One of the most controversial areas of laparoscopic colorectal surgery is that of rectal cancer. This is due to the perception that it is more challenging than colon resection in achieving radical cancer surgery within the complex anatomy of the pelvis. Furthermore, the lack of data on large patient series with longer follow-up partly accounts for surgeons’ reluctance to adopt this method in patients with rectal cancer. The purpose of this prospectively conducted study is to report the perioperative and late results, with emphasis on complications, and long-term survival in 194 patients who underwent a “totally” laparoscopic resection for rectal cancer at our department. Furthermore, published papers dealing with laparoscopic rectum resection are reviewed.

Materials and Methods

Between November 1992 and July 2003, a total of 1000 colorectal laparoscopic procedures were performed at our institution; 601 patients had a benign colorectal disease and 399 cases were resected laparoscopically for colorectal malignancy. Of these, 200 cases had a colon carcinoma and the remaining 194 had rectal cancer. The tumor was considered as primary rectal carcinoma if it was located in the lower third (0–6 cm from the anal verge), middle third (7–12 cm), or upper third of the rectum (12–18 cm) as measured by rigid rectosigmoidoscopy. All patients underwent preoperative tumor staging by contrast medium enema, rectoscopy and colonoscopy with biopsies of the tumor, endorectal ultrasonography, abdominal ultrasound, and chest x-ray.

Patients with locally advanced rectum carcinoma (uT3/uT4) and no evidence of distant metastases were candidates for neoadjuvant chemoradiation including the following schedule: 30–50 Gy radiotherapy and 5-fluouracil (5-FU) in combination with folinic acid over 5 weeks. The operation was carried out 2–3 weeks after completion of the multimodality treatment. Adjuvant treatment was administered to patients with UICC stage II/III disease and consisted of six cycles of 5-FU/folinic acid.

All patients were followed up prospectively by means of clinical examination, tumor markers (CEA, CA19-9), rectoscopy, and ultrasonography of the abdomen every 3 months for the first 2 years, every 6 months after 2 years, and then every year after 5 years in order to evaluate late complications and tumor recurrence. The survival probability analysis was performed using the Kaplan-Meier method.

Surgical technique

All operations were performed by one surgeon (E. Bärlehner). The patients were placed in a steep Trendelenburg position. We utilize five ports: four 10 mm and one 12 mm. The specimen is removed through a small (5-cm) incision in the left lower quadrant port site. The operation is performed with a harmonic scalpel (Ultracision, Ethicon Endosurgery). We begin with the mobilization of the sigmoid, descending colon, and splenic flexure on the Gerota fascia up to the mid transverse. The gonadal vessels and the left ureter are identified. The retroperitoneum is incised on the promontory, and the inferior mesenteric vessels are identified and divided. The artery is taken 1–2 cm anterior to the aorta and the vein is divided close to the pancreas. During mobilization of the mesorectum care must be taken to avoid any damage to the underlying hypogastric nerve plexus. Dissection is continued ventrally in front of the Denonvilliers’ fascia. Following that, we perform a deep or intersphincteric stapling of the rectum with a linear cutter. The rectum and sigmoid colon are extracted through a plastic wound protector in a small left lower quadrant incision. The anvil of the circular stapler is positioned in the descendens colon. The circular stapler is then inserted through the rectum and an end-to-end descendo-rectostomy is performed. The integrity of the stapled anastomosis is checked by the inspection of an intact double donut and an underwater bubble test. A drain is placed at the end of the operation in the pelvis. In cases of abdomino-perineal resection the mobilized rectum together with the whole mesorectum is retrieved through the perineal incision in the traditional fashion. The perineal wound is closed primarily and a terminal colostomy is fashioned at the left lower quadrant site.

Registry database

Clinical, surgical, and histopathological data were recorded prospectively in a registry database (PC). Data supplied for each patient included gender, age, body mass index, ASA score, tumor localization, preoperative and postoperative multimodality therapy, type of resection and anastomosis, operating time, intraoperative and postoperative complications, number of resected lymph nodes, length of removed specimen, UICC tumor stage, late complications, tumour recurrence, overall survival, cancer-related survival, and stage-specific survival. All data were analyzed by a statistician (B. Schicke) with the SPSS 11.5 software for Windows (SPSS, Chicago, IL).

Results

Patients

Patients’ baseline data are listed in Table 1. The unselected patient coort comprised 125 men and 69 women whose average age was 65 years. Of the patients, 28% were 70 years and older. The mean body mass index was 26.4 kg/m2. It was >25 kg/m2 in 102 cases. The ASA score was III–IV in >70% of the patients. Tumors were located in the upper rectum in 44 patients, the midrectum in 86 patients, and the lower rectum in 64 cases. A total of 102 patients received preoperative radiation and/or chemotherapy; 125 patients received this treatment modality postoperatively.

Table 1 Baseline characteristics of patients (n = 194)
Full size table

Procedure performed

The most common procedure was low anterior resection with total mesorectum excision in 127 patients and colonic J-pouch in 28 cases (Table 2). A total of 49 patients had resection of the rectosigmoid junction, and 16 had abdomino-perineal resection. One patient with ulcerous colitis and rectal carcinoma received a proctocolectomy and another one a Hartmann procedure. The double stapling technique was the most common anastomosis type (90%), followed by colostomy in 9% of the cases and two hand-performed coloanal anastomoses. Average operative time was 174 min.

