30-Day Outcomes of Laparoscopic Versus Open Total Proctocolectomy with Ileoanal Anastomosis in Children and Young Adults: A Combined Analysis of the National Surgical Quality Improvement Project Pediatric and Adult Databases

Abstract

Background:

Studies comparing pediatric laparoscopic and open total proctocolectomy with ileoanal anastomosis (TPC-IAA) are limited in size and number. This study utilized the adult and pediatric databases of the National Surgical Quality Improvement Project (NSQIP) to evaluate 30-day outcomes of these two techniques.

Materials and Methods:

Patients younger than 21 years who underwent TPC-IAA from 2012 to 2016 were identified in both NSQIP databases. Simple and multivariate logistic regression was used to compare risk of reoperation, readmission, and postoperative occurrences between laparoscopic and open groups. Cox regression was used to evaluate length of stay (LOS).

Results:

A total of 440 cases were identified, of which 421 (95.7%) were elective. Median age in the elective group was 15.8 years (interquartile range 13–18). Diagnoses included inflammatory bowel disease (47%), benign neoplasm (42%), and Hirschsprung disease (6%). The laparoscopic group (67.5%, n = 139) had shorter median postoperative LOS (6 versus 8 days, P < .001) and decreased incidence of pulmonary complications (risk ratio [RR] 0.09; CI: 0.01–0.80, P = .031) and superficial surgical site infections (SSI) (RR 0.30; 95% CI: 0.10–0.88, P = .028). Median operative time was shorter (4.6 versus 5.1 hours, P = .013) and risk of organ space SSI was lower (RR = 0.11, 95% CI: 0.01–0.80, P = .037) in the open group (n = 282). Rates of 30-day readmission and reoperation were similar between groups.

Conclusions:

In the first study to utilize data from both the pediatric and adult NSQIP databases, resulting in the largest pediatric sample of TPC-IAA to date, we found that 67.5% of elective cases were performed laparoscopically, the highest reported in a multi-institutional pediatric study, indicating increasing comfort with advanced laparoscopic techniques among pediatric surgeons. The laparoscopic approach resulted in shorter postoperative LOS and decreased risk of superficial SSI, whereas the open approach was associated with shorter operative time and lower risk of organ space SSI.

Introduction

Despite increasing utilization of minimally invasive techniques in colorectal surgery, a significant proportion of elective pediatric colectomies and proctocolectomies in the United States are performed by using an open approach.^1–4 Most studies comparing laparoscopic and open total proctocolectomy with ileoanal anastomosis (TPC-IAA) in children are limited to single institutions and are characterized by relatively small sample sizes.^5–8 One study that utilized a multi-institutional database to compare laparoscopic and open proctocolectomy in children with inflammatory bowel disease (IBD) did not limit its sample to those who underwent concurrent ileoanal anastomosis.⁹ Another multi-institutional study that compared the two approaches in patients with familial adenomatous polyposis (FAP) consisted primarily of adults.¹⁰ By contrast, the purpose of our study was to utilize a multi-institutional database to compare 30-day outcomes of laparoscopic versus open TPC-IAA performed in children.

Materials and Methods

After IRB acknowledgement (IRB #00149969), we identified subjects by current procedural terminology (CPT) code in the 2012–2016 adult and pediatric databases of the American College of Surgeons National Surgical Quality Improvement Project (NSQIP). Patients younger than 21 years of age who underwent laparoscopic (CPT 44211) and open total abdominal proctocolectomy with ileoanal anastomosis with (CPT 44158) or without (CPT 44157) creation of ileal reservoir were included. Given the anticipated differences in baseline physiology and comorbidity profiles between elective and non-elective cases, these two groups were examined separately. Descriptive statistics were derived to compare demographics, comorbidities, preoperative laboratory values, and 30-day clinical outcomes between open and laparoscopic groups. For purposes of multivariate analysis, comorbidities and risk factors were combined into generalized categories (Appendix 1). The presence of an ostomy at the time of proctocolectomy was determined either by explicit documentation of a pre-existing ostomy in the database (when available) or by description of concurrent ostomy closure or revision.

