A systematic review of sacral nerve stimulation for low anterior resection syndrome

AIM

The efficacy of sacral nerve stimulation (SNS) in low anterior resection syndrome (LARS) is largely undocumented. A review of the literature was carried out to study this question.

METHOD

Pubmed, Medline and Cochrane databases were searched for relevant articles up to August 2014. Studies were included if they evaluated the use of SNS following rectal resection and assessed at least one of the following end-points: bowel function, quality of life and ano-neorectal physiology. No restrictions on language or study size were made.

RESULTS

Seven papers were identified including one case report and six prospective case series. These included 43 patients with a median follow-up of 15 months. After peripheral nerve evaluation definitive implantation was carried out in 34 (79.1%) patients. Overall, 32 (94.1%) of the 34 patients experienced improvement of symptoms which, based on intention to treat, was 32/43 (74.4%).

CONCLUSION

The review suggests that SNS for faecal incontinence in LARS has success rates comparable to its use for other forms of faecal incontinence.

Is robotic ventral mesh rectopexy better than laparoscopy in the treatment of rectal prolapse and obstructed defecation? A meta-analysis

Ventral mesh rectopexy is an approach in the treatment of internal and external rectal prolapse and rectocele. Our aim was to assess whether robotic surgery confers any significant advantages over laparoscopy, and the associated complication rate. Two reviewers performed a literature search using MEDLINE and PubMed databases for studies comparing robotic versus laparoscopic surgery. Five prospective, non-randomised studies were identified and included. A total of 244 patients (101 robotic and 143 laparoscopic) were included in the analysis. Operative time was shorter with laparoscopic surgery, mean weighted difference 27.94 [confidence interval (CI) 19.30-36.57; p < 0.00001]. The conversion rate was not significantly different between groups. There was a trend towards a reduction in length of inpatient stay and early post-operative complications in the robotic group; however, these did not reach statistical significance. Recurrence rates were similar between groups (odds ratio 0.91, CI 0.32-2.63; p = 0.87). Functional results were comparable between groups. Early studies show that robotic ventral rectopexy is a safe option compared to the laparoscopic approach, with overall comparable results. There appeared to be a trend towards a reduction in length of inpatient stay and post-operative complications. These perceived benefits may offset the longer operative times and outlay costs. Larger randomised controlled trials are needed to further evaluate clinical value and cost-effectiveness.

Ventral colporectopexy for overt rectal prolapse and obstructed defaecation syndrome: a systematic review

AIM

Laparoscopic ventral rectopexy (VR) with the use of prosthesis has been advocated for both overt rectal prolapse (ORP) and obstructed defaecation syndrome (ODS). The present study reviews the short-term and functional results of laparoscopic VR.

METHOD

A search was performed of MEDLINE, EMBASE, Ovid and Cochrane databases on all studies reporting on VR for ORP, ODS and other anatomical abnormalities of the pelvic floor from 2004 until February 2013. No language restrictions were made. All studies on VR were reviewed systematically. The main outcomes were intra-operative complications, conversion, procedure duration, short-term mortality and morbidity, length of stay, recurrence of ORP, recurrence of anatomical disorder, faecal incontinence and constipation, quality of life (QoL) score and patient satisfaction. Quality assessment and data extraction were performed independently by three observers.

RESULTS

Twenty-three studies including 1460 patients were eligible for analysis. The conversion rate ranged from 0 to 14.3%. No mortality was reported. The immediate postoperative morbidity rate was 8.6%. Length of stay ranged from 1 to 7 days. A significant improvement in constipation and incontinence symptoms was observed in the postoperative period for both ORP and ODS (chi-square test, P < 0.0001).

CONCLUSION

Laparoscopic VR is a safe and effective procedure for ORP and ODS. Longer follow-up is required, and studies comparing VR with standard rectopexy and stapled transanal rectal resection are not yet available.

