Surgery in Crohn’s Disease: Past Performance Does Predict Future Results

In the article, “Postoperative Disease Recurrence Risk and Optimal Biologic Timing After Temporary Diversion Following Ileocolic Resection,”¹ we are presented with a large cohort of patients undergoing surgery in the setting of Crohn’s disease (CD) from an extremely experienced group when it comes to managing these patients. Despite this, there are perhaps just as many questions raised by this study as answers provided. Who should start biologics after stoma takedown? When should they start them? The authors suggest that there may be some benefit to initiating biologics prior to stoma takedown in diverted patients, but do not really give us the data to support or refute that suggestion.

Within this study, we see that severity of CD at the time of surgery likely becomes a self-fulfilling prophecy. Patients undergoing diversion at the time of surgery were more likely to have been on biologics prior to and at the time of surgery, more likely to have penetrating and perianal disease, and were more likely to be malnourished and anemic. This generally makes sense but also points to a more severe disease phenotype than those not undergoing diversion at the time of surgery. We also see that those undergoing diversion are more likely to develop surgical recurrence, although interestingly, they are at a similar risk of developing radiologic and endoscopic recurrence—implying that when these patients’ disease returned, it returned more aggressively.

The use of biologics in this cohort is highly selective and it is not entirely clear what prompted their use within the population. Interestingly, not being on biologics at all seems protective when compared with biologic exposure. Even after the multivariate regression, we see the hazard ratio for radiographic recurrence is lower for the no biologic group than it is for any of the patients who start or resume biologics postoperatively. Here again, this seems more likely to be a marker of disease severity, as all things being equal, one would expect biologics to be protective against recurrent disease in a larger population. It is hard to know why the physicians in this group selected biologics for patients after surgery, but one might imagine it is either because the patient was on a biologic prior to surgery, or because they had more significant disease. This implies that once again we see that patients with worse disease prior to stoma takedown manifest worse outcomes after, as well.

In looking at whether starting biologics before or after stoma takedown is the optimal approach, it is hard to draw any real conclusions, as well. It does seem like the majority of patients who started biologics prior to takedown were on these medications prior to surgery (23 of 25), and thus this may just be a continuation of their preoperative biologic. The real challenge here is that we do not know if these patients’ biologics were altered from prior to surgery or if they continued the same dose or same medication from before, information which is crucial to the clinician when considering what the next step is for their patient. The number of patients who were on biologics prior to stoma takedown is also extremely small, about 15% of the overall cohort, making it extremely difficult to compare these outcomes to the much larger groups receiving therapy after takedown or not at all. While the general percentages of patients with disease recurrences are lower in the group who initiated the biologic prior to stoma takedown, the numbers are clearly skewed by this small sample size, which is demonstrated by the fact that there is no statistical difference between the groups.

Regardless of whether the primary end point is supported by the data, there is a lot to take away from this article. As should go without saying, retrospective, single-site, database studies are inherently riddled with selection biases. We do what we can to overcome these by using complex statistical techniques, but these studies are unable to settle questions the way expensive randomized control trials can. They can, however, lay the groundwork to develop the hypotheses we wish to prove with prospective studies. The groundwork has been laid here to ask the question about timing of biologic initiation after diversion in this setting. Given the recent glut of studies such as PUCCINI² and others^3,4 demonstrating minimal risk for septic complications after surgery in the setting of active biologic or small molecule therapy, ileostomy takedown can be performed safely for those receiving advanced therapies to control their disease. Based on this, perhaps there are many who would say (including myself) that biologics may best be initiated as soon as is feasible in the postoperative recovery of a patients with CD. This would allow the patient to reach an adequate therapeutic level prior to restoration of continuity and potentially prevent early low-grade recurrence. The main question for the detractors of this approach is will it provide any benefit to the patient, and if it does not, is it worth several months of exposure to expensive immunosuppression if the patient is not deriving any benefit? Evidence is still mixed in regards to the use of biologics after surgery in the setting of CD and to date has generally been guided by risk factors and endoscopic surveillance.⁵ Although an argument can certainly be made that if the disease was severe enough to warrant surgery, patients are more likely than not to recur at some point if it is not managed.

There are many challenges to us advancing the science from studies such as this. The first is the heterogeneity of this population. Even within this study, there is significant variability across the population between their disease phenotype, the complications that arose as a result of the disease, and how it was managed surgically. A 30% ileostomy rate might be considered high within a single institution, which could indicate that this cohort does not represent the general CD population nationally. To gain this detailed understanding, we either must have a larger cohort from which we can clearly analyze subsets of patients or a much narrower cohort that phenotypically appears similar. To get these populations, we need to be pushing multi-institutional studies from diverse locations with highly standardized approaches to the management of this population.

Until we are able to carry out these studies, I think the jury is still out on exactly when biologics should be instituted after ileocolic resection for Crohn’s disease. Most can agree that it is certainly safe to start biologic after resection but before stoma takedown and that it is also safe to perform stoma takedown in the setting of biologic management. The question is will this provide a benefit to the patient, and while it may, it is hard to know for sure based on existing data. Perhaps a more significant takeaway from this article is that if your patient has severe enough disease to warrant diversion at the time of their procedure, then they are at high risk for significant recurrence in the future, and they should be counseled and managed accordingly.

Funding Source

None.

Conflicts of Interest

S.E. has received consulting fees from Takeda and research support from the Crohn’s and Colitis Foundation.

References