The primary outcome of interest was any 30-day complication, including mortality, anemia requiring blood transfusion, pulmonary complication, deep venous thrombosis, sepsis, urinary tract infection, or surgical site infection (SSI). We defined pulmonary complications as pneumonia, pulmonary embolism, or unplanned reintubation. Secondary outcomes included postoperative length of stay (LOS), operative time, unplanned readmission, unplanned reoperation, and specific complications. When calculating operative time, subjects undergoing additional procedures at the time of the primary operation were excluded from analysis. Categorical independent variables were evaluated by using Fisher's exact test. Continuous independent variables were assessed by using Wilcoxon rank-sum test. Risk of complications, readmission, and reoperation were compared between groups by using simple and multivariate logistic regression. We used Cox regression to evaluate for differences in LOS.¹¹ Variables for simple logistic regression were selected a priori based on a review of the literature and availability in the database. ^12–15 With the exception of age and sex, which were included a priori in the multivariate model, independent variables for multivariate logistic regression were selected based on the evidence of association with the outcome of interest on simple logistic regression, defined as P < .1. The same criteria were used for inclusion of independent variables in the adjusted Cox regression, with the exception that diagnosis was also included a priori to account for the effects of disease type on LOS. Odds ratios derived from multivariate logistic regression were converted to risk ratios (RR) by using the method described by Zhang and Yu.¹⁶ An alpha level of 0.05 was used for all statistical tests.

Results

A total of 440 subjects were included in the study, 316 from the pediatric database and 124 from the adult database. The vast majority of cases were elective (95.7%), and the results described in the next three paragraphs pertain specifically to this group. Only 19 patients (4.3%) underwent nonelective procedures, and the results pertaining to this subgroup are discussed separately in the final paragraph of this section.

Of the 421 elective cases, a slight majority were male (52.7%). Median age was 15.8 years (interquartile range [IQR] 13–18) and was slightly higher in the laparoscopic group (15.9 versus 15.2, P = .023). Median weight was 55.8 kg (IQR 40.3–68.9). The majority of patients underwent a laparoscopic procedure (67.5%). The most common disease processes that necessitated the procedure were IBD (46.8%), benign neoplasm of the colon (FAP) (41.8%), and Hirschsprung disease (HD) (6.2%). The proportion of children with a diagnosis of HD was higher in the open procedure group (10.2% versus 4.2%, P = .028), as was the proportion of children with a preceding stoma (16.1% versus 5.3%, P = .001). Moreover, the proportion of children who underwent an open procedure was higher among those with a preceding stoma (59%) as compared with those without a preexisting stoma (26%) and those for whom stoma history was unknown (31%) (P = .001). Demographic characteristics, comorbidities, and diagnosis distribution were similar between open and laparoscopic groups with the exception of those mentioned earlier (Table 1).

Table 1.

Demographics, Diagnosis, and Comorbidities of Children Undergoing Elective Open Versus Laparoscopic Total Proctocolectomy with Ileoanal Anastomosis