Complete mesocolic excision in colorectal cancer: a systematic review

AIM

Several studies have suggested an increased lymph node yield, reduced locoregional recurrence and increased disease-free survival after complete mesocolic excision (CME) for colorectal cancer. This review was undertaken to assess the use of CME for colon cancer by evaluating the technique and its clinical outcome.

METHOD

A literature search of publications was performed using PubMed and Medline. Only studies published in English were included. Studies assessed for quality and data were extracted by two independent reviewers. End-points included number of lymph nodes per patient, quality of the plane of mesocolic excision, postoperative mortality and morbidity, 5-year locoregional recurrence and 5-year cancer-specific survival.

RESULTS

There were 34 articles comprising 12 retrospective studies, nine prospective studies and 13 original articles including case series, observational studies and editorials. Of the prospective studies, four reported an increased lymph node harvest and a survival benefit. The others reported an improvement in the quality of the specimen as assessed by histopathological examination. Laparoscopic CME has the same oncological outcome as open surgery but completeness of excision during laparoscopy may be compromised for tumours in the transverse colon.

CONCLUSION

Studies demonstrate that CME removes significantly more tissue around the tumour including maximal lymph node clearance. There is little information on serious adverse events after CME and a long-term survival benefit has not been proved.

Modified right colon inversion technique as a salvage procedure for colorectal or coloanal anastomosis

AIM

A tension-free well vascularized colorectal or coloanal anastomosis is not always possible following rectal or sigmoid resection. The study reports on the short-term and long-term outcome of a modified right colon inversion technique as a means of facilitating a low colorectal or coloanal anastomosis.

METHOD

All patients who underwent right colonic inversion, a modified Deloyers' procedure, were identified retrospectively from the prospective database of the Colorectal Department of the Royal Marsden Hospital from October 2008 to December 2013.

RESULTS

There were 14 (nine male) patients of median age 58.7 (45-75) years. The main indication was extensive diverticular disease (50%) and previous colonic surgery (21.4%). A defunctioning stoma was performed in 64.3% which was reversed in all within 3-6 months. Three (21.4%) patients developed postoperative complications (Clavien-Dindo 1-2) and none required reoperation. The median duration of follow-up was 11 months. One (7.2%) patient had one bowel movement per day, 10 (71.4%) patients had two bowel movements per day and three (21.4%) patients had three per day.

CONCLUSION

The modified right colonic inversion technique is safe and achieves intestinal continuity with a tension-free well vascularized anastomosis. Good function and low morbidity show that the procedure is a credible alternative to ileorectal or ileoanal anastomosis.

Does rectal cancer height influence the oncological outcome?

AIM

The influence of the height of rectal cancer from the anal verge on the oncological outcome is controversial. This study aimed to determine the influence of the height of the tumour on the survival of patients treated in a specialized rectal cancer unit.

METHOD

Patients undergoing surgery for primary rectal cancer from 2006 to 2013 were identified from a prospectively maintained rectal cancer database. Those requiring total or multicompartmental pelvic exenteration were excluded. Low cancer was defined as tumour < 5 cm from the anal verge, as assessed by endoscopy and/or digital rectal examination. The primary outcome was 3-year disease-free survival (DFS).

RESULTS

Of 340 patients, 203 (59.7%) had low cancer. There were 302 (89%) restorative and 38 (11%) nonrestorative procedures. The rate of positive circumferential resection margin was similar for low compared with high cancer (3.4% vs 2.9%, P = 1.0) and for restorative compared with nonrestorative procedures in low cancer only (3.0% and 5.3%, P = 0.619). Low compared with high anterior resection was associated with increased anastomotic leakage (8.5% vs 2.2%, P = 0.023). Three-year DFS was similar for low and high resection (82% vs 86%, P = 0.305) and between restorative vs nonrestorative procedures in low cancer only (88% vs 77%, P = 0.215). In an adjusted model, low height did not lead to worse survival outcome (3-year DFS hazard ratio 0.54, 95% CI 0.24-1.24, P = 0.147).

