doi:10.1111/codi.12381
Original article
Completion surgery following transanal endoscopic microsurgery: assessment of quality and short- and long-term outcome R. Hompes*, R. McDonald*, C. Buskens†, I. Lindsey*, N. Armitage‡, J. Hill§, A. Scott¶, N. J. Mortensen* and C. Cunningham* on behalf of the Association of Coloproctology of Great Britain and Ireland Transanal Endoscopic Microsurgery Collaboration *Department of Colorectal Surgery, Oxford University Hospitals, Oxford, UK, †Department of Surgery, Academic Medical Center, Amsterdam, The Netherlands, ‡Department of Colorectal Surgery, Nottingham University Hospitals, Nottingham, UK, §Department of Colorectal Surgery, Central Manchester University Hospitals, Manchester, UK and ¶Department of Colorectal Surgery, University Hospitals of Leicester, Leicester, UK Received 9 January 2013; accepted 21 April 2013; Accepted Article online 12 August 2013
Abstract Aim Patients with unfavourable pathology after transanal endoscopic microsurgery (TEM) should be offered completion surgery (CS) if appropriate. The aim of this retrospective cohort study was to assess the short-term outcome and long-term oncological results of CS and identify factors compromising the quality of resection specimens. Method Data were retrieved and analysed on patients who underwent CS from a comprehensive national TEM database (1992–2008) and the institutional prospective database from the Oxford University Hospitals (2008–2011). Results There were 36 patients eligible for analysis. Postoperative complications occurred in 19 and were minor (grade I–II) in 13 and major (grade III–V) in six patients. The quality of the resected specimen was graded as good in 23 (64%), moderate in six (16.6%) and poor in seven (19.4%). Full-thickness excision by TEM (P = 0.03), an interval to CS greater than 7 weeks (P = 0.05) and distally located lesions (P = 0.04) were associated with increased risk for an inferior surgical specimen. Overall survival after CS was
Introduction Rectal cancer management has been transformed in the last two decades, with emphasis on a detailed understanding of pelvic anatomy, meticulous surgery and the targeted use of chemoradiation [1]. More recently the inherent drawbacks of radical surgery including morbidity, mortalCorrespondence to: Roel Hompes, Department of Colorectal Surgery, Oxford University Hospitals, Old Road, Headington, Oxford OX3 9DU, UK. E-mail: [email protected]
e576
91% at 1 year and 83% at 5 years. Patients with a ‘good’ TME specimen had significantly improved disease-free survival compared with patients with an ‘inferior’ specimen (100 vs 51%, P = 0.001). Conclusion Patients having full-thickness TEM excision, distally placed lesions and a long interval (> 7 weeks) to CS were likely to have an inferior TME specimen. The results confirm that CS after TEM does not negatively influence local recurrence and survival, but the reduced disease-free survival in patients with an inferior specimen is of concern. Keywords Cancer, transanal endoscopic microsurgery (TEM), rectum, completion surgery, recurrence What does this paper add to the literature? Reports of completion surgery after transanal endoscopic microsurgery are infrequent and predominantly focus on oncological outcome. We report on the feasibility and quality of total mesorectal excision surgery after transanal endoscopic microsurgery. Prognostic factors that have an impact on the difficulty of completion surgery are described and the impact on long-term outcome is explored.
ity and poor function and the growing numbers of patients with early rectal cancer coming through the national screening programmes has stimulated interest in surgeons and patients to consider local excision more often [2–5]. Transanal endoscopic microsurgery (TEM) has proved to be safe and effective for lesions throughout the rectum [6]. Any method of local excision inevitably involves a balance between oncological and clinical factors, the latter including avoidance of peri-operative
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
R. Hompes et al.
morbidity, long-term functional sequelae and, in some patients, a permanent stoma. In order to minimize oncological compromise, in particular local failure, patients with adverse pathological features following local excision should have recourse to completion surgery (CS). Completion surgery after local excision for rectal cancer does not appear to compromise oncological outcome compared with radical surgery performed as a primary treatment [7–11]. However, it is unknown to what extent a previous local excision by TEM disrupts the anatomy of the tissue planes and the implications for the quality and outcome of CS. This study aimed to describe the outcome of patients undergoing CS after TEM and to identify factors compromising the quality of resection specimens. We also describe short-term morbidity and long-term oncological outcome.
Method Definition of completion surgery
We have used the term ‘completion surgery’ (CS) to define surgery carried out after local excision to complete surgical treatment of the primary tumour. This is applied to patients with an inadequate or unclear resection margin after local excision, unfavourable pathology according to current standards and patients with a low risk cancer who still wish to proceed to radical surgery after counselling. Data collection
Data were retrieved from a comprehensive national TEM database (http://www.temsurgery.co.uk). This was supplemented by a comprehensive chart review by one of the authors who visited all centres willing to participate in the study. Additional patients from one participating centre (Oxford) were included in the analysis and data were gathered from the institution’s prospective TEM database. Thirty-six patients were eligible for analysis. Of the patient cohort (n = 487) within the UK TEM database (1992–2008), comprehensive data on CS could be gathered on 29 (46%) out of a total of 63 patients undergoing CS. For the remaining 34 patients, no detailed surgical information (no detailed operation note available) and no pathological information (no information regarding the quality of the resection specimen) could be retrieved and these patients were excluded. This was mainly due to CS being undertaken in a centre other than that where the TEM was performed. An additional seven
Completion surgery after transanal endoscopic microsurgery
patients were identified in the Oxford patient cohort (2008–2011). Other data collected included details on the completion surgical procedure, morbidity and mortality according to the Clavien–Dindo classification, hospital stay and histopathological characteristics of the specimen. The long-term oncological outcome was also analysed. TEM technique and CS (total mesorectal excision)
The standard Richard Wolf (Knittlingen, Germany) equipment was used for TEM [7,12]. The type of CS (anterior resection or abdominoperineal excision) was selected on the basis of the tumour location, the appearance of the previous resection site and the surgeon’s preference. In general, CS was performed when the TEM site had completely healed, confirmed by digital rectal examination and luminal assessment, although sometimes surgery within 2 or 3 weeks of local excision was carried out. An open or laparoscopic approach was undertaken according to the surgeon’s preference. Based on subjective assessment of the operation note by two of the authors blinded to other variables, the procedure was labelled as ‘standard’ (no difficulty at any stage of the operation) or ‘difficult’ (difficulty as specified in the operation note). The quality of mesorectal excision was assessed according to definitions used in the MRC CR07 trial [13]. Follow-up
Follow-up was carried out for all patients at the time of data collection. Patients were followed according to local policy or National Institute for Health and Care Excellence guidelines on rectal cancer. Data were often recorded by specialist colorectal cancer nurses. Statistics
Statistical analysis was performed using PASW STATISTICS for Windows (version 18, IBM Corporation, Armonk, NY, USA). Continuous data were given as mean standard deviation or as median and range for non-parametric data. Categorical data were given as frequency or percentage. For dichotomous outcomes, groups were compared by the v2 or Fisher’s exact test. The t test was used to compare means for parametric data and the Mann– Whitney U test was used for continuous, not non-parametric, data. The cumulative survival rates were calculated using the Kaplan–Meier method, and differences in relapse rates were analysed by the log-rank test. A P value of ≤ 0.05 was considered to be statistically significant.
