Open Versus Laparoscopic Right Hemicolectomies in Pediatric Patients with Crohn's Disease

Abstract

Introduction:

Surgery remains an important treatment modality for the management of pediatric Crohn's disease (CD). The objective of this study was to perform a comparative analysis of open right hemicolectomy (ORH) and laparoscopic right hemicolectomy (LRH) for the management of pediatric CD.

Materials and Methods:

The Kids' Inpatient Database (KID) was queried (2009–2012) for ICD-9 procedure codes for ORH (45.73) and LRH (17.33) in patients with CD (ICD-9 codes: 555.0, 555.1, 555.2, 555.9). Open and laparoscopic procedures were compared using propensity score (PS)-matched analysis (PSMA) of 41 variables.

Results:

Overall 889 patients were identified and after PS matching, there were 380 ORHs and 380 LRHs. There were zero in-hospital deaths (0/821). ORH patients were more likely to have septicemia, respiratory compromise, pneumonia, perforation and/or laceration, complications, and require blood transfusions (all, P < .05). Although LRH patients were more likely to develop postoperative nausea/vomiting/diarrhea (P < .0001), they had a shorter hospital length of stay (P < .0001) and lower overall hospital charges and cost (P < .001).

Conclusion:

ORH and LRH in KID have similar low in-hospital mortality in pediatric CD. However, ORH was associated with higher morbidity including an increased risk for respiratory complications, surgical complications, need for blood transfusions, and increased resource utilization than patients who had laparoscopic procedures. In select patients, LRH is safe, feasible, and potentially superior to ORH.

Introduction

Crohn's disease (CD) is a chronic relapsing inflammatory bowel disease (IBD) with a multifactorial etiology of immune, bacterial, and environmental factors in the background of genetically susceptible individuals.¹ In the pediatric population, CD is under the umbrella category of pediatric IBDs, and at diagnosis these patients tend to have more extensive anatomical involvement with more rapid progression than their adult counterparts.² Although the etiology remains unclear, the worldwide incidence of CD has significantly increased over the past 50 years and up to 25% of the new CD cases are diagnosed in childhood or adolescence.^1,3

Although medical advancements have improved immunomodulation for patients with CD, surgical management often becomes necessary in the pediatric subset due to the aggressive nature of the disease. One-third of pediatric CD patients will develop fistula, stricture, obstruction, and/or complications to their medical therapy for which they eventually require surgical management.⁴ The core goal of any surgical procedure for CD is bowel length preservation while rectifying sequelae such as bowel strictures and fistulization. Up to 50% of pediatric CD patients have terminal ileal involvement, with its associated complications,⁵ and consequently partial or total right hemicolectomy is a common surgical procedure in this population.

Owing to the chronic nature of this disease, surgical management may need to be repeated. In adults after ileal or ileocecal resection, previous reports have demonstrated a subsequent operation in 5% after 1 year, 11%–32% after 5 years, and 20%–44% after 10 years.^6,7 Thus, a minimally invasive approach is appealing. Better outcomes with a laparoscopic approach have been demonstrated in the adult literature,^8,9 and von Allmen et al. have also shown superior outcomes in laparoscopic-assisted bowel resection in pediatric CD with respect to length of stay (LOS), use of parenteral narcotics, and return to regular diet.¹⁰ Likewise, a total intracorporeal laparoscopic approach has been shown to be safe and effective.¹¹ However, there are currently few nationwide analyses comparing the surgical approach (open versus laparoscopic) with outcomes and health care utilization for this surgical procedure in this patient population.

Owing to the high disease burden in pediatric CD, higher likelihood of surgical management, and longer lifespan overall than adult counterparts, we set out to perform a comparative analysis of open right hemicolectomy (ORH) versus laparoscopic right hemicolectomy (LRH) in the pediatric population. We hypothesized that a laparoscopic approach may be associated with lower postoperative complications and health care utilization.

Materials and Methods

The Kids' Inpatient Database (KID) is maintained by the Healthcare Cost and Utilization Project (HCUP), Agency for Healthcare Quality and Research. The database, released triennially, contains information from pediatric discharges in 44 states, and in the 2012 release, contained 3 million unweighted pediatric discharges (6.7 million weighted). Diagnoses and procedures from each hospitalization are coded using the International Classification of Disease, 9th Revision (ICD-9).