Demographics, diagnosis, and comorbidities	Total (n = 421)	Open (n = 137)	Lap (n = 284)	P
Age	15.8 (12–18)	15.2 (11–18)	15.9 (13–18)	.023
Sex (male)	222 (52.7)	73 (53.3)	149 (52.5)	.875
Race				.610
White	343 (81.5)	112 (81.8)	231 (81.3)	1.000
Black	44 (10.5)	12 (8.8)	32 (11.3)	.499
Other/unknown	34 (8.1)	13 (9.5)	21 (7.4)	.452
Weight (kilograms)	55.8 (40.3–68.9)	54.0 (33.6–66.0)	56.1 (41.6–69.2)	.134
Diagnosis				<.001
Benign neoplasm of colon and rectum	176 (41.8)	50 (36.5)	126 (44.4)	.140
Inflammatory bowel disease	197 (46.8)	59 (43.1)	138 (48.6)	.299
Hirschsprung disease	26 (6.2)	14 (10.2)	12 (4.2)	.028
Other	22 (5.2)	14 (10.2)	8 (2.8)	.004
Steroid therapy	81 (19.2)	19 (13.9)	62 (21.8)	.064
Nutritional support	21 (5.0)	9 (6.6)	12 (4.2)	.341
Pulmonary risk factor	23 (5.5)	11 (8.0)	12 (4.2)	.114
Cardiac risk factor	9 (2.1)	2 (1.5)	7 (2.5)	.724
Renal risk factor	2 (0.5)	2 (1.5)	0 (0)	.094
Neurocognitive risk factor	18 (4.3)	7 (5.1)	11 (3.9)	.383
Neuromuscular disorder	3 (0.7)	2 (1.5)	1 (0.4)	.136
Congenital malformation	49 (11.6)	19 (13.9)	30 (10.6)	.160
ASA class ≥3	117 (27.1)	38 (27.7)	76 (26.8)	.907
SIRS within 48 hours before surgery	8 (1.9)	1 (0.7)	7 (2.5)	.447
Inotropic support	2 (0.5)	0 (0)	2 (0.7)	.115
Hematologic disorder	38 (9.0)	13 (9.5)	25 (8.8)	.818
Immune disorder	21 (5.0)	6 (4.4)	15 (5.3)	.586
Wound infection	2 (0.5)	1 (0.7)	1 (0.4)	.545
Preoperative laboratory values
Hematocrit	38 (35–42)	38 (34–41)	38 (35–42)	.292
Platelets	315 (255–380)	302 (251–361)	321 (258–388)	.071
INR	1.0 (1.0–1.1)	1.0 (1.0–1.0)	1.0 (1.0–1.1)	.042
Albumin	4.0 (3.5–4.3)	4.0 (3.4–4.3)	4.0 (3.5–4.3)	.294
Sodium	139 (137–141)	139 (137–141)	139 (137–140)	.114
Preceding stoma				.001
Yes	37 (8.8)	22 (16.1)	15 (5.3)	.001
No	80 (19.0)	21 (15.3)	59 (20.8)	.233
Unknown	304 (72.2)	94 (68.6)	210 (73.9)	.296

Continuous variables presented as median (interquartile range); categorical variables presented as frequency (percentage). P values <0.050 are in bold.

Lap, laparoscopic; ASA, American Society of Anesthesiologists; SIRS, systemic inflammatory response syndrome; IBD, inflammatory bowel disease; INR, international normalized ratio.

The overall incidence of any 30-day complication after an elective procedure was 17.4% (Table 2). On univariate analysis, there was no difference in the rate of complications between open and laparoscopic groups (19.0% versus 16.6%, P = .583). With respect to secondary outcomes, median postoperative LOS was 2 days shorter in the laparoscopic group, and this trend remained when adjusting for age, sex, and comorbidities by using the Cox proportional hazards model (Table 3). Median operative time was 30 minutes shorter in the open group (4.6 versus 5.1 hours, P = .013). On multivariate analysis, laparoscopic surgery was associated with decreased incidence of postoperative superficial SSI (RR 0.30; 95% CI: 0.10–0.88; P = .028) and pulmonary complications (RR 0.09; CI: 0.01–0.80, P = .031). Conversely, risk of organ space SSI was lower in the open group (RR 0.11, 95% CI: 0.01–0.79, P = .028). Risk of unplanned reoperation and risk of unplanned readmission were similar between laparoscopic and open groups. There was only one mortality, resulting in an overall 30-day mortality risk of 0.2%.

Table 2.

30-Day Outcomes After Elective Open Versus Laparoscopic Total Proctocolectomy with Ileoanal Anastomosis

30-day outcomes	Total (n = 421)	Open (n = 137)	Laparoscopic (n = 284)	P
Median operative time (hours)	5.0 (3.8–6.5)	4.6 (3.4–6.0)	5.1 (4.0–6.7)	.013
Median post-op LOS (days)	6 (5–9)	8 (6–10)	6 (4–8)	<.001
Any complication	74 (17.6)	26 (19.0)	48 (16.6)	.588
Mortality	1 (0.2)	0 (0)	1 (0.4)	1.000
Blood transfusion	17 (4.0)	6 (4.4)	11 (3.9)	.796
Pulmonary complication	6 (1.4)	5 (3.7)	1 (0.4)	.015
Pneumonia	2 (0.5)	2 (1.5)	0 (0)	.105
Unplanned reintubation	3 (0.7)	2 (1.5)	1 (0.4)	.248
Pulmonary embolism	1 (0.2)	1 (0.7)	0 (0)	.325
Venous thromboembolism	4 (1.3)	1 (0.9)	3 (1.5)	.687
Sepsis	15 (3.6)	4 (2.9)	11 (3.9)	.782
Urinary tract infection	13 (3.1)	4 (2.9)	9 (3.2)	1.000
Wound complication	39 (9.3)	13 (9.5)	26 (9.2)	1.000
SSI	37 (8.8)	12 (8.8)	25 (8.8)	1.000
Superficial SSI	14 (3.3)	9 (6.6)	5 (1.8)	.017
Deep SSI	4 (1.0)	2 (1.5)	2 (0.7)	.599
Organ space SSI	19 (4.5)	1 (0.7)	18 (6.3)	.010
Wound dehiscence	4 (1.0)	1 (0.7)	3 (1.1)	1.000
Unplanned reoperation	53 (12.6)	16 (11.7)	37 (13.0)	.756
Unplanned readmission	85 (20.2)	27 (19.7)	58 (20.4)	.898
Related to operation	70 (16.6)	23 (16.8)	47 (16.6)	1.000
Unrelated to operation	15 (3.6)	4 (2.9)	11 (3.9)	.782