CONCLUSION

With careful planning and a multidisciplinary approach, equivalent oncological outcome can be achieved for patients with rectal cancer who undergo curative surgery regardless of differences in tumour characteristics, location and operation performed.

A systematic review and meta-analysis evaluating the role of laparoscopic surgical resection of transverse colon tumours

OBJECTIVES

A meta-analysis of published literature comparing outcomes after laparoscopic resection (LR) with open resection (OR) for transverse colon tumours.

METHODS

Medline, PubMed, CINAHL, EMBASE and Cochrane were searched from inception to October 2013. The text words "minimally invasive", "keyhole surgery" and "transverse colon" were used in combination with the medical subject headings "laparoscopy" and "colon cancer". Outcome variables were chosen based upon whether the included articles reported results. A meta-analysis was performed to obtain a summative outcome.

RESULTS

Six comparatives involving 444 patients were analysed. Of them 245 patients were in the LR group and 199 in the OR group. There was a significant increase in operative time in the LR group compared with the OR group [random effects model: SMD = -0.65, 95% CI (-1.01, -0.30), z = -3.60, p < 0.001] but there was significant heterogeneity amongst trials (Q = 15.51, df = 5, p = 0.008, I(2) = 68). There was less blood loss in the LR group [fixed effects model: SMD = 0.70, 95% CI (0.47, 0.93), z = 6.01, p < 0.001] and patients returned to oral diet earlier [random effects model: SMD = 0.78, 95% CI (0.40, 1.16), z = 4.01, p < 0.001] and had a reduced time to functioning bowel [fixed effects model: SMD = 0.86, 95% CI (0.60, 1.11), z = 6.63, p < 0.001]. No difference was seen for overall morbidity (p = 0.76) or mortality (p = 0.58).

CONCLUSIONS

LR of transverse colon tumours is a safe and effective technique. Although there is an increase in operating time, operative and clinical outcomes of intraoperative blood loss and faster recovery are seen with laparoscopic procedures.

The role of epithelial mesenchymal transition and resistance to neoadjuvant therapy in locally advanced rectal cancer

AIM

Nonresponse to neoadjuvant therapy is a significant challenge for clinicians managing solid cancers. This study aimed to determine whether epithelial mesenchymal transition (EMT) was associated with nonresponse to neoadjuvant therapy in patients with locally advanced rectal cancer.

METHOD

Representative tissue specimens from the tumour-invasive front of consecutive patients undergoing resection of rectal cancer from 2009 to 2011 were used. Patients with marked regression to neoadjuvant therapy were classified as responders and the remainder were classified as nonresponders. Markers of EMT included reduced immunohistochemical expression of membranous E-cadherin, increased nuclear beta-catenin expression and tumour budding. In-situ hybridization was used to assess the expression of microRNA-200c (mir200c), an upstream master-regulator of EMT.

RESULTS

Of 103 patients undergoing resection of rectal cancer, 69 received neoadjuvant chemoradiotherapy; 65% of these were nonresponders. Reduced expression of mir200c was significantly associated with a higher T grade. Reduced membranous E-cadherin, increased nuclear beta-catenin and tumour budding individually predicted the presence of extramural vascular invasion. Reduced E-cadherin, nucleic beta-catenin, reduced expression of mir200c and tumour budding were all significantly associated with nonresponse to neoadjuvant therapy (all P < 0.001). Reduced E-cadherin and expression of mir200c were both associated with reduced cancer-specific survival (log-rank P-values 0.036 and 0.009, respectively).

CONCLUSION

Targeted biomarkers of EMT were associated with nonresponse to neoadjuvant therapy and reduced survival in advanced rectal cancer. EMT may provide a practical clinical biomarker and a novel therapeutic target to improve the proportion of patients who respond to neoadjuvant therapy.

Survival outcome of operated and non-operated elderly patients with rectal cancer: A Surveillance, Epidemiology, and End Results analysis

BACKGROUND

With an ageing population, surgery is increasingly offered to elderly patients with rectal cancer, although outcomes for the oldest patients remain poorly defined. This study aimed to determine whether operative intervention improves outcome in elderly patients.

