CROHNS-01018; No of Pages 10 Journal of Crohn's and Colitis (2014) xx, xxx–xxx

Available online at www.sciencedirect.com

ScienceDirect

Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States☆ Ing Shian Soon a , Jennifer C.C. deBruyn b , James Hubbard a , Iwona Wrobel b , Reg Sauve a,b , David L. Sigalet c , Gilaad G. Kaplan a,d,⁎ a

Department of Community Health Sciences, University of Calgary, Calgary, Canada Department of Paediatrics, University of Calgary, Calgary, Canada c Department of Surgery, University of Calgary, Calgary, Canada d Department of Medicine, University of Calgary, Calgary, Canada b

Received 6 February 2014; received in revised form 19 April 2014; accepted 11 May 2014 KEYWORDS Ulcerative colitis; Colectomy; Children; Postoperative complications; Kids' Inpatient Database

Abstract Background and aims: In children with ulcerative colitis, data on temporal colectomy trends and in-hospital post-colectomy complications are limited. Thus, we evaluated time trends in colectomy rates and post-colectomy complications in children with ulcerative colitis. Methods: We identified all children (≤ 18 years) with a diagnosis code of ulcerative colitis (ICD-9: 556.X) and a procedure code of colectomy (ICD-9: 45.8 and 45.7) in the Kids' Inpatient Database for 1997, 2000, 2003, 2006 and 2009. The incidence of colectomies for pediatric ulcerative colitis was calculated and Poisson regression analysis was performed to evaluate the change in colectomy rates. In-hospital postoperative complication rates were assessed and predictors for postoperative complications were evaluated using multivariate logistic regression. Results: The annual colectomy rate in pediatric ulcerative colitis was 0.43 per 100,000 person-years, which was stable throughout the study period (P N .05). Postoperative complications were experienced in 25%, with gastrointestinal (13%) and infectious (9.3%) being the most common. Postoperative complication rates increased significantly by an annual rate of 1.1% from 1997 to 2009 (P = .01). However, other independent predictors of postoperative complications were not identified.

Abbreviations: CI, confidence interval; HCUP, Healthcare Cost and Utilization Project; ICD-9-CM, International Classification of Diseases, 9th Revision, Clinical Modification; IRR, incidence rate ratio; KID, Kids' Inpatient Database; NIS, Nationwide Inpatient Sample; OR, odds ratio; Q1, first quartile; Q3, third quartile; TNF, tumor necrosis factor; UC, ulcerative colitis; US, United States. ☆ Conference presentation: Part of this study was presented at the North American Society of Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) annual meeting in Orlando, United States, in October of 2011. ⁎ Corresponding author at: Departments of Medicine and Community Health Sciences, University of Calgary, Teaching Research and Wellness Center, 3280 Hospital Drive NW, 6D17, Calgary, AB T2N 4N1, Canada. Tel.: + 1 403 592 5025; fax: + 1 403 592 5050. E-mail address: [email protected] (G.G. Kaplan).

http://dx.doi.org/10.1016/j.crohns.2014.05.002 1873-9946/© 2014 Published by Elsevier B.V. on behalf of European Crohn's and Colitis Organisation. Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

2

I.S. Soon et al. Patients with postoperative complications had significantly longer median length of stay (14.3 days vs 8.2 days; P b .001) and higher median hospital charges per patient (US $81,567 vs US $55,461; P b .001) compared to those without complications. Conclusion: Colectomy rates across the United States in children with ulcerative colitis have remained stable between 1997 and 2009; however, in-hospital postoperative complication rates have increased. © 2014 Published by Elsevier B.V. on behalf of European Crohn's and Colitis Organisation.