For this study, KID was analyzed for the triennial data set releases 2009 and 2012 utilizing ICD-9 procedure codes for ORH (45.73) and LRH (17.33) in patients with CD (ICD-9 codes: 555.0, 555.1, 555.2, 555.9). These particular ICD-9 codes capture ileocecectomy as well as total right colectomy. Patients with dispositions coded as “transfer to short-term hospital” and “other transfers, including skilled nursing facility, intermediate care, and other type of facility” were excluded from analyses to avoid inclusion of duplicate cases. Cases were only included if they were coded specifically as “laparoscopic” or “open” and were excluded if there was mention of both procedure types being utilized simultaneously. Cases were weighted to allow national estimates. All analyses were limited to available data.

In-hospital postoperative complications used in analyses were previously defined for the ICD-9 coding scheme by Tuinen et al.¹² Continuous variables were compared using Student's t tests, Mann–Whitney U tests, or ANOVA as appropriate. Categorical variables were compared using chi-square tests or Fischer's exact tests, as appropriate. A subanalysis was performed for costs associated with hospitalizations. Cost conversions were derived from charge data using cost-to-charge files, available from HCUP from 2009 to 2012. Analyses involving total charges (TCs) and cost were adjusted to 2012 U.S. dollar values according to inflation rates determined by the U.S. Department of Labor.¹³ All analyses were performed using SPSS Statistics, version 24.0 (IBM, Armonk, NY).

We compared in-hospital outcomes and health care utilization between LRHs and ORHs in pediatric CD patients using a propensity score (PS)-matched technique. This method matches LRH and ORH on a ratio based on comorbid and demographic factors and allows for fair and risk-adjusted comparisons derived from a large national cohort. For each type of operative procedure, a data set containing a total of 77 variables (including demographics, hospital factors, and associated diagnoses) was matched according to 41 covariates to construct a matched cohort. PS values were assigned through multivariate logistic regression, according to demographic (age, gender, and race), socioeconomic (payer status and median income quartile), and hospital characteristics (bed size, location/teaching status, region, and type), as well as procedure type (laparoscopic, open, elective, and nonelective). Additional risk adjustment was performed using the Elixhauser method, which has been validated in previous retrospective outcome studies.^14–17

Results

Demographics

In total, 889 pediatric CD patients underwent right hemicolectomies during the study period. Half (n = 448) underwent ORH while the other half underwent LRH (n = 441). Demographic and clinical data are given in Table 1. Males comprised 63% and 52% of the ORH and LRH cohorts, respectively, and most patients in both groups were Caucasian (both >70%). A higher percentage of cases underwent LRH (61%) than ORH (45%) in later years, P = .001. More than 60% of patients with nonelective admissions had ORH, while 62% of patients admitted electively underwent LRH, P = .001. There were zero inpatient hospital deaths for both ORH and LRH.

Table 1.

Demographic and Clinical Characteristics of Open Versus Laparoscopic Right Hemicolectomy, Kids' Inpatient Database 2009–2012, in Pediatric Crohn's Disease Patients