METHOD

Patients aged 18+ years diagnosed with rectal adenocarcinoma between 1998 and 2009 were identified from the Surveillance, Epidemiology, and End Results database. The primary endpoint was adjusted hazard ratios (HR) for 5-year cancer specific survival (CSS); the secondary endpoint was 5-year overall survival (OS).

RESULTS

With increasing age, patients were less likely to undergo surgery, receive a complete stage or receive neoadjuvant radiotherapy. CSS and OS increasingly diverged with age in patients undergoing surgery. Those aged 80+ had reduced CSS compared to those aged 70-79 years (stages I-III, respective adjusted HR 2.14, 1.58, 1.48, all p < 0.001). However, stage II patients aged 80+ treated with resection and neoadjuvant therapy had similar survival to those aged 70-79 years (adjusted HR 1.26, p = 0.149). For only patients aged 80+ years, those treated non-operatively had lower survival than those undergoing surgery, who in turn had the best survival when treated with neoadjuvant radiotherapy (adjusted HR 0.74, p = 0.001).

CONCLUSION

Contrary to common expectation, in patients aged over 80 with rectal cancer, surgery with or without other modalities was associated with better survival than non-operative treatment. Despite selection bias in this observational study, these findings support consideration of maximal therapy regardless of age in selected patients deemed to be fit, since this leads to outcomes equivalent to younger patients.

Prognostic significance of extramural vascular invasion in T4 rectal cancer

AIM

Extramural vascular invasion (EMVI) has been proposed as an adverse prognostic indicator in colorectal cancer, although its use remains both variable and controversial. This study aimed to determine the survival effect of EMVI in T4 rectal cancer.

METHOD

Patients undergoing surgery with curative intent for primary T4 rectal cancer between 1971 and 2011 were included from two prospectively collected rectal cancer databases. The main end-point was 3-year survival.

RESULTS

From 1142 patients undergoing resection of rectal cancer during the study period, 126 (11.0%) had T4 rectal cancer and were included in the study group. Sphincter preservation was performed in 61 (48%) and a pathologically negative resection margin (R0) was achieved in 104 patients (82.5%). EMVI was present in 51 patients (40.5%) and was an independent predictor of positive lymph node status (adjusted odds ratio 2.66, P = 0.013). Considering all patients, EMVI was associated with reduced overall survival (P = 0.007) and disease-free survival (P = 0.002), but not local recurrence-free survival (P = 0.198). In only those undergoing R0 resection, EMVI predicted reduced disease-free survival with positive nodal status (P = 0.021); it did not predict survival with negative nodal status.

CONCLUSION

Overall, EMVI was a strong prognostic marker of survival. However, after complete surgical resection in patients with node-negative disease, EMVI did not predict local recurrence.

A systematic review and meta-analysis of laparoscopic vs open restorative proctocolectomy

AIM

The benefits of a laparoscopic approach to restorative proctocolectomy (RPC) are controversial. The aim of this meta-analysis was to compare the outcome following laparoscopic and open RPC, with particular attention to adverse events and long-term function.

METHOD

A systematic search of the MEDLINE, EMBASE and Ovid databases was performed for studies published until March 2012. The primary end-point was long-term function. Secondary end-points were intra-operative details, short-term postoperative outcome and postoperative adverse events. Weighted mean difference (WMD) and odds ratio (OR) were calculated using fixed/random effect meta-analytic techniques.

RESULTS

The final analysis included 27 comparative studies of 2428 patients, of whom 1097 (45.1%) underwent laparoscopic surgery. A laparoscopic approach was associated with a significantly longer operation time (WMD 70.1 min, P < 0.001), shorter length of hospital stay (WMD -1.00 day, P < 0.001), reduced intra-operative blood loss (WMD -89.10 ml, P < 0.001) and a lower incidence of wound infection (OR 0.60, P < 0.005). No significant differences were observed in the rate of pouch failure. Although there was no significant difference in the number of daily bowel movements (OR 0.04, P = 0.950), laparoscopic surgery led to fewer nocturnal bowel movements (WMD -1.14, P < 0.001) and reduced pad usage during the day (OR 0.22, P < 0.001) and night (OR 0.33, P < 0.001). The post hoc power to detect differences in adverse event rates ranged from 5% to 42%.

