Anastomotic Troubleshooting

Abstract

Anastomotic leak remains a critical and feared complication in colorectal surgery. The development of a leak can be catastrophic for a patient, resulting in overall increased morbidity and mortality. To help mitigate this risk, there are several ways to assess and potentially validate the integrity of a new anastomosis to give the patient the best chance of avoiding this postoperative complication. A majority of anastomoses will appear intact with no obvious sign of anastomotic dehiscence on gross examination. However, each anastomosis should be interrogated before the conclusion of an operation. The most common method to assess for an anastomotic leak is the air leak test (ALT). The ALT is a safe intraoperative method utilized to test the integrity of left-sided colon and rectal anastomoses and most importantly allows the ability to repair a failed test before concluding the operation. Additional troubleshooting is sometimes needed due to technical difficulties with the circular stapler. Problems, such as incomplete doughnuts and stapler misfiring, do occur and each surgeon should be prepared to address them.

Anastomotic leak is one of the most dreaded complications in colorectal surgery. The development of a leak can be catastrophic for a patient, resulting in significantly increased morbidity and mortality. ^{1 2} Anastomotic leak is the complication with the greatest effect on the incidence of 30-day end-organ dysfunction and contributes the most to postoperative readmission and reoperation in patients who have undergone elective colorectal surgery. ³ While some patient and operative risk factors have been identified, many leaks still develop with no known etiology, making it difficult to predict which patients will develop this complication. ^{4 5 6} Unfortunately, the subjective assessment of the integrity of anastomoses by surgeons at the end of an operation has been shown to have low predictive precision at determining who will develop a leak. ⁷ Therefore, it is incumbent upon surgeons to closely evaluate the quality of their anastomoses intraoperatively to decrease the incidence of an anastomotic leak.

Intraoperative Anastomotic Assessment

In relatively few cases, a defunct anastomosis is visually apparent. Evident signs of impending anastomotic failure include tissue ischemia or an obvious defect between the stitches or staples that have been placed. If the anastomosis is visibly compromised, it is obligatory that the surgeon do a full reconstruction of the anastomosis if possible.

High-Risk Patients

There is a subset of high-risk patients who, although may appear to have an intact anastomosis, may be of greater risk of anastomotic leak due to their comorbidities and perioperative factors. Risk can be categorized into three groups: patient factors, pathologic factors, and technical factors. Patient factors that place patients at higher risk include obesity, steroid use, elevated Charlson's comorbidity score, and having a low lying tumor. ^{6 8 9 10} Pathologic factors include if the lesion is malignant and if the patient received preoperative radiation. ¹¹ Finally, technical factors include intraoperative transfusion, duration of operation, perfusion of the anastomotic tissue, and the amount of tension on a low anastomosis. ^{10 12 13} High-risk patients may benefit from a planned diverting loop ileostomy (DLI), if the threat of leak is concerning enough. This concern should be especially taken into consideration for patients undergoing low rectal anastomoses. Multiple studies have confirmed that low-lying anastomoses have a significantly higher risk of developing anastomotic leak when compared with higher sitting anastomoses. ^{14 15 16} The use of a DLI is to temporarily divert stool as the anastomosis heals, which may be beneficial in this situation.

Air Leak Test

A majority of anastomoses will appear intact with no obvious sign of anastomotic leak on gross examination. However, these patients may still be at risk of having a leak and require closer evaluation. The most common method to troubleshoot an anastomotic leak is the air leak test (ALT). The ALT is a safe method to intraoperatively test the integrity of left-sided colon and rectal anastomoses before the conclusion of the operation and most importantly allows the ability to repair a failed test during the initial procedure.

How to Perform an Air Leak Test

There are many variations on how surgeons perform the ALT but the general principal remains the same. This procedure is best performed with two surgeons, be that two attendings or an attending supervising a resident. Once the anastomosis has been made, either stapled or handsewn, the pelvis is filled with 1 L of normal saline to fully submerge the extraluminal aspect of the new anastomoses ( Fig. 1A ). One surgeon breaks sterility and gently inserts a flexible endoscope into the rectum and the bowel is slowly insufflated with air. The scope is carefully advanced to a position just distal to the anastomosis. The bowel proximal to the anastomosis should be gently occluded, either by a hand in an open operation or with an atraumatic grasper in a laparoscopic or robotic case. Once the upstream bowel is occluded by the sterile surgeon, the contaminated surgeon will continue to slowly fill the bowel with air from the endoscope until the edges of the anastomoses are visualized intraluminally ( Fig. 1B ). The sterile surgeon who is occluding the bowel should observe the pelvis closely while the insufflation is occurring to look for bubbles exiting from the anastomosis. The presence of bubbles is indicative of a positive ALT. Insufflation of the bowel should take no longer than 10 to 15 seconds to limit barotrauma to the new anastomosis. Once satisfied, the sterile surgeon may remove their hand or instrument from the upstream occlusion. The endoscope should now be carefully withdrawn with the application of pulsed suction to remove the air out of the bowel as it is withdrawn.

