A call for clarity: a scoping review of predictors of poor outcome after emergency abdominal surgery for inflammatory bowel disease

The medical management of inflammatory bowel disease (IBD) is rapidly progressing; however, many patients with the disease still require surgery. Often this is done as an emergency. Initiatives such as the National Emergency Laparotomy Audit have shown how evidence‐based emergency surgery improves outcomes for the patient. The aim of this scoping review is to describe the current evidence base on risk stratification in emergency abdominal surgery for IBD.

demonstrated an improvement in survival and other outcome metrics for adult patients [4].A recommendation of NELA has been the identification and formal documentation of high-risk adult patients preoperatively.There are several guidelines for the surgical management of IBD in the elective setting, including the World Society of Emergency Surgery, the Association of Coloproctology of Great Britain and Ireland and the British Society of Gastroenterologists [3,5,6].There is a lack of evidence-based guidance regarding risk stratification in the emergency setting.IBD patients who present as an emergency do so with a variety of disease phenotypes.
Interpretation of data is challenging due to the heterogeneity of the patient group and the procedures performed.
The aim of this scoping review is to describe the current evidence base on risk stratification in emergency abdominal surgery for IBD for the purpose of consolidating what is known, identifying areas suitable for meta-analysis and identifying areas for future study.

ME THODS
This review follows the methodological protocol proposed by Arksey and O'Malley [7].The research question identified was 'What is known from the existing literature about factors which affect the outcomes of emergency laparotomy/laparoscopy in inflammatory bowel disease?'A librarian-reviewed search strategy was drawn up and online databases were searched between 1 January 1974 and 5 January 2022.Embase and MEDLINE were searched by JB based on the following terms, adjusted for specific Medical Subject Headings (MeSH): Colitis, Crohn's, Inflammatory bowel disease, IBD, urgent, unplanned, emergency, laparotomy, laparoscopy and surgery.The EU Clinical Trials Register, Clini caltr ials.gov and the Cochrane Library were also searched, as were reference lists.Results were limited by resources to the English language; the full search strategies can be found in Appendix S1.
Exclusion criteria for abstracts were cases prior to 1998 (the start of the biologic era), operations in children, peri-anal procedures alone, case reports, pilot or preliminary studies reporting novel techniques or outcomes, elective populations, pregnancy, narrative reviews, opinion articles and non-clinical outcome measures.Abstracts were reviewed using the online software Rayyan by JB and JW [8].
Full papers identified by the above search criteria were excluded if IBD outcomes were not reported as a separate cohort from other emergency patients, emergency laparotomy outcomes were not reported as a separate cohort from elective laparotomy patients, the impact of postoperative medication alone was described, or papers were descriptive lacking a control or analysis of factors affecting outcome.The aim of this review was to identify predictors of poor outcome, as only such papers reporting adjusted odds or risk ratios with confidence intervals were included.The remaining papers included in this review were organized according to subject theme and charted by JB as per Ritchie and Spencer using Microsoft Excel into groups according to which clinically applicable factor was being discussed [9].The PRISMA-SCR checklist is available in Appendix S2.
To present the evidence on this topic most comprehensively, the inclusion criteria do not include specific measures of poor outcome.
All studies reporting negative outcomes have been considered.The aim is that the results will not present the authors' opinion on what is considered a poor outcome but rather a summary of what has been published in the literature as a poor outcome.

RE SULTS
The literature search produced 5682 abstracts, of which 341 remained after the abstract screening stage.After full paper review, 17 Of the included 17 studies, two reported on CD and UC separately and so have been charted twice.As a result, we have charted 19 studies, 13 reporting on outcomes in UC, three in CD and three not distinguishing between IBD subtypes (referred to as IBD).The 19 studies can be seen in Table 1.All included papers are cohort studies.To aid clinical application, the studies are reported by each reported factor affecting the outcome of emergency laparotomy in IBD patients.

Definition of an emergency operation
Of the 16 included studies, six stated how an emergency operation was defined in this cohort of patients.These are shown in Table 2.

Time to theatre
Six studies comment on the effect of timeliness to theatre on patient outcomes.Bartels et al., observing a population of 71 UC patients, found a significant association between increased time to theatre and postoperative complications (OR 5.5, 95% CI 1.6-19.0,P = 0.007) [10].