CONCLUSION

Laparoscopic and open approaches to RPC produced equivalent adverse event rates and long-term functional results. However, the present evidence is underpowered to detect true differences in adverse event rates.

Comparison of long-term survival outcome of operative vs nonoperative management of recurrent rectal cancer

AIM

Complete surgical resection is considered the best treatment for recurrent rectal cancer (RRC). The aim of the study was to compare survival outcomes from operative and nonoperative patients presenting with RRC.

METHOD

Patients with RRC whose management was discussed by a tertiary referral specialist multidisciplinary team between January 2007 and August 2011 were identified from a prospectively maintained database. The primary end-point was 3-year overall survival.

RESULTS

Of 127 patients with RRC, it was isolated to the pelvis in 105 and associated with distant disease at presentation in 22. From the time of primary surgery to first recurrence, 1-, 3-, 5- and 10-year local recurrence rates were 22%, 72%, 85% and 96%, respectively. The number of operated patients available at 1, 2 and 3 years' follow-up was 53, 34 and 23, respectively. Of 70 patients who underwent pelvic resection for recurrence, 64% received R0, 20% received R1 and 16% received R2 resections. Corresponding 3-year overall survival rates were 69%, 56% and 20% (P=0.011). There was no significant difference in survival between R2 resection and those managed nonoperatively (hazard ratio=1.258; P=0.579). Those undergoing surgery for pelvic recurrence affecting one or more compartments had a worse prognosis than those with single-compartment involvement (hazard ratio=2.640; P=0.027). Three-year local recurrence-free survival was 80% with R0 resection vs 60% with R1 resection.

CONCLUSION

Most recurrences occur within 5 years of primary surgery, although some occur up to 10 years later. R0 resection is the treatment of choice. There was no survival benefit of R2 resection over nonresected recurrences.

Meta-analysis of survival based on resection margin status following surgery for recurrent rectal cancer

AIM

To determine the presence and duration of survival advantages was investigated for resection margin status (R0, R1 or R2) following surgery for locally recurrent rectal cancer (LRRC).

METHOD

A systematic review of the literature was performed for studies comparing resection margin status for LRRC. Weighted mean differences and meta-analysis of hazard ratios were used as a measure of median and overall cumulative survival.

RESULTS

Twenty-two studies were included, providing outcome for 1460 patients undergoing surgery for LRRC. 57% underwent an R0 resection, 25% an R1 resection and 11% an R2 resection. The most commonly performed operations were abdominoperineal excision (35%), exenteration (23%) and anterior resection (21%). The range of median survival per resection margin was R0 28-92 months, R1 12-50 months, R2 6-17 months. Patients undergoing an R0 resection survived on average for 37.6 (95% confidence interval: 23.5-51.7) months longer than those undergoing R1 resection and 53.0 (31.2-74.8) months longer than those undergoing R2 resection. This correlated to a hazard ratio of 2.03 (1.73-2.38) for R0 vs R1 and 3.41 (2.21-5.25) for R0 vs R2. Patients undergoing R1 resection survived on average 13.3 (7.23-19.4) months longer than those undergoing R2 resection [hazard ratio of 1.68 (1.33-2.12)].

CONCLUSION

Patients undergoing R0 resection have the greatest survival advantage following surgery for recurrent rectal cancer. There is a survival advantage for R1 over R2 resection, but there may be no benefit of R2 resection over palliative treatment.