	Open right hemicolectomy (n = 448)		Laparoscopic right hemicolectomy (n = 441)		P
Age, years, mean ± SD	16.7 ± 2.3		16.1 ± 2.6		.001
LOS, days, mean ± SD	11 ± 10		8.1 ± 6.3		.001
Total cost, USD, mean ± SD	26,787.4 ± 30,482		21,216.1 ± 17,369		.002
	n	%	n	%
Mortality	^*	<0.01	^*	<0.01	n.s.
Year
2009	247	55.1	171	38.8	.001
2012	201	44.9	270	61.2	.028
Gender					.001
Male	284	63.4	228	51.7
Female	162	36.6	210	48.3
Race					n.s.
Caucasian	281	71.1	282	74.8
Black	72	18.2	56	14.9
Hispanic	18	4.6	14	3.7
Asian/Pacific Islander	^*	<0.01	^*	<1.0
Native American	^*	<1.0	^*	<0.01
Other	20	5.1	21	5.6
Admission type					.001
Elective	171	38.4	271	61.7
Nonelective	274	61.6	168	38.3
Payer					.02
Medicare	^*	<1.0	^*	<0.01
Medicaid	97	21.7	68	15.5
Self-pay	^*	<2.0	^*	<1.0
Private insurance	316	70.9	353	80.2
No charge	^*	<1.0	^*	<0.01
Other	20	4.5	15	3.4
Hospital type					.002
Nonchildren's hospital	103	45.8	46	28.6
Children's hospital	31	13.8	26	16.1
Children's unit in hospital	91	40.4	89	55.3
Hospital location/teaching					n.s.
Rural	20	4.6	^*	<1.0
Urban nonteaching	81	18.7	59	13.
Urban teaching	332	76.7	373	85.7
Median household income national quartile					.001
First quartile	106	24.0	57	13.1
Second quartile	100	22.6	93	21.4
Third quartile	96	21.7	97	22.4
Fourth quartile	140	31.7	187	43.1

Cells marked with an asterisk (^*) represent actual values censored from publication in accordance with the HCUP use agreement.

HCUP, Healthcare Cost and Utilization Project; LOS, length of stay; n.s., not significant, SD, standard deviation; USD, U.S. dollars.

Hospital type

Operations performed in the rural setting were 87% ORH, whereas urban-teaching hospitals had a more balanced distribution between open (47%) and laparoscopic operations (53%), P = .001. A higher percentage of cases underwent LRH in children's hospitals (46%) than in the children's unit in a general hospital (32%) and in a nonchildren's hospital (31%), n = 0.002.

Admission type

Subset analysis was then performed by admission type to analyze results of LRH versus ORH in those undergoing elective or nonelective operations (Table 2). In the nonelective setting, ORH was associated with higher rates of sepsis (5% versus 1%) and blood transfusion (9% versus 3%) than LRH, whereas postoperative nausea/vomiting/diarrhea was more common in LRH (5% versus 2%). For the elective procedures, the only difference remained higher nausea/vomiting/diarrhea in LRH (5% versus 0.2%). With regard to health care utilization, ORH was associated with longer LOS in both the elective and nonelective settings in Table 3. Hospital cost and TCs were also significantly higher, especially in the nonelective admissions.

Table 2.

Analysis of Admission Type in Open Versus Laparoscopic Right Hemicolectomy for Pediatric Crohn's Disease Kids’ Inpatient Database 2009–2012

	Open right hemicolectomy (n = 271)		Laparoscopic right hemicolectomy (n = 168)		P
	n	%	n	%	P
Nonelective procedure
Complications
Septicemia	20	5.0	^*	<1.0	.01
Wound complications	13	3.3	^*	<2.0	n.s.
pRBC	34	8.5	12	3.0	.02
Nausea, vomiting, diarrhea	^*	<2.0	19	4.8	.002
Requires postoperative procedure	10	2.5	^*	<0.5	n.s.
Elective procedure
Complications
Septicemia	^*	<0.01	^*	<0.2	n.s.
Wound complications	^*	<1.2	^*	<1.2	n.s.
pRBC	^*	<2.1	10	2.4	n.s.
Nausea, vomiting, diarrhea	^*	<0.2	19	4.5	.004
Requires postoperative procedure	^*	<0.01	^*	<0.2	n.s.

Cells marked with an asterisk (^*) represent actual values censored from publication in accordance with the HCUP use agreement.

HCUP, Healthcare Cost and Utilization Project; n.s. not significant; pRBC, packed red blood cells.

Table 3.

Analysis of Elective Versus Nonelective Admission and Health Care Utilization After Right Hemicolectomy for Pediatric Crohn's Disease, Kids' Inpatient Database 2009–2012

	Elective admissions, Mean ± SD		Nonelective admissions, Mean ± SD		P
	ORH (n = 171)	LRH (n = 274)	ORH (n = 271)	LRH (n = 168)	P
LOS, days	7.2 ± 5.7	5.69 ± 3.6	13.7 ± 12.8	12.95 ± 11.0	<.001
TCs, USD	54,705 ± 51,113	53,108 ± 36,243	106,652 ± 115,830	91,068 ± 61,120	<.001
Cost, USD	18,162 ± 18,755	16,651 ± 12,091	33,280 ± 37238	28,799 ± 21,645	<.001

LOS, length of stay; LRH, laparoscopic right hemicolectomy; ORH, open right hemicolectomy; SD, standard deviation; USD, U.S. dollars; TCs, total charges.