Minimally invasive surgery (MIS)
Five studies report the effect of MIS on patient outcomes.Singh et al. found a significant reduction in postoperative complications in patients having colectomy via MIS (OR 0.237, 95% CI 0.082-0.68,P = 0.007) [16].This was corroborated by Skancke et al.

TA B L E 2
The studies which reported their definition of emergency or how they classified urgency within their study.during weekdays.In UC, they found significantly increased risk of postoperative complications (OR 1.71, 95% CI 1.01-2.9,P = 0.03) and reoperation within the same hospital episode (OR 11.5, 95% CI 2.32-57.1,P = 0.008) in those having an operation at the weekend [20].However, in the same cohort, they found no association between having an operation at the weekend and increased mortality (OR 1.31, 95% CI 0.42-4.05,P = 0.64) [20].

Subspecialty of surgeon
MacFarlane et al. investigated the impact of the specialty interest of the operating surgeon on the outcomes of all IBD patients having bowel resection using hospital episode statistics data in the UK.

Disease and operation characteristics
Uchino et al. were the only authors to report on the impact of disease duration, wound contamination and duration of surgery and the association with surgical site infections.They found a significant increase in surgical site infections in patients with contaminated wounds (OR 10.04, 95% CI 2.91-34.63,P < 0.001) [23].However, they identified no significant association between increased disease duration and the incidence of surgical site infections (OR 0.73, 95% CI 0.18-2.91,P = 0.65) or duration of surgery and incidence of surgical site infections (OR 0.34, 95% CI 0.3-5.95,P = 0.85) [23].Leeds et al. (2018) reported on the impact of sepsis on outcomes.They found sepsis to be a significant predictor of postoperative complications (OR 2.18, 95% CI 1.4-3.38,P = 0.001) but not of mortality (OR 1.64, 95% CI 0.52-5.2,P = 0.397) [12].

Comorbidity
Five papers reported the effect of comorbidities on mortality; these are shown in Table 3.In addition to these, Leeds et al. (2017) found a significant association between increasing comorbidity index and postoperative complications (OR 1.36, 95% CI 1.17-1.58,P < 0.001) [11].

American Society of Anesthesiologists (ASA) grade
Three studies reported the effect of increasing ASA grade on outcomes.Uchino

Age
Seven studies reported on the impact of increasing age on outcomes, with Lowe et al. [22] reporting CD and UC separately.
Table 4 shows the studies reporting the effect of increasing age on increasing risk of mortality.King et al. [13] were the only ones not to report significant association.Leeds

Patient sex
Several authors reported on the effect of female patient sex on postoperative outcomes.These are listed in Table 5.In addition, MacFarlane et al. reported the effect of male sex and found significantly reduced mortality (OR 0.41, 95% CI 0.18-0.96,P = 0.041) [21].

Medication
Three studies reported on the impact of steroids on outcomes.
Uchino et al. [23] found that the incidence of surgical site infections in UC patients undergoing colectomy was not significantly elevated in those taking steroids (OR 3.06, 95% CI 0.94-10, P = 0.06).Leeds et al.

Malnutrition
The effect of malnutrition on postoperative outcomes was measured in two studies.

Biochemical and haematological abnormalities
Two studies reported the effect of hypoalbuminaemia on outcomes.Abbreviations: CD, Crohn's disease; IBD, inflammatory bowel disease; UC, ulcerative colitis.

TA B L E 5
The studies reporting the predictive value of female patient sex on mortality, in comparison to male sex.They also investigated the effect of preoperative red blood cell transfusion and found no significant association with mortality (OR 0.52, 95% CI 0.1-2.71,P = 0.441) or postoperative complications (OR 1.49, 95% CI 0.79-2.79,P = 0.218) [12].

Respiratory insufficiency
Reporting on their cohort of IBD patients undergoing colectomy for toxic colitis, Dayama et al. found respiratory insufficiency to be a predictor of increased mortality (OR 2.713, 95% CI 1.052-7.001,P = 0.039) [24].