The tumour biology of synchronous and metachronous colorectal liver metastases: a systematic review

Forty to fifty percent of colorectal cancer (CRC) patients develop colorectal liver metastases (CLM) that are either synchronous or metachronous in presentation. Clarifying whether there is a biological difference between the two groups of liver metastases or their primaries could have important clinical implications. A systematic review was performed using the following resources: MEDLINE from PubMed (1950 to present), Embase, Cochrane and the Web of Knowledge. Thirty-one articles met the inclusion criteria. The review demonstrated that the majority of studies found differences in molecular marker expression between colorectal liver metastases and their respective primaries in both the synchronous and metachronous groups. Studies investigating genetic aberrations demonstrated that the majority of changes in the primary tumour were 'maintained' in the colorectal liver metastases. A limited number of studies compared the primary tumours of the synchronous and metachronous groups and generally demonstrated no differences in marker expression. Although there were conflicting results, the colorectal liver metastases in the synchronous and metachronous groups demonstrated some differences in keeping with a more aggressive tumour subtype in the synchronous group. This review suggests that biological differences may exist between the liver metastases of the synchronous and metachronous groups. Whether there are biological differences between the primaries of the synchronous and metachronous groups remains undetermined due to the limited number of studies available. Future research is required to determine whether differences exist between the two groups and should include comparisons of the primary tumours.

Outcome of abdominosacral resection for locally advanced primary and recurrent rectal cancer

Background

The aim was to assess the indications for and outcomes of abdominosacral resection for patients with locally advanced primary and recurrent rectal cancer.

Methods

Consecutive patients undergoing abdominosacral resection between January 2006 and December 2011 were identified from a prospectively maintained database. The main endpoints were 3-year local recurrence-free (LRFS) and disease-free (DFS) survival.

Results

Thirty patients underwent abdominosacral resection, 22 for recurrent rectal cancer and eight for locally advanced primary cancer. Sacrectomy was performed at S1/2 in five, S3 in 11 and S4/5 in 14 patients. R0 resection was achieved in 23 patients; all seven positive margins (R1) were in patients with recurrent disease. There were no deaths in hospital or within 30 days. S1/2 sacrectomy was associated with the highest rate of major complications (60 per cent versus 27 and 29 per cent for S3 and S4/5 respectively) and long-term complications (60, 36 and 14 per cent). Overall 3-year LRFS was 66 per cent and 3-year DFS was 55 per cent. Both were significantly better in patients with negative compared with positive margins (LRFS: 87 versus 0 per cent, P &lt; 0·001; DFS: 71 versus 0 per cent, P = 0·033).

Conclusion

Abdominosacral resection was associated with long-term survival in carefully selected patients with advanced rectal cancer. Postoperative complications were common and often multiple. Sacrectomy for locally advanced primary rectal cancer was associated with a low margin-positive rate and should be considered as an acceptable treatment. Margin-positive resection was associated with poor survival outcomes and should be avoided.

Relevance of magnetic resonance imaging-detected pelvic sidewall lymph node involvement in rectal cancer

BACKGROUND

The significance of magnetic resonance imaging (MRI)-suspected pelvic sidewall (PSW) lymph node involvement in rectal cancer is uncertain.

METHODS

Magnetic resonance images were reviewed retrospectively by specialist gastrointestinal radiologists for the presence of suspicious PSW nodes. Scans and outcome data were from patients with biopsy-proven rectal cancer and a minimum of 5 years' follow-up in the Magnetic Resonance Imaging and Rectal Cancer European Equivalence Study. Overall disease-free survival (DFS) was analysed using the Kaplan-Meier product-limit method and stratified according to preoperative therapy. Binary logistic regression was used to match patients for propensity of clinical and staging characteristics, and further survival analysis was carried out to determine associations between suspicious PSW nodes on MRI and survival outcomes.

