Conservative Approaches to Salvaging Patients With Chronic Anastomotic Leaks

ABBREVIATIONS

CRP

C-reactive protein

CT

computed tomography

ECM

biologic extracellular matrices

EUA

examination under anesthesia

OTSC

over-the-scope clipsTAMIS, transanal minimal invasive surgery

VAC

vacuum-assisted closure

INTRODUCTION

Anastomosis of the bowel anywhere in the gastrointestinal tract restores function after resection or bypass. However, with every anastomosis comes the possibility of an anastomotic leak, and the escape of intestinal content outside the bowel is usually problematic. Leaks, and their consequences, vary in severity and their effects on the patients, so options for their management are also varied. These options include observation with antibiotics and possibly fecal diversion, drainage of the leak by interventional radiology or surgery, repair of the leak or redo anastomosis at reoperation, secondary closure of a contained leak by laying it open, and closure of the leak using endoscopic clips or sutures. The choice of options is shown in Fig. 27.1 and depends on the location (and hence the accessibility) of the anastomosis, the degree to which the contamination from the leak is sealed off from the peritoneal cavity, the timing of the leak, and the clinical sequelae of the leak. Intraperitoneal anastomotic leaks generally need surgery unless the sepsis around the anastomosis is contained and presents as a drainable abscess or a stable enterocutaneous fistula. Such quarantined leaks are usually late developments and occur during a time of dense postoperative adhesions when reoperation may not be easy and is sometimes impossible. With effective counter drainage, leaks may heal spontaneously, especially below a diversion, and the options of waiting to see if closure occurs, or trying an endoscopic control of the defect are most attractive. As the last resort, an elective resection and redo of the anastomosis is indicated.

Leaks from anastomoses in the rectum are most amenable to endoscopic control, particularly in the low rectum. Here, the anastomosis is in the pelvis, quarantined from the peritoneal cavity by the omentum and small intestine, and easily reachable with a variety of rigid and flexible proctoscopes. In this chapter, a conservative approach to leaks will be discussed, focusing on leaks from a colorectal anastomosis.

CAUSES, RISK FACTORS, AND TESTING THE ANASTOMOSIS

A leaking colorectal anastomosis may be the result of one or several factors. These factors can be grouped into those relating to the patient, their disease, the stapling instruments, and the surgeon. Patient factors include gender, especially with rectal anastomoses, as the narrow male pelvis creates technical issues with exposure and puts operative skill and experience at a premium. Body habitus is important, as obese patients face a challenge because of limited exposure, especially if the tumor is low in the pelvis. Risky patient comorbidity includes anemia, immunosuppression, smoking, and malnutrition, while recent data suggest that a patient’s gut microbiota may be implicated in the pathogenesis of anastomotic leaks, meaning that there is a potential for mechanical and antibiotic bowel preparation to modify this risk. Disease-related factors include prior use of radiation and concomitant sepsis, which impair wound healing and argue against performing an anastomosis in a high-risk patient. ^, The location of the cancer, which determines the level of the anastomosis, affects the risk of a leak, with low anastomoses more risky than high ones. Instrument failure is an occasional cause of leaks and is unpredictable and surprising. If a misfire is detected, the anastomosis can be redone or reinforced, either with another stapler or by hand. Technical issues are a more common cause of an anastomotic leak than stapler dysfunction and include ischemia due to aggressive clearing of the bowel edges or mishaps with the mesenteric vessels, creation of an enterotomy proximal to the anastomosis either de novo or at the site of a diverticulum, excessive tension at the anastomosis due to inadequate mobilization of the bowel, and mismatched diameters of bowel ends leading to the inability of the wider end to fit into a small circular stapler. Many potential leaks can be prevented by using good judgment in deciding whether to anastomose or divert and good technique in the construction of the anastomosis. After the anastomosis, examination of donuts and air testing of the staple line for low anastomoses are important ways of making sure that there is at least not an immediate gross leak. ^, There is also the advantage of finding a leak intraoperatively when it can be repaired and retested.

Some patients with a colorectal or coloanal anastomosis will have had neoadjuvant chemoradiation, and some will have been given a loop ileostomy to protect from an anastomotic leak. The chances of an anastomotic leak under these circumstances vary from a low of 3.5% to a high of 18%. ^,

Some anastomotic leaks are never discovered and simply heal. These are subclinical. The consequences of an anastomotic leak depend on whether the stool has been diverted by a stoma, whether there is peritoneal contamination, the size of the leak, the time it happened relative to the surgery, the time it was discovered, and the adequacy of the initial management. Clinical leaks are those causing symptoms. A large, undiverted, early leak will probably cause peritonitis and need reoperation with lavage and diversion. A large, diverted leak is less likely to cause peritonitis and may be managed with antibiotics. Small leaks are less symptomatic and less likely to cause peritonitis, may be partially or fully sealed, and may present with a para-anastomotic abscess or cavity.

DIAGNOSIS

Anastomotic leaks are usually diagnosed because the patient has some perturbation in one of their postoperative parameters, such as elevated heart rate, temperature, white blood cell count, or C-reactive protein (CRP). Testing with gastrografin enema or computed tomography (CT) scan with rectal contrast is a good way of detecting the leak and may be followed up with an examination under anesthesia (EUA) to confirm it. ^,

MANAGEMENT

If the leak is small and discovered early and if it has minimal impact on the patient (either because it is very small or because the anastomosis has been diverted), antibiotics alone may be adequate. This is especially the case if there is not an abscess in the perirectal spaces. If the leak is due to ischemic bowel ends, then no conservative treatment will result in healing, but at least the redo anastomosis can be elective. If the leak is large and discovered late, with fecal peritonitis and obvious sepsis and a loaded colon above the leak, the anastomosis will need to be taken down and a proximal end stoma created. The aims of treatment are primarily to manage the sepsis and prevent death and then to preserve the anastomosis if possible so that ultimately per anal defecation can result. Here, we focus on less calamitous situations, where the integrity of the anastomosis can be preserved. Endoscopic and transanal techniques for managing an anastomotic leak will be summarized and reviewed ( Table 27.1 ).

Table 27.1

Summary of Conservative Techniques for Managing Anastomotic Leaks

Technique	Equipment	Application
Drainage	Mushroom catheters, malecot catheters	All leaks as needed
Lay open/marsupialization	Stapler, surgery	Established anastomotic sinuses
Glue/ECM	Vistaseal, Ethicon, Johnson and Johnson. Tisseel, Baxter (Acell Inc., Columbia, Republic of Moldova Republica)	Narrow anastomotic sinuses
Flap repair	Surgery	Small, shallow defects in the anastomosis
Endo-SPONGE	(B. Braun Medical Inc.)	Large, fresh defects in a colorectal anastomosis
Clips	(Over-the-scope clips, Ovesco Endoscopy USA, Inc. Cary, New Caledonia)	Small defects with supple edges
Sutures	(Apollo Endosurgery Inc, Austin, TX)	Any defects that can be reached
Stents	WallFlex Colonic Stents—Boston Scientific	Mid and upper rectal anastomotic defects