Table 2 Data related to surgery and histopathological results (n = 194)
Full size table

Histopathological results

Average number of lymph nodes removed was 25.4 (ranging from 11 to 84) and length of specimen resected was 27.6 cm (range 11–68 cm). Resection was curative in 145 patients and palliative in 49 cases. The distribution of patients according to the UICC classification was as follows: stage I: 49 patients (25.2%), stage II: 53 (27.3%), stage III: 59 (30.4%), and stage IV: 33 patients (17%).

Surgery-related complications

Intraoperative complications occurred rarely, <1% for lesions of the ureter, urinary bladder, and deferent duct. Conversion to conventional surgery was necessary in two cases (1 %). Blood transfusion was required in only two patients (Table 3).

Table 3 Surgery-related complications (n = 194)
Full size table

The most common postoperative complication was the anastomotic leakage in 13.5% of the patients. The frequency of other complications such as bleeding, rectovaginal fistula, ileus, and infections was <2%. There was no postoperative mortality. A relaparoscopy was necessary in 5.1% of the patients and a laparotomy in 6.2%. Follow-up evaluation ranged from 1 to 128 months with a mean of 46.1 months. The records of 97% of the patients could be investigated prospectively.

Late complications such as anastomotic stenosis, ileus, and incontinence occurred rarely. The most common late complication was incisional hernia in 3.6% of the patients. Port site metastasis occurred in one patient (0.5%) with stage IV disease (peritoneal carcinosis) who underwent a palliative resection.

Oncological results and survival

After a mean follow-up of 46.1 (1–128) months, tumor recurrence occurred in 23 of the 145 curatively resected patients (11.7% distant metastases and 4.1% local recurrence). Overall local recurrence rate was 6.7% (4.1% after curative resection and 14.3% after palliative resection) (Table 4). Overall survival rate was 90.6% at 1 year, 74.5% at 3 years, and 66.3% at 5 years. Overall 5-year survival rate was 76.9% after curative resection and 31.8% after palliative resection (Fig. 1). Cancer-related survival rate was 94% at 1 year, 82.4% at 3 years, and 78.9% at 5 years. It was 87.7% at 5 years after curative resection and 48.5% at 5 years after palliative resection (Fig. 2). At 5 years, the survival rate was 100% for stage I, 94.4% for stage II, 66.6% for stage III, and 44.6% for stage IV (Fig. 3).

Table 4 Data related to tumor recurrence
Full size table
Figure 1
figure 1

Kaplan-Meier estimates of overall survival for curative and palliative laparoscopic rectum resection (n = 194) [mean follow-up 46.1 (1–128) months]

Full size image
Figure 2
figure 2

Kaplan-Meier estimates of cancer-related survival for curative and palliative laparoscopic rectum resection (n = 194) [mean follow-up 46.1 (1–128) months]

Full size image
Figure 3
figure 3

Kaplan-Meier estimates of stage-specific survival for laparoscopic rectum resection (n = 194) [mean follow-up 46.1 (1–128) months]

Full size image

Discussion

Historically, the three major and decisive strides that have been made in the last two decades in the treatment of rectal cancer were the establishment of multimodality therapy, the introduction of total mesorecrum excision, and the application of laparoscopy. Neoadjuvant therapy made it possible for many patients with locally advanced tumors who were previously deemed unresectable to undergo potentially curative surgical resection without colostomy, preserving the anal sphincter and maintaining continence. Furthermore, it has been clearly demonstrated that adjuvant treatment has the potential of improving prognosis not only in terms of local recurrence, but also in terms of overall survival [11]. A further milestone in the outcome of rectal cancer has been the introduction of total mesorectal excision (TME). It is only historically interesting to mention that the concept of TME had already been elucidated in an impressive way by Heinrich Westhues from Erlangen, Germany, in 1934 [31]. He showed in meticulous pathologic studies that the complete en bloc removing of the tumor together with the “perirectal and retrorectal tissue” is the way to reduce pelvic recurrence, and he applied this radical procedure in 46 patients with rectum carcinoma. After a follow-up of more than 2 years, no tumor recurrence had occurred in his patient group. At that time ~50% of the patients developed local recurrence within 1 year after rectum resection. In the early 1980 s, Heald and colleagues substantiated the principles of TME [9]. With this technique, they reported a local recurrence rate of 4% at 10 years, and a disease-free survival of 80% at 5 years and 78% at 10 years after curative resection. In light of these and other results [15], TME-based operations have become the new standard technique of care for mobile rectal cancer. With the rapid application of advanced laparoscopy, many investigations have demonstrated benefits for laparoscopic colorectal surgery in terms of faster postoperative recovery, less pain, less morbidity and mortality, and earlier return to full activity. Furthermore, many reports have repeatedly shown the feasibility and partly the efficacy of laparoscopic rectum resection for malignancy [1, 4–8, 10, 12–14, 16, 18–22, 24, 26–30, 32, 33]. However, the effectiveness of this new method is, as yet, unclear. The main concern regarding this technique is that most trials involve only small series that are based on retrospective reviews. Only the two studies reported by Tang et al. from Singapore and Leung et al. from Hong Kong were prospective randomized [14, 27]. The trial by Tang et al. was performed in parallel with the UK Medical Research Council Conventional versus Laparoscopic-assisted Surgery in Colorectal Cancer (CLASSIC) trial. The local component of this collaboration investigated the immune stress response following open and laparoscopic resection for cancer. Thus, oncological data are not available [27]. In the study by Leung et al., 403 patients with rectosigmoid carcinoma were randomized to receive either laparoscopic assisted (n = 203) or conventional open (n = 200) resection of the tumor [14].

