Tech Coloproctol (2004) 8:S79–S81 DOI 10.1007/s10151-004-0119-8
I. Kanellos • K. Vasiliadis • S. Angelopoulos • T. Tsachalis • M.G. Pramateftakis • I. Mantzoros • D. Betsis
Anastomotic leakage following anterior resection for rectal cancer
Abstract Background The aim of this study is to present the incidence of anastomotic leakage after anterior resection for rectal cancer and to demonstrate the therapeutic approach for the treatment of this complication. Patients and methods During the last ten years, 93 patients underwent anterior resection of the rectum for rectal cancer. Low anterior resection with total mesorectal excision (TME) was performed in 72, and high anterior resection in 21 patients. The definition of the anastomotic leakage was based on clinical features, peripheral blood investigations and abdominal CT scan. Results Clinically apparent anastomotic leakage developed in 9 patients (9.7%). Four patients were managed conservatively and five operatively. Postoperative mortality among the patients with anastomotic leakage was not recorded. Conclusions The incidence of anastomotic leakage after anterior resection of the rectum for rectal cancer is relatively low. It remains however the most serious complication following rectal resection for cancer. Key words Rectal cancer • Anterior resection • Anastomotic leakage
I. Kanellos • K. Vasiliadis • S. Angelopoulos • T. Tsachalis M.G. Pramateftakis • I. Mantzoros • D. Betsis 4th Surgical Department Aristotle University, Thessaloniki, Greece I. Kanellos () Antheon 1, Panorama 55236 Thessaloniki, Greece E-mail:
[email protected]
Introduction Anastomotic leakage is one of the most feared and lifethreatening early complications following rectal anastomosis and continues to represent a major clinical problem. The reported clinical leakage rate after anterior resection of the rectum for rectal cancer varies from 3% to 21% and generally results in an important increase of postoperative morbidity [1, 2]. The aim of this study is to present the incidence of anastomotic leakage after anterior resection for rectal cancer and to demonstrate the therapeutic approach for the treatment of this complication.
Patients and methods During the last ten years, 93 patients underwent anterior resection of the rectum for rectal cancer. There were 53 males and 40 females, the average age being 66.7 years (range 43–88). Low anterior resection with total mesorectal excision (TME) was performed in 72 and high anterior resection in 21 patients. Low anterior resection was considered anastomosis of the colon to the extraperitoneal rectum. Tumour location was in the rectum in 87 and in the rectosigmoid junction in 6. All the patients underwent an elective operation after preoperative bowel irrigation one day prior to surgery. Antibiotic prophylaxis (1 g cefotaxim and 0.5 g metronidazole) was administrated at the time of induction of anaesthesia. The surgical procedure started with a median laparotomy in lithotomy position. The splenic flexure was taken down routinely in order to avoid any tension to the anastomosis. In patients with rectal cancer of the middle or lower third of the rectum the mesorectal was routinely totally excised, down to the pelvic floor. Thirteen anastomoses were handsewn (14%) and were fashioned using the interrupted serosubmucosal technique whereas 80 (86%) anastomoses were performed using a circular stapler of appropriate size. The integrity of the anastomoses was tested by betadine solution injected through the rectum. The procedure was followed by an insertion of easyflow drainage into the small pelvis. Additionally, low anterior
S80
I. Kanellos et al.: Anastomotic leakage following anterior resection
resection was completed by the separation of the anastomosis and sacral drain from the abdominal cavity, suturing the parietal peritoneum to the colon and mesocolon. None of the 93 anterior resections was covered by a defunctioning stoma. The definition of the anastomotic leakage was based on clinical features, peripheral blood investigations and abdominal CT scan. In particular, anastomotic leakage was considered to be present when a patient presented pelvic pain, unexplained pyrexia, often with tachycardia, and leucocytosis and additionally symptoms and signs simulating an acute cardiorespiratory event. Indications of anastomotic leakage were also faecal discharge from the drain, septicaemia or peritonitis. CT scan of the abdomen and pelvis with rectal contrast enema verified the diagnosis. Anastomotic leaks were classified as major and minor. Major leaks were defined as peritonitis and septicaemia due to leakage, and a necessary surgical intervention in the case of broad leakage whereas minor leaks were defined as those characterised by a small amount of extravasations accompanied by mild clinical signs and no signs of sepsis. Minor leaks were managed conservatively by total parenteral nutrition and antibiotic therapy for a minimum of 8 days.
Table 1 Details of patients with anastomotic leakage
Results A total of 9 (9.7%) anastomotic leaks were confirmed. These were 8 males and 1 female, the average age being 68.8 years (range 60–78). All of them underwent low anterior resection of the rectum. Rectal resection was completed with stapled in 8 and hand-sewn anastomosis in one. Eight resections were curative and one palliative, due to multiple liver metastases. One patient with locally advanced rectal cancer was treated by extended resection of the tumour en bloc with invaded organs, which were portion of urinary bladder and
Diagnostic parameters
n
%
Leakage in CT Pelvic pain Fever/tachycardia Leucocytosis Faeculent drain Symptoms simulating cardiorespiratory event Incomplete ileus Rectovescical fistula
9/9 8/9 8/9 7/9 5/9 5/9
100 89 89 78 55 55
5/9 1/9
55 11
the left ureter. Distribution according the Dukes classification was BI lesion in 4, BII in 2, CI in one, CII in one and D in one patient. The breakdown by TNM classification was 2 patients stage I, four stage II, two stage III and finally one had stage IV disease. The mean distance of the anastomoses from the anal verge was 6.2 cm (range 3–9 cm). Differentiation of the tumour according to the histological examination was moderate in all patients. None of the patients received preoperative radio-chemotherapy. Table 1 provides an overview of the diagnostic parameters in the patients with anastomotic leakage. Reoperation was carried out in five patients with major leaks. One of these developed pelvic abscess and rectovaginal fistula and one rectovescical fistula. None of these patients presented peritonitis or septicaemia. The anastomotic leaks were postoperatively confirmed in a mean period of 6.9 days (range 5–8 days). In particular, 3 patients were treated by drainage of the pelvic cavity and loop transversostomy while 2 had pelvic cavity drainage, the anastomosis was removed, the rectal stump oversewn and an end colostomy
Fig. 1 Diagnostic and therapeutic approach of patients with anastomotic leakage
I. Kanellos et al.: Anastomotic leakage following anterior resection
S81
performed. Four patients with minor leaks were treated conservatively by nutritional support and antibiotic therapy (Fig. 1). Postoperative mortality among the patients with anastomotic leakage was not recorded.
