Early and Late Closure of Loop Ileostomies: A Retrospective Comparative Outcomes Analysis

The optimal timing of loop ileostomy reversal remains largely unknown, but evidence that delayed ileostomy closure may increase postoperative complication rates is increasing. Purpose: Retrospective research was conducted to compare outcomes between patients who had early (<6 months) or late (>6 months) loop ileostomy closure. Methods: Records of patients >18 years of age who underwent circumstomal reversal of a loop ileostomy over a period of 5 years in 1 hospital’s colorectal unit were abstracted and analyzed. Data from patients who had a planned or conversion to laparotomy, a concurrent bowel resection, reversal of double-barrel small bowel and colonic stomas, or closure of an end ileostomy or patients whose records were incomplete were excluded. Demographic information, American Society of Anesthesiologists (ASA) grade, primary operation indication, surgery and inpatient dates, readmission within 30 days of discharge, reasons for readmission, complication type, and Clavien-Dindo classification were extracted and compared between early and late closure groups using independent-sample t test and Fisher’s exact test. Results: Among the 75 study participants, 25 had an early closure (mean age 68.6 [range 26 – 93] years, mean time since primary surgery 3.8 months) and 50 had a late closure procedure (mean age 71.6 [range 46 – 93] years, mean time since primary surgery 12.8 months). Gender distribution, ASA grades, primary surgery indication, and total number of readmissions were similar between the 2 groups. Hospital length of stay was significantly shorter (5.5 days vs 9.4 days; P = .01) and average number of complications was significantly lower (0.33 vs 0.61; P = .04) in the early closure group. Rates of postoperative ileus, anastomotic bleed, and wound-related complications were not significantly different. Conclusion: Hospital length of stay and average number of postoperative complications following circumstomal loop ileostomy closure were significantly lower in the early than in the late closure group. Additional studies are warranted to help guide practice.

The loop ileostomy is a type of stoma created to divert the flow of intestinal content away from a distal colorectal anastomosis. It defunctions an anastomosis and limits the clinical impact of an anastomotic leak, which is one of the most feared complications in colorectal surgery.¹ According to retrospective and prospective studies,^1,2 a loop ileostomy is favored over loop colostomy for ease of construction and lower complication profile. Additionally, the efficacy of a loop ileostomy in reducing the consequence of anastomotic dehiscence and improving outcomes in distal colorectal anastomoses is well documented in prospective cohort studies.^3,4 However, a systematic review⁵ has shown nearly 20% of patients develop a variety of ileostomy-related complications that can negatively affect quality of life and lead to hospital readmissions. For these reasons, loop ileostomies tend to be temporary and many surgeons aim to reverse them within 3 months, although the optimal time for reversal of ileostomy remains unknown.

The reversal operation generally is regarded as safe, with reported mortality rates as low as 0.4%.⁵ This surgery is commonly performed through a circumstomal incision followed by full mobilization of the ileostomy, formation of ileo-ileal anastomosis, and closure of the fascia and wound. In some instances, a full laparotomy is required. However, the literature suggests that up to one third of loop ileostomies may never be reversed.⁵ A number of factors have been shown in a retrospective study⁶ of 964 patients with ileostomies to be implicated in the delay or failure of stoma closure; these factors included older age, comorbidities, delayed recovery after the initial operation, complications such as anastomotic dehiscence, and the need for adjuvant chemotherapy. It is hypothesized that the time delay may have an impact on postoperative outcomes when the loop ileostomy is eventually closed.

 The aim of this study was to compare outcomes between patients treated in a single colorectal unit who underwent early (<6 months) or late (>6 months) circumstomal closure of a loop ileostomy in order to determine whether timing of reversal influences postoperative outcomes.

A retrospective analysis of data from a single colorectal institution was undertaken. Electronic general surgical records were used to obtain a list of patients that had undergone closure or ileostomy reversal between June 2012 and June 2017. Approval for the collection of patient data was obtained from the Board of King’s College Hospital NHS Foundation Trust.

Patient criteria. Patients were considered for inclusion if they were >18 years of age and underwent circumstomal approach of reversal of loop ileostomy. Patients requiring a planned or conversion to laparotomy for reversal of ileostomy, reversal of ileostomy with concurrent bowel resection, reversal of double-barrel small bowel and colonic stomas, and closure of end ileostomies, as well as patients for whom information needed for the analysis was unavailable or incomplete, were excluded.

Patient data. Patient data abstracted included demographics, American Society of Anesthesiologists (ASA) grade, primary operation where the ileostomy was formed, indication of primary operation, date of reversal of loop ileostomy, interval between primary operation and date of reversal of loop ileostomy, length of inpatient stay, readmission within 30 days of discharge, reasons for readmission within 30 days of discharge, and complication types and numbers (see Table 1).