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
e577
R. Hompes et al.
Completion surgery after transanal endoscopic microsurgery
Table 2 Completion surgery characteristics in 36 patients.
Results Patient and tumour/TEM characteristics
The clinico-pathological characteristics of the 36 patients are summarized in Table 1. The indications for CS were adenocarcinoma (11) and adenoma (25). No patient with adenocarcinoma had received any form of chemoradiotherapy. A full-thickness excision was performed in all malignant lesions with sampling of the mesorectum in one case. For the 25 presumed benign lesions a full-thickness excision was performed in 14, and a partial thickness and a combination of partial and full thickness was carried out in seven and four patients. The final histology of the TEM specimen revealed one (3%) patient with a large severely dysplastic adenoma, 16 (43%) with a pT1 lesion and 19 (54%) with a more advanced lesion. Completion surgery
Based on the current guidelines proposed in the UK and USA, two patients had CS for an early rectal cancer that had undergone a potentially curative local excision, a so-called low risk tumour [7,14]. The other pT1 tumours all contained one or more accepted high risk features and these patients were counselled to have completion major surgery, as were the patients with a pT2–3 tumour. For the single patient with a severely dysplastic adenoma, CS was performed owing to the possibility of malignant invasion (Table 2). In 31 (86%) patients a sphincter-preserving procedure was performed, with a Table 1 Patient and initial tumour characteristics at TEM in 36 patients subsequently requiring completion surgery. Sex, M:F Age, years* BMI, kg/m2* Tumour distance from anal verge, cm* Tumour location (cm) Lower rectum (0–6) Mid rectum (7–11) Upper rectum (12–16) Tumour surface, cm2* Specimen surface, cm2* Tumour stage after TEM T0 T1 T2 T3
19:17 64 (39–84) 27 (21–40) 5 (14%) 9 25 6 7.2 18
(2–15) (69%) (17%) (0.5–57) (3.1–142.5)
1 16 12 7
(3%) (43%) (33%) (21%)
TEM, transanal endoscopic microsurgery; BMI, body mass index. *Value or median number with range in parentheses.
e578
Interval (TEM to CS), months* Procedures Ultra-low AR Open AR Laparoscopic AR APE Hartmann Operation time, min* Blood loss, ml* Morbidity/mortality (Dindo–Clavien) Minor (I–II) Major (III–IV) Mortality (V) Hospital stay, days* Final staging after CS† Stage I Stage II Stage IIIa Stage IIIb Stage IIIc
2 (0.5–8.7) 1 18 12 4 1 200 (120–360) 300 (120–1800) 13 5 1 10 (6–76) 19 4 5 7 0
TEM, transanal endoscopic microsurgery; CS, completion surgery; AR, anterior resection; APE, abdominoperineal excision. *Values are median number with ranges in parentheses. †American Joint Committee on Cancer Staging system.
defunctioning loop ileostomy in 25/31. In one patient severe intra-operative bleeding led to conversion of an open low anterior resection to a low Hartmann’s procedure. The procedure was graded as ‘difficult’ in 53% (19/ 36) of cases, mostly due to bleeding and/or fibrosis/ adhesions at the TEM site. Operation time and total blood loss were significantly higher in this group of patients compared with the ‘standard’ cases (230.3 62.3 min vs 187.5 27.2 min, P = 0.02; and 721.9 562.4 ml vs 286.9 139.7 ml, P = 0.008, respectively). Three (9.7%) patients developed a clinically significant postoperative anastomotic leak and underwent Hartmann’s procedure. Two patients returned to theatre for postoperative bleeding and revision of a retracted ileostomy. The median hospital stay was 10 (6–76) days and six patients were readmitted within 30 days for pulmonary embolus (n = 1), small bowel obstruction (n = 3) and stoma related problems (n = 2). The final histology report of the total mesorectal excision specimen showed residual tumour in 17 (47%) patients. These included seven with residual tumour and involved lymph nodes, five with isolated tumour and five with isolated lymph node involvement. In all but two patients a microscopically clear excision margin resection (R0) was obtained. A total mesorectal excision Grade 3 specimen (good) was achieved in 23 (64%) patients, a Grade 2 specimen in six (16.6%) and a Grade 1 specimen (poor) in seven (19.4%) Although there was
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
R. Hompes et al.
Completion surgery after transanal endoscopic microsurgery
no correlation between grade of specimen and the type of resection, specimens with a lower grade were more frequently found after full-thickness TEM, in distal lesions and after an interval from TEM to CS of more than 7 weeks (Table 3). All 13 ‘inferior’ specimens were in the ‘difficult’ surgery category. Outcome
Adjuvant chemotherapy or radiotherapy was administered in 11 (30.5%) patients after CS. The overall surTable 3 Patient, tumour and procedure related characteristics.