Continuous variables presented as median (interquartile range); categorical variables presented as frequency (percentage). Variables with statistically significant difference between groups are displayed in bold.

LOS, length of stay; SSI, surgical site infection.

Table 3.

Factors Affecting Likelihood of Earlier Discharge After Elective Pediatric Total Proctocolectomy

Risk factor	Unadjusted HR (95% CI)	P	Adjusted HR (95% CI)	P
Laparoscopic approach	1.36 (1.10–1.67)	.004	1.46 (1.18–1.82)	.001
Sex (male)	0.92 (0.76–1.11)	.387	0.93 (0.76–1.11)	.469
Age	1.01 (1.10–1.67)	.499	1.01 (0.99–1.04)	.351
Race (quartile)
White	—	.968
Black	1.00 (0.73–1.37)	.999
Other/unknown	1.05 (0.73–1.50)	.797
Diagnosis		.922
Benign neoplasm	1.03 (0.64–1.63)	.917	0.80 (0.49–1.29)	.354
IBD	0.98 (0.62–1.55)	.927	0.89 (0.56–.43)	.631
Hirschsprung disease	1.12 (0.62–2.00)	.710	1.17 (0.61–2.24)	.638
Other	—
Steroid therapy	0.80 (0.63–1.03)	.078	0.79 (0.59–1.05)	.106
Nutritional deficiency	0.62 (0.39–0.97)	.038	0.75 (0.46–1.22)	.243
Pulmonary risk factors	1.07 (0.69–1.64)	.773
Cardiac risk factors	0.78 (0.38–1.57)	.478
Renal risk factors	1.14 (0.28–4.60)	.849
Neurocognitive risk factors	0.83 (0.52–1.34)	.456
Neuromuscular risk factors	1.71 (0.55–5.35)	.354
ASA class ≥3	0.98 (0.79–1.22)	.886
SIRS before surgery	0.95 (0.47–1.91)	.880
Inotropic support	0.90 (0.22–3.63)	.884
Hematologic risk factors	0.85 (0.60–1.19)	.342
Immune risk factors	0.62 (0.39–0.99)	.045	0.63 (0.39–1.02)	.062
Wound infection	0.48 (0.12–1.92)	.298

HR and P values derived by using the Cox proportional hazards model with discharge as the event of interest. Age, sex, diagnosis, and surgical approach included in the adjusted model a priori—other variables selected based on evidence of association with earlier discharge in the unadjusted model as demonstrated by P < .10.

CI, confidence interval; IBD, inflammatory bowel disease; HR, hazard ratios; ASA, American Society of Anesthesiologists.

On subgroup analysis, children with a diagnosis of IBD (n = 197) had an overall incidence of postoperative complications of 21.3%. There were no significant differences in individual or aggregate complications between laparoscopic and open groups (Table 4). Children with IBD in the laparoscopic group (n = 138) were more likely to be discharged earlier (hazard ratios 1.46, CI: 1.06–2.02, P = .20) with a median postoperative LOS of 6 days as compared with 8 days in the open group. Among children with FAP (n = 176), the rate of complications was 13.1%, and there was a lower incidence of superficial SSI in the laparoscopic group. Median LOS for those with FAP was shorter in the laparoscopic group, but the difference was not significant when adjusting for other factors by using the Cox proportional hazards model.

Table 4.

