Ileocecectomy as an Acceptable Alternative to Right Hemicolectomy in Trauma: A Propensity Score Matched TQIP Analysis

Abstract

Background

Ileocecectomy (IC) as an alternative to right hemicolectomy (RH) for traumatic indication is controversial with limited comparison data. We sought to compare IC vs RH for traumatic injury and hypothesized there was no difference in outcomes.

Methods

The Trauma Quality Improvement Program database was queried from 2017 to 2022 with IC and RH patients evaluated using propensity score matching (PSM) in a 1:3 ratio. Presenting factors and outcomes were analyzed.

Results

After PSM, 558 RH and 186 IC patients had no difference in age, penetrating mechanism, time to OR and injury severity score. RH patients were more likely to have unplanned ICU admission (8.6% vs 3.8%, P = 0.03) and organ space infection (7.5% vs 3.2%, P = 0.04) but similar rates of acute kidney injury (P = 0.5), unplanned OR (P > 0.9) and mortality (P = 0.08) compared to IC patients.

Conclusions

Traumatic colon injury managed by ileocecectomy had similar outcomes compared to right hemicolectomy and should be considered an acceptable alternative when anatomically feasible.

Keywords

ileocecectomy right hemicolectomy trauma colon injury

Key Takeaways

• Ileocecectomy is rarely performed in the trauma population, with only 2.9% of the analyzed population undergoing this procedure.

• Patients selected to undergo ileocecectomy exhibited non-inferior outcomes, including infection, unplanned OR, unplanned ICU, LOS, and mortality, when compared to those who underwent right hemicolectomy.

• Clinical judgment remains pivotal and further investigation of the role of segmental resection in traumatic colon injuries is warranted

Introduction

The management of traumatic colon injuries has been scrutinized and continuously reviewed since World War I, when William H. Ogilvie advocated for the mandatory exteriorization of all colon wounds.¹ This belief has been increasingly challenged, particularly over the past 40 years, with evaluations focusing on the mechanism of injury, blood loss and transfusion requirements, patient comorbidities, level of contamination, and metabolic disturbances.^2-7 Management of traumatic colon injuries no longer obligates diversion; instead, it is considered only when specific conditions are met.

Most penetrating colon injuries occur on the right side with destructive injuries requiring colon resection in most settings. Historically, right colon injuries were associated with decreased rates of anastomotic leak and abdominal complications when compared to left sided colon injuries, with diversion infrequently performed.^5,8,9 As a result, treatment of right-sided colon injuries involved primary repair or resection with anastomosis; however, surgeons are still faced with the decision to perform right hemicolectomy (RH) or ileocecectomy (IC). IC is a controversial alternative with limited evaluation in the trauma literature.

Critics of IC argue against its efficacy, citing increased risk for anastomotic leak and colon-related morbidity given inconsistent vasculature and lack of well-established and predictable collateral circulation.^10,11 Advocates of the procedure point to success in patients with Crohn’s disease, where IC is the most commonly performed resection with favorable results.^12,13 The purpose of this study is to compare outcomes of patients who underwent RH vs IC for traumatic RCI by using the national American College of Surgeons Trauma Quality Improvement Program data set (ACS-TQIP). We hypothesize there is no difference in outcomes.