Propensity score

On PS-matched analysis (380 ORH versus 380 LRH), ORH patients were associated with higher likelihood to develop septicemia, respiratory compromise, perforation and/or laceration, complications, and require blood transfusions (all, P < .05). Conversely, LRH patients were associated with higher likelihood to develop postoperative nausea/vomiting/diarrhea (P < .001); however, these patients had overall shorter hospital LOS (P < .001) and lower overall hospital charges and cost (P < .001) (Table 4).

Table 4.

Propensity Score-Matched Comparison of Open Versus Laparoscopic Right Hemicolectomy for Pediatric Crohn's Disease Patients, Kids' Inpatient Database 2009–2012

	Open right hemicolectomy (n = 380)		Laparoscopic right hemicolectomy (n = 380)		P
	n	%	n	%	P
Mortality	^*	<0.01	^*	<0.01	—
Complications
Wound infection	15	3.9	10	2.5	n.s.
Wound disruption	^*	<2.0	^*	<1.0	n.s.
Reopening surgical site	^*	<1.0	^*	<0.01	n.s.
pRBC transfusion	43	11.3	17	4.5	.001
Perforation or laceration	^*	<1.0	^*	<0.01	.04
Septicemia	20	6.0	^*	<1.0	<.001
Pneumonia	^*	<2.0	^*	<1.0	n.s.
Other infections	13	3.4	20	5.3	n.s.
Diseases of respiratory system	^*	<3.0	^*	<0.01	<.001
Acute GI ulcer or bleed	^*	<2.0	12	2.7	n.s.
Nausea, vomiting, diarrhea	10	2.6	30	8.0	<.001
Procedure-specific complications	^*	<2.0	^*	<0.01	.01
Require postoperative procedure	10	2.6	^*	<1.0	.03
Resource utilization
LOS, days	11.10 ± 11		8.1 ± 6.3		<.001
TCs, USD	85,740 ± 98,379		67,636 ± 50,697		<.001
Cost, USD	26,930 ± 31,744		21,216 ± 17,369		<.001

PS values were assigned, and ORH versus LRH cohorts underwent risk-adjusted matched comparison using 41 covariates. Resource utilization variables presented as mean ± SD. Cells marked with an asterisk (^*) represent actual values censored from publication in accordance with the HCUP use agreement.

GI, gastrointestinal; HCUP, Healthcare Cost and Utilization Project; LOS, length of stay; n.s., not significant at α = 0.05; pRBC, packed red blood cell; PS, propensity score; SD, standard deviation; USD, U.S. dollars; TCs, total charges.

Discussion

Our analysis demonstrates several advantages of performing LRH over ORH in children with CD. Both operations have a very low in-patient mortality and thus LRH is a safe option, congruent with previous reports.^18–20 Our analysis suggests that ORH is associated with more in-hospital complications than LRH. A previous study of pediatric CD patients by Mahida et al. failed to demonstrate a difference between postoperative complications; however, the study was not sufficiently powered to elicit a statistical difference between the cohorts.²¹ Our current study, conversely, utilized a PS-matched analysis in lieu of a randomized controlled trial (RCT) to control for potential confounding factors and demonstrated statistically significantly lower rates of wound disruption, perforation/laceration, procedure-specific complications, pneumonia, sepsis, and requirement for blood transfusions in the LRH cohort. This analysis demonstrating the laparoscopic approach is associated with fewer postoperative complications is consistent with a smaller population-based study²² and an RCT that also demonstrated increased morbidity with open operations in CD patients.²³