Fig. 1.

Intraoperative colorectal anastomosis air leak test (ALT). ( A ) Intraabdominal view of pelvis filled with normal saline. The new anastomosis is submerged in the saline and the proximal bowel is gently occluded with a blunt grasper. ( B ) A sigmoidoscope allows direct visualization of the anastomosis and provides air for gentle insufflation to check for bubbles in the ALT.

Variations of this procedure exist in the method of how air is introduced intraluminally. Other surgeons have described the use of 60-cc syringes to instill air into the rectum ( Fig. 2 ). With this method, the tip of the syringe is carefully introduced into the rectum and air within the syringe is manually expressed into the bowel, again with upstream occlusion by the sterile surgeon. The same principle applies in that the operating team is looking for bubbles to leak from the anastomosis into the saline that fills the pelvis. Our practice favors use of an endoscope to gently instill air because if 60-cc air is used incorrectly, it can result in a false-negative ALT. In the hands of a trainee or if not inserted close enough to the anastomoses, the air can dissipate resulting in no passage of bubbles through a defunct anastomosis. Meanwhile, gentle air insufflation from the scope has higher reliability of bubbles escaping into the pelvis if a defect is present. Additionally, visualization from the endoscope can allow for safe insertion that keeps the anastomosis in view to minimize risk of inadvertent iatrogenic trauma and also allows for the detection of other anastomotic complications, such as an anastomotic bleed, intraluminal ischemia, or compromise of the anastomosis, which would have otherwise been missed.

Fig. 2.

Air leak test (ALT) with use of a 60-cc bulb syringe. Reprinted with permission of Cleveland Clinic Foundation; All Rights Reserved.

While the absence of bubbles does not guarantee an intact anastomosis, studies have shown that the rate of clinical leak (CL) is lower than if the anastomosis was not tested. A randomized trial by Beard and colleagues randomly allocated 145 consecutive patients who were undergoing an operation with a colorectal anastomosis to either receive an ALT or not. ¹⁷ A total of 74 patients received an ALT while 71 patients did not. In the test group, 18 leaks (25%) were detected and repaired intraoperatively. Following the operation, there were 3 (4%) CL in the ALT group and 10 (14%) CL in the non-ALT group ( p  = 0.043). This study also analyzed rates of radiologic leaks (RL) by performing water-soluble contrast enemas on postoperative day 10. There were 8 (11%) RL in the ALT group and 20 (29%) RL in the non-ALT group ( p  = 0.006). This prospective work concluded that performing an ALT is a safe and feasible way to detect an anastomotic leak intraoperatively and significantly decreases the risk of anastomotic leak. To address safety concerns that arose due to the novelty of the ALT at the time of publication, Beard and colleagues described that the use of “air insufflation from a sigmoidoscope is unlikely to disrupt an anastomosis because, unlike saline, air is compressible, and so a rapid rise in pressure with increasing volume does not occur.” ¹⁷

How to Manage a Positive Air Leak Test

Should an ALT be positive, there are three options on how to proceed ( Fig. 3A ). Dependent on the size of the anastomotic leak and on surgeon preference, the anastomosis may be oversewn, diverted, or reconstructed. In many situations, if a small anastomotic leak is detected after an ALT, a surgeon may choose to oversew the anastomosis at the site of the defect. This is done with interrupted, absorbable suture. Following this repair, another ALT must be performed to ensure the effectiveness of the repair demonstrated by a negative ALT. Another option is to utilize a DLI to allow the compromised anastomoses to heal. Surgeons may choose to use a DLI alone or in combination with oversewing of the defect ( Fig. 3B ). If oversewn, an ALT must be repeated until negative. The third option is to redo the entire anastomosis. This is done by resecting the incompletely stapled portion of the bowel. All prior staples must be removed, so as to not interfere with the second attempt at stapling the bowel. Once two fresh ends of bowel are obtained, the anastomoses may be reconstructed, either in a stapled or handsewn fashion, as per surgeon preference. Once reconstructed, an ALT must again be performed. This method may be difficult in an already low rectal anastomosis as there may not be much length remaining for a new anastomosis.

