A Response to ‘Guidelines for Avoiding Bile Duct Injury: Beware’

We wish to respond to The Great Debate between Drs. Guy Voeller and Leo Gordon (April 2021) in follow-up to the prior debate on the role of intraoperative cholangiography (IOC)/imaging during laparoscopic cholecystectomy (December 2020).

In the most recent debate, Dr. Voeller raises a number of points, in particular about the consensus conference’s clinical practice guidelines on the prevention of bile duct injuries (BDIs) during cholecystectomy (Ann Surg 2020;272[1]:3-23; co-publication Surg Endosc 2020;34[7]:2827-2855), the purpose of which was to address the infrequent but ongoing problem of BDI in this common procedure, and decries their use. We respond both in general and to some of the specific points he makes.

First, the main purpose of a clinical practice guideline is to use the best available evidence to formulate recommendations based on that evidence, and to provide practitioners with suggested and recommended strategies they can use to enhance clinical care and improve patient outcomes. A secondary, but very important, purpose is to determine the shortcomings in current knowledge and develop recommendations for further studies that can fill in the gaps in our knowledge. Key steps in creating a guideline are reviewing a vast amount of literature, rating the quality of the evidence in that literature, and making recommendations largely based on the quality of that evidence.

Dr. Voeller states “an individual surgeon’s algorithm in this challenging setting [when anatomic clarity is not achievable] should be consistent and based on their expertise.” But expertise is not simply based on an individual surgeon’s personal experience in surgery. It is also based on the literature, and guidelines have become a very important, high-level summary of the literature.

Guidelines are not, as suggested by Dr. Voeller, “mandated” or required, nor are they in any way “enforced.” The consensus guideline on prevention of BDI during cholecystectomy was a multisociety effort—including the Society of American Gastrointestinal and Endoscopic Surgeons, the Americas Hepato-Pancreato-Biliary Association, the International Hepato-Pancreato Biliary Association, the Society for Surgery of the Alimentary Tract, and the European Association of Endoscopic Surgery—and involved a rigorous and transparent methodological process, taking more than 2.5 years, to address key questions on this topic and provide the best recommendations possible given the evidence. It involved numerous diverse stakeholders including both high-volume academic and community surgeons who perform cholecystectomy and surgeons who are expert in repairing BDIs. The consensus meeting also involved an expert panel of 25 highly experienced surgeons who were not involved in developing the guideline recommendations, but were asked to comment and critique them. A high bar was set for approving a recommendation (80% agreement by the expert panelists).

Five of the 18 guideline questions did not have recommendations due to a lack of evidence. It is true that most of the recommendations that were made were derived from studies with low or very low quality of evidence, which reflects the quality of studies in the surgical literature and the rigor with which the GRADE methodology for forming a guideline assesses quality, rather than an invalidation of putting forth guideline recommendations. And it should be noted that low quality of evidence does not mean the “absence” of evidence. In developing the recommendations, considerations of not only the evidence but also feasibility and acceptability by surgeons were substantially taken into account.

Any guideline should have a dissemination plan, which in this case has primarily involved publicizing this guideline through the sponsoring societies to their members, peer-reviewed publications and panel sessions at national meetings, and has not to date involved hospitals, insurers and other groups. The recommendations were also carefully worded with regard to any potential medicolegal implications for surgeons.

Here we address some of the specific guideline recommendations commented on by Dr. Voeller.

1. Use of the critical view of safety (CVS). Dr. Voeller comments on a lack of high-quality evidence about the recommendation that the CVS be used as the primary method of ductal identification. A prospective, randomized trial to address this issue is not feasible because of the very large number of patients (20,000 or more) who would be needed to detect a significant difference, given the low rate of BDI. However, it is based on considerable case series evidence as well as the fact that in studies that look at videos of BDIs, the CVS method was never used. We have also personally seen multiple videos of lap chole cases in which a BDI occurred and have never seen the CVS obtained in any of those. The CVS is also based on the proven method of safe ductal identification from the era of open cholecystectomy (Ann Surg 2017;265[3]:464-465), and difficulty in achieving it should alert the surgeon to a potentially dangerous situation in which the approach may need to be altered. From a medicolegal standpoint, surgeon use of the CVS should be protective and should lower the potential for putting him or her in jeopardy. Also, as Dr. Voeller serves as a defense expert, he should be aware of the prior perspective published in the Journal of the American College of Surgeons that addresses some of his concern (2010;211[1]:132-137).