Clinico-pathological parameters Age (mean, years)* 64.9 10.7 Sex Male 11 Female 12 Location lesion Anterior 5 Posterior 7 Lateral 11 Anterior vs posterior 5 vs 18 and lateral Distance anus Upper 4 Mid 17 Lower 2 Lower vs upper and 2 vs 21 mid TEM thickness Partial 7 Full 16 Time to surgery 9.5 7.2 (mean, weeks)* ≤ 7 weeks 12 > 7 weeks 11 Closure TEM defect No 14 Yes 9 TEM complication No 18 Yes 5 Type of surgery AR 22 APE 1 Hartmann 0
Inferior specimen (n = 13)
P value
62.8 14.2
0.6
Discussion 8 5 6 3 4 6 vs 7
2 7 4 4 vs 9
0.4
0.1
0.07
Large data sets on CS after TEM are lacking and the outcome focuses predominantly on survival with acceptable local/distant recurrence rates [7–10]. Anecdotal evidence suggests that this type of surgery is difficult and may result in a higher permanent stoma rate [15,16]. The data presented here show that CS after TEM is feasible with acceptable postoperative morbidity and low mortality. In addition, the reported leak rate of under 10% is comparable with that reported for anterior
0.2
1.0
0.04
0 13 11.7 9.5
0.03
3 10
0.05
10 3
0.3
11 2
0.6
9 3 1
0.08
0.2
Probability of disease-free survival
Good specimen (n = 23)
vival at 1 year was 91 and 83% at 5 years. After a median follow-up of 49.2 (3–137) months, the relapse rate was 16.7% (6/36) with 1- and 5-year disease-specific survival rates of 91 and 74%. Only one patient (3%) developed local recurrence (at 39 months after CS) and five (14%) developed distant metastases without signs of local recurrence at a mean follow-up of 33.2 months. Patients with a ‘good’ total mesorectal excision specimen had significantly improved 5-year disease-free survival compared with patients with an ‘inferior’ specimen (100 vs 51%, P = 0.001, log-rank test) (Fig. 1). Although the number of patients was too small to perform a multivariate analysis, no significant correlation between other prognostic parameters (e.g. pT, pN and Dukes staging) and reduced disease-free survival could be demonstrated.
0.8
0.6
0.4
0.2
0.0
TEM, transanal endoscopic microsurgery; AR, anterior resection; APE, abdominoperineal excision. Bold font denotes significant values. *Values are mean SD.
0.00
2.00
4.00 6.00 8.00 Follow up in years
10.00 12.00
Figure 1 Kaplan–Meier curves of 36 patients having completion surgery after transanal endoscopic microsurgery. There is a significant difference in disease-free survival between patients with ‘good’ (solid line) or ‘inferior’ (dotted line) resection specimens (100 vs 51%, P = 0.001, log-rank test).
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
e579
Completion surgery after transanal endoscopic microsurgery
resections [17]. In all but one patient the procedure could be completed according to the preoperative plan and in the majority (86%) of patients a sphincterpreserving procedure was achieved. In over half of the cases the dissection was difficult owing to the disruption of the normal tissue planes. This is reflected by the significantly higher intra-operative blood loss and operating time. Platell et al. [15] found that, in patients proceeding to a more radical procedure within 2–6 weeks after TEM, disruption of the normal pelvic dissection planes added significantly to the technical difficulty. The compromised mesorectal plane and the weakened rectal wall also increase the risk of disruption of the surgical specimen. Similar to our results, Levic et al. [8] reported an intra-operative rectal perforation rate of 20%, all at the previous TEM site. Thus, violation of the surgical planes not only potentially renders radical surgery after TEM more difficult but also significantly increases the risk of producing an inferior specimen. However, these risks have to be considered against the potential benefits of full-thickness TEM as a means of offering cure for early stage rectal neoplasia compared with submucosal excision. The reasons for this decreased survival remain speculative. Unfortunately, the patient group was too small to perform a multivariate analysis, but no association with T-stage and N-stage could be demonstrated. The strong correlation between survival and an inferior specimen suggests a potential direct oncological effect related to difficult surgery (e.g. increased blood loss: 0.7 l vs 0.3 l, increased tumour spill and circulating tumour cells, decreased immune system). Overall, the long-term outcome of patients after CS is acceptable in this population, with a 1-year and 5-year disease-free survival of 91 and 74% respectively. It is striking that only one patient developed local recurrence and that almost all patients with disease relapse had metastatic disease. Doornebosch et al. [18] described a similar outcome in patients following salvage surgery for local recurrence after TEM for T1 rectal cancers; only one patient had a further local recurrence but seven had distant metastases. They speculated that these patients represent a different biological group and that salvage therapy should be intensified with the inclusion of neoadjuvant radiotherapy and/or adjuvant chemotherapy in an effort to improve outcomes. In the study by Levic et al. [8] one patient developed distal metastasis and no local recurrence was observed in 25 patients, although the median follow-up period of 25 months was short. Various factors were analysed that may contribute to difficult surgery and an inferior resection specimen. As expected, an association between a full-thickness TEM
e580
R. Hompes et al.
and an inferior total mesorectal excision specimen could be demonstrated. The same association was found for distally located lesions with a similar tendency towards an increased chance of obtaining an inferior specimen in patients with an anterior lesion. This is in keeping with the anatomical characteristics of the distal mesorectum, described as the bare area by Morson et al. [19]. The mesorectal fat peters out distally as it does for anterior located lesions. These observations suggest that there might be a greater need for abdominoperineal excision after TEM of a tumour in the lower third of the rectum. This might reduce the chance of a poor or perforated surgical specimen. An alternative approach to avoid a difficult dissection from the abdomen is to attempt a perianal disconnection via a lower third intersphincteric dissection. Clearly, whatever approach is chosen, potentially difficult surgery should be anticipated and an experienced surgical team should be available. The risk of converting a potential sphincter-preserving procedure (as primary treatment) into an abdominoperineal excision needs to be part of the discussion with the patient before considering TEM. There is no consensus regarding the timing of CS. Examining the impact of various TEM to CS intervals, we found that an interval longer than 7 weeks significantly increased the risk for an inferior specimen. The cut-off point at 7 weeks is more likely to reflect infection or dehiscence at the TEM site. One of the weaknesses of this study is the inevitable selection bias of the study population. Only 36 patients were eligible for analysis out of a potential study population of 70 patients. The retrospective nature of the study with strict inclusion criteria led to the exclusion of 34 patients. Our findings are thus based on a small group of patients and there is therefore a possibility of type II error. Furthermore the absence of a common protocol for the management of these patients and the impact of surgical experience (with TEM and CS) of each centre will need to be considered when interpreting the results. While these shortcomings may hamper strong conclusions, this series is one of the largest series available and helps to identify risk factors for less favourable outcomes after CS following TEM. Transanal endoscopic microsurgery has been a relatively niche subspecialty in colorectal surgery; however, the increasing incidence of early stage rectal cancer and the greater availability of transanal minimally invasive surgery are likely to see it enter mainstream practice over the next few years as clinicians and patients opt for organ preservation in favour of radical surgery. In addition, various national randomized controlled trials comparing radical surgery with local excision for rectal