Methods

The ACS-TQIP database was queried from 2017 to 2022. The TQIP database is a national registry of trauma centers throughout the United States. It utilizes International Classification of Diseases, tenth edition (ICD-10) procedure, and diagnosis codes to organize deidentified data compiled from participating centers. Patients with right-sided colon injuries were identified with ICD-10 codes using a substring word search. ICD-10 procedure codes were used to identify patients who underwent right hemicolectomy (0DBF0ZZ and 0DTF0ZZ) or ileocecectomy (0DTC0ZZ and 0DBC0ZZ). Patient demographics, presenting vitals, Glasgow Coma Scale (GCS), Injury Severity Score (ISS), blood product requirement, concomitant injuries, and mechanism of injury (MOI) were recorded. Outcomes analyzed included total days on ventilator, intensive care unit (ICU) stay, acute kidney injury (AKI), infection, unplanned ICU admission, return to operating room (OR), and mortality. Infection, as defined by the 2024 TQIP database dictionary, includes surgical site infection (SSI) characterized by purulent drainage from the surgical incision, deep surgical site infection (DSI) involving infection of the muscle or fascia, and organ space infection (OSI), which refers to infections occurring deeper than the fascia that was opened or manipulated during a surgical procedure. Due to the deidentified nature of the TQIP database, this study was exempt from review by the institutional review board. We adhered to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for cohort studies.¹⁴

Multiple imputations by chained equations were performed for missing covariate information. Missing data was treated as missing at random. Five data sets were imputed prior to propensity score matching. A logistic regression was used to generate a propensity score for each patient based on potential confounding factors. The average propensity score for each patient was used to match patients completing the right hemicolectomy to ileocecectomy in a 3:1 ratio for age, systolic blood pressure (SBP), heart rate (HR), ISS, MOI, time to procedure, concomitant injuries to the small bowel, stomach, or rectum, and blood transfusion requirement within the first four hours of presentation. Continuous data were reported using medians and interquartile ranges (IQRs) and compared using Mann-Whitney U Tests. Categorical data were reported using relative frequencies and percentages and compared using Fisher’s exact tests or Chi-square tests depending on normality. Univariate analyses were performed to compare patients who underwent right hemicolectomy vs ileocecectomy. A P-value of <0.05 was considered statistically significant. All statistical analyses were performed using R (version 4.4.0, R Foundation for Statistical Computing, Vienna, Austria).

Results

During the study period, we identified 6327 patients who underwent resection for traumatic right-sided colon injuries, with 186 (2.9%) completing IC and 6141 (97.1%) completing RH (Figure 1). Median age was 34 years (IQR 24-51) with 78% male and 49.3% white. The majority of patients presented after sustaining a blunt trauma (57.1%).

Patients who underwent RH and IC did not have a significant difference in presenting HR (98 beats per minute (bpm) (IQR 82-116) vs 98 bpm (IQR 83-117), P = 0.7), SBP (120 millimeters of mercury (mmHg) (IQR 100-140) vs 117 mmHg (IQR 98-137), P = 0.3), total GCS (15 (IQR 14-15) vs 15 (IQR 14-15), P = 0.7), ISS (18 (IQR 11-17) vs 19 (IQR 13-25), p= 0.3), or time to OR (2 hours (1-27) vs 1 hour (1-14), P = 0.9). Patients received similar rate of blood product transfusion within the first four hours of presentation (RH 2 (0-700) vs IC 2 (0-700, P > 0.9). RH patients were more likely to present after sustaining penetrating trauma; however, this was not significantly different than the IC group (43% vs 37%, P = 0.08). When concerning concomitant injuries, patients undergoing RH were more likely to have an associated gastric injury (11% vs 8.6%, P = 0.3) while significantly less likely to have an associated small bowel (70% vs 85%, P < 0.001) or rectal injury (2% vs 6.5%, P < 0.001). Patients completing initial IC were significantly more likely to undergo second laparotomy (20% vs 27%, P = 0.017); however, only 2 (1.1%) required unplanned return for completion RH (Table 1).

Table 1.

Pre-match Demographics and Injury Parameters.

