with Dr. Romina Pena
Anastomotic leaks are among the most devastating complications following colorectal resection. Despite improvements in preoperative preparation, surgical technique and postoperative care, complications varying from surgical site infection (SSI) to anastomotic leakage are still a heavy burden in colorectal surgery.
As to why they occur, surgeons have been studying for decades the factors involved in anastomotic healing. A problem in the connection engineering, type of suture material or mechanical devices, patient-related causes such as a poor nutritional status. and neoadjuvant therapy have all been the culprit at some point in history.
In an effort to reduce postoperative complications such as surgical site infections, preoperative digestive decontamination was first introduced in the seventies.
Mechanical bowel preparation (MBP) was once routinely used and thought to improve outcomes, but evidence from randomized controlled trials and Cochrane reviews published nearly two decades ago suggested otherwise. Nevertheless, results from several large retrospective series challenged non-bowel preparation (NBP) and suggested that MBP plus oral antibiotics (MOABP) decrease the rate of SSI and overall complications. These studies compelled four prominent societies to change their recommendations: ASCRS, SAGES, ERAS and the Perioperative Quality Initiative, which have since recommended MOABP over NBP. Nowadays, more studies are being published on the subject.
Due to gaps in evidence-based information and the need for standardization, the World Health Organization (WHO) decided to publish guidelines for the prevention of SSI in 2016. Indications were provided on preoperative, intraoperative and postoperative care. The use of MOABP for elective colorectal surgery in adult patients was suggested, and the use of MBP alone was strongly recommended against, based on the result of several randomized controlled trials. Nevertheless, the quality of evidence was moderate, and the protocols differed across trials. Intravenous antibiotic prophylaxis was recommended in addition to MOABP.
Later on, CDC released its own prevention guidelines in 2017. General recommendations on prophylactic antibiotics, normoglycemia, normothermia and an increased fraction of inspired oxygen for the prevention of SSIs were given, but there were no comments regarding colorectal surgery and MBP specifically.
To support these guidelines, a large number of studies was taken into consideration. Many of them focused on the effect of digestive decontamination on surgical site infection as well as anastomotic leakage as main outcomes. We present some of the most recent ones, starting with observational studies:
This study was published in the Annals of Surgery in 2018. It included more than 30000 patients who underwent elective colorectal surgery. It was an observational study with a propensity score analysis. This study aimed to analyze the effect of MBP on surgical site infection, but among the results, it stated that when compared to the non-preparation group, the MBP and MOABP groups had a lower risk of anastomotic leak.
The next study was published in the Journal of Surgical Oncology in 2019, and included almost 500 patients who underwent surgery for colorectal cancer. The aim was to evaluate the effect of oral antibiotics in preventing anastomotic leak. Results showed that the leak rate was lower when oral antibiotics were administered. This statement coincides with the WHO recommendations.
We move on to a higher level of evidence.
In February 2019, the SELECT trial was published. It was a RCT that included 455 patients and concluded that selective digestive decontamination (SDD) with oral antibiotics (oral colistin, tobramycin and amphotericin B) reduces infectious complications after colorectal cancer resection but did not significantly reduces anastomotic leakage. SDD was intended to eliminate potentially pathogenic microorganisms in the bowel. The load of Proteobacteria and Enterobacteriaceae was significantly reduced in the SDD group. Nevertheless, MBP was only given to patients who underwent left-sided colectomies, or sigmoid and low anterior resections.
In August 2019, the MOBILE trial was published, stating that MOABP does not reduce SSIs or overall morbidity compared to NBP for colectomy patients. It was a RCT, multicenter study, including 417 patients. SSI was detected in 7% of the patients assigned to MOABP versus 7% of those assigned to NBP. This study also addressed anastomotic leaks, reporting a 4% rate in both groups. The authors suggested that recommendations on the use of MBP for colectomies should be revisited.
A meta-analysis published in the World Journal of Gastroenterology in 2018 included 21568 patients from 36 studies. There were 23 randomized controlled trials and 13 observational studies. MBP was not only not associated with anastomotic leaks, but it was not associated with surgical site infection, abdominal collection, or length of stay either. This paper states that due to its potential adverse effects and patient discomfort, MBP should not be administered in elective colorectal surgery.
This new review published in 2019 aimed to update current evidence on MBP for colorectal anastomotic leaks. It included 7 randomized controlled studies and one observational study with a propensity score analysis, all of them published since 2010 and involving a total of 1065 patients. According to the authors, there was no evidence that MBP is associated with a lower rate of colorectal anastomotic leak compared to the non-preparation group. The quality of the evidence was low according to GRADE assessments, but the results coincide with other reviews and meta-analyses that were previously published.
The most important conclusion on this matter might be that there is paradoxical evidence in the literature, as high evidence studies have yielded opposite results. We should take several points into consideration in this matter:
First of all, protocols differ across trials, including different dosage and types of oral antibiotics, timing of application and MBP agents. This makes the comparison and extrapolation of the results more difficult. On a more physiological level, studies suggest that there might be a difference between the effect of the MBP on the colon and the rectum. We already know that colon cancer and rectal cancer have different behaviors and surgical treatments, so stratification should be applied accordingly.
More importantly, intestinal antisepsis is extensively administered, with little understanding of its mechanism. It reduces fecal bulk, allowing for an easier handling of the bowel during surgery and palpation of small lesions if necessary. It also diminishes the bacterial burden, thus lowering the SSI rates amongst other infectious complications. Nevertheless, it severely affects the colorectal microbiota.
There is extensive literature on the colorectal microbiota and how it is affected by MBP, antibiotics and surgery. The process breaks the host-microbiota balance and deprives the colon of healthy flora and nutrients. This results in a dysbiosis and the bloom of opportunistic pathogens.
This new flora might be responsible for postoperative complications, such as anastomotic leak and cancer recurrence.
The role of MBP in the prevention of anastomotic leak is still controversial, as evidence is contradictory, although the most recent evidence does show a tendency toward recommending against MBP.
Nevertheless, the addition of oral antibiotics to MBP does seem to be key in lowering infectious complications, including leak rates.
Change in clinical practice might be difficult, as the benefit of MBP on surgical site infection rates should be taken into consideration.
There are more aspects to be studied (ie. colorectal microbiota) regarding anastomotic leak rate to fully understand the role of MBP in its healing.
Share it with your network and help spread clinical knowledge.
