Excisional Hemorrhoidectomy: Safe in Patients With Crohn’s Disease?

Abstract

INTRODUCTION

Nearly one third of patients with Crohn’s disease (CD) have symptomatic perianal disease.¹ Population-based data report that hemorrhoids represent only 1%–2% of all perianal CD; far more common is perianal fistulizing disease, comprising 50% of all perianal disease.¹ Optimal treatment of symptomatic hemorrhoids in patients with Crohn’s disease (CD) remains unclear, largely due to paucity of available evidence in the scientific literature. In fact, the majority of available data is from case series spanning over the last 50 years.^2–6 Current practice guidelines outlined by the American Society of Colon and Rectal Surgeons (ASCRS) recommend that surgeons take extreme caution in both the selection and execution of hemorrhoidectomy in patients with CD.⁷ This recommendation stems from reports from the 1970s and 1980s of inevitable proctectomy after hemorrhoidectomy in patients with CD due to impaired wound healing.^{4, 8}

A recent review of the available literature reported that the incidence of complications after hemorrhoidectomy in CD ranged 0%–100% and stated that the incidence of complications was higher in CD (17.1%) as compared with patients with ulcerative colitis (UC) (5.5%). The authors proposed that minimally invasive surgical techniques, such as transanal hemorrhoidal dearterialization (THD) and the use of biological agents, may account for the lower complication rates observed in recent years.² However, excisional hemorrhoidectomy remains the gold standard for advanced medically refractory hemorrhoids and is recommended for grade 3 and 4 internal hemorrhoids with or without the presence of external hemorrhoids based on practice parameters by ASCRS in 2011⁵ and again in an updated version in 2018.⁹

Regardless of these recommendations, most physicians avoid an excisional hemorrhoidectomy in patients with CD due to the perception of increased complications. However, there are very little data investigating whether these claims are relevant in the current era of CD treatment. We therefore sought to assess whether excisional hemorrhoidectomy in patients with CD was safe and effective for the management of internal and/or external hemorrhoids.

METHODS

After institutional review board (IRB) approval, hospital records and clinical charts from 2 inflammatory bowel disease referral centers (Cleveland Clinic Ohio and Cedars-Sinai Medical Center) were queried for patients with an a priori diagnosis of CD who underwent both medical therapy and surgical intervention in the form of excisional hemorrhoidectomy for symptomatic hemorrhoids over a 24-year period between January 1, 1995, and January 1, 2019. Exclusion criteria included patients with a diagnosis of indeterminate colitis or ulcerative colitis or patients without intestinal continuity.

Patient factors included gender, age at surgery, indication for surgery (based on the patient’s predominant symptoms), prior treatment of hemorrhoids including nonoperative therapy (eg, injection sclerotherapy and/or application of topical steroids) and surgical therapy (eg, prior hemorrhoidectomy); factors were tabulated. In addition, data regarding the disease profile of CD including age of diagnosis of CD, location of the disease (ileum, colon, and rectum), concomitant perianal disease and exposure to medical therapy (defined as exposure to corticosteroids or immunomodulators within 4 weeks of surgery) and biologics (infliximab, adalimumab, certolizumab pegol, vedolizumab, ustekinumab within 12 weeks of surgery) were also collated. Both short-term (within 30 days after surgery) and long-term complications (>30 days after surgery) were assessed. Short-term complications included bleeding, perianal sepsis, and need for operative re-intervention. Long-term complications included nonhealing wounds, need for temporary fecal diversion, or proctectomy. Summary statistics to describe the data were represented by the median (range) or the absolute numbers (percent) where appropriate.