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
R. Hompes et al.
cancer will also increase patient numbers (TREC trial, UK; CARTS study, The Netherlands; GRECCAR 2 trial, France). This should be good news for patients, but one must be cognizant of the pitfalls and challenges of local excision. These approaches will need to be part of a strategy for early rectal cancer management that include CS, careful surveillance and salvage surgery along with the use of neoadjuvant and other adjuvant therapy. In conclusion, CS after TEM is feasible in most patients without significant morbidity and acceptable permanent stoma rates. Our results confirm those of other studies that CS after TEM does not negatively influence local recurrence and survival, but the reduced disease-free survival in patients with an inferior specimen is of concern. Completion surgery after fullthickness TEM excision within the distal and/or anterior rectum deserves special consideration. In this group there may be a role for neoadjuvant and adjuvant treatment strategies combined with CS in an effort to improve survival.
Author contributions The concept for this paper was designed by the lead author and C. Cunningham. Analysis of the data was performed by C. Buskens and interpretation thereafter was done by the lead author, C. Cunningham and C. Buskens. The other authors were crucial for accurate and detailed data collection and on revision made suggestions regarding the interpretation of the data. The main draft was written by the lead author with major input from C. Cunningham. All other authors critically revised the content and approved the final version of the paper.
Funding None declared.
References 1 Kapiteijn E, Marijnen CA, Nagtegaal ID et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 2001; 345: 638–46. 2 Nesbakken A, Nygaard K, Bull-Njaa T, Carlsen E, Eri LM. Bladder and sexual dysfunction after mesorectal excision for rectal cancer. Br J Surg 2000; 87: 206–10. 3 You YN, Baxter NN, Stewart A, Nelson H. Is the increasing rate of local excision for stage I rectal cancer in the United States justified?: a nationwide cohort study from the National Cancer Database. Ann Surg 2007; 245: 726–33.
Completion surgery after transanal endoscopic microsurgery
4 Hompes R, Cunningham C. Extending the role of transanal endoscopic microsurgery (TEM) in rectal cancer. Colorectal Dis 2011; 13(Suppl 7): 32–6. 5 Lezoche G, Guerrieri M, Baldarelli M et al. Transanal endoscopic microsurgery for 135 patients with small nonadvanced low rectal cancer (iT1–iT2, iN0): short- and long-term results. Surg Endosc 2011; 25: 1222–9. 6 de Graaf EJ, Doornebosch PG, Tetteroo GW, Geldof H, Hop WC. Transanal endoscopic microsurgery is feasible for adenomas throughout the entire rectum: a prospective study. Dis Colon Rectum 2009; 52: 1107–13. 7 Bach SP, Hill J, Monson JR et al. A predictive model for local recurrence after transanal endoscopic microsurgery for rectal cancer. Br J Surg 2009; 96: 280–90. 8 Levic K, Bulut O, Hesselfeldt P, Bulow S. The outcome of rectal cancer after early salvage surgery following transanal endoscopic microsurgery seems promising. Dan Med J 2012; 59: A4507. 9 Hahnloser D, Wolff BG, Larson DW, Ping J, Nivatvongs S. Immediate radical resection after local excision of rectal cancer: an oncologic compromise? Dis Colon Rectum 2005; 48: 429–37. 10 Borschitz T, Heintz A, Junginger T. The influence of histopathologic criteria on the long-term prognosis of locally excised pT1 rectal carcinomas: results of local excision (transanal endoscopic microsurgery) and immediate reoperation. Dis Colon Rectum 2006; 49: 1492–506. 11 Lee W, Lee D, Choi S, Chun H. Transanal endoscopic microsurgery and radical surgery for T1 and T2 rectal cancer. Surg Endosc 2003; 17: 1283–7. 12 Bretagnol F, Merrie A, George B, Warren BF, Mortensen NJ. Local excision of rectal tumours by transanal endoscopic microsurgery. Br J Surg 2007; 94: 627–33. 13 Quirke P, Steele R, Monson J et al. Effect of the plane of surgery achieved on local recurrence in patients with operable rectal cancer: a prospective study using data from the MRC CR07 and NCIC-CTG CO16 randomised clinical trial. Lancet 2009; 373: 821–8. 14 Mellgren A, Goldberg S, Rothenberger DA. Local excision: some reality testing. Surg Oncol Clin N Am 2005; 14: 183–96. 15 Platell C. Transanal endoscopic microsurgery. ANZ J Surg 2009; 79: 275–80. 16 Bujko K, Richter P, Kolodziejczyk M et al. Preoperative radiotherapy and local excision of rectal cancer with immediate radical re-operation for poor responders. Radiother Oncol 2009; 92: 195–201. 17 Paun BC, Cassie S, MacLean AR, Dixon E, Buie WD. Postoperative complications following surgery for rectal cancer. Ann Surg 2010; 251: 807–18. 18 Doornebosch PG, Ferenschild FT, de Wilt JH, Dawson I, Tetteroo GW, de Graaf EJ. Treatment of recurrence after transanal endoscopic microsurgery (TEM) for T1 rectal cancer. Dis Colon Rectum 2010; 53: 1234–9. 19 Morson BC, Vaughan EG, Bussey HJ. Pelvic recurrence after excision of rectum for carcinoma. Br Med J 1963; 2: 13–8.