	Right Hemicolectomy (n = 6141)	Ileocecectomy (n = 186)	P-value
Age (IQR)	34 (24,51)	34 (24,50)	0.8
Sex
Male (%)	4748 (78)	142 (76)	0.7
Female (%)	1393 (22)	44 (24)
Race
White (%)	3028 (49)	91 (49)	>0.9
Non-White (%)	3113 (51)	95 (51)
Penetrating trauma (%)	2632 (43)	67 (37)	0.08
Heart rate - BMP (IQR)	98 (82-116)	98 (83-117)	0.7
Systolic blood pressure - mmHg (IQR)	120 (100-140)	117 (98-137)	0.3
GCS (IQR)	15 (14-15)	15 (14-15)	0.7
Injury severity score (IQR)	18 (11-27)	19 (13-25)	0.3
Time to OR - hours (IQR)	2 (1-27)	1 (1-14)	0.9
Transfusion requirement (<4 hrs) (IQR)	2 (0-700)	2 (0-700)	>0.9
Concomitant injuries
Gastric (%)	667 (11)	16 (8.6)	0.3
Small bowel (%)	4318 (70)	159 (85)	< 0.001
Rectal (%)	121 (2)	12 (6.5)	< 0.001
Second laparotomy (same admission) (%)	1215 (20)	50 (27)	0.02

Of the 6327 patients, 744 patients were matched in a 3:1 ratio to compare patients who underwent RH (n = 558) and IC (n = 186). There was no difference between the RH and IC groups regarding presenting age, sex, white ethnicity, HR, SBP, GCS, ISS, MOI, or time to OR. Patients undergoing IC experienced significantly higher rates of associated small bowel injury (85% vs 74%, P < 0.001) and rectal injury (6.5% vs 2%, P = 0.002) (Table 2).

Table 2.

Post-match Demographics and Injury Parameters.

	Right Hemicolectomy (n = 558)	Ileocecectomy (n = 186)	P-value
Age (IQR)	35 (24-51)	34 (24-50)	0.7
Sex
Male (%)	426 (76)	142 (76)	>0.9
Female (%)	132 (24)	44 (24)
Race
White (%)	261 (47)	91 (49)	0.6
Non-White (%)	267 (53)	95 (51)
Penetrating trauma (%)	227 (41)	67 (37)	0.3
Heart rate - BMP (IQR)	98 (83-114)	98 (83-117)	0.6
Systolic blood pressure - mmHg (IQR)	122 (100-140)	117 (98-137)	0.2
GCS (IQR)	15 (14-15)	15 (14-15)	0.10
Injury severity score (IQR)	18 (12-26)	19 (13-25)	>0.9
Time to OR - hours (IQR)	1 (1-33)	1 (1-14)	0.4
Transfusion requirement (<4 hrs) (IQR)	0 (0-350)	2 (0-700)	0.2
Concomitant injuries
Gastric (%)	64 (11)	16 (8.6)	0.3
Small bowel (%)	411 (74)	159 (85)	< 0.001
Rectal (%)	11 (2)	12 (6.5)	0.002
Second laparotomy (same admission) (%)	128 (23)	50 (27)	0.3

Following resection, patients undergoing RH vs IC had similar rates of DSI (3.6% vs 4.3%, P = 0.7), SSI (0.7% vs 1.1%, P = 0.6) and incidence of severe sepsis (7.2% vs 4.8%, P = 0.3). Patients who underwent RH were significantly more likely to experience organ space infections (7.5% vs 3.2%, P = 0.04) and unplanned ICU admissions (9.7% vs 3.8%, P = 0.03). However, there were no significant differences between the RH and IC cohorts in ICU length of stay (6 days [IQR 6-14] vs 7 days [IQR 4-15], P = 0.3), unplanned return to the operating room (14% vs 16%, P = 0.6), or mortality (15% vs 11%, P = 0.2) (Table 3). Two patients that originally had IC had an unplanned return to the operating room with subsequent RH.

Table 3.

Outcomes of Patients.