1. Introduction Ulcerative colitis (UC) is a chronic inflammatory condition of the colon that imparts a significant burden to patients and society.1 A meta-analysis of population-based studies has shown that the 10-year risk of colectomy in children with UC is 22%.2 Population-based time trend analyses of colectomy rates have shown that the 10-year risk of colectomy has decreased over the past several decades in adults.2 However, similar studies in children are lacking. A small populationbased study of 30 children with UC who underwent a colectomy over a 27 year period reported a stable colectomy rate of 0.06 per 100,000 person-years.3 In contrast, a study conducted in the same health region demonstrated that colectomy rates in adults significantly decreased by 4% per year over a 15-year period.4 The difference in colectomy rates over time between children and adults may be due to earlier adoption and greater utilization of immunomodulators and/or biologics in adults with UC as compared to children. Alternatively, non-significant time trend results may be attributable to the smaller sample size of pediatric studies. In children with UC, the reported incidence of in-hospital or early postoperative complications ranges between 30 and 55%,3,5–7 with infections (e.g., abscess formation) and gastrointestinal complications (e.g., bowel obstruction) being the most common.3,5,6,8 In contrast to the adult literature, pediatric studies have not reported mortality following colectomy for UC.5,6,9,10 One population-based study reported higher postoperative complication rates in children with emergency colectomy (90% vs 50%) and delayed colectomy of N 2 weeks after admission (60% vs 0%).3 However, this study was restricted to one geographic area and limited by the small sample size. Hence, the objective of the present study was to evaluate the nationwide colectomy rates and in-hospital complications following colectomy in children with UC in the United States (US), using the Kids' Inpatient Database (KID) from 1997 to 2009.

2. Materials and methods 2.1. Data source The KID, part of the Healthcare Cost and Utilization Project (HCUP), is the largest all-payer pediatric inpatient database with 2 to 3 million pediatric discharges per year. Stratified random sampling is done to ensure that the database is representative of the US population and accounts for approximately 90% of all pediatric hospitalizations in US.17

The database contains information on demographic characteristics, admission type (emergent, urgent or elective), up to 25 diagnostic and 15 procedure codes based on the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM), outcomes (length of stay and hospital charges), and hospital characteristics.11 KID assigns individual-level discharge weights that allow estimation of national case rates and trends. It has been used to study hospitalization and surgical intestinal resection in pediatric IBD.12,13

2.2. Study population All discharges for children (≤ 18 years) with a diagnosis of UC (ICD-9-CM: 556.X) and a procedure code of colectomy (ICD-9-CM: 45.8 & 45.7) were identified in the HCUP KID for 1997, 2000, 2003, 2006 and 2009. Validation studies have shown that administrative data correctly identified UC patients who underwent colectomy in 96% of children3 and 86% of adults. To minimize misclassification with Crohn's disease, we excluded 25 discharges of patients who were coded with both UC and Crohn's disease (ICD-9-CM: 555.X).

2.3. Variables Our primary factors of interest were the admission type and the number of preoperative days between admission and colectomy among emergently admitted patients. Admission type was categorized into emergent and urgent/elective as this was previously shown in a validation study to increase the specificity of identifying patients admitted for emergent colectomy.14 An emergent admission refers to unplanned admissions to hospital for a serious medical condition. In contrast, elective and urgent admissions are both planned admissions with urgent admission referring to a medical condition that requires immediate attention for treatment. Covariates included patients' age; sex; race/ethnicity (white or non-white); patients' residence (urban or rural); admission source (non-hospital or hospital transfer); health care insurance status (Private, Medicaid/Medicare, or other); hospital region (Northwest, Midwest, South or West); hospital location (urban or rural); hospital type [teaching or nonteaching; pediatric or non-pediatric (pediatric hospital is defined as either a children's hospital or a general hospital with a children's unit)]; hospital size (small, medium or large); and creation of pouch at time of colectomy. A year variable was included to indicate the year the data was collected to test for trends over time.

Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

Colectomy in children with ulcerative colitis

2.4. Outcomes The primary outcomes were colectomy rates for pediatric UC and the in-hospital postoperative complications following colectomy. Postoperative complications were defined as unexpected medical events that occurred between the time of colectomy and discharge from the hospital. Postoperative complication was defined as one of the following: wound complication, infection, urinary, pulmonary, gastrointestinal, cardiovascular, complication during the surgical procedure, reopening of a recent laparotomy site, and death. The specific postoperative complications that compose each category and their corresponding ICD-9-CM codes are found in Appendix A. The postoperative complication codes have been previously validated.14

2.5. Statistical analyses 2.5.1. Pediatric UC colectomy rates The annual colectomy rates in pediatric UC were calculated by dividing the annual total number of colectomies in children ≤ 18 years by the annual US population ≤ 18 years obtained from US Bureau of the Census.15 The overall age and sex standardized annual colectomy rate was reported using the 2000 US Census population. The average annual percentage change in the colectomy rate from 1997 to 2009 was estimated using a generalized log-linear model that assumed a Poisson distribution. Poisson regression analysis was performed to evaluate the change in colectomy rates in pediatric UC. Age and sex were assessed as potential predictors of colectomy rates, and adjusted incidence rate ratios (IRRs) with 95% confidence interval (CI) were reported.

3 with an urgent admission was performed to compare the postoperative complications among patients with emergent and elective admissions. The primary outcome was in-hospital postoperative complications; whereas secondary outcomes of interest included the length of hospital stay and hospital charges per patient. Hospital charges per patient in US dollars were inflation adjusted to 2010 US dollars, using the US Consumer Price Index for Medical Care.16 2.5.3. Missing values Variables with missing data values N 10% of the total cohort included race/ethnicity, admission type, duration between admission to colectomy and admission source. Postoperative complication rates among patients with and without missing values were compared to ensure that patients with and without missing values were similar with respect to their outcomes. The variable with N 20% missing data values, i.e. race/ethnicity, was not included in our primary multivariate analysis model, but was assessed in a subgroup analysis including only patients with information on race/ethnicity. All statistical analyses were conducted using STATA version 11.0 (StataCorp LP, College Station, TX) to adjust for the complex sampling design of the KID.17 P values less than .05 were considered to be statistically significant. All reported estimates were national estimates calculated using the statistical discharge weights unless otherwise noted. Data for cells ≤ 10 were presented as “≤ 10” as HCUP Data Use Agreement prohibits the presentation of data for cells ≤ 10. This research was approved by the University of Calgary's Conjoint Health Research Ethics Board.

3. Results 2.5.2. In-hospital postoperative complications The proportion of patients with in-hospital postoperative complications was calculated. Logistic regression was performed to evaluate the change in postoperative complication rates over time. Association between study variables and postoperative complications following colectomy was assessed using the Pearson chi-square test to compare proportions and the Wilcoxon rank sum test for continuous variables, expressed as medians with first quartile (Q1) and third quartile (Q3). Multivariate logistic regression was performed to assess predictors of postoperative complications. Variables with P value ≤ .10 in the univariate analysis and preselected variables including age, sex, and admission type were included in the model, as emergency colectomy was found to be a strong predictor of in-hospital postoperative complications in a previous study.3 A subgroup analysis was performed on the cohort of UC patients who had an emergency admission (n = 308). In this cohort, the number of days from admission to colectomy was assessed as a potential predictor of postoperative complications. This variable was a priori selected because a previous study identified time from admission to colectomy as a predictor of postoperative complications.3 In a post-hoc sensitivity analysis, emergently admitted patients who had colectomy within the first two days of admission were excluded in an attempt to exclude those with fulminant colitis or complications such as toxic megacolon and bowel perforation. A sensitive analysis excluding children with UC

During the five study years, 1997, 2000, 2003, 2006 and 2009, a nationwide weighted estimate of 328 children with UC underwent colectomy in the US every year, giving a total of 1641 children during the five-year period. The baseline characteristics of the cohort are presented in Table 1. The majority of patients were white (81%), lived in an urban area (82%), and had private health insurance (78%). The median age at colectomy was 14.3 years (Q1, Q3: 11.2 years, 16.2 years). The median length of stay was 9.4 days (Q1, Q3: 6.1 days, 17.3 days) and the median hospital charges were US $59,117 per patient (Q1, Q3: US $39,323, US $98,344).