Our analysis not only demonstrated an overall decreased in-hospital morbidity with laparoscopic operations, but also decreased resource utilization in the laparoscopic surgery cohort. LRH patients spent an average of 3 days less in the hospital than their ORH counterparts, which was found to be statistically significant. This is concordant with a previous study that also demonstrated a decreased LOS for patients undergoing laparoscopic bowel resection with primary anastomosis.²¹ Previously unreported, our analysis reflected a cost-effectiveness in the laparoscopic approach as both hospital charges and total cost were higher in the ORH cohort, with a mean difference in cost of $5,714. This is important because although general consensus would argue that the laparoscopic equipment is more costly and operative costs are higher, the overall cost of the hospitalization remains lower in the LRH group, suggesting that operating room expenditures do not translate to an increased overall cost. Several studies in the adult population have made this conclusion as well.^23–25

The benefits of LRH include improved cosmesis, decreased LOS, less use of parenteral narcotics, and earlier return to regular diet^10,26,27 In addition, a significant proportion of patients require repeat operations throughout their lifetime for recurrent disease.^6,7 Although previous reports demonstrate a 44% risk of subsequent operation at 10 years, this does not take into account modern treatment algorithms and novel immunomodulatory agents, such as early and aggressive use of biologic agents before and after surgical management. This may decrease the risk of recurrence and the need for subsequent surgical treatment. Unfortunately, long-term data regarding the effects of novel therapies on reoperative rates are not yet available.

This study is not without limitations. First, the study design is a retrospective review of the KID, which lends itself to errors in data sampling, collection measures, usage of the ICD-9 coding scheme, and possible administrative errors during data entry. Important perioperative factors such as operative time and operative difficulty, current medical therapies utilized (biologic agents and corticosteroids), and postoperative analgesia required are not recorded in KID and, therefore, are not included in this analysis. Owing to the nature of KID and its accrual of data, information regarding possible readmissions and longer term complications is not available and thus are outside the scope of this study.

Conclusion

ORH and LRH in KID have similar low in-hospital mortality for pediatric patients with CD. However, open procedures are associated with higher in-hospital morbidity including an increased risk for respiratory complications, surgical complications, need for blood transfusions, and increased resource utilization than patients who had laparoscopic procedures. This study suggests that laparoscopic surgery for the management of pediatric CD is safe, feasible, and potentially superior to open resections in select patients.

Footnotes

Acknowledgment

This study was exempt from full Institutional Review Board review as it was retrospective in nature and utilizes the KID, which contains deidentified patient information.

Authors' Contributions

The final draft of the article was written by H.Q.

Disclosure Statement

No competing financial interests exist.

Funding Information

Funding was received from intramural funding from the Department of Pediatric Surgery at the University of Miami Miller School of Medicine.

References

Cho

, Brant

. Recent insights into the genetics of inflammatory bowel disease. Gastroenterology, 2011; 140:1704–1712.

Wilson

, Russell

. Overview of paediatric IBD. Semin Pediatr Surg, 2017; 26:344–348.

Amil-Dias

, Kolacek

, Turner

, et al. Surgical management of Crohn disease in children: Guidelines from the paediatric IBD Porto Group of ESPGHAN. J Pediatr Gastroenterol Nutr, 2017; 64:818–835.

Stewart

. Surgical care of the pediatric Crohn's disease patient. Semin. Pediatr. Surg, 2017; 26:373–378.

Schaefer

, Machan

, Kawatu

, et al. Factors that determine risk for surgery in pediatric patients with Crohn's disease. Clin Gastroenterol Hepatol, 2010; 8:789–794.

Nos

, Domenech

. Postoperative Crohn's disease recurrence: A practical approach. World J Gastroenterol, 2008; 14:5540–5548.

, Kamm

. Management of postoperative Crohn's disease. Am J Gastroenterol, 2008; 103:1029–1035.

Soop

, Larson

, Malireddy

, Cima

, Young-Fadok

, Dozois

. Safety, feasibility, and short-term outcomes of laparoscopically assisted primary ileocolic resection for Crohn's disease. Surg Endosc, 2009; 23:1876–1881.

Lesperance

, Martin

, Lehmann

, Brounts

, Steele

. National trends and outcomes for the surgical therapy of ileocolonic Crohn's disease: A population-based analysis of laparoscopic vs. open approaches. J Gastrointest Surg, 2009; 13:1251–1259.

