Anastomotic Leak/Pelvic Abscess

Fig. 32.1
Postoperative diagnostic and therapeutic algorithm. PCD percutaneous drainage

Type I: Generalized Peritonitis

Approximately 40–45 % of all anastomotic leaks present in this fashion. Patients complain of severe abdominal pain and have a high fever, tachycardia, marked leukocytosis, and signs of generalized peritonitis such as rebound abdominal tenderness and/or rigidity. In these cases, imaging is not required to make a diagnosis of an anastomotic leak. Initial efforts should focus on broad-spectrum antibiotics and aggressive fluid resuscitation.

With the patient in a modified Lloyd-Davies position, an examination under anesthesia will allow for an assessment of the anastomosis, as well as access to the rectum in the event that rectal washout is indicated after abdominal–pelvic exploration. The operative exploration aims to identify the site and extent of leakage, contain further leakage, and aggressively lavage the entire abdominal/pelvic cavity. In cases where the anastomotic disruption and contamination are minimal, pelvic drainage and a proximal diverting stoma may be all that is required. Depending on the long-term plans, a temporary loop ileostomy is a quick option, but the possibility of high output and associated dehydration needs to be taken into consideration in the context of the patient’s age and overall comorbidities. In cases where dehydration is of concern and/or a permanent proximal diversion is envisioned, a left upper quadrant, end-loop (Prasad–Abcarian) colostomy is an option that prevents further contamination, yet affords passage of mucus from the efferent limb should a distal obstruction from a stricture at the anastomosis develop. If the colon is full of feces and therefore a source of further ongoing contamination, a proximal end or end-loop colostomy with distal limb rectal washout is indicated. When creating an end-loop colostomy, great care must be exercised to avoid damage to the marginal artery, which perfuses the afferent limb of the anastomosis, and may be the only source of perfusion in cases where the left colic artery is sacrificed during the initial resection. In cases where a large anastomotic disruption and contamination have occurred, a redo low anterior resection is not a viable option, and these cases are best managed by bringing the afferent limb out as an end colostomy and attempting to suture/staple shut the efferent limb; one should then create a Hartmann’s pouch and assure adequate drainage of the pelvis with drains.

Type II: Localized Pelvic Abscess

About 30–40 % of patients with an anastomotic leak present with vague abdominal pain, prolonged ileus, mild fever, leukocytosis, and abdominal distension associated with ileus and localized peritoneal signs. Such patients warrant an abdominal pelvic CT scan with oral, IV, and rectal contrast, when possible. When no extravasation of contrast is noted, it is sometimes difficult to distinguish between an anastomotic leak and a postoperative abscess. If the abscess is of reasonable size and walled off, it should be drained percutaneously under radiologic guidance. Success rates for CT-guided placement of a percutaneous drainage catheter are about 80 % [25]. Following successful drainage, some patients go on to develop a chronic enterocutaneous fistula or sinus, which can be managed conservatively. Although most of these patients do not usually require a reexploration, some fail this conservative approach and develop symptoms late after discharge from the hospital. Long-term close follow-up is therefore warranted.

Type III: Fistula

A fistula may be a long-term sequelae in upward of 25–30 % of anastomotic leaks following a low anterior resection. A thorough evaluation includes a careful physical examination as well as optimal contrast imaging with a CT scan and fistulogram in order to delineate a fistula track and rule out distal obstruction. Endoscopy may be helpful in evaluating the orifices of a fistula. If the fistula is well drained and the overall nutritional condition of the patients is adequate, a conservative approach including optimal skin care with a stoma appliance, low suction device, with or without antibiotics, and somatostatin are often sufficient to control the fistula. In cases with a persistent, poorly drained, or intolerable fistula, a surgical approach should be considered. Local therapy with fibrin glue or plugs is rarely effective in this setting. Surgical options include creation of proximal diverting stoma versus primary closure of a rectovaginal fistula via a transanal approach with an advancement flap and temporary stoma or redo low anterior resection.

Long-Term Outcome

In addition to the immediate impact on postoperative morbidity and mortality, anastomotic leakage also impacts long-term functional outcome and possibly even long-term prognosis of the rectal cancer patient.