3.1. Pediatric UC colectomy rates The overall annual colectomy rate for pediatric UC in the US was 0.43 per 100,000 person-years (95% CI: 0.38–0.47) from 1997 to 2009, and it ranged from 0.34 (95% CI: 0.26–0.42) in 2006 to 0.52 (95% CI: 0.42–0.62) in 2009 (Fig. 1). The ageand sex-standardized annual colectomy rate for pediatric UC was 0.43 per 100,000 (95% CI: 0.40–0.45). After adjusting for age and sex, the UC colectomy rate remained relatively stable from 1997 to 2009 with an adjusted IRR of 1.01 (95% CI: 0.99–1.03; P = .25). The adjusted IRR for males compared to females was 0.91 (95% CI: 0.76–1.09) and the adjusted IRRs for age groups 5–9, 10–14 and 15–18 as compared to age group 0–4 were 2.51 (95% CI: 1.47–4.28),

Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

4

I.S. Soon et al.

7.88 (95% CI: 4.88–12.7) and 15.8 (95% CI: 9.85–25.5), respectively.

3.2. In-hospital postoperative complications In-hospital postoperative complications were experienced in 416 children (25%). The overall in-hospital postoperative complication rate was 25.3 (95% CI: 22.2–28.7) per 100

Table 1

colectomies, ranging from 16.9 (95% CI: 11.1–24.9) in 1997 to 31.0 (95% CI: 25.7–36.9) in 2009 (Fig. 2). The average annual postoperative complication rates increased significantly by 1.1% (P = .01) from 1997 to 2009. An in-hospital mortality rate of ≤0.6% was reported in this cohort. The exact number cannot be reported as HCUP Data Use Agreement prohibits the presentation of data for cells ≤10. A total of 613 events were coded in 416 children with postoperative complications (Table 2). The most frequent

Characteristics of children with ulcerative colitis who underwent colectomy stratified by postoperative complications.

Age at colectomy (years) Median (Q1, Q3) Age group (%, 95% CI) 0–4 years 5–9 years 10–14 years 15–18 years Sex (%, 95% CI) Female Male Race/ethnicity (%, 95% CI) White Non–white Patients' residence (%, 95% CI) c Urban Rural Admission type (%, 95% CI) Emergency Urgent/elective Admission source (%, 95% CI) Non-hospital Hospital transfer Health insurance (%, 95% CI) Private Medicaid/Medicare Other Hospital region (%, 95% CI) Northeast Midwest South West Hospital location (%, 95% CI) Urban Rural Hospital type (%, 95% CI) Teaching Non-teaching Pediatric Non-pediatric Hospital size (%, 95% CI) Small Medium Large

Total cohort n = 1641 a

Postoperative complications n = 416

No postoperative complications n = 1225

P value

14.3 (11.2, 16.2)

14.5 (11.6, 16.5)

14.2 (11.1, 16.1)

0.29 b

4.1 (3.0–5.6) 10.3 (8.3–12.8) 32.7 (29.4–36.2) 52.9 (49.1–56.7)

4.4 (2.5–7.7) 8.9 (5.9–13.2) 32.4 (25.9–39.7) 54.3 (47.2–61.2)

4.0 (2.7–5.9) 10.8 (8.3–13.8) 32.8 (29.2–36.6) 52.4 (48.2–56.6)

0.85

51.1 (47.7–84.4) 48.9 (45.5–52.3)

46.0 (39.5–52.7) 54.0 (47.3–60.5)

52.8 (48.9–56.8) 47.2 (43.2–51.1)

0.09

81.2 (77.4–84.4) 18.8 (15.6–22.6)

72.9 (65.7–79.1) 27.1 (20.9–34.3)

84.1 (79.9–87.6) 15.9 (12.4–20.1)