Data collection. Data were collected from patient electronic and paper files by the lead author. Data were extracted using hospital numbers only and stored in a secured password-protected file.

Data analysis. The following variables were calculated in Excel: time interval in months from primary operation to reversal of ileostomy, length of stay, and number of complications. The categorization of complications are shown in Table 1. The Clavien-Dindo classification was used for comparison of the postoperative complications. This validated grading system for postoperative complications was first described by Dindo et al⁷ in 2004 and is now widely used for grading postoperative complications in an objective and reproducible manner. All postoperative complications were manually graded using the Clavien-Dindo grading tool as shown in Table 2.⁷ Analyses were carried out on the total number of patients as well as 2 subgroups of patients, categorized in terms of early (<6 months) and late (>6 months) reversal of loop ileostomy, using IBM SPSS Statistics, version 23 (IBM Corp, Armonk, NY). Variables were analyzed using independent sample t tests for the difference in means between the 2 groups. Categorical data were reported as the number of patients/percentage of patients and compared using Fisher’s exact and chi-squared tests. A P value of <.05 was considered statistically significant.

To ascertain the results of this study in comparison to published data, a manual electronic search of PubMed (Medline) and Embase for literature published between January 1, 1987 and December 31, 2017, was performed using the search terms ileostomy, reversal, outcomes, and complications to find research that compared outcomes in patients undergoing ileostomy closure at different time intervals. All citations identified were reviewed independently by the lead author to assess suitability for comparison to study findings; 6 original studies were identified (see Table 3).

Of the 101 patients that underwent reversal of loop ileostomy during the study years, 75 (25 with early and 50 with late reversal) met the inclusion criteria and their data were included in the study. Mean age was 68.6 (range 26–93) years for the early group and 71.6 (range 46–93) years for the late group. Gender distribution, ASA grades, and indication for primary operation were similar between the 2 groups (see Table 1).

The majority (76%) of loop ileostomies were constructed due to cancer; the remainder were created as a result of benign diseases such as diverticular disease (19%), inflammatory bowel disease (4%), and rectal trauma (1%). Within 30 days of discharge after reversal of ileostomy, 14% of the total number of patients had 1 or more hospital readmissions, 4 in the early group and 7 in the late group (see Table 1). Mean length of hospital stay following ileostomy reversal was 4.5 (range 0–16) days in the early group versus 9.4 (range 0–39) days in the late group (P = .01) (see Table 1 and Figure). The total number of postoperative complications was higher in the late group (35) than the early group (9), although consideration must be given to the fact that the number of patients in the late group is double that in the early group. Some patients developed no postoperative complications (42, 56%); 7 patients developed more than 1 postoperative complication (9.3%). The mean number of complications was 0.33 in the early and 0.66 in the late group (P = .04). Ileus was the most common complication in the early (4/9, 44%) and late (11/35, 31.4%) closure groups, but the difference was not statistically significant (see Table 1).

In terms of Clavien-Dindo classification, the late group experienced more grade I, II, III, and V complications, although these numbers were too small for statistical calculations (see Table 2).

Systematic reviews and randomized controlled trials^8-10 have shown defunctioning loop ileostomies effectively reduce the rate of symptomatic anastomotic leaks and the need for reoperation in such cases. Restoration of gastrointestinal continuity by reversing loop ileostomies requires a second operation, the timing of which remains controversial. Furthermore, the morbidity rate following closure of ileostomy may be close to 20%,⁵ and systematic reviews and retrospective studies^5,11,12 have shown mechanical small bowel obstruction, ileus, and wound-related problems are common complications.

Because the optimal timing of ileostomy closure remains unknown, interest has been growing in the assessment of the association between timing of closure and postoperative outcomes. Prospective and retrospective reports^13-15 are available in the literature comparing ileostomy closures as early as 8 days following the primary operation to the traditionally timed closures at 12 weeks. Two (2) retrospective studies^15-16 and 1 prospective randomized trial¹⁴ compared closures at various time intervals: 1 month to after 6 months, 3 months to after 6 months, and within 3 months to after 3 months, with variable conclusions (see Table 3).

The current study compares the outcomes of ileostomy closures within 6 months to those after 6 months, the first published comparison of outcomes at these specific time intervals. A key finding from the current data is that patients who underwent early closure had a significantly shorter length of hospital stay than patients who had late closure (4.5 days vs 9.4 days, respectively; P = .01). Similar findings were reported in a prospective randomized trial by Alves et al¹⁷ that compared outcomes in 186 patients randomized into 2 groups that underwent either early (within 8 days) or late closure (at 60 days) of a loop ileostomy. The reported length of stay in the early group was significantly lower than the late group (16 vs 18 days, respectively; P = .013). Similarly, a retrospective study of 93 patients by Omundsen et al¹⁸ that compared ileostomy closures within 10 days and at 90 days found length of hospital stay in the early group was significantly lower than in the late group (14 vs 17 days, respectively; P = .05).