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
e581
Original article
Completion surgery following transanal endoscopic microsurgery: assessment of quality and short- and long-term outcome R. Hompes*, R. McDonald*, C. Buskens†, I. Lindsey*, N. Armitage‡, J. Hill§, A. Scott¶, N. J. Mortensen* and C. Cunningham* on behalf of the Association of Coloproctology of Great Britain and Ireland Transanal Endoscopic Microsurgery Collaboration *Department of Colorectal Surgery, Oxford University Hospitals, Oxford, UK, †Department of Surgery, Academic Medical Center, Amsterdam, The Netherlands, ‡Department of Colorectal Surgery, Nottingham University Hospitals, Nottingham, UK, §Department of Colorectal Surgery, Central Manchester University Hospitals, Manchester, UK and ¶Department of Colorectal Surgery, University Hospitals of Leicester, Leicester, UK Received 9 January 2013; accepted 21 April 2013; Accepted Article online 12 August 2013
Abstract Aim Patients with unfavourable pathology after transanal endoscopic microsurgery (TEM) should be offered completion surgery (CS) if appropriate. The aim of this retrospective cohort study was to assess the short-term outcome and long-term oncological results of CS and identify factors compromising the quality of resection specimens. Method Data were retrieved and analysed on patients who underwent CS from a comprehensive national TEM database (1992–2008) and the institutional prospective database from the Oxford University Hospitals (2008–2011). Results There were 36 patients eligible for analysis. Postoperative complications occurred in 19 and were minor (grade I–II) in 13 and major (grade III–V) in six patients. The quality of the resected specimen was graded as good in 23 (64%), moderate in six (16.6%) and poor in seven (19.4%). Full-thickness excision by TEM (P = 0.03), an interval to CS greater than 7 weeks (P = 0.05) and distally located lesions (P = 0.04) were associated with increased risk for an inferior surgical specimen. Overall survival after CS was
Introduction Rectal cancer management has been transformed in the last two decades, with emphasis on a detailed understanding of pelvic anatomy, meticulous surgery and the targeted use of chemoradiation [1]. More recently the inherent drawbacks of radical surgery including morbidity, mortalCorrespondence to: Roel Hompes, Department of Colorectal Surgery, Oxford University Hospitals, Old Road, Headington, Oxford OX3 9DU, UK. E-mail: [email protected]
e576
91% at 1 year and 83% at 5 years. Patients with a ‘good’ TME specimen had significantly improved disease-free survival compared with patients with an ‘inferior’ specimen (100 vs 51%, P = 0.001). Conclusion Patients having full-thickness TEM excision, distally placed lesions and a long interval (> 7 weeks) to CS were likely to have an inferior TME specimen. The results confirm that CS after TEM does not negatively influence local recurrence and survival, but the reduced disease-free survival in patients with an inferior specimen is of concern. Keywords Cancer, transanal endoscopic microsurgery (TEM), rectum, completion surgery, recurrence What does this paper add to the literature? Reports of completion surgery after transanal endoscopic microsurgery are infrequent and predominantly focus on oncological outcome. We report on the feasibility and quality of total mesorectal excision surgery after transanal endoscopic microsurgery. Prognostic factors that have an impact on the difficulty of completion surgery are described and the impact on long-term outcome is explored.
ity and poor function and the growing numbers of patients with early rectal cancer coming through the national screening programmes has stimulated interest in surgeons and patients to consider local excision more often [2–5]. Transanal endoscopic microsurgery (TEM) has proved to be safe and effective for lesions throughout the rectum [6]. Any method of local excision inevitably involves a balance between oncological and clinical factors, the latter including avoidance of peri-operative
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
R. Hompes et al.
morbidity, long-term functional sequelae and, in some patients, a permanent stoma. In order to minimize oncological compromise, in particular local failure, patients with adverse pathological features following local excision should have recourse to completion surgery (CS). Completion surgery after local excision for rectal cancer does not appear to compromise oncological outcome compared with radical surgery performed as a primary treatment [7–11]. However, it is unknown to what extent a previous local excision by TEM disrupts the anatomy of the tissue planes and the implications for the quality and outcome of CS. This study aimed to describe the outcome of patients undergoing CS after TEM and to identify factors compromising the quality of resection specimens. We also describe short-term morbidity and long-term oncological outcome.
Method Definition of completion surgery
We have used the term ‘completion surgery’ (CS) to define surgery carried out after local excision to complete surgical treatment of the primary tumour. This is applied to patients with an inadequate or unclear resection margin after local excision, unfavourable pathology according to current standards and patients with a low risk cancer who still wish to proceed to radical surgery after counselling. Data collection
Data were retrieved from a comprehensive national TEM database (http://www.temsurgery.co.uk). This was supplemented by a comprehensive chart review by one of the authors who visited all centres willing to participate in the study. Additional patients from one participating centre (Oxford) were included in the analysis and data were gathered from the institution’s prospective TEM database. Thirty-six patients were eligible for analysis. Of the patient cohort (n = 487) within the UK TEM database (1992–2008), comprehensive data on CS could be gathered on 29 (46%) out of a total of 63 patients undergoing CS. For the remaining 34 patients, no detailed surgical information (no detailed operation note available) and no pathological information (no information regarding the quality of the resection specimen) could be retrieved and these patients were excluded. This was mainly due to CS being undertaken in a centre other than that where the TEM was performed. An additional seven
Completion surgery after transanal endoscopic microsurgery
patients were identified in the Oxford patient cohort (2008–2011). Other data collected included details on the completion surgical procedure, morbidity and mortality according to the Clavien–Dindo classification, hospital stay and histopathological characteristics of the specimen. The long-term oncological outcome was also analysed. TEM technique and CS (total mesorectal excision)
The standard Richard Wolf (Knittlingen, Germany) equipment was used for TEM [7,12]. The type of CS (anterior resection or abdominoperineal excision) was selected on the basis of the tumour location, the appearance of the previous resection site and the surgeon’s preference. In general, CS was performed when the TEM site had completely healed, confirmed by digital rectal examination and luminal assessment, although sometimes surgery within 2 or 3 weeks of local excision was carried out. An open or laparoscopic approach was undertaken according to the surgeon’s preference. Based on subjective assessment of the operation note by two of the authors blinded to other variables, the procedure was labelled as ‘standard’ (no difficulty at any stage of the operation) or ‘difficult’ (difficulty as specified in the operation note). The quality of mesorectal excision was assessed according to definitions used in the MRC CR07 trial [13]. Follow-up
Follow-up was carried out for all patients at the time of data collection. Patients were followed according to local policy or National Institute for Health and Care Excellence guidelines on rectal cancer. Data were often recorded by specialist colorectal cancer nurses. Statistics
Statistical analysis was performed using PASW STATISTICS for Windows (version 18, IBM Corporation, Armonk, NY, USA). Continuous data were given as mean standard deviation or as median and range for non-parametric data. Categorical data were given as frequency or percentage. For dichotomous outcomes, groups were compared by the v2 or Fisher’s exact test. The t test was used to compare means for parametric data and the Mann– Whitney U test was used for continuous, not non-parametric, data. The cumulative survival rates were calculated using the Kaplan–Meier method, and differences in relapse rates were analysed by the log-rank test. A P value of ≤ 0.05 was considered to be statistically significant.