	Right Hemicolectomy (n = 558)	Ileocecectomy (n = 186)	P-value
Acute kidney injury (%)	55 (9.9)	16 (8.6)	0.5
Presence of infection
Deep (%)	20 (3.6)	8 (4.3)	0.7
Organ space (%)	42 (7.5)	6 (3.2)	0.04
Surgical site (%)	4 (0.7)	2 (1.1)	0.6
Severe sepsis (%)	40 (7.2)	9 (4.8)	0.3
Unplanned ICU admission (%)	48 (8.6)	7 (3.8)	0.03
Unplanned OR return (%)	86 (15)	29 (16)	>0.9
Total ICU length of stay - days (IQR)	7 (3-14)	7 (4-15)	0.7
Mortality/Hospice (%)	79 (14)	17 (9.1)	0.08

Discussion

This is the first study to evaluate IC vs RH for a traumatic indication, to our knowledge. Overall, the rate of IC in the setting of a right colon injury requiring resection was 2.9% in the 2019 to 2022 TQIP patient cohort. Patients in both groups had comparable presenting demographics and injury parameters; however, following propensity-matched analysis, RH patients had higher rates of unplanned ICU admission and organ space infection and did not demonstrate superiority in clinical outcomes, including ICU length of stay, unplanned OR, and mortality. Reflecting the ongoing adjustments in traumatic colon management, we found IC as a practical and safe procedure for RCI when feasible.

Decreased utilization of IC is likely related to the notion that IC is an inferior operation compared to RH due anatomical considerations of the blood supply to the right colon, including the ileocolic and right colic artery.^11,15 Traditionally, the right colic artery was thought to arise from the superior mesenteric artery; however, in review of over 1000 cadaveric studies, Haywood and colleagues found increased variation in origin, location, and presence—with almost 10% of cases having an absent right colic artery.¹⁰ Furthermore, the right colon lacks the robust collateral circulation as seen with the left colon vasculature system, and is regarded as an end-organ system akin to the renal and cardiac systems.^11,15 Detractors are apprehensive of IC due to the subsequent vascular supply of the anastomosis, which was considered tenuous and at increased risk of compromise during periods of exacerbated low flow states, as seen with dialysis, cardiac failure, and shock.^11,15 This line of thinking extends beyond colon resections and is evident in the preference for using the left colon for interposition grafts after esophagectomy, despite comparable success rates with right-sided interpositions.^16-18

Failure to ensure adequate blood supply for an anastomosis can result in leakage or breakdown presenting as failure to progress, intra-abdominal infection, or uncontrolled sepsis requiring subsequent resection and diversion.¹⁹ In this study, patients completing IC had similar rates of surgical site and deep site infection, sepsis, and rate of unplanned return to the operating room when compared to the RH group. Notably, IC demonstrated significantly lower rates of organ space infections and unplanned ICU admissions, which may indicate reduced anastomotic leak rates and associated complications compared to RH but more likely represents a lower injury burden not well captured by a database study. Further investigation is warranted to validate these observations.

Despite IC being infrequently performed in trauma patients, it remains one of the most common resections completed in patients with Crohn’s disease (CD). In their analysis of a prospectively maintained database, 139 patients with medically refractory CD underwent IC with no reported anastomotic failures.¹² Interestingly, 68% of patients were current smokers while 46% were found to have advanced disease on pathology. Furthermore, in a propensity-matched analysis of the National Surgical Quality Improvement Program (NSQIP), patients who were considered to have malnutrition (hypoalbuminemia <3.5 g/dL, weight loss >10% in the preceding 6 months, or body mass index (BMI) < 18.5 kg/m²) had increased risk of bleeding and infection but not leak when compared to patients with adequate nutrition following IC.¹³ Finally, in a multicenter study evaluating 538 patients with CD, only 3% of patient experienced a leak after IC despite over 90% of patients having advanced disease and 42% having ongoing sepsis at time of surgery.²⁰ Across the 3 studies, 50-60% of patients were on chronic steroids at the time of surgery. Despite facing multiple barriers to medical optimization before surgery, including immunosuppression, chronic inflammation, and the presence of infection at the time of operation, colon-related morbidity remained low following IC, showcasing the durability of this form of colon resection.