30-Day Outcomes After Open Versus Laparoscopic Elective Total Proctocolectomy with Ileoanal Anastomosis in Children with Familial Adenomatous Polyposis and Inflammatory Bowel Disease

Outcomes in children with FAP	Total (n = 176)	Open (n = 50)	Laparoscopic (n = 126)	P
Median operative time (hours)	5.5 (4.2–6.9)	5.1 (3.6–6.4)	5.6 (4.3–7.5)	.045
Median post-op LOS (days)	6 (5–8)	8 (6–9)	6 (4–8)	<.001
Any complication	23 (13.1)	7 (14.0)	16 (12.7)	.808
Mortality	0 (0)	0 (0)	0 (0)	—
Blood transfusion	2 (1.1)	1 (2.0)	1 (0.8)	.489
Pulmonary complication	0 (0)	0 (0)	0 (0)	—
Venous thromboembolism	1 (0.6)	0 (0)	1 (0.8)	1.000
Sepsis	3 (1.7)	0 (0)	3 (2.4)	.559
Urinary tract infection	5 (2.8)	1 (2.0)	4 (3.2)	1.000
SSI	14 (8.0)	5 (10.0)	9 (7.1)	.544
Superficial SSI	5 (2.8)	4 (8.0)	1 (0.8)	.023
Deep SSI	1 (0.6)	1 (2.0)	0 (0)	.284
Organ space SSI	8 (4.6)	0 (0)	8 (6.4)	.107
Unplanned reoperation	18 (10.2)	5 (10.0)	13 (10.3)	1.000
Unplanned readmission	28 (15.9)	6 (12.0)	22 (17.5)	.494
Related to operation	23 (13.1)	6 (12.0)	17 (13.5)	1.000

Outcomes in children with IBD	Total (n = 197)	Open (n = 59)	Laparoscopic (n = 138)	P
Median operative time (hours)	4.9 (3.1–6.1)	4.9 (3.9–6.0)	4.6 (3.4–6.3)	.518
Median post-op LOS (days)	6 (5–9)	8 (5–11)	6 (5–8)	.013
Any complication	42 (21.3)	12 (20.3)	30 (21.7)	1.000
Mortality	1 (0.5)	0 (0)	1 (0.7)	1.000
Blood transfusion	12 (6.1)	2 (3.4)	10 (7.3)	.516
Pulmonary complication	4 (2.0)	3 (5.1)	1 (0.7)	.081
Venous thromboembolism	4 (2.0)	1 (1.7)	3 (2.2)	1.000
Sepsis	9 (4.6)	2 (3.4)	7 (5.1)	.727
Urinary tract infection	8 (4.1)	3 (5.1)	5 (3.6)	.698
SSI	18 (9.1)	4 (6.8)	14 (10.1)	.593
Superficial SSI	6 (3.1)	2 (3.4)	4 (2.9)	1.000
Deep SSI	3 (1.5)	1 (1.7)	2 (1.5)	1.000
Organ space SSI	9 (4.6)	1 (1.7)	8 (5.8)	.284
Unplanned reoperation	30 (15.2)	9 (15.3)	21 (15.2)	1.000
Unplanned readmission	54 (27.4)	18 (30.5)	36 (26.1)	.601
Related to operation	44 (22.3)	14 (23.7)	30 (21.7)	.852

FAP, familial adenomatous polyposis; IBD, inflammatory bowel disease; LOS, length of stay; SSI, surgical site infection.

When considering the group of children who underwent nonelective surgery (n = 19), only 47.4% of cases were performed laparoscopically, as compared with 67.5% in the elective group (P = .083). As anticipated, the prevalence of preoperative comorbidities was higher among children undergoing nonelective surgery than among those undergoing elective surgery. Postoperative morbidity was likewise higher in the nonelective group, with an overall complication rate of 42.1% as compared with 17.4% in the elective cohort (P = .013). There were no appreciable differences in demographics or comorbidities between the laparoscopic and open group among nonelective cases, and with the exception of median operative time, which was significantly longer in the laparoscopic group (5.1 versus 2.2 hours, P = .003), outcomes were likewise similar.