Need for a Permanent Stoma

Following an anastomotic leak, some patients may be left with a permanent stoma because the leakage and associated contamination may result in such profound fibrosis and scarring of the pelvis and residual rectum so as to prohibit a re-resection and creation of a supple and functional primary colorectal or coloanal anastomosis. In addition, because of fear of further complications, the surgeon and patient alike may choose to not pursue further major surgery. In these cases, the patient is left with either the initially created protecting proximal stoma or the stoma created during the reoperation following the anastomotic leak. In these situations, a reoperation may nevertheless be required in order to convert an ileostomy into a colostomy resulting in fewer evacuations and less volume loss. The reported rates for the need of a permanent stoma in these circumstances are 2.9 ~ 19 % [2629].

Stenosis or Stricture

The rate of clinically significant stenosis or stricturing of a colorectal anastomosis ranges from 3 to 30 %, depending on the criteria employed. The most common cause of anastomotic stenosis/stricture is anastomotic leakage [30, 31]. Management, which depends on the severity of the stenosis, includes simple dilatation (digital or Hagar), balloon dilatation (radiologic or endoscopic), proximal diverting stoma, Hartmann procedure with/or without resection of stricture site, and redo low anterior resection. Because local recurrence is the second most common cause of anastomotic stricture after sphincter-preserving surgery for rectal cancer, a biopsy of the stricture should be obtained whenever feasible in order to exclude cancer.

Anorectal Dysfunction and Quality of Life

Following a sphincter-preserving low anterior resection, anorectal dysfunction in the form of fecal incontinence, evacuation problems, and clustering of bowel movements is frequently noted [32]. Anastomotic leakage has been identified as a predictive factor of anorectal dysfunction. Although there are few reports addressing bowel function and quality of life after an anastomotic leak of a low anterior resection (Table 32.1) [6, 33, 34], it is generally agreed upon that long-term function is impaired in patients with anastomotic leakage.

Table 32.1
Impact of anastomotic leakage after low anterior resection for rectal cancer on bowel function and quality of life




No of pts

F/U period

Leakage rate (%)


Results and conclusion
Bowel function


Hallböök et al. [8]




30 months


There was no difference in sphincter function
Neorectal volume, compliance, urgency, and MTV were significantly reduced in patients with leakage
Long-term functional outcome may be impaired

Nesbakken et al. [6]




2 years

Manometry questionnaire
Reduced neorectal capacity, more evacuation problem, and a trend toward more fecal urgency and incontinence
Long-term anorectal function had been impaired

Ashburn et al. [34]




3.2 years



Cleveland Global QoL

One year: worse physical and mental component scores, more frequent daytime and nighttime bowel movements, and worse control of solid stool
Short-Form 36

Recent: worse mental component scores and increased use of perineal pads
Early adverse consequences on bowel function and QoL

FISI fecal incontinence severity index, MTV maximal tolerable volume

Local Recurrence

Great controversy exists on whether anastomotic leakage following a rectal cancer resection is a prognostic factor for local recurrence and/or survival (Table 32.2) [16, 3542]. Conflicting results may be due to varied definitions of anastomotic leaks, patient selection, heterogeneity of cases (colon and rectal cancer versus rectal cancer, alone), and variability in the management of the anastomotic leak.

Table 32.2
Impact of anastomotic leakage after low anterior resection for rectal cancer on oncologic outcomes




# pts

F/U period

Leak rate

Local recurrence

Cancer-specific survival

Overall survival

AL (+)

AL (−)

p value

AL (+)

AL (−)

p value

AL (+)

AL (−)

p value

Merkel et al. [35]


Single center


90 months








Bell et al. [36]


Single center






Eriksen et al. [16]


National cohort


7.6 years





< div class='tao-gold-member'>

Only gold members can continue reading. Log In or Register to continue

Jun 28, 2016 | Posted by in GASTOINESTINAL SURGERY | Comments Off on Anastomotic Leak/Pelvic Abscess
Premium Wordpress Themes by UFO Themes