0.003

81.8 (78.0–85.0) 18.2 (15.0–22.0)

79.3 (72.5–84.7) 20.7 (15.3–27.5)

82.8 (78.4–86.4) 17.2 (13.6–21.6)

0.57

21.4 (18.2–24.9) 78.6 (75.1–81.8)

25.7 (20.0–32.4) 74.3 (67.6–80.0)

19.9 (16.5–23.7) 80.1 (76.3–83.5)

0.08

89.9 (87.0–92.2) 10.1 (7.8–13.0)

88.8 (83.2–92.7) 11.2 (7.3–16.8)

90.3 (86.8–92.9) 9.7 (7.1–13.2)

0.60

77.7 (74.4–80.6) 17.3 (14.7–20.2) 5.0 (3.4–7.3)

72.7 (66.3–78.4) 22.3 (17.2–28.4) 5.0 (2.8–8.9)

79.4 (75.7–82.6) 15.6 (12.8–18.9) 5.0 (3.1–7.9)

0.11

21.0 28.2 26.4 24.4

(16.6–26.1) (22.7–34.4) (22.0–31.4) (20.1–29.3)

21.1 (15.3–28.4) 26.2 (19.6–34.1) 26.9 (20.7–34.2) 25.8 (19.7–33.0)

20.9 28.8 26.3 24.0

(16.1–26.7) (22.6–36.0) (21.4–31.8) (19.3–29.3)

0.92

98.1 (95.6–99.2) 1.9 (0.8–4.4)

97.6 (94.2–99.1) ≤ 2.4 d

98.2 (95.1–99.4) 1.8 (0.6–4.9)

0.64

87.2 12.8 73.7 26.3

(82.8–90.6) (9.4–17.2) (69.1–77.9) (22.1–30.9)

86.0 (79.5–90.7) 14.0 (9.3–20.5) 70.9 (63.4–77.4) 29.1 (22.6–36.6)

87.5 12.5 74.7 25.3

0.62

12.9 (9.8–16.9) 25.1 (20.2–30.8) 61.9 (56.5–67.1)

12.1 (8.0–17.7) 23.2 (17.3–30.4) 64.7 (57.2–71.6)

13.2 (9.8–17.6) 25.8 (20.4–32.0) 61.0 (54.9–66.8)

(82.5–91.3) (8.7–17.5) (69.5–79.2) (20.8–30.5)

0.34

0.65

Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

Colectomy in children with ulcerative colitis

5

Table 1 (continued)

Creation of pouch (%, 95% CI) Yes No Length of stay (days) Median (Q1, Q3) Hospital charges per patient (US $) Median (Q1, Q3) Duration between admission and colectomy (days) e Median (Q1, Q3) Duration between admission and colectomy (%, 95% CI) e ≤ 1 week N 1 week

Total cohort n = 1641 a

Postoperative complications n = 416

No postoperative complications n = 1225

P value

30.7 (26.3–35.5) 69.3 (64.5–73.7)

28.5 (22.4–35.5) 71.5 (64.5–77.6)

31.5 (26.4–37.1) 68.5 (62.9–73.6)

0.46

9.4 (6.1, 17.3)

14.3 (8.6, 26.1)

8.2 (5.6, 14.7)

b 0.001 b

59,117 (39,323, 98,344)

81,567 (48,167, 165,297)

55,461 (36,474, 84,603)

b 0.001 b

8.4 (1.8, 14.7)

4.6 (0.6, 12.3)

9.4 (2.8, 15.4)

0.02

44.0 (35.5–52.9) 56.0 (47.1–64.5)

59.4 (44.5–72.8) 40.6 (27.2–55.5)

37.5 (27.9–48.1) 62.5 (51.9–72.1)