Nonetheless, most data reported must be considered preliminary and medium-term results at best. These reports should be interpreted and compared very carefully. Including our series, a total of 1818 patients worldwide have been reported as receiving a laparoscopic resection for rectum malignancy (Table 5). The most commonly performed operation was anterior resection in ~68% of the cases. The operative time ranged from 88 to 600 min. The reported morbidity rate was 0–55% including an anastomotic insufficiency rate of 0–27%. Conversion to open surgery varied from 0 to 50%. The postoperative mortality rate was <3% and zero in most reported series. It would seem that laparoscopic rectal surgery is associated in some series with a somewhat higher morbidity rate [2, 23, 25, 34]. The high prevalence of anastomotic breakdown is particularly worrisome, although the leak rate is comparable to that of conventional anterior resection. The overall insufficiency rate in our series was 13.5% (24/177 patients). Of these 24 patients three received rectosigmoid resection and 21 underwent low anterior resection. Only two of these 21 patients had a diverting loop ileostomy. As a matter of fact, laparoscopic low anterior resection is a very demanding operation requiring advanced surgical skills in laparoscopic colorectal surgery [17]. Nevertheless, the importance of surgical technique should be highlighted since a strong association between leakage from a colorectal anastomosis and the development of local recurrence after potentially open curative resection for rectal cancer has been recently reported [3]. No local recurrence was diagnosed after a mean follow-up of 46.1 months in our patients with anastomotic leakage. The mean survival time was 67.6 months in this subgroup of patients.

Table 5 Perioperative results after laparoscopic rectum resection: Literature review
Full size table

Despite acceptable postoperative data, evaluation of oncological adequacy is not only based on distal margin clearance or on the number of lymph nodes removed but also on long-term outcomes in terms of local recurrence and 5-year survival rates. The reported medium-term oncological data are altogether comparable to those after open rectum resection (Table 6). It is interesting and very important to note the very low incidence of port-site metastases. Only two cases of a total of 1818 patients who underwent laparoscopic rectum resection developed this complication (0.11%). Both patients had primary metastatic disease. Reservations must be expressed with regard to the high local recurrence rates reported by Weaver (15%) and Feliciotti (21%) [30, 6]. In the prospective randomized trial by Leung et al., laparoscopic resection did not worsen survival and disease control for patients with rectosigmoid cancer compared with open resection, and its benefits in reducing pain and allowing earlier postoperative recovery were confirmed. Conspicuous in this study is the high conversion rate (23.2%) [14]. Appropriate surgical technique together with experience in laparoscopy will ensure low incidence of morbidity rate, conversion, and local recurrence once the learning curve is passed. Our results originate from a center with extensive experience in all fields of minimally invasive surgery with more than 1100 colorectal resections performed to date and all laparoscopic rectum resections were performed by one surgeon. This reflects among other things the exceptional low abdominoperineal resection rate of 8% and the high low anterior resection rate with TME in 65.5% of the patients.

Table 6 Oncological results after laparoscopic rectum resection; Literature review
Full size table

In our experience with this technique, the rates of local recurrence, distant metastases, and patient survival are comparable with those established by the larger series of patients with rectal carcinoma. Laparoscopic rectum resection allows accurate magnification and identification of all fine and important anatomic structures in the narrow pelvis, e.g., the hypogastric nerves, and the “holy planes.” Dissection under laparoscopic view allows minimal blood loss and avoids tumor cell seeds in the pelvis. Of course, laparoscopic surgery is technically more demanding than the open approach, and the learning curve is somewhat protracted. Nevertheless, what is laparoscopically difficult may also be technically arduous with the conventional counterpart.

Conclusion

Rectum surgery has not reached a plateau beyond which it will not progress. Laparoscopic resection for rectal malignancy might be the next step in the evolution of modern rectal cancer surgery. Our results and the literature review clearly demonstrate that laparoscopic resection for rectal carcinoma is not associated with a higher morbidity or mortality. Established oncological and surgical principles are respected, and the few reported data on long-term outcomes seem to equal to those after open surgery. The fundamental feasibility and efficacy of this method have been proven. Now it should be examined to what extent these favorable results can be reproduced in the setting of a multicenter Phase 3 study. Nonetheless, at the present time, laparoscopic rectum resection for cancer cannot be yet recommended for routine use.