|Home | About | Journals | Submit | Contact Us | Français|
Incomplete specimens resulting in residual mesorectum in the patient and an increased risk of local recurrence remains a problem. We have introduced transanal-total mesorectal excision (Ta-TME) in our department to potentially overcome this problem due to more direct access to the lower pelvis in patients undergoing TME for rectal cancer and this article presents our initial experience with the new procedure.
Patients with a T1-T3 mid or low rectal cancer eligible for TME or intersphincteric abdominoperineal excision were selected for a combined transanal and transabdominal laparoscopic resection. The primary aim of the study was to evaluate the feasibility and efficacy of the method with a special focus on the quality of the specimen.
During a 9-month period, 11 patients were operated with this technique. All procedures resulted in complete or nearly complete specimen. We did, however, find the procedure technically demanding and experienced several complications with three anastomotic leaks (all with preserved intestinal continuity) and a urethral lesion.
Ta-TME is feasible and might be the answer to obtaining good quality specimens and overcome some of the technical difficulties that can be encountered in the obese narrow male pelvis. The procedure however is technically demanding.
The introduction of total mesorectal excision (TME) in combination with advanced neoadjuvant and adjuvant oncological treatment regimens has improved the overall survival and local recurrence in rectal cancer patients. Although minimal invasive procedures are used more and more frequently and generally adds to this positive development, the dissection in the pelvic cavity still remains a technical challenge. Concerns about adequate distal resection margin, bulky tumours and inadequate view of the operative field in the low pelvis, may negatively affect the surgical as well as the oncological quality of the procedure.
Thus, the problem of acquiring a complete Grade 1 specimen, according to the definition by Quirke et al., remains in at least 10% of the cases.[1,3,4] Post-operative magnetic resonance imaging (MRI) scans have shown inadvertent residual mesorectum in 36% of patients following TME, which can be correlated to increased risk of local recurrence as does an incomplete plane of surgery (muscularis propria plane).
The transanal-TME (Ta-TME) dissection technique was initially described by Funahashi et al., who performed the dissection as an open procedure. Sylla et al. presented a case report in 2010, using the transanal endoscopic microsurgery (TEM) equipment to overcome the challenges in the pelvic dissection. Several following case reports and series have demonstrated the surgical feasibility and oncological safety of this procedure.[9,10,11,12]
The possible benefits of the Ta-TME procedure might be a better and more direct access to the small pelvis, especially in males with mid-rectal tumours, improved quality of specimens, direct visualisation of the distal resection margin and trans-anal extraction of the specimen. Furthermore, anastomosing without multiple stapler firings, associated with increased risk of anastomotic leaks, can be avoided.
The aim of this study was to demonstrate the implementation of laparoscopy-assisted Ta-TME in a specialised high volume colorectal cancer unit. The primary outcome was the oncological and pathological quality of the specimens and secondary outcomes included surgical complications and morbidity.
From April 2014 to January 2015, 11 patients with rectal cancer were operated with the Ta-TME technique. Only patients with low- or mid-rectal T1-T3 tumours on MRI, a planned TME procedure and ability to give an informed consent were eligible. We primarily aimed to include patients who did not have former pelvic radiotherapy other than neoadjuvant chemoradiotherapy, extra-peritoneal pelvic surgery or Stage IV disease. All patients underwent a standard clinical examination including rigid proctoscopy, MRI of the rectum and a thoraco-abdominal computed tomography (CT) scan.
The final treatment plan was decided at the multidisciplinary team conference.
All procedures were performed at a single institution by a surgeon (HO) working in a dedicated surgical team of certified colorectal surgeons, with extensive experience in TEM through more than 10 years and in laparoscopic colorectal surgery.
The training phase included cadaver dissection, technical and practical courses and participation as assistant in a human Ta-TME procedure.
All patients received standard mechanical bowel preparation and followed a standard post-operative enhanced recovery program for colorectal resection.
After the patient was under general anaesthesia, a proctoscopy was performed to ensure an adequate pre-operative bowel preparation and to perform a wash out of the rectum with cetrimide.
An Olympus Exera II, 10 mm 30° Endoeye and a standard Olympus insufflator was used for the entire procedure.