leaks were detected in a mean period of 6.9 days. The most usual clinical presentations were pelvic pain and unexplained elevation in temperature with tachycardia. A CT scan verified the diagnosis in all cases. Our approach to management was immediate surgery in cases of major leaks characterised by peritonitis or/and septicaemia or broad leaks. Minor leaks were managed conservatively with successful outcome. The conservative treatment required close monitoring of the patients that provided the option of immediate surgery in cases of sepsis. Furthermore, extraperitonealisation of the anastomosis appeared to mitigate the consequences of leakage and represents an advantage for the conservative treatment of minor leaks [7]. The reported postoperative mortality rate associated with anastomotic complications ranges from 6% to 22% [3]. In particular, anastomotic leakage after low anterior resection is accompanied by a mortality rate as high as 30% and is known to be the most frequent cause of death after low anterior resection of the rectum. The main causes of fatal outcome are peritonitis and septic complications [1, 3]. However, in our study postoperative mortality among the patients with anastomotic leakage was not recorded. In conclusion, the incidence of anastomotic leakage after anterior resection of the rectum is relatively low although it remains the most serious and life-threatening early complication of rectal surgery.
Discussion Despite the recent advances in rectal surgery, anastomotic leakage continues to pose a major clinical problem. The reported incidence of clinical leakage rate after anterior resection for rectal cancer varies from 3% to 21% [1]. In our study, the clinical leakage rate was 9.7% and concerned patients who underwent low anterior resection with TME. TME is regarded as a significant risk factor increasing anastomotic leak rate [3]. Further, considering the cause analysis of anastomotic leakage after anterior resection for rectal cancer, many parameters such as tension of the anastomosis, bacterial contamination, anastomotic blood flow and surgical techniques are reported to be responsible for having adverse effects on the healing of anastomoses. However, the most important risk factor for leakage is the height of the anastomosis from the anal verge. The lower the anastomosis, the higher the risk. Furthermore, the high-risk level for anastomotic leakage is reported to be between 5 and 7 cm from the anal verge [1, 4]. In our study, among the patients with anastomotic leakage the mean distance of the anastomoses from the anal verge was 6.2 cm (range 3–9 cm), a level of anastomosis that according to the above mentioned data represents the most important risk factor for leakage. It is generally believed that covering stoma in patients that have to undergo low anterior resection does not abolish leakage but mitigates the consequence. However, controversy persists as to whether a prophylactic defunctioning stoma should be performed routinely. In our study, none of the 72 low anterior resections was covered by a defunctioning stoma, as long as it remains controversial largely due to a lack of adequate controlled studies [5]. Early detection of anastomotic leakage is possible by careful clinical monitoring and involves a high index of suspicion. In addition, early diagnosis of a leakage, and urgent therapeutic intervention are required to avert lifethreatening conditions. The usual presentation of anastomotic leakage is unexplained pyrexia often with tachycardia [1, 6]. Abdominal signs are initially absent in the extraperitoneal anastomosis. Prior to peritonitis, early detection is possible by careful clinical monitoring together with a CT scan with rectal contrast enema. In our study,
References 1. Alberts JCJ, Parvaiz A, Moran BJ (2003) Predicting risk and diminishing the consequences of anastomotic dehiscence following rectal resection. Colorectal Dis 5:478–482 2. Karanjia ND, Cored AP, Bearn P, Heald RJ (1994) Leakage from stapled low anastomosis after total mesorectal excision for carcinoma of the rectum. Br J Surg 81:1224–1226 3. Scott N, Jackson P, al Jaberi T (1995) Total mesorectal excision and local recurrence: a study of tumor spread in the mesorectum distal to rectal cancer. Br J Surg 82:1031–1033 4. Rullier E, Laurent C, Garrelon JL (1998) Risk factors for anastomotic leakage after anterior resection of rectal cancer. Br J Surg 85:355–358 5. Moran BJ, Heald RJ (2000) Anastomotic leakage after colorectal anastomosis. Semin Surg Oncol 18:244–248 6. Sutton CD, Marshall LJ, Williams N, Berry DP, Thomas WM, Kelly MJ (2003) Colorectal anastomotic leakage often masquerades as a cardiac complication. Colorectal Dis 6:21–22 7. Jatzko GR, Lisborg PH, Wette VM (1996) Extraperitonealization of the anastomosis and sacral drain in restorative surgery for rectal carcinoma: a safely mechanism in the absence of a covering stoma. Surg Today 26:591–596