RESULTS

Of 325 patients, 38 (11·7 per cent) had MRI-identified suspicious PSW nodes on baseline scans. Such nodes were associated with poor outcomes. Five-year DFS was 42 and 70·7 per cent respectively for patients with, and without suspicious PSW nodes (P < 0·001). Among patients undergoing primary surgery, MRI-suspected PSW node involvement was associated with worse 5-year DFS (31 versus 76·3 per cent; P = 0·001), but the presence of suspicious nodes had no impact on survival among patients who received preoperative therapy. After propensity matching for clinical and tumour characteristics, the presence of suspicious PSW nodes on MRI was not an independent prognostic variable.

CONCLUSION

Patients with suspicious PSW nodes on MRI had significantly worse DFS that appeared improved with the use of preoperative therapy. These nodes were associated with adverse features of the primary tumour and were not an independent prognostic factor.

A systematic review of cancer related patient outcomes after anterior resection and abdominoperineal excision for rectal cancer in the total mesorectal excision era

PURPOSE

It is a widely held view that anterior resection (AR) for rectal cancer is an oncologically superior operation to abdominoperineal excision (APE). However, some centres have demonstrated better outcomes with APE. We conducted a systematic review of high-quality studies within the total mesorectal excision (TME) era comparing outcomes of AR and APE.

METHODS

A literature search was performed to identify studies within the TME era comparing AR and APE with regard to the following: circumferential resection margin (CRM) status, tumour perforation rates, specimen quality, local recurrence, overall survival (OS; 3 or 5 year), cancer-specific survival (CSS) and disease-free survival (DFS). Additional data regarding patient demographics and tumour characteristics was collected.

RESULTS

Twenty four studies fulfilled the eligibility criteria with Newcastle-Ottawa scores of six or greater. Where a significant difference was found, all studies reported lower and more advanced tumours for APE and 4/5 studies observed more frequent use of neoadjuvant and adjuvant therapies in APE patients. Tumour perforation rates and CRM involvement where reported, were significantly greater for APE. 8 out of 10 studies showing significant differences in local recurrence reported higher rates for APE but no differences were observed with distant recurrence. Where differences were noted, AR was reported to have increased DFS, CSS and OS compared to APE.

CONCLUSIONS

Patients treated with AR have lower rates of tumour perforation and CRM involvement and tend to have better outcomes with regard to disease recurrence and survival. However, tumours treated by APE are lower and more locally advanced.

One millimetre is the safe cut-off for magnetic resonance imaging prediction of surgical margin status in rectal cancer

BACKGROUND

A pathologically involved margin in rectal cancer is defined as tumour within 1 mm of the surgical resection margin. There is no standard definition of a predicted safe margin on magnetic resonance imaging (MRI). The aim of this study was to assess which cut-off (1, 2 or 5 mm) was the best predictor of local recurrence based on preoperative MRI assessment of the circumferential resection margin (CRM).

METHODS

Data were collected prospectively on the distance between the tumour and mesorectal fascia for patients with documented radiological margin status in the MERCURY study. Positive margin and local recurrence rates were compared for MRI distances from the tumour to the mesorectal fascia of 1 mm or less, more than 1 mm up to 2 mm, more than 2 mm up to 5 mm, and more than 5 mm. The Cox proportional hazard regression method was used to determine the effect of level of margin involvement on time to local recurrence.

RESULTS

Univariable analysis showed that, relative to a distance measured by MRI of more than 5 mm, the hazard ratio (HR) for local recurrence was 3·90 (95 per cent confidence interval 1·99 to 7·63; P < 0·001) for a margin of 1 mm or less, 0·81 (0·36 to 1·85; P = 0·620) for a margin of more than 1 mm up to 2 mm, and 0·33 (0·10 to 1·08; P = 0·067) for a margin greater than 2 mm up to 5 mm. Multivariable analysis of the effect of MRI distance to the mesorectal fascia and preoperative treatment on local recurrence showed that a margin of 1 mm or less remained significant regardless of preoperative treatment (HR 3·72, 1·43 to 9·71; P = 0·007).

CONCLUSION

For preoperative staging of rectal cancer, the best cut-off distance for predicting CRM involvement using MRI is 1 mm. Using a cut-off greater than this does not appear to identify patients at higher risk of local recurrence.