Fig. 3.

( A ) Positive air leak test (ALT) of a colorectal anastomosis after sigmoid resection. The bubbles in the saline (Jacuzzi Sign) indicate a defect in the staple line that requires further evaluation. ( B ) Primary repair of the anastomotic defect which was protected by a diverting loop ileostomy.

Benefits of an Air Leak Test

Multiple studies have analyzed the efficacy of the ALT with a general consensus that it is necessary and beneficial to the patient. Allaix and colleagues compared 398 patients who underwent an ALT to 379 patients who did not have an ALT (non-ALT) done. ¹⁸ There was a lower rate of CL in the ALT group compared with the non-ALT group. A total of 20 patients (5%) in the ALT group had a positive test. To remedy this positive test, 14 of the patients were given a DLI with 2 of the 14 also receiving a total reconstruction of the anastomoses, while the other 6 patients underwent only a sutured repair of the leak site. None of the 20 patients had a CL postoperatively. Of all the patients in the study, the CL rate was 2.5% in the ALT group and 5.8% in the non-ALT group ( p  = 0.03). The use of an ALT was found to be independently associated with a decreased rate of postoperative anastomotic leak.

No difference has been found in the risk of leak between stapled and handsewn anastomoses. Ricciardi and colleagues surmised this after analyzing a cohort of 998 patients who underwent left-sided colorectal anastomoses. ¹⁹ A CL occurred in 48 of the 998 patients (4.8%) with no significant difference found in leak rates between stapled and handsewn anastomoses. An ALT was performed in 825 (82.7%) of the patients, of whom 65 (7.9%) had a positive leak test. There was a higher rate of CL in the group that did no undergo an ALT. And despite the positive leak tests being repaired, the rate of CL in repaired anastomoses was still higher than in those who were initially air tight. Their discussion argues that diversion or redo of the anastomosis may be better than simple suture repair.

Technical Difficulties

The use of a circular stapler has now become more common than the traditional handsewn technique to make a colorectal anastomosis. The choice of technique is left to the operating surgeon; however, should a circular stapler be used, the operator must be aware of certain technical difficulties that have been described. These difficulties exist no matter the type of device used, with the most commonly used circular staplers in the United States being the Covidien EEA (Minneapolis, MN) and the Ethicon circular stapler (Bridgewater, NJ). Despite increased usage and familiarity with this tool, there is still approximately a 20% rate of technical error with the device. In a study of 349 operations, there were 67 errors with the most common errors being positive leak tests ( n  = 19), operator error ( n  = 18), and incomplete doughnuts ( n  = 13). ²⁰ Patients who experienced a technical error had higher rates of gastrointestinal bleed, ileus, and were twice as likely to have an unplanned DLI.

Stapler Misfire

Device failure can manifest in various ways. There are times when the circular stapler may not disengage from the bowel. This situation occurs when the tissue “doughnuts” have been cut and the staples fired, but the device will not completely disengage from the bowel that had been anastomosed together, which prevents the withdrawal of the device. Often, either slow rotation of the stapler or gentle advancement of the stapler, 1- to 2-cm proximal to the anastomosis will disengage the device from the tissue. One study reports disconnecting the main handle of the stapler from the anvil and extracting the handle from the rectum, while leaving the anvil within the bowel. ²¹ They conclude that doing so will prevent any additional injury to the tissue that may occur from pulling or manipulating a stapler if it is stuck. A sigmoidoscope can then be inserted into the rectum and hot forceps may be used to dissect the piece of tissue still attaching the anvil to the bowel. An additional approach to extract a separated anvil is to use either a transanal endoscopic microsurgery (TEM) or transanal minimally invasive surgery (TAMIS) approach. Regardless of the method used, performing an ALT is pertinent in this scenario to ensure an intact anastomosis.

An additional type of staple misfire is when the stapling device cuts the tissue but does not staple. In this situation, the stapling device has only completed half of its designed function. After placing the stapler within the rectum, connecting with the anvil and engaging the tissue, the first function of the device is to cut the excess tissue held between its jaws at the site of the new anastomosis. This serves to remove any potentially ischemic ends of the bowel and also prevents the existence of excess tissue within the bowel lumen. The cut function is then followed by the staple function in which two rows of staples are fired uniformly to secure the new anastomosis. Occasionally, the stapling device will perform the first function but fail to perform the second. Without the placement of staples, this results in a failure of the anastomosis to seal together and also results in bleeding from the cut edge of the tissue. The only option in this situation is to redo the anastomosis.