0.02

CI, confidence interval; Q1, first quartile; Q3, third quartile; US, United States. a Number of missing data: race/ethnicity (n = 395), patients' residence (n = 23), admission type (n = 201), admission source (n = 318), health insurance (n = 7), hospital location (n = 59), hospital type: teaching (n = 59) and pediatric (n = 87), hospital size (n = 59), and hospital charges per patient (n = 50). b Analyses performed using actual count instead of statistical weights. c Data only available for year 2003, 2006 and 2009. d HCUP Data Use Agreement prohibits the presentation of data for cells ≤ 10. e Analyses restricted to patients who had emergency admission.

postoperative complications were gastrointestinal related [n = 211 (13%)] and infections [n = 152 (9.3%)]. Patients with postoperative complications had significantly longer median length of stay (14.3 days vs 8.2 days; P b .001) and higher median hospital charges per patient (US $81,567 vs US $55,461; P b .001) (Table 1). In the univariate analysis, age, sex, patients' residence, admission type, admission source, health care insurance status, and the hospital region, location, type and size were not predictors of postoperative complications following colectomy for pediatric UC. For variables with missing data values N 10% of the total cohort, postoperative complication rates were similar among patients with or without missing values. After adjusting for age, sex and admission type, year of data collection remained a significant predictor of postoperative complications following colectomy, with an adjusted OR of 1.05 (95% CI: 1.01–1.09; P = .04) (Table 3). In a subgroup analysis including only patients with information on race/ethnicity, race/ethnicity was not a significant predictor after adjusting for year, age, sex, and admission type [adjusted OR (non-white vs white) of 1.56; 95% CI: 0.95–2.55]. Among emergently admitted patients, the overall inhospital postoperative complication rate was 30.4 (95% CI: 24.0–37.7) per 100 colectomies. In this cohort, the time to colectomy following admission was an independent predictor of postoperative complications after adjusting for year, age and sex [adjusted OR (≤ 1 week vs N 1 week) of 2.44; 95% CI: 1.20–5.00] (Table 3). In a post-hoc sensitivity analysis excluding emergently admitted patients who had colectomy within the first two days of their admission time from admission to colectomy was no longer associated with

postoperative complications (adjusted OR 2.22; 95% CI: 0.85–5.88; P = 0.10). In the sensitive analysis excluding UC patients with urgent admission, the admission type was not a significant predictor of postoperative complications after adjusting for year, age and sex (adjusted OR of 1.39; 95% CI: 0.95–2.03) (Table 3).

4. Discussion This is the largest population-based study to assess time trends in colectomy and in-hospital postoperative complication rates in children with UC. The annual incidence of colectomy in pediatric UC in the US has remained relatively stable over the past decade suggesting that advances in medical management including the increased utilization of immunomodulators and biologics have not resulted in a corresponding decrease in the colectomy rate in children with UC. This is consistent with the stable rates reported in a pediatric study conducted in Canada.3 Several studies have shown that the risk of colectomy in children with UC is higher than compared to adults with UC.18–20 At diagnosis children with UC often have greater disease extent (i.e. higher proportion of pancolitis at diagnosis) and disease severity when compared to adults diagnosed with UC,21 which may contribute to a higher colectomy rate. Alternatively, children with UC may have had reduced access to anti-TNF therapy because infliximab approval in children with UC only occurred in 2011. A previous study demonstrated that infliximab utilization in adults for UC increased considerably after 2006 and that the rate of colectomy from 2006 to 2009 was nearly half the rate observed from 1997 to 2005.4 In

Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

6

I.S. Soon et al. Table 2 Postoperative complications following colectomy in pediatric ulcerative colitis. Postoperative complications (n = 613)

Figure 1 Colectomy rates (with 95% confidence interval) of pediatric ulcerative colitis in United States, 1997 to 2009.

the Active Ulcerative Colitis Trials (ACT 1 & 2) of adults with moderate to severe active UC, infliximab treatment was associated with a 7% absolute risk reduction of colectomy at one year.22 However, a recent study in adults with UC showed that anti-TNF therapy was not associated with a decreased long-term risk of colectomy.23 The effect of infliximab in reducing colectomy rates in pediatric UC patients should be evaluated in future studies. The most common postoperative complications were gastrointestinal-related and infectious, which were similar to previously reported complications following colectomy in children with UC.7,8 Postoperative complications were associated with a significant burden to the healthcare system as evidence by longer length of stay and hospital charges per patient compared to those without postoperative complications. Postoperative mortality in this cohort was rare (≤ 0.6%) and considerably lower than the mortality rate (2.3%) reported in US adults using the Nationwide

Figure 2 In-hospital post-colectomy complication rates (with 95% confidence interval) of pediatric ulcerative colitis in United States, 1997 to 2009.