2. Top–down approach versus subtotal cholecystectomy. Dr. Voeller states he has used the top–down approach in “impossibly diseased gallbladders” thousands of times. It would be helpful if he could explain how he manages to avoid any injuries in such cases, since the “impossibly diseased” language he uses implies a hostile hepatocystic triangle that cannot be safely dissected. Certainly, there are situations in which the top–down approach can be useful, and the recommendation does not abrogate that. However, the rare but deadly “extreme vascular biliary injury” seems to occur when a shrunken gallbladder is dissected in a top–down fashion, where it can be difficult if not impossible to ascertain when the dissection from above is approaching the bile duct, portal vein and hepatic artery, which are drawn into the zone of severe inflammation (HPB [Oxford] 2012;14[1]:1-8) To wit, a “very memorable gallbladder” case was recently posted on ACS communities in which a dome–down approach in a very difficult inflamed gallbladder resulted in a common BDI. And what about surgeons in the early part of their career who don’t have a vast experience? Is it not safer to utilize subtotal cholecystectomy, which, if properly done, should have a low rate of serious complications and reoperation, or risk a severe injury?

In a study of more than 8,000 laparoscopic cholecystectomies from 166 hospitals in the United Kingdom, 779 patients had grade 4 inflammation, which is described as “cystic pedicle—impossible to clarify” (note the similar use of the word “impossible”) (Surg Endosc 2019;33[1]:110-121). Thirteen patients in that group had major BDIs for an injury rate of 1.7%, which is more than five times higher than the overall BDI rate for laparoscopic cholecystectomy, and this represents the best data from a Western country on the current status of BDI when there is severe inflammation. That is a key reason why more general surgery residents are being taught how to do the much safer operation of subtotal fenestrating cholecystectomy in the selected subset of patients with severe adverse conditions.

3. Use of IOC (or other imaging) to reduce the risk for BDI—the topic of the original debate between Drs. Brunt, Taylor Riall and Raul Rosenthal. Dr. Voeller implies that because of the guideline recommendation, this means that surgeons who do not perform IOC are “violating the standard of care.” This is an inflammatory, misleading and erroneous statement. First, the guideline recommendation around IOC has two aspects: 1) to suggest its use in acute cholecystitis or a history of acute cholecystitis, based on substantial, albeit low-quality, national database studies; and 2) a strong recommendation to use IOC in cases of intraoperative uncertainty of anatomy or suspected biliary injury to mitigate the risk for BDI.

The latter can result in intraoperative recognition and avoidance of a higher level of injury in which a portion of the bile duct is excised. No recommendation was made in elective non-acute cholecystectomy due to the absence of evidence. Dr. Voeller also omits from his critique the concluding statements from Dr. Brunt in the aforementioned General Surgery News debate article, which states that he would not—nor would we—presume to take the position that all surgeons should perform IOC routinely (nor do the guidelines state this): “However, to perform IOC rarely or not at all or not in a significant percentage of one’s cases is a missed opportunity to identify unsuspected pathology, potentially reduce biliary injury, provide training for residents, and maintain an important skill set.”

Finally, Dr. Voeller states that “it is time to reel in the guideline madness for the benefit of all concerned.” His personal series of 5,000 cases without a BDI is remarkable and undoubtedly reflects his experience and considerable skill as a surgeon, and perhaps some good fortune, but it is irrelevant to the issue of BDI and the surgical community at large. In addition, in his criticisms he offers no constructive suggestions whatsoever for how to reduce the estimated 2,500 to 3,000 major BDIs that occur in the United Sates annually. In fact, he seems to take the nihilistic view that BDIs will occur and nothing can be done to prevent them—a position we could not disagree with more vigorously. He also almost completely ignores the most important constituent here: the patients. Many of the patients who suffer a BDI are young and more often women than men. The “real-world” impact for the patient can be devastating and diminish their quality of life long term, adversely affect their personal finances, and can even result in liver transplantation or death. And those of us who have experienced a BDI in our careers know all too well how it adversely affects us as surgeons as well.

Like it or not, we live in an era of evidence-based medicine. Out in the real world, surgeons need guidelines. They don’t need dictums or mandates, but they do need guidelines. They need something upon which they can lean when they tell a patient or family member or referring physicians why they did not complete the cholecystectomy or left a drain tube. Summaries of evidence and guideline recommendations are also not rigid and are never meant to supplant surgeon judgment in a given case, which any astute reader of these guidelines should recognize. However, if we, as a profession, collectively do not do something to try and move the needle on BDI during cholecystectomy in the right direction, then shame on us. Perhaps it is worth remembering the words of George Santayana: “Those who do not remember the past are condemned to repeat it.”

Respectfully submitted,

This article is from the June 2021 print issue.