Degree of injury, contamination and comorbidities continued to guide management of type and degree of resection for traumatic colon injuries.^4,6 In this series, we found that patients were young (34 years IQR (24-51)), had a median GCS of 15 and were severely injured with a median ISS 18. Presenting median SBP and HR in both groups did not indicate a greater than class II hemorrhagic shock and we did not find a significant difference in blood product requirement between the two groups, suggesting similar presentation in hemodynamics. However, this cannot be completely characterized. Diastolic Blood Pressure is not available in the database; thus, pulse pressure could not be used to characterize class of hemorrhage further. Previous assessments of traumatic colon resections identified risk factors such as massive transfusion and physiological derangements as barriers to performing anatomic resection with primary anastomosis, and even more so for segmental resection such as IC.⁴ In reviews addressing the management of traumatic colon injuries, ileocecectomy is neither explicitly mentioned as a form of resection nor thoroughly analyzed, implying a potential bias against this approach.^4,5,8,9,21 However, as advancements in triage, hemorrhage control, and damage control resuscitation continue to progress, we anticipate more favorable conditions for resection and subsequent anastomosis. IC demonstrated successful outcomes, with similar rates of complication, ICU length of stay, and mortality in this cohort compared to RH; however, clinical judgment remains paramount when determining degree of resection.

This study has several limitations. Coding errors, inherent to all large registries, have the potential to bias results. Abbreviated injury scales and Injury Severity Score serve as surrogates for injury burden but do not provide granular information regarding concomitant injuries. Hemodynamic and physiologic patient information is restricted to a set of admission vitals in this TQIP time period. Prehospital data and administration of specific medications is not available. The specific reasons for ICU admission and return to the operating room were not further specified. Damage control laparotomy is not specifically identified. The presence of anastomotic leak or breakdown was inferred from complications such as infection, sepsis, and return to the operating room; however, precise clinical and radiographic confirmation could not be established.

The exact location and extent of colon injury could not be determined, which may influence the physician’s decision to choose one surgical approach over another. Propensity score matching was employed to mitigate potential selection bias by aligning demographics, physiological parameters, associated injuries, and injury severity; however, it does not comprehensively address all factors influencing the decision to perform RH vs IC. Furthermore, any regression technique only attempts to account or control for known and/or measured confounders and does not control for unknown confounders which is limitation of any study that is not a randomized control trial. Future prospective studies would be necessary to validate these findings.

Conclusion

Ileocecectomy when feasible was an acceptable form of resection in patients sustaining right-sided colon injuries in this TQIP analysis. Patients selected to undergo ileocecectomy exhibited non-inferior outcomes, including infection, unplanned OR, unplanned ICU, LOS, and mortality, when compared to those who underwent right hemicolectomy. Clinical judgment remains pivotal and further investigation of the role of segmental resection in traumatic colon injuries is warranted.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Devanshi D. Patel

Andrew M. Fleming

References

Ogilvie WH . Forward surgery in modern war. Butterworth & co. ltd.; etc., etc., 1944;iii:96.

Bugis

Blair

Letwin

. Management of blunt and penetrating colon injuries. Am J Surg. 1992;163(5):547-550. doi:10.1016/0002-9610(92)90408-j

Chappuis

Frey

Dietzen

Panetta

Buechter

Cohn

Jr . Management of penetrating colon injuries. A prospective randomized trial. Ann Surg. 1991;213(5):492-497. doi:10.1097/00000658-199105000-00015

Sharpe

Magnotti

Fabian

Croce

. Evolution of the operative management of colon trauma. Trauma Surg Acute Care Open. 2017;2(1):e000092. doi:10.1136/tsaco-2017-000092

Steele

Wolcott

Mullenix

, et al.

Colon and rectal injuries during Operation Iraqi Freedom: are there any changing trends in management or outcome?