The procedure started with the laparoscopic transabdominal part of the dissection according to TME principles and high tie with central ligation of the inferior mesenteric artery. To ease the transanal extraction of the specimen, the splenic flexure was mobilised entirely to the midline in all patients. The mobilisation continued caudally/posteriorly to the sacral promontory and anteriorly to the peritoneal reflection of the rectum. Anteriorly, we thus identified the top of the vagina/seminal vesicles, respectively.
We then commenced by inserting an Applied Medical GelPOINT Path Trans Anal Access Platform™ and closed the rectum with a purse string suture under direct visualisation of the tumour. This provided optimal conditions for identifying the appropriate distal margin at least 1 cm below tumour. Given the more direct angel and shorter anal canal in female patients, the purse string can be applied through the transanal port in an openly fashion. After purse string placement, the rectum was washed again.
A full thickness incision in the entire circumference of the rectum now provided access to the embryonic planes. The mobilisation of the mesorectum started posteriorly and was then continued anteriorly. Finally, the dissection progressed laterally, to avoid the specimen blocking the operative field during the posterior dissection and to ensure the correct lateral plane to avoid nerve injury.
When communication to the peritoneal cavity was achieved and the trans-anal and trans-abdominal dissection planes were united, one surgeon returned to the abdominal site and assisted the final mobilisation of the rectum. Depending on the tumour size, the specimen was extracted either via a Pfannenstiel incision or transannaly, using the Alexis Wound Protector™.
The remaining part of the colonic mesentery was divided extra-corporally above the division of the inferior mesenteric artery, testing the marginal artery for sufficient blood flow by division and subsequent ligation. The colon was now divided over a purse string forceps and the anvil of the circular stapler was inserted and fastened. The end of the colon could now be repositioned into the abdomen and the anal canal was closed with a purse string suture around a surgical tube drain. The center rod of the EEA stapeling device was inserted with the aforementioned surgical drain as a guide and finally the anvil was connected and stapling performed around the anal purse string suture. The anastomosis was performed as an end-to-end or side-to-end anastomosis using a 28–33 mm circular stapling device depending on the colonic caliber.
All anastomoses were protected by a diverting loop-ileostomy. We did not perform any leak test, but the excellent trans-anal access allowed for a full visualisation of the anastomosis. In case of a very low anastomosis, we applied reinforcement sutures.
Anastomotic leaks were graded according to the classification system proposed by the International Study Group of Rectal Cancer and complications by The Clavien-Dindo (CD) Classification.
All 11 Ta-TME procedures were performed for adenocarcinoma of the rectum. None of the patients had undergone former pelvic surgery. Patient characteristics and details of the procedures are shown in Table 1 and short-term and oncological outcomes are presented in Table 2. The median follow-up time was 263 days (range 127–370).
Due to our selection criteria, only one patient (number 5) had neoadjuvant chemoradiotherapy and most tumours were located in the middle rectum. Four patients, two with low rectal cancers and two with mid-rectal cancers, had inter-sphincteric abdominoperineal excision (APE) performed either due to tumour localisation (radicality of resection) or pre-operative information of expected post-operative functional outcome and risk of LARS.
AJCC stage was distributed between 5 Stage I/II and 6 Stage III.
One patient (number 3) was converted to open surgery at the end of the transanal procedure due to a suspected lesion of the ureter, which was not confirmed at open exploration.
In the seven patients with a subsequent anastomosis and thus a preserved anal canal, four of the specimens were extracted trans-anally. This was due to specimen size.
No patients underwent reoperation, but one leak was treated under general anaesthesia and thus CD Grade IIIb. The 30-day mortality was 0%. We did not experience extensive bleeding (range 20–600 mL) in any procedures.
Overall surgical complications were observed in four patients (Patient 1, 2, 3 and 8) as described below and one patient had prolonged ileus (Patient 7) and treated with total parenteral nutrition and thus CD Grade II.
Intra-operative lesions comprised one urethral lesion in patient 3, treated with suturing and a urinary catheter and one intra-operative perforation of the anal canal 3 cm below the tumour in Patient 5 (the second APE performed). Patient 3 is now suffering from permanent urinary dysfunction with incontinence and need to wear a uridome.