Discussion

Among patients younger than 21 years in the pediatric and adult NSQIP databases who underwent elective TPC-IAA, this analysis found that the laparoscopic approach resulted in shorter postoperative LOS and decreased risk of superficial SSI when compared with the open approach. The open approach conferred a modest decrease in operative time and lower risk of organ space SSI. When considering nonelective cases, the only appreciable difference between laparoscopic and open techniques was increased operative time in the former, although the small number of nonelective cases in our sample substantially limited statistical analysis of this group, and larger studies are needed to determine the true impact of surgical approach in this population. This study is unique insofar as it is the first study to utilize data from both the pediatric and adult NSQIP databases, which resulted in the largest sample size to date among studies comparing laparoscopic and open techniques for elective TPC-IAA in the pediatric population. It is also remarkable insofar as greater than two-thirds of the proctocolectomies were performed laparoscopically, which is the highest proportion of any multi-institutional pediatric study on the subject.⁴ This would seem to indicate a trend, similar to that seen in the adult colorectal literature, of increased comfort and facility with the laparoscopic approach among pediatric surgeons.^1,3,10

Shorter postoperative LOS with the laparoscopic approach has been reported in both the pediatric and adult colorectal surgery literature and may be related to lower levels of postoperative pain and decreased time to oral feeding following the laparoscopic approach.^17,18 Although lower rates of wound complications in children undergoing laparoscopic TPC-IAA have been reported, this is the first study to show decreased rates of superficial SSI in this population.⁹ Moreover, this is the first study to demonstrate an increased risk of deep organ space infection with the laparoscopic approach to TPC-IAA in children. The reason for this increased risk is not readily apparent. Additional clinical data, such as use of preoperative mechanical bowel prep or a diverting ostomy, would have been an interesting adjunct to the analysis, but were unfortunately not available in the database. Given the relatively high rate of SSI among children undergoing TPC-IAA (8.8%), further study regarding perioperative optimization of these children is warranted.¹⁹

Two previous studies used data from NSQIP Pediatric to examine this topic. Dukleska and colleagues examined data from 2012 to 2015 in a study that included 260 children with FAP or ulcerative colitis.⁴ As in our study, they found that the laparoscopic approach was associated with an increase in operative time. This is consistent with other studies in both pediatric and adult populations.^10,20 They also found that preoperative steroid use and obesity were associated with reoperation. The risk of morbidity was higher in our study (17.4 versus 11.5%), which is, in part, due to a more liberal definition of morbidity in our study (we included postoperative transfusion of any volume in our definition whereas the aforementioned study only included transfusion greater than 25 mL/kg in the first 72 hours) and, in part, due to the inclusion of patients from the adult database, among whom risk for complication was 19.8%.

Mahida and colleagues studied 140 children from NSQIP Pediatric 2012–2013 who underwent proctocolectomy (103 laparoscopic, 37 open) for ulcerative colitis.⁹ The study included patients who did not undergo ileoanal anastomosis at the time of proctocolectomy, whereas our study excluded these patients. Similar to our findings, they found that postoperative LOS was shorter in the laparoscopic group (6 versus 8 days, P = .025), although they were unable to demonstrate this on multivariate analysis, likely due to a small sample size. The same study reported a lower risk of “any incisional complication” in the laparoscopic group (14% versus 24% P = .026) on univariate analysis, although once again, small sample size precluded multivariate analysis. Among the patients in our study with a diagnosis of IBD, there was not a significant difference in the rate of SSI between open and laparoscopic groups, although the incidence of organ space SSI was higher in the laparoscopic group (5.8% versus 1.7%, P = .284), a trend that achieved statistical significance when patients of all diagnoses were included in the analysis, as described earlier. The same trend was seen in patients with a diagnosis of FAP (Table 4).

The more frequent use of the open approach among children with a preceding stoma may be related to the possibility of performing the open procedure through the same incision required for stoma reversal. Among children with HD, for example (the diagnostic group with the highest incidence of preceding stoma), 80% of those with a preceding stoma had an open surgery, as compared with only 16.7% of HD children without a stoma and 50% of HD patients in whom stoma history was unknown (P = .054). Unfortunately, a definitive explanation for these findings remains elusive given the absence of operative reports and inconsistency with which preceding stoma is documented in the database.