10.

von Allmen

, Markowitz

, York

, Mamula

, Shepanski

, Baldassano

. Laparoscopic-assisted bowel resection offers advantages over open surgery for treatment of segmental Crohn's disease in children. J Pediatr Surg, 2003; 38:963–965.

11.

Dutta

, Rothenberg

, Chang

, Bealer

. Total intracorporeal laparoscopic resection of Crohn's disease. J Pediatr Surg, 2003; 38:717–719.

12.

Tuinen

, Elder

, Link

, Li

, Song

, Pritchett

. Advances in patient safety surveillance of surgery-related adverse events in Missouri using ICD-9-CM codes. In: Henriksen K, Battles JB, Marks ES, Lewin DI (eds). Advances in Patient Safety: From Research to Implementation (Volume 1: Research Findings). Rockville, MD: Agency for Healthcare Research and Quality (US), 2005.

13.

Clark DE Turner

, Hahn

. ICDPIC: Stata module to provide methods for translating International Classification of Diseases (Ninth Revision) diagnosis codes into standard injury categories and/or scores. Chestnut Hill MA: Statistical Software Components S457028, Boston College Department of Economics, 2009. (Accessed September 1, 2018).

14.

Elixhauser

, Steiner

, Harris

, Coffey

. Comorbidity measures for use with administrative data. Med Care, 1998; 36:8–27.

15.

Stone

, Lapar

, Kane

, Rasmussen

, McGahren

, Rodgers

. The effect of race and gender on pediatric surgical outcomes within the United States. J Pediatr Surg, 2013; 48:1650–1656.

16.

Stone

, LaPar

, Mulloy

, et al. Primary payer status is significantly associated with postoperative mortality, morbidity, and hospital resource utilization in pediatric surgical patients within the United States. J Pediatr Surg, 2013; 48:81–87.

17.

Tashiro

, Perez

, Sola

. Reduced hospital mortality with surgical ligation of patent ductus arteriosus in premature, extremely low birth weight infants: A propensity score-matched outcome study. Ann Surg, 2016; 263:608–614.

18.

Simon

, Orangio

, Ambroze

, Schertzer

, Armstrong

. Laparoscopic-assisted bowel resection in pediatric/adolescent inflammatory bowel disease: Laparoscopic bowel resection in children. Dis Colon Rectum, 2003; 46:1325–1331.

19.

Huang

, Koleilat

, Lee

. Single surgeon experience with laparoscopic surgery in pediatric patients with inflammatory bowel disease. J Laparoendosc Adv Surg Tech A, 2013; 23:61–64.

20.

Mattioli

, Pini-Prato

, Barabino

, et al. Laparoscopic approach for children with inflammatory bowel diseases. Pediatr Surg Int, 2011; 27:839–846.

21.

Mahida

, Asti

, Deans

, Minneci

, Nwomeh

. Laparoscopic bowel resection for pediatric inflammatory bowel disease. J Surg Res, 2015; 199:130–136.

22.

Frolkis

, Kaplan

, Patel

, et al. Postoperative complications and emergent readmission in children and adults with inflammatory bowel disease who undergo intestinal resection: A population-based study. Inflamm Bowel Dis, 2014; 20:1316–1323.

23.

Maartense

, Dunker

, Slors

, et al. Laparoscopic-assisted versus open ileocolic resection for Crohn's disease: A randomized trial. Ann Surg, 2006; 243:143–149; discussion 150–143.

24.

Duepree

, Senagore

, Delaney

, Brady

, Fazio

. Advantages of laparoscopic resection for ileocecal Crohn's disease. Dis Colon Rectum, 2002; 45:605–610.

25.

Shore

, Gonzalez

, Bondora

, Vickers

. Laparoscopic vs conventional ileocolectomy for primary Crohn disease. Arch Surg, 2003; 138:76–79.

26.

White

, Jenkins

, Coomber

, Clark

, Phillips

, Kennedy

. Outcomes of laparoscopic and open restorative proctocolectomy. Br J Surg, 2014; 101:1160–1165.

27.

Dasari

, McKay

, Gardiner

. Laparoscopic versus open surgery for small bowel Crohn's disease. Cochrane Database Syst Rev, 2011:Cd006956.