RESULTS

A total of 36 patients with symptomatic hemorrhoidal disease were included in the study cohort; 21 were from institution A and 15 from institution B. Sixteen were males (44%), and the median age was 49 (range, 21–77; interquartile range [IQR] 32.2–60) years (Table 1). Predominant hemorrhoidal symptoms included pain (n = 16; 44%), bleeding (n = 12; 33%), and prolapse (n = 8; 22%). Sixteen patients (44%) had nonoperative therapy before attempting excisional hemorrhoidectomy, including topical medication and sclerotherapy. Two thirds of the patients (n = 24) had other perianal disease, including perianal skin tags (n = 12; 33%), perianal fistula (n = 6; 17%), anal fissure (n = 6; 17%) and condyloma acuminata (n = 1; 3%) at the time of hemorrhoidectomy. Patients were diagnosed with CD for a median of 8 (range, 2–31) years before surgery, with disease limited to the colon and rectum (n = 19; 53%), ileum (n = 9; 25%), both small and large bowel (n = 6; 17%), or perianal fistulizing disease alone (n = 2; 6%). At the time of hemorrhoidectomy, 9 (25%) patients were exposed to corticosteroids, 8 (22%) to immunomodulators, and 9 (25%) to biologics.

During a median follow-up time of 31.5 (range, 1–255; IQR 2–47) months after hemorrhoidectomy, a total of 4 patients (11%) developed long-term complications potentially related to the hemorrhoidectomy: 1 developed an anal stricture, 1 developed a nonhealing wound, 1 developed a perianal abscess/fistula, and 1 had intermittent bleeding due to recurrent hemorrhoidal disease treated with injection sclerotherapy. No patients in this series required proctectomy or fecal diversion as a result of their hemorrhoidal disease or excisional hemorrhoidectomy; 3 patients had a loop ileostomy for several perianal fistulizing disease (n = 2) and an ileal pouch revision (n = 1).

DISCUSSION

Despite including data from 2 large inflammatory bowel disease referral centers, we only identified 36 CD patients who underwent excisional hemorrhoidectomy in the previous 24 years. This may reflect a widespread reluctance of colorectal surgeons to perform excisional hemorrhoidectomy in CD patients, fueled by decades-old case series reporting a high rate of subsequent proctectomy.^{4, 6} However, in the era of biologic therapy for CD, our results demonstrate promising findings that challenge this notion. In our case series of 36 patients, none required proctectomy or fecal diversion over a median follow-up of nearly 3 years for complications related to their excisional hemorrhoidectomy. This suggests that excisional hemorrhoidectomy for symptomatic hemorrhoidal disease may now be safer in CD patients due to improved luminal disease control with more widespread use of biologics. However, it is important to keep in mind that these conclusions are limited by the small number of patients included.

A recent study from Italy by D’Ugo³ et al reported the results of surgical treatment of hemorrhoids in 15 CD patients (open hemorrhoidectomy [n = 11], closed hemorrhoidectomy [n = 3], stapled hemorrhoidopexy [n = 1]). The overall complication rate was reported to be 41%. Interestingly, 8 of 15 patients had surgery without a prior diagnosis of CD, leading the authors to conclude that patients who underwent hemorrhoidectomy without a prior diagnosis of CD were at a significantly higher risk of complications. Similarly, a recent systematic review by Cracco et al² including 11 retrospective studies with 135 patients found that the rate of complications was much higher in CD patients with an unknown diagnosis at the time of skin tag removal or excisional hemorrhoidectomy compared with a known diagnosis (50% vs 9.8%). This is likely due to poorly controlled luminal disease and lack of ongoing medical therapy. The authors concluded that the combination of biologic and immunomodulator medications could minimize complications after hemorrhoid surgery in CD.² Biologic drugs have revolutionized the treatment of perianal fistulizing CD, with complete remission in up to 30% of patients.³ It is possible that, in a similar fashion, perioperative biologic therapy aids in healing wounds after excisional hemorrhoidectomy. The low complication rate we identified in our own cohort may reflect a high rate of concurrent biologic use in our patients. Thus, our findings should be interpreted within this context of patients with a known diagnosis of CD, whose luminal symptoms are being addressed with immunosuppressive medical therapy.