Mechanical wound Postoperative hematoma Dehiscence of operative wound Persistent postoperative fistula Infectious Postoperative infection (skin, wound, intra-abdominal abscess) Bacteremia or sepsis Urinary (urinary retention, urinary tract infection, acute renal failure) Pulmonary Postoperative atelectasis and pneumonia Acute respiratory failure Postoperative pulmonary edema Gastrointestinal (small bowel obstruction, nausea, vomiting, pancreatitis, ileus, pancreatitis, hepatic failure) Cardiovascular Deep vein thrombosis Cardiac arrest or heart failure Acute myocardial infarction Postoperative shock Cardiac arrest Complication during surgical procedure Accidental puncture or laceration Foreign body accidentally left during procedure Hemorrhage complicating procedure Reopening of a recent laparotomy site Death

Total number of patients N = 1641 (%) 14 (0.9%) 17 (1.0%) ≤ 10 (≤ 0.6%) a 152 (9.3%) 80 (4.9%) 86 (5.2%) 19 (1.2%)

20 (1.2%) 42 (2.6%) ≤ 10 (≤ 0.6%) a 211 (13%)

14 (0.9%) ≤ 10 (≤ 0.6%) a ≤ 10 (≤ 0.6%) a ≤ 10 (≤ 0.6%) a ≤ 10 (≤ 0.6%) a 21 (1.3%) ≤ 10 (≤ 0.6%) a 25 (1.5%) 30 (1.8%) ≤ 10 (≤ 0.6%) a

a

HCUP Data Use Agreement prohibits the presentation of data as percentages for cells ≤ 10 (≤ 0.6%).

Inpatient Sample (NIS) database.24 The lower mortality rate in children is likely due to the age difference and the absence of comorbidities that were predictors of in-hospital mortality in adults.24,25 In our cohort, the postoperative complication rates have significantly increased over the past decade. This observation is important because several studies have explored whether the increased utilization of salvage agents (e.g. cyclosporine and infliximab) are associated with increased risk of postoperative complications. Preoperative cyclosporine use was not associated with an increased post-colectomy complication in both children and adults with UC.26,27 A recent study in children with UC reported a higher postoperative complication rate in patients who received infliximab prior to ileal pouch anal anastomosis.28 In contrast, preoperative infliximab use was not associated with increased risk of short-term postoperative complications in a meta-analysis of primarily adult UC patients.29 Our cohort was limited by the lack of medications in the KID database and thus, the etiology of complication rates should

Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

Colectomy in children with ulcerative colitis Table 3

7

Predictors of postoperative complications following colectomy in pediatric ulcerative colitis. Total cohort

Emergency and elective only

Adjusted OR (95% CI) Year of colectomy Age group 0–4 years 5–9 years 10–14 years 15–18 years Sex Female Male Admission type Urgent/elective c Emergency Duration between admission and colectomy N 1 week ≤ 1 week

a

Adjusted OR (95% CI)

a

Emergency admission only Adjusted OR (95% CI) b

1.05 (1.01–1.09)

1.04 (0.99–1.09)

1.05 (0.96–1.13)

Reference 0.77 (0.28–2.08) 1.08 (0.47–2.50) 1.05 (0.47–2.37)

Reference 0.82 (0.26–2.65) 1.05 (0.38–2.90) 1.11 (0.41–3.02)

Reference 0.16 (0.02–1.61) 0.26 (0.04–1.55) 0.26 (0.04–1.58)