Our study has several limitations in addition to those mentioned earlier, some of which are inherent to the use of a multi-institutional database. The generalizability of our study is limited insofar as the data are derived from hospitals participating in NSQIP, which may not be representative of all hospitals that perform pediatric colorectal surgery. The significant proportion of pediatric colorectal procedures that are performed by using an open approach may suggest that not all centers or surgeons are comfortable about utilizing advanced laparoscopic reconstructive techniques in children. It is likewise important to note that our study excludes patients undergoing subtotal colectomy. We elected to include only patients undergoing ileoanal reconstruction, for the reason that ileoanal anastomosis introduces a unique set of challenges that are distinct from those encountered with subtotal colectomy.

An additional limitation of our study is the inability to examine clinical data that are not included in NSQIP. Such data include operative details such as bowel prep, method of anastomosis, type of ileal reservoir, whether or not a diverting ileostomy was performed, and whether or not the procedure was performed in a staged fashion. The impact of each of these factors on outcomes after pediatric proctocolectomy warrants further investigation. The database also lacks data on postoperative outcomes such as time to weaning from parenteral pain medicine, time to oral feeding, or risk for small bowel obstruction (SBO). In their series of seven children undergoing TPC-IAA, Flores and colleagues found that children in the laparoscopic group (n = 3) tolerated oral feeding earlier and were weaned from opiate analgesics sooner than those in the open group, but their small sample size limits the generalizability of their conclusions.⁵ Linden and colleagues followed 117 children with ulcerative colitis who underwent total proctocolectomy and ileal pouch anal anastomosis at their institution. Compared with children in the open group (n = 39), those in the laparoscopic-assisted group had lower risk for SBO at the time of 1-year follow-up.⁶ Although our analysis does include risk of 30-day reoperation and readmission, which would potentially detect early SBO requiring either reoperation or readmission, it does not provide information regarding SBO occurring beyond 30 days from operation. Likewise, although our analysis does not include specific data on time to oral feeding or postoperative pain control, postoperative LOS may be considered a crude surrogate for these insofar as pain control and oral feeding are often the limiting factors for hospital discharge.

In conclusion, our study of 421 elective cases of TPC-IAA in patients younger than 21 years found that 67.5% were performed laparoscopically, the highest reported in a study of this kind, suggesting a trend toward greater comfort with advanced laparoscopic techniques among pediatric surgeons. We also found that the laparoscopic approach is associated with shorter postoperative LOS and decreased risk of superficial SSI as compared with the open approach. The open approach conferred lower risk of organ space SSI and was associated with shorter operative time. Future studies should consider methods for increasing the efficiency and safety of minimally invasive colorectal surgery in pediatric patients, with specific attention to strategies for preventing SSI.

Footnotes

Acknowledgments

The American College of Surgeons National Surgical Quality Improvement Program and the hospitals participating in the ACS NSQIP are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.

Disclosure Statement

No competing financial interests exist.

Appendix 1.

Risk Factor and Comorbidity Definitions

Risk factor/comorbidity	Definition
Pulmonary risk factor	Asthma, cystic fibrosis, chronic lung disease, chronic obstructive pulmonary disease, dyspnea with moderate exertion, structural pulmonary abnormality, oxygen dependence, active treatment for pneumonia, tracheostomy at time of surgery, and need for ventilator-assisted respiration at any time during the 48 hours preceding surgery
Cardiac risk factor	Pre-existing cardiac conditions, compromise of cardiac function requiring medication, history of myocardial infarction or angina, history of percutaneous coronary intervention or cardiac surgery, or hypertension requiring mediation.
Nutritional deficiency	Dependence on intravenous total parenteral nutrition, enteral feeding support, or >10% weight loss in the 6 months before surgery
Neurocognitive risk factor	History of central nervous system (CNS) tumor, acquired CNS abnormality, cerebral palsy, impaired cognitive status, or seizure disorder
Neuromuscular disorder	History of congenital or acquired degenerative neuromuscular disorder, quadriplegia or paraplegia
Hematologic risk factor	Bleeding disorder or congenital hematologic disorder
Renal risk factor	Acute rise in creatinine above 2.0 mg/dL in the 7 days before surgery or renal failure requiring dialysis within 2 weeks before surgery
Steroid therapy	Administration of corticosteroid in the 30 days before surgery

References

Davis

, Shirkey

, Moore

, Gaglani

, Du

, Bailey

, Cusick

. Trends in laparoscopic colorectal surgery over time from 2005–2014 using the NSQIP database. J Surg Res, 2018; 223:16–21.