In our modern era cohort of 36 patients, only 1 patient developed a nonhealing wound, and 1 developed a stricture requiring self-anal dilation and ongoing biologic therapy to attempt luminal remission. The patient with the nonhealing wound required a return to the operating room for a wound revision and eventual fistulotomy; the patient’s perianal disease also worsened, which resulted in a subsequent loop colostomy 21 months later. However, no patient had an intestinal diversion or proctectomy for a nonhealing wound itself. Another patient developed a perianal abscess and fistula thought to be a result of the patient’s active perianal disease rather than the hemorrhoidectomy, but the timing was within 3 months of hemorrhoidectomy, so it not possible to say with certainty whether the 2 events were unrelated. Another recent report by McKenna et al¹⁰ of 49 Crohn’s patients with skin tags or hemorrhoids also found surgical treatment of hemorrhoidal disease in CD was safe. However, only 16% of their Crohn’s cohort underwent excisional hemorrhoidectomy, whereas the rest underwent rubber band ligation (36%), skin tag excision (28%), or incision/excision of a thrombosed external hemorrhoid (14%). Our cohort included only patients who underwent excisional hemorrhoidectomy, a more invasive operative intervention. Our study, therefore, further underscores the safety in performing surgical intervention for hemorrhoids in CD patients. Summarizing these results, we postulate that patients with an established diagnosis of CD—started on appropriate immunosuppressive and biologic therapy thereby allowing remission of mucosal inflammation, especially in the anal canal and rectum—have better outcomes than previously expected. Therefore, surgeons should be cautious to follow through with a proper evaluation of any patient with concern for undiagnosed CD before proceeding with hemorrhoidectomy with pre-intervention endoscopy. In addition, surgery should be reserved for symptomatic disease that has failed less invasive nonoperative options as every perianal operation in CD has potential risk.

Although we did not investigate the role of transanal hemorrhoidal dearterialization (THD), several controlled studies have shown that compared with excisional hemorrhoidectomy, THD is associated with decreased postoperative pain and faster return to work but with a slightly higher rate of hemorrhoid recurrence.^11–13 As a minimally invasive technique, THD has been performed in a small series of patients with CD (n = 13) without proctitis CD who suffered from symptomatic grade 3 hemorrhoids.¹⁴ Recurrence rate was 23% (3 of 13) at 18-month follow-up, and there were no major complications in their prospective study. Although these results suggest that THD may be an attractive minimally invasive option for refractory hemorrhoids in patients without proctitis, we do not know if these results can be extrapolated to patients with proctitis. In addition, their rate of hemorrhoidal recurrence was higher than our 1 patient (3%), although both are small series and not comparative studies. A multicenter controlled trial comparing excisional haemorrhoidectomy and THD in CD patients with and without anorectal disease would provide robust data to guide colorectal surgeons managing these patients. Subset analysis of the role of various biologics would also shed light on the effect of these drugs on postoperative healing.

One study limitation is its retrospective design and the inherent selection bias of performing excisional hemorrhoidectomy only on carefully selected CD patients with mild perianal disease. Although this is a large series of excisional hemorrhoidectomy in CD, patient numbers remain small and preclude the ability to draw definitive conclusions about outcomes with varying medical regimens, influence of skin tags, perianal fistulizing disease or anal canal disease, and patients who have intestinal diversion at the time of operative intervention. For this reason, we still recommend exuding caution when performing perianal surgery in Crohn’s disease or referring to an inflammatory bowel disease center when unsure of the pathology or optimal management. In addition, given the retrospective design, we did not compare nonoperative management such as injection sclerotherapy or rubber band ligation with excisional hemorrhoidectomy. Lastly, we did not include patients who underwent THD to compare the results of a less invasive approach with excisional hemorrhoidectomy as they are far too few in number.

CONCLUSION

Our data suggest that excisional hemorrhoidectomy may be performed safely in select CD patients without active proctitis who have failed nonoperative hemorrhoidal therapy. Hemorrhoidectomy should not be excluded as a possible therapeutic option solely because the patient has CD. Careful assessment and optimal medical management with biologics may be a pivotal therapeutic “game changer” in this group of patients.

Conflicts of Interest: AL is a consultant for Takeda.

REFERENCES