulio Mari.
1. How P, Shihab O, Tekkis P, et al. A systematic review of cancer related patient outcomes after anterior resection and abdominoperineal excision for rectal cancer
in the total mesorectal excision era. Surg Oncol. 2011; 20:e149–e155
2. Nayeri M, Iskander O, Tabchouri N, et al. Low tie compared to high tie vascular ligation of the inferior mesenteric artery in rectal cancer
surgery decreases postoperative complications without affecting overall survival. Anticancer Res. 2019; 39:4363–4370
3. Mari GM, Crippa J, Cocozza E, et al. Low ligation of inferior mesenteric artery in laparoscopic anterior resection for rectal cancer
reduces genitourinary dysfunction: Results from a randomized controlled trial (HIGHLOW Trial). Ann Surg. 2019; 269:1018–1024
4. Wang Q, Zhang C, Zhang H, et al. Effect of ligation level of inferior mesenteric artery on postoperative defecation function in patients with rectal cancer
. Zhonghua Wei Chang Wai Ke Za Zhi. 2015; 18:1132–1024
5. Matsuda K, Hotta T, Takifuji K, et al. Randomized clinical trial of defaecatory function after anterior resection for rectal cancer
with high versus low ligation of the inferior mesenteric artery. Br J Surg. 2015; 102:501–508
6. Fujii S, Ishibe A, Ota M, et al. Randomized clinical trial of high versus low inferior mesenteric artery ligation during anterior resection for rectal cancer
. BJS Open. 2018; 2:195–202
7. Matsuda K, Yokoyama S, Hotta T, et al. Oncological outcomes following rectal cancer
surgery with high or low ligation of the inferior mesenteric artery. Gastrointest Tumors. 2017; 4:45–52
8. Pezim ME, Nicholls RJ. Survival after high or low ligation of the inferior mesenteric artery during curative surgery for rectal cancer
. Ann Surg. 1984; 200:729–733
9. Slanetz CA Jr, Grimson R. Effect of high and intermediate ligation on survival and recurrence rates following curative resection of colorectal cancer. Dis Colon Rectum. 1997; 40:1205–1218. discussion 1218
10. Singh D, Luo J, Liu XT, et al. The long-term survival benefits of high and low ligation of inferior mesenteric artery in colorectal cancer surgery. Medicine (United States). 2017; 94:e8520
11. Yang Y, Wang G, He J, et al. High tie versus low tie of the inferior mesenteric artery in colorectal cancer: A meta-analysis. Int J Surg. 2018; 52:20–24
12. National Comprehensive Cancer Network. Rectal Cancer
/Surveillance, version 1. 2015. www.nccn.org
13. Mari G, Maggioni D, Costanzi A, et al. “High or low inferior mesenteric artery ligation in laparoscopic low anterior resection: study protocol for a randomized controlled trial” (HIGHLOW trial). Trials. 2015; 16:21
14. Quirke P, Dixon MF, Durdey P, et al. Local recurrence of rectal adenocarcinoma due to inadequate surgical resection. Histopathological study of lateral tumour spread and surgical excision. Lancet. 1986; 2:996–999
15. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: Five-year experience. Ann Surg. 2009; 250:187–196
16. Fleshman J, Branda ME, Sargent DJ, et al. Disease-free survival and local recurrence for laparoscopic resection compared with open resection of stage II to III rectal cancer
. Ann Surg. 2019; 269:589–595
17. Stevenson ARL, Solomon MJ, Brown CSB, et al.; Australasian Gastro-Intestinal Trials Group (AGITG) ALaCaRT investigators. Disease-free survival and local recurrence after laparoscopic-assisted resection or open resection for rectal cancer
: the Australasian Laparoscopic Cancer of the Rectum Randomized Clinical Trial. Ann Surg. 2019; 269:596–602
18. Rouffet F, Hay JM, Vacher B, et al. Curative resection for left colonic carcinoma: hemicolectomy vs. segmental colectomy. A prospective, controlled, multicenter trial. French Association for Surgical Research. Dis Colon Rectum. 1994; 37:651–659
19. Dimitriou N, Felekouras E, Karavokyros I, et al. High versus low ligation of inferior mesenteric vessels in rectal cancer
surgery: a retrospective cohort study. J BUON. 2018; 23:1350–1361
20. Kim CS, Kim S. Oncologic and anastomotic safety of low ligation of the inferior mesenteric artery with additional lymph node retrieval: a case-control study. Ann Coloproctol. 2019; 35:167–173
21. Kanemitsu Y, Hirai T, Komori K, et al. Survival benefit of high ligation of the inferior mesenteric artery in sigmoid colon or rectal cancer
surgery. Br J Surg. 2006; 93:609–615
22. Titu LV, Tweedle E, Rooney PS. High tie of the inferior mesenteric artery in curative surgery for left colonic and rectal cancers: a systematic review. Dig Surg. 2008; 25:148–157
23. Bonnet S, Berger A, Hentati N, et al. High tie versus low tie vascular ligation of the inferior mesenteric artery in colorectal cancer surgery: impact on the gain in colon length and implications on the feasibility of anastomoses. Dis Colon Rectum. 2012; 55:515–521
24. Uehara K, Yamamoto S, Fujita S, et al. Impact of upward lymph node dissection on survival rates in advanced lower rectal carcinoma. Dig Surg. 2007; 24:375–381
25. Gouvas N, Georgiou PA, Agalianos C, et al. Does conversion to open of laparoscopically attempted rectal cancer
cases affect short- and long-term outcomes? A systematic review and meta-analysis. J Laparoendosc Adv Surg Tech A. 2018; 28:117–126
26. Bucher P, Wutrich P, Pugin F, et al. Totally intracorporeal laparoscopic colorectal anastomosis using circular stapler. Surg Endosc. 2008; 22:1278–1282