Dis Colon Rectum. 2007;50(6):870-877. doi:10.1007/s10350-007-0235-4

Stewart

Fabian

Croce

Pritchard

Minard

Kudsk

. Is resection with primary anastomosis following destructive colon wounds always safe? Am J Surg. 1994;168(4):316-319. doi:10.1016/s0002-9610(05)80156-4

Stone

Fabian

. Management of perforating colon trauma: randomization between primary closure and exteriorization. Ann Surg. 1979;190(4):430-436. doi:10.1097/00000658-197910000-00002

Sharpe

Magnotti

Weinberg

, et al. Impact of location on outcome after penetrating colon injuries. J Trauma Acute Care Surg. 2012;73(6):1428-1432. doi:10.1097/TA.0b013e31825bff06

Steele

Maykel

Johnson

. Traumatic injury of the colon and rectum: the evidence vs dogma. Dis Colon Rectum. 2011;54(9):1184-1201. doi:10.1007/DCR.0b013e3182188a60

10.

Haywood

Molyneux

Mahadevan

Srinivasaiah

. Right colic artery anatomy: a systematic review of cadaveric studies. Tech Coloproctol. 2017;21(12):937-943. doi:10.1007/s10151-017-1717-6

11.

Landreneau

Fry

. The right colon as a target organ of nonocclusive mesenteric ischemia. Case report and review of the literature. Arch Surg. 1990;125(5):591-594. doi:10.1001/archsurg.1990.01410170037007

12.

Cullen

O'Toole

Keegan

Sheahan

Hyland

O'Donoghue

D P

. Long-term clinical results of ileocecal resection for Crohn's disease. Inflamm Bowel Dis. 2007;13(11):1369-1373. doi:10.1002/ibd.20220

13.

Kodia

Huerta

Alnajar

, et al. Outcomes among malnourished patients with Crohn's disease undergoing Elective ileocecectomy: a Nationwide analysis. Am Surg. 2023;1:31348231209866. doi:10.1177/00031348231209866

14.

Vandenbroucke

von Elm

Altman

, et al. Strengthening the reporting of observational studies in Epidemiology (STROBE): explanation and elaboration. Int J Surg. 2014;12(12):1500-1524. doi:10.1016/j.ijsu.2014.07.014

15.

Ludwig

Quebbeman

Bergstein

Wallace

Wittmann

Aprahamian

. Shock-associated right colon ischemia and necrosis. J Trauma. 1995;39(6):1171-1174. doi:10.1097/00005373-199512000-00029

16.

Akutsu

Fujita

Kajiyama

, et al. Operative outcomes and long-term survival of patients undergoing colon interposition after esophagectomy for cancer. Thorac Cancer. 2022;13(6):844-852. doi:10.1111/1759-7714.14332

17.

Chirica

Veyrie

Munoz-Bongrand

, et al. Late morbidity after colon interposition for corrosive esophageal injury: risk factors, management, and outcome. A 20-years experience. Ann Surg. 2010;252(2):271-280. doi:10.1097/SLA.0b013e3181e8fd40

18.

Sanchez

Alicuben

Luketich

Sarkaria

. Colon interposition for esophageal cancer. Thorac Surg Clin. 2022;32(4):511-527. doi:10.1016/j.thorsurg.2022.07.006

19.

Hyman

Manchester

Osler

Burns

Cataldo

. Anastomotic leaks after intestinal anastomosis: it's later than you think. Ann Surg. 2007;245(2):254-258. doi:10.1097/01.sla.0000225083.27182.85

20.

de Buck van Overstraeten

Eshuis

Vermeire

, et al. Short- and medium-term outcomes following primary ileocaecal resection for Crohn's disease in two specialist centres. Br J Surg. 2017;104(12):1713-1722. doi:10.1002/bjs.10595

21.

Yamamoto

Logue

Muir

. Colon trauma: evidence-based practices. Clin Colon Rectal Surg. 2018;31(1):11-16. doi:10.1055/s-0037-1602175