DISCUSS ION
This scoping review has summarized the current evidence base for predictors of poor outcome in patients having emergency abdominal surgery for IBD.The poor outcomes described by the included studies are mortality, postoperative length of stay, readmission, reoperation and postoperative complications including surgical site infections and sepsis.
This review included papers explicitly reporting on the emergency IBD cohort using broad literature search terms to capture as many studies as possible.Despite this, we were only able to include 17 papers in this review.During the full paper screening phase, many papers were excluded because they did not clearly define their cohort as emergency, or unplanned, with regard to the timing of surgery.Terms such as 'medically refractory disease' were commonly used, but the reviewer's perception was that many papers included a mix of elective or semi-elective patients with those who were truly surgical emergency cases.
Some authors stated that their cohort included both emergency and elective cases but did not report separately on the two groups.The likelihood is that there are more than 17 papers reporting outcomes for IBD emergency operations, but it is not possible to identify these results accurately due to descriptive terms used in the papers.For example, we identified three papers reporting adjusted odds ratios of poor outcome from emergency surgery after taking steroids in the preoperative period.
We could identify no studies specifically reporting on the effects of biologic drugs or anti-tumour necrosis factor α medication on emergency surgery.
Of the 16 papers included, six clearly stated how they defined an emergency operation, as shown in Table 2, with a wide variety of definitions.The variation in how a patient is classified as an emergency was also found by Udholm et al. in a systematic review of the difference in outcome comparing emergency and elective operations for CD [26].The range of urgencies in these six papers, classified as 'an emergency', was from an operation within 12 h of admission to any time within an unplanned admission.The patients represented in this spectrum may be very different.

An operation within 12 h of admission is defined by the National
Confidential Enquiry into Patient Outcome and Death (NCEPOD) as 'urgent' or 'immediate' and is required for patients with peritonitis, septic shock or major haemorrhage [27].An operation described as 'within the same unplanned admission' may include this unstable patient cohort, but also includes a category described by NCEPOD as 'expedited'.These operation timings are suitable for stable patients in whom there is no threat to life, for example in those stable patients with small bowel obstruction, with no concerning features, or those with medically refractory colitis.
The NCEPOD urgency classification system has been adopted and adapted by the NELA and is recommended for acute hospitals in the UK [4].Every case booked for an emergency operation has a clinical urgency registered as requiring an operation within 2, 2-6, 6-18 or 18-24 h.NCEPOD recommend that this urgency is decided by the consultant performing the procedure.Whilst there is scope for bias within this subjective system, it may also capture intangible patient factors, the 'end of the bed' assessment for example.Clinical urgency at the time of booking helps facilitate prioritization of emergency theatre and has been widely adopted in the UK.A benefit of this classification which has not yet been explored is how it might improve the reporting of outcomes in IBD patients having 'emergency' surgery.
Within a single cohort study, multivariate regression analysis can adjust for differences in individual patient clinical presentation, but the greater value of cohort studies is the pooling of data for meta-analysis.For this to be accurate the cohorts must be comparable, and for them to be comparable the patient cohort must be comprehensively described.The STROBE guidelines on reporting of cohort studies include the characteristics of study participants [28].
This review highlights a particular challenge in following STROBE guidelines for cohort studies of IBD patients as there is no single widely accepted way of defining an emergency IBD operation in the reviewed literature.
We recommend consistency of approach in reports of outcomes in the IBD emergency surgery patient group.We particularly recommend a clear description of the urgency of the surgery and would suggest that the NELA modification of the NCEPOD categories is a requirement of all such publications.This would facilitate more accurate classification of procedure urgency, patient subtype and analysis of results.

Race
Leeds et al. (2017) were the only authors to report the effect of patient race on their postoperative outcomes.They found no significant difference in outcome in Caucasian patients (OR 1.06, 95% CI 0.82-1.36,P = 0.647)[11].
The included studies showing publication date, country, number of patients included, data collection period, procedure and IBD subtype reported on.
Ananthakrishnan et al.were the only authors to report the outcomes of patients undergoing emergency surgery at the weekend versusTA B L E 1Abbreviations: CD, Crohn's disease; IBD, inflammatory bowel disease; UC, ulcerative colitis.
The studies reporting the predictive value of increasing patient comorbidity on mortality.
Note: Tøttrup et al. reported mortality rate ratio (MRR) as denoted by the asterisk.P < 0.05 was considered significant in all studies.Abbreviations: CD, Crohn's disease; IBD, inflammatory bowel disease; UC, ulcerative colitis.
The studies reporting the predictive value of increasing patient age on mortality.Note: Tøttrup et al. reported mortality rate ratio (MRR) as denoted by the asterisk.P < 0.05 was considered significant in all studies.
[23]no et al.reported no significant association with surgical site infections (OR 2.74, 95% CI 0.63-12.0,P=0.18)[23].This was supported by Galata et al. who found no significant association TA B L E 4