The current study also found the mean number of complications was significantly lower in the early group as compared to the late group (0.33 vs 0.61, respectively; P = .04), findings supported by a randomized controlled trial of 112 patients by Danielsen et al¹⁹ in which significantly fewer complications occurred in the early group as compared to the late group (1.2 vs 2.9, respectively; P <.0001). A retrospective study by Rubio-Perez et al²⁰ concluded the delay in ileostomy closure was associated with a significant increase in postoperative complications, specifically wound infections (13%; P = .007) and pseudomembranous colitis (4%; P = .003).

In the current study, the number of cases of ileus were higher in the late group (11) than the early group (4), but the difference was not statistically significant.

Williams et al¹³ analyzed the changes that occur in the distal, inactive limb of a loop ileostomy and found a significant reduction in the ileal smooth muscle contractility following a period of diversion of intestinal content. Based on this finding, the authors concluded such changes may result in reduced compliance and contractility of the defunctioned bowel segment, leading to the obstructive symptoms that may be observed following ileostomy closure. This may help explain the higher rate of ileus observed in the current late group, an observation also supported by the study by Alves et al.¹⁷

In order to objectively assess postoperative complications, study data were categorized according to the Clavien-Dindo classification (see Table 2). All postoperative complications were manually graded using the Clavien-Dindo grading tool as shown in Table 2. When comparing the 2 groups, the current authors found the late group had more grade I (14/35), II (16/35), III (3/35), and V (2/35) complications compared to the early group. However, these numbers were too small to determine statistical significance.

Some studies did not demonstrate significant differences in studied outcomes between early and late ileostomy closures. In their prospective controlled trial of ileostomy closures within 1 month and after 6 months, Zhen at al¹⁴ found rates of closure-related complications (20.9% vs 18.6%, respectively; P = .637) and length of hospital stay (7.94 vs 7.97 days, respectively; P = .588) were not statistically different between the early and late groups.

A prospective study by Zhou et al¹⁵ of early defunctioning ileostomy closure among 123 patients found no significant difference between the group that had ileostomy closure within 90 days and the group that underwent ileostomy closure after 180 days in relation to stoma-related complications (early group 10%, late group 16%; P = .387), anastomotic leakage (early group 1%, late group 2%; P = .691), ileus (early group 4%, late group 9%; P = .245), and peristomal hernia (early group 5%, late group 4%; P = .190).

A large retrospective study by Li and Ozuner¹⁶ on ileostomy reversal among 358 patients also reported no differences between the early (<3 months) and late (>3 months) ileostomy closure groups in terms of rates of ileus (12.3% vs 13.4%, respectively; P >.87), small bowel obstruction (15.6% vs 15.1%, respectively; P = .5), wound infection (2.2% vs 1.7%, respectively; P = .99), and surgery-related readmission rate (0.56% vs 0%, respectively; P = .85) (see Table 3).

The current study demonstrated that early ileostomy closure may lead to a significant reduction in length of hospital stay and average number of postoperative complications. These findings are supported by some of the aforementioned studies^17,19-20 and could encourage surgeons to expedite reversal of loop ileostomies whenever possible. This study is the first to compare outcomes of ileostomy closure within these particular time intervals; larger prospective studies are required to confirm these findings and help guide practice.

This study is limited by its retrospective design and relatively small numbers. The difference in the group sizes also was a consideration during data analysis. A future follow-up study will have a larger number of participants more equally divided between the comparison groups and fewer confounders such as comorbidity status. The authors suggest a prospective, randomized study design that also will explore the effect of timing of ileostomy closure on parameters such as health-related costs and quality of life.

Although the optimal timing of closure of loop ileostomy is unknown, there is evidence to support early rather than late closure. The current study demonstrates the length of hospital stay and average number of postoperative complications were significantly lower in the early (<6 months) than in the late (>6 months) closure group. The current study supports results of previous studies and warrants prospective research to validate these findings.

The authors thank the general surgery theatre manager and colorectal clinical nurse specialist for their help with the provision of the data and Samantha Matin for her assistance with data analysis.

Ms. Abdalla is a Specialist Registrar in General Surgery; and Dr. Scapinata is a Consultant Colorectal Surgeon, Princess Royal University Hospital, King’s College NHS Foundation Trust, Orpington, UK.

Please address correspondence to: Sala Abdalla, BSc, MBBS, MRCS, Princess Royal University Hospital, King’s College NHS Foundation Trust, Farnborough Common, Orpington, BR6 8ND, UK: email: sala3001@doctors.org.uk.