Reference 1.29 (0.92–1.80)

Reference 1.46 (1.00–2.13)

Reference 1.91 (0.86–4.25)

Reference 1.40 (0.98–2.00)

Reference 1.39 (0.95–2.03)

NA

NA

NA

Reference 2.44 (1.20–5.00)

CI, confidence interval; OR, odds ratio; NA, not applicable. a Multivariable logistic regression adjusted for year, age, sex, and admission type. b Multivariable logistic regression adjusted for year, age, sex, and duration between admission and colectomy. c For the total cohort emergency admissions were compared to urgent or elective admissions. For the emergency and elective only cohort (i.e. urgent admissions not included) emergency admissions were compared to elective admissions.

be studied in a well-designed study that includes data on medication use prior to colectomy. In contrast to previously reported predictors of postoperative complications including admission type3,10,24 and insurance status,24 these were not predictors of postoperative complications following colectomy in our pediatric cohort. In a subgroup of emergently admitted patients, the time from admission to colectomy was inversely associated with postoperative complications. This contradicted the results of our previous study in children3 as well as those in adults10,24,26 that reported a higher morbidity and mortality when colectomy was delayed. This may be due to the fact that emergently admitted patients who had colectomy shortly after admission (for example, due to fulminant colitis or complications such as toxic megacolon or bowel perforation) had greater disease severity and thus, more likely to experience a postoperative complication. This was explored in a post-hoc sensitivity analysis that excluded emergently admitted patients who had their colectomy within the first two days of their admission. In this sensitivity analysis, time from admission to colectomy was no longer associated with postoperative complications. This data should be interpreted cautiously as clinical data on the patients who had a colectomy within two days of admission were not available. Several limitations are inherent to the study. First, the KID database cannot determine the prevalence of children with UC. Due to the lack of nationwide data on the prevalence of UC in the US, we used the annual US population to calculate rates of colectomy in children overtime. Reports from regional areas in the US have shown a stable incidence and a slight increase in the prevalence of pediatrics UC in the US.30,31 Thus, our analysis may have overestimated the rates of colectomy in children with UC in areas in the US with prevalence rising faster

than annual US population rates. Second, misclassification of the outcome and the exposure is possible in all studies using administrative databases.32 Validation studies have shown that administrative data correctly identified UC patients who underwent colectomy.3,14 Postoperative complications have been assessed in multiple studies using administrative databases,14,33–35 and the positive predictive value of diagnostic codes for these complications has been shown to be higher than 80%.14,36 Third, some postoperative complication codes mix serious and less serious complications. For example, ICD-9 code 997.3 captures both postoperative pneumonia and atelectasis. However, prior validation studies have shown that administrative databases predominantly record serious postoperative complications that influence hospital stay.14 Fourth, KID contains only discharge-level records, not patient-level records. This means that individual patients who are hospitalized multiple times would be recorded as separate patients in the KID database. However, this is unlikely to alter the result of the study given that colectomy is a definitive surgery. Because the KID database extracts data from discharge abstracts, this prevents investigators from following patient outcomes post-discharge and limits our study to in-hospital complications. Fifth, the study is limited by the lack of data on relevant clinical data such as disease extent and severity. We attempted to indirectly control for disease severity by stratifying our analysis by admission type (emergent versus urgent/elective). Finally, the database cannot be linked to a pharmacy database to allow us to evaluate the effect of medications such as corticosteroids, infliximab, and antibiotics on outcomes. In conclusion, we used a nationwide population database to demonstrate an increase in postoperative complications despite a stable colectomy rate in children with UC over the

Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

8

I.S. Soon et al.

past decade. Postoperative complications were associated with a significant burden to the healthcare with a longer length of stay and hospital charges per patient. Increase in postoperative complication rates over time may be due to increase use of salvage therapies prior to colectomy. To minimize postoperative complications, optimal management of UC requires close monitoring and timely decisions on colectomy. Finally, our data support future research