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
e577
R. Hompes et al.
Completion surgery after transanal endoscopic microsurgery
Table 2 Completion surgery characteristics in 36 patients.
Results Patient and tumour/TEM characteristics
The clinico-pathological characteristics of the 36 patients are summarized in Table 1. The indications for CS were adenocarcinoma (11) and adenoma (25). No patient with adenocarcinoma had received any form of chemoradiotherapy. A full-thickness excision was performed in all malignant lesions with sampling of the mesorectum in one case. For the 25 presumed benign lesions a full-thickness excision was performed in 14, and a partial thickness and a combination of partial and full thickness was carried out in seven and four patients. The final histology of the TEM specimen revealed one (3%) patient with a large severely dysplastic adenoma, 16 (43%) with a pT1 lesion and 19 (54%) with a more advanced lesion. Completion surgery
Based on the current guidelines proposed in the UK and USA, two patients had CS for an early rectal cancer that had undergone a potentially curative local excision, a so-called low risk tumour [7,14]. The other pT1 tumours all contained one or more accepted high risk features and these patients were counselled to have completion major surgery, as were the patients with a pT2–3 tumour. For the single patient with a severely dysplastic adenoma, CS was performed owing to the possibility of malignant invasion (Table 2). In 31 (86%) patients a sphincter-preserving procedure was performed, with a Table 1 Patient and initial tumour characteristics at TEM in 36 patients subsequently requiring completion surgery. Sex, M:F Age, years* BMI, kg/m2* Tumour distance from anal verge, cm* Tumour location (cm) Lower rectum (0–6) Mid rectum (7–11) Upper rectum (12–16) Tumour surface, cm2* Specimen surface, cm2* Tumour stage after TEM T0 T1 T2 T3
19:17 64 (39–84) 27 (21–40) 5 (14%) 9 25 6 7.2 18
(2–15) (69%) (17%) (0.5–57) (3.1–142.5)
1 16 12 7
(3%) (43%) (33%) (21%)
TEM, transanal endoscopic microsurgery; BMI, body mass index. *Value or median number with range in parentheses.
e578
Interval (TEM to CS), months* Procedures Ultra-low AR Open AR Laparoscopic AR APE Hartmann Operation time, min* Blood loss, ml* Morbidity/mortality (Dindo–Clavien) Minor (I–II) Major (III–IV) Mortality (V) Hospital stay, days* Final staging after CS† Stage I Stage II Stage IIIa Stage IIIb Stage IIIc
2 (0.5–8.7) 1 18 12 4 1 200 (120–360) 300 (120–1800) 13 5 1 10 (6–76) 19 4 5 7 0
TEM, transanal endoscopic microsurgery; CS, completion surgery; AR, anterior resection; APE, abdominoperineal excision. *Values are median number with ranges in parentheses. †American Joint Committee on Cancer Staging system.
defunctioning loop ileostomy in 25/31. In one patient severe intra-operative bleeding led to conversion of an open low anterior resection to a low Hartmann’s procedure. The procedure was graded as ‘difficult’ in 53% (19/ 36) of cases, mostly due to bleeding and/or fibrosis/ adhesions at the TEM site. Operation time and total blood loss were significantly higher in this group of patients compared with the ‘standard’ cases (230.3 62.3 min vs 187.5 27.2 min, P = 0.02; and 721.9 562.4 ml vs 286.9 139.7 ml, P = 0.008, respectively). Three (9.7%) patients developed a clinically significant postoperative anastomotic leak and underwent Hartmann’s procedure. Two patients returned to theatre for postoperative bleeding and revision of a retracted ileostomy. The median hospital stay was 10 (6–76) days and six patients were readmitted within 30 days for pulmonary embolus (n = 1), small bowel obstruction (n = 3) and stoma related problems (n = 2). The final histology report of the total mesorectal excision specimen showed residual tumour in 17 (47%) patients. These included seven with residual tumour and involved lymph nodes, five with isolated tumour and five with isolated lymph node involvement. In all but two patients a microscopically clear excision margin resection (R0) was obtained. A total mesorectal excision Grade 3 specimen (good) was achieved in 23 (64%) patients, a Grade 2 specimen in six (16.6%) and a Grade 1 specimen (poor) in seven (19.4%) Although there was
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
R. Hompes et al.
Completion surgery after transanal endoscopic microsurgery
no correlation between grade of specimen and the type of resection, specimens with a lower grade were more frequently found after full-thickness TEM, in distal lesions and after an interval from TEM to CS of more than 7 weeks (Table 3). All 13 ‘inferior’ specimens were in the ‘difficult’ surgery category. Outcome
Adjuvant chemotherapy or radiotherapy was administered in 11 (30.5%) patients after CS. The overall surTable 3 Patient, tumour and procedure related characteristics.