Cheng

, Rezac

. The role of robotics in colorectal surgery. BMJ, 2018; 360:j5304.

Yeo

, Isaacs

, Abelson

, Milsom

, Sedrakyan

. Comparison of open, laparoscopic, and robotic colectomies using a large national database: Outcomes and trends related to surgery center volume. Dis Colon Rectum, 2016; 59:535–542.

Dukleska

, Berman

, Aka

, Vinocur

, Teeple

. Short-term outcomes in children undergoing restorative proctocolectomy with ileal-pouch anal anastomosis. J Pediatr Surg, 2018; 53:1154–1159.

Flores

, Bailez

, Cuenca

, Fraire

. Comparative analysis between laparoscopic (UCL) and open (UCO) technique for the treatment of ulcerative colitis in pediatric patients. Pediatr Surg Int, 2010; 26:907–911.

Linden

, Bairdain

, Zurakowski

, Shamberger

, Lillehei

. Comparison of laparoscopic-assisted and open total proctocolectomy and ileal pouch anal anastomosis in children and adolescents. J Pediatr Surg, 2013; 48:1546–1550.

Tan

, Jaffray

. A comparison of open and laparoscopic restorative proctocolectomy in children. Pediatr Surg Int, 2009; 25:877–879.

Sheth

, Jaffray

. A comparison of laparoscopic and open restorative proctocolectomy in children. J Pediatr Surg, 2014; 49:262–264; discussion 264.

Mahida

, Asti

, Deans

, Minneci

, Nwomeh

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

10.

Konishi

, Ishida

, Ueno

, Kobayashi

, Hinoi

, Inoue

, Ishida

, Kanemitsu

, Yamaguchi

, Tomita

, et al. Feasibility of laparoscopic total proctocolectomy with ileal pouch-anal anastomosis and total colectomy with ileorectal anastomosis for familial adenomatous polyposis: Results of a nationwide multicenter study. Int J Clin Oncol, 2016; 21:953–961.

11.

Cox

. Regression models and life-tables. J Royal Stat Soc Series B-Stat Methodol, 1972; 34:187–220.

12.

McKenna

, Habermann

, Glasgow

, Mathis

, Lightner

. Risk factors for readmission following ileal pouch-anal anastomosis: An american college of surgeons national surgical quality improvement program analysis. J Surg Res, 2018; 229:324–331.

13.

Noh

, Han

, Cho

, Hur

, Min

, Lee

, Kim

. Factors affecting pouch-related outcomes after restorative proctocolectomy. PLoS One, 2017; 12:e0186596.

14.

Ozdemir

, Kiran

, Erem

, Aytac

, Gorgun

, Magnuson

, Remzi

. Functional outcomes and complications after restorative proctocolectomy and ileal pouch anal anastomosis in the pediatric population. J Am Coll Surg, 2014; 218:328–335.

15.

Rangel

, Islam

, St Peter

, Goldin

, Abdullah

, Downard

, Saito

, Blakely

, Puligandla

, Dasgupta

, et al. Prevention of infectious complications after elective colorectal surgery in children: An american pediatric surgical association outcomes and clinical trials committee comprehensive review. J Pediatr Surg, 2015; 50:192–200.

16.

Zhang

, Yu

. What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA, 1998; 280:1690–1691.

17.

Page

, Sashittal

, Chatzizacharias

, Davies

. The role of laparoscopic surgery in the management of children and adolescents with inflammatory bowel disease. Medicine (Baltimore), 2015; 94:e874.

18.

Moghadamyeghaneh

, Hwang

, Hanna

, Phelan

, Carmichael

, Mills

, Pigazzi

, Stamos

. Risk factors for prolonged ileus following colon surgery. Surg Endosc, 2016; 30:603–609.

19.

Feng

, Sidhwa

, Cameron

, Glass

, Rangel

. Rates and burden of surgical site infections associated with pediatric colorectal surgery: Insight from the national surgery quality improvement program. J Pediatr Surg, 2016; 51:970–974.

20.

Ahmed Ali

, Keus

, Heikens

, Bemelman

, Berdah

, Gooszen

, van Laarhoven

. Open versus laparoscopic (assisted) ileo pouch anal anastomosis for ulcerative colitis and familial adenomatous polyposis. Cochrane Database Syst Rev, 2009; 21:CD006267.