Incomplete Anastomotic Doughnut

One indication that a stapler has misfired or has fired incompletely is when the anastomotic “doughnut” is incomplete. When the circular stapler fires, it creates two rings of tissue that look akin to a doughnut which are removed along with the stapler. To completely staple two ends of full-thickness mucosa together without defect, the resultant rings should be intact. Dr. Francis Nance was the first to report use of the new EEA stapler to create gastrointestinal anastomoses in 1979. ²² He warned that “an incomplete doughnut should be regarded as a sign of a defective anastomosis. The anastomosis should be resected unless careful inspection shows it to be intact.” Therefore, it is standard practice by many surgeons to check the integrity of the doughnuts that are removed along with the circular stapler. Two studies show that incompleteness of doughnut was consistently associated with an increased risk of a positive ALT. ^{16 23} However, this was not associated with an increased incidence of CL, likely because the anastomoses were repaired after the positive ALT was noted. Griffith and Hardcastle concluded that reliance should not be met by obtaining complete doughnuts but an ALT should still be performed each time. ²³ It should be mentioned that, to properly assess the entire circumference of the tissue doughnut, the purse-string suture must be cut and removed, as this stitch can artificially make the doughnut appear intact ( Fig. 4 ). ^{16 24}

Fig. 4.

Tissue doughnuts after use of a circular stapler. ( A ) Examples of a complete distal and proximal doughnut with purse string in place. ( B ) Incomplete doughnut.

In the largest cohort analyzed, a series of 1,000 stapled anastomoses were retrospectively reviewed. Detry and colleagues noted that the most common problems were hematoma ( n  = 45), muscular breaches ( n  = 30), incomplete doughnuts ( n  = 29), or full-thickness defects ( n  = 20). ¹⁵ The patients with incomplete doughnuts were treated with repair sutures over the presumed site of defect and/or a diverting colostomy. The full-thickness defects were repaired with suture and nine cases received a concurrent diverting colostomy and seven defects were wrapped with omentum. Of the 20 full-thickness defects, only one developed a CL.

Anastomotic Doughnut Characteristics

Surgeons have attempted to describe additional morphological characteristics of anastomotic doughnuts to better predict which patients will develop an anastomotic leak. Aside from completeness of the doughnut ring, minimal conclusive findings have been reported. However, one study by Cauchy and colleagues described the significance of the minimal height of the resected colonic doughnut as a predictive indicator of anastomotic integrity. ²⁵ From a cohort of 154 patients, the intraoperative measurements of maximal diameter and maximal and minimal heights and widths of both the colonic and rectal doughnut were reported. The only measurement significantly associated with risk of anastomotic leak was the minimal height of the colonic doughnut. A minimal height of <4.5 mm had an odds ratio (OR) of 5.74 for development of anastomotic leak with an anastomotic leak rate of 20.9% below this height and 3.4% above it. The practicality for a surgeon to perform these intraoperative measurements has yet to be determined, given that this is a single report, yet additional work in anastomotic doughnut characteristics could prove beneficial.

Staple Line Bleed

Another anastomotic complication is the occurrence of a staple-line bleed. Typically, the bleed will occur intraluminally at the cut edge of the tissue. With proper use of the circular stapler this should ideally not occur. By holding firm, continuous pressure on the stapling device handle for a minimum of 30 to 60 seconds, this ensures the completed deployment of all staples which results in an anastomotic connection in addition to hemostasis at the previously cut edge. However, with improper operator use of the device or anticoagulated patients, bleeding from the cut edges of tissue can occur. The design of the circular stapler results in the staples and cut edges residing on the intraluminal side of the anastomosis which is most often where the bleeding is located. This is an additional reason why our group utilizes a flexible sigmoidoscope when performing the ALT as the scope allows for the visualization of any bleeding that can be immediately addressed. Bleeding can often be controlled through use of the placement of hemoclips or through transanal techniques. Excessive bleeding typically only occurs when the device fails to staple after cutting the tissue in which case the entire anastomosis has to be redone, as mentioned above. This complication is rarer when handsewn bowel anastomoses are performed.

Conclusion

Anastomotic leak is a devastating complication of colorectal surgery. However, there are methods to troubleshoot your anastomoses in an effort to prevent the occurrence of a leak. Every new anastomosis should undergo an ALT to find any acute, technical defects. While a negative ALT will not guarantee that a CL will not develop, the evidence clearly indicates that many potential leaks are found and troubleshooted this way, resulting in improved patient outcomes. Surgeons should additionally be well trained in proper stapling technique, should they chose to use a circular stapler and be aware of the potential technical problems that can occur with use of this device.