Clinico-pathological parameters Age (mean, years)* 64.9 10.7 Sex Male 11 Female 12 Location lesion Anterior 5 Posterior 7 Lateral 11 Anterior vs posterior 5 vs 18 and lateral Distance anus Upper 4 Mid 17 Lower 2 Lower vs upper and 2 vs 21 mid TEM thickness Partial 7 Full 16 Time to surgery 9.5 7.2 (mean, weeks)* ≤ 7 weeks 12 > 7 weeks 11 Closure TEM defect No 14 Yes 9 TEM complication No 18 Yes 5 Type of surgery AR 22 APE 1 Hartmann 0
Inferior specimen (n = 13)
P value
62.8 14.2
0.6
Discussion 8 5 6 3 4 6 vs 7
2 7 4 4 vs 9
0.4
0.1
0.07
Large data sets on CS after TEM are lacking and the outcome focuses predominantly on survival with acceptable local/distant recurrence rates [7–10]. Anecdotal evidence suggests that this type of surgery is difficult and may result in a higher permanent stoma rate [15,16]. The data presented here show that CS after TEM is feasible with acceptable postoperative morbidity and low mortality. In addition, the reported leak rate of under 10% is comparable with that reported for anterior
0.2
1.0
0.04
0 13 11.7 9.5
0.03
3 10
0.05
10 3
0.3
11 2
0.6
9 3 1
0.08
0.2
Probability of disease-free survival
Good specimen (n = 23)
vival at 1 year was 91 and 83% at 5 years. After a median follow-up of 49.2 (3–137) months, the relapse rate was 16.7% (6/36) with 1- and 5-year disease-specific survival rates of 91 and 74%. Only one patient (3%) developed local recurrence (at 39 months after CS) and five (14%) developed distant metastases without signs of local recurrence at a mean follow-up of 33.2 months. Patients with a ‘good’ total mesorectal excision specimen had significantly improved 5-year disease-free survival compared with patients with an ‘inferior’ specimen (100 vs 51%, P = 0.001, log-rank test) (Fig. 1). Although the number of patients was too small to perform a multivariate analysis, no significant correlation between other prognostic parameters (e.g. pT, pN and Dukes staging) and reduced disease-free survival could be demonstrated.
0.8
0.6
0.4
0.2
0.0
TEM, transanal endoscopic microsurgery; AR, anterior resection; APE, abdominoperineal excision. Bold font denotes significant values. *Values are mean SD.
0.00
2.00
4.00 6.00 8.00 Follow up in years
10.00 12.00
Figure 1 Kaplan–Meier curves of 36 patients having completion surgery after transanal endoscopic microsurgery. There is a significant difference in disease-free survival between patients with ‘good’ (solid line) or ‘inferior’ (dotted line) resection specimens (100 vs 51%, P = 0.001, log-rank test).
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
e579
Completion surgery after transanal endoscopic microsurgery
resections [17]. In all but one patient the procedure could be completed according to the preoperative plan and in the majority (86%) of patients a sphincterpreserving procedure was achieved. In over half of the cases the dissection was difficult owing to the disruption of the normal tissue planes. This is reflected by the significantly higher intra-operative blood loss and operating time. Platell et al. [15] found that, in patients proceeding to a more radical procedure within 2–6 weeks after TEM, disruption of the normal pelvic dissection planes added significantly to the technical difficulty. The compromised mesorectal plane and the weakened rectal wall also increase the risk of disruption of the surgical specimen. Similar to our results, Levic et al. [8] reported an intra-operative rectal perforation rate of 20%, all at the previous TEM site. Thus, violation of the surgical planes not only potentially renders radical surgery after TEM more difficult but also significantly increases the risk of producing an inferior specimen. However, these risks have to be considered against the potential benefits of full-thickness TEM as a means of offering cure for early stage rectal neoplasia compared with submucosal excision. The reasons for this decreased survival remain speculative. Unfortunately, the patient group was too small to perform a multivariate analysis, but no association with T-stage and N-stage could be demonstrated. The strong correlation between survival and an inferior specimen suggests a potential direct oncological effect related to difficult surgery (e.g. increased blood loss: 0.7 l vs 0.3 l, increased tumour spill and circulating tumour cells, decreased immune system). Overall, the long-term outcome of patients after CS is acceptable in this population, with a 1-year and 5-year disease-free survival of 91 and 74% respectively. It is striking that only one patient developed local recurrence and that almost all patients with disease relapse had metastatic disease. Doornebosch et al. [18] described a similar outcome in patients following salvage surgery for local recurrence after TEM for T1 rectal cancers; only one patient had a further local recurrence but seven had distant metastases. They speculated that these patients represent a different biological group and that salvage therapy should be intensified with the inclusion of neoadjuvant radiotherapy and/or adjuvant chemotherapy in an effort to improve outcomes. In the study by Levic et al. [8] one patient developed distal metastasis and no local recurrence was observed in 25 patients, although the median follow-up period of 25 months was short. Various factors were analysed that may contribute to difficult surgery and an inferior resection specimen. As expected, an association between a full-thickness TEM
e580
R. Hompes et al.
and an inferior total mesorectal excision specimen could be demonstrated. The same association was found for distally located lesions with a similar tendency towards an increased chance of obtaining an inferior specimen in patients with an anterior lesion. This is in keeping with the anatomical characteristics of the distal mesorectum, described as the bare area by Morson et al. [19]. The mesorectal fat peters out distally as it does for anterior located lesions. These observations suggest that there might be a greater need for abdominoperineal excision after TEM of a tumour in the lower third of the rectum. This might reduce the chance of a poor or perforated surgical specimen. An alternative approach to avoid a difficult dissection from the abdomen is to attempt a perianal disconnection via a lower third intersphincteric dissection. Clearly, whatever approach is chosen, potentially difficult surgery should be anticipated and an experienced surgical team should be available. The risk of converting a potential sphincter-preserving procedure (as primary treatment) into an abdominoperineal excision needs to be part of the discussion with the patient before considering TEM. There is no consensus regarding the timing of CS. Examining the impact of various TEM to CS intervals, we found that an interval longer than 7 weeks significantly increased the risk for an inferior specimen. The cut-off point at 7 weeks is more likely to reflect infection or dehiscence at the TEM site. One of the weaknesses of this study is the inevitable selection bias of the study population. Only 36 patients were eligible for analysis out of a potential study population of 70 patients. The retrospective nature of the study with strict inclusion criteria led to the exclusion of 34 patients. Our findings are thus based on a small group of patients and there is therefore a possibility of type II error. Furthermore the absence of a common protocol for the management of these patients and the impact of surgical experience (with TEM and CS) of each centre will need to be considered when interpreting the results. While these shortcomings may hamper strong conclusions, this series is one of the largest series available and helps to identify risk factors for less favourable outcomes after CS following TEM. Transanal endoscopic microsurgery has been a relatively niche subspecialty in colorectal surgery; however, the increasing incidence of early stage rectal cancer and the greater availability of transanal minimally invasive surgery are likely to see it enter mainstream practice over the next few years as clinicians and patients opt for organ preservation in favour of radical surgery. In addition, various national randomized controlled trials comparing radical surgery with local excision for rectal