We observed three Grade 2 anastomotic leaks in the seven anastomoses performed. The leaks were diagnosed as perianastomotic fluid collections by CT scans with rectal contrast in patient 1 and 2. Neither patients had leak of contrast into the collections. In patient 8 a minor defect in the anastomotic ring with pus discharge was diagnosed clinically without findings on CT scan.
However, all leaks were treated with antibiotics, transanal puncture and lavage and all anastomoses were preserved. Thus, two leaks were CD Grade I and one CD Grade IIIb.
At follow-up, 5 of 7 patients with a diverting stoma have had their intestinal continuity restored (Patient 1, 2, 6, 7 and 11). Patient 8 and 10 still awaits ileostomy closure. In patient number 10 and 11, we applied the abovementioned reinforcement sutures and no leaks were observed.
One patient (number 7) had a prolonged stay due to paralytic ileus and two patients (number 4 and 6) had urinary retention and were discharged with a urinary catheter in place, but with subsequent removal at follow up.
There were no medical complications during the first 30 post-operative days. One patient was readmitted at post-operative day 17 due to a Grade 2 anastomotic leak (Patient 1).
All patients had a R0 resections and the specimen quality was Grade 1 in 8 out of the 11 cases, according to Quirke's classification and no specimens were incomplete (muscularis propria plane).
Our primary aim for the initiation of the Ta-TME technique was to achieve experience in this novel surgical approach, especially in low- and mid-rectal cancers, as we believe the technique will potentially improve the surgical and oncologic outcome in selected cases (obesity, males, narrow pelvis).
Overall, we achieved good quality specimens, but we experienced several complications including a high leak rate and a urethral lesion.
There are indeed pitfalls in performing the Ta-TME procedure, illustrated by one urethral lesion, one conversion and two cases of urinary retention. An obese male with a narrow pelvis certainly represents a challenge even from below. Our first case was a woman with a mid-rectal tumour and the wider pelvis in women and the shorter and more direct angle of the anal canal made this a fairly easy case, which can be recommended as a first case.
The quality of our specimens seems equal to the series by Atallah et al. with 85% complete or nearly complete and by Velthuis et al. with a 100% complete or nearly complete. Short-term results were satisfactory with a low morbidity and consequently a median length of stay of 6 days and no mortality or reoperations.
The high leak rate can to some degree be explained by adherence to a strict definition of anastomotic leaks counting all pelvic abscesses and fluid collections as leaks. They have all been Grade 2 and with preserved intestinal continuity.
In the COLOR II study, laparoscopic surgery proved superior to open surgery in respect to circumferential resection margin positivity in low rectal cancers. It was speculated that this was due to improved visualisation in the laparoscope.
The results of our initial cases presented in this paper were obtained during our learning curve and as we continue to learn by each procedure performed, we do not expect to have reached the top of the curve yet. Our initial experience has thus confirmed our impression during the preparation phase that the technique presents many challenges both operatively and in the use of the necessary equipment. Having tried out the procedure in more recent cases with an insufflator that provides continuous flow, we found that it eases the procedure remarkably.
Another area of development is the incorporation of the benefits of robotic surgery, by applying the more flexible instruments of the robot from below. This has been performed in other centres[15,16] and could prove advantageous with further development.
The Ta-TME technique has been considered a promising improvement in the treatment of rectal cancer and published series have shown the method's feasibility.[9,10,11,12] Randomised studies as well as long-term results are lacking, before the method's final place in the surgical treatment of rectal cancer can be defined in relation the standard transabdominal approach. In conclusion, based on our presented initial experience with 11 patients, the Ta-TME procedure seems to be safe and feasible, but the technique have pitfalls during the learning curve, despite a dedicated and experienced colorectal team. However, it appears that the technique can add potential benefits for this patient group. We will encourage to the development of a national Ta-TME databases to proper data registration for continuing quality and outcome assessment. As national and international centres have overcome the learning curve, a randomised trial comparing the combined laparoscopic and Ta-TME technique with standard laparoscopy should be initiated.
There are no conflicts of interest.