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
R. Hompes et al.
cancer will also increase patient numbers (TREC trial, UK; CARTS study, The Netherlands; GRECCAR 2 trial, France). This should be good news for patients, but one must be cognizant of the pitfalls and challenges of local excision. These approaches will need to be part of a strategy for early rectal cancer management that include CS, careful surveillance and salvage surgery along with the use of neoadjuvant and other adjuvant therapy. In conclusion, CS after TEM is feasible in most patients without significant morbidity and acceptable permanent stoma rates. Our results confirm those of other studies that CS after TEM does not negatively influence local recurrence and survival, but the reduced disease-free survival in patients with an inferior specimen is of concern. Completion surgery after fullthickness TEM excision within the distal and/or anterior rectum deserves special consideration. In this group there may be a role for neoadjuvant and adjuvant treatment strategies combined with CS in an effort to improve survival.
Author contributions The concept for this paper was designed by the lead author and C. Cunningham. Analysis of the data was performed by C. Buskens and interpretation thereafter was done by the lead author, C. Cunningham and C. Buskens. The other authors were crucial for accurate and detailed data collection and on revision made suggestions regarding the interpretation of the data. The main draft was written by the lead author with major input from C. Cunningham. All other authors critically revised the content and approved the final version of the paper.
Funding None declared.
References 1 Kapiteijn E, Marijnen CA, Nagtegaal ID et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 2001; 345: 638–46. 2 Nesbakken A, Nygaard K, Bull-Njaa T, Carlsen E, Eri LM. Bladder and sexual dysfunction after mesorectal excision for rectal cancer. Br J Surg 2000; 87: 206–10. 3 You YN, Baxter NN, Stewart A, Nelson H. Is the increasing rate of local excision for stage I rectal cancer in the United States justified?: a nationwide cohort study from the National Cancer Database. Ann Surg 2007; 245: 726–33.
Completion surgery after transanal endoscopic microsurgery
4 Hompes R, Cunningham C. Extending the role of transanal endoscopic microsurgery (TEM) in rectal cancer. Colorectal Dis 2011; 13(Suppl 7): 32–6. 5 Lezoche G, Guerrieri M, Baldarelli M et al. Transanal endoscopic microsurgery for 135 patients with small nonadvanced low rectal cancer (iT1–iT2, iN0): short- and long-term results. Surg Endosc 2011; 25: 1222–9. 6 de Graaf EJ, Doornebosch PG, Tetteroo GW, Geldof H, Hop WC. Transanal endoscopic microsurgery is feasible for adenomas throughout the entire rectum: a prospective study. Dis Colon Rectum 2009; 52: 1107–13. 7 Bach SP, Hill J, Monson JR et al. A predictive model for local recurrence after transanal endoscopic microsurgery for rectal cancer. Br J Surg 2009; 96: 280–90. 8 Levic K, Bulut O, Hesselfeldt P, Bulow S. The outcome of rectal cancer after early salvage surgery following transanal endoscopic microsurgery seems promising. Dan Med J 2012; 59: A4507. 9 Hahnloser D, Wolff BG, Larson DW, Ping J, Nivatvongs S. Immediate radical resection after local excision of rectal cancer: an oncologic compromise? Dis Colon Rectum 2005; 48: 429–37. 10 Borschitz T, Heintz A, Junginger T. The influence of histopathologic criteria on the long-term prognosis of locally excised pT1 rectal carcinomas: results of local excision (transanal endoscopic microsurgery) and immediate reoperation. Dis Colon Rectum 2006; 49: 1492–506. 11 Lee W, Lee D, Choi S, Chun H. Transanal endoscopic microsurgery and radical surgery for T1 and T2 rectal cancer. Surg Endosc 2003; 17: 1283–7. 12 Bretagnol F, Merrie A, George B, Warren BF, Mortensen NJ. Local excision of rectal tumours by transanal endoscopic microsurgery. Br J Surg 2007; 94: 627–33. 13 Quirke P, Steele R, Monson J et al. Effect of the plane of surgery achieved on local recurrence in patients with operable rectal cancer: a prospective study using data from the MRC CR07 and NCIC-CTG CO16 randomised clinical trial. Lancet 2009; 373: 821–8. 14 Mellgren A, Goldberg S, Rothenberger DA. Local excision: some reality testing. Surg Oncol Clin N Am 2005; 14: 183–96. 15 Platell C. Transanal endoscopic microsurgery. ANZ J Surg 2009; 79: 275–80. 16 Bujko K, Richter P, Kolodziejczyk M et al. Preoperative radiotherapy and local excision of rectal cancer with immediate radical re-operation for poor responders. Radiother Oncol 2009; 92: 195–201. 17 Paun BC, Cassie S, MacLean AR, Dixon E, Buie WD. Postoperative complications following surgery for rectal cancer. Ann Surg 2010; 251: 807–18. 18 Doornebosch PG, Ferenschild FT, de Wilt JH, Dawson I, Tetteroo GW, de Graaf EJ. Treatment of recurrence after transanal endoscopic microsurgery (TEM) for T1 rectal cancer. Dis Colon Rectum 2010; 53: 1234–9. 19 Morson BC, Vaughan EG, Bussey HJ. Pelvic recurrence after excision of rectum for carcinoma. Br Med J 1963; 2: 13–8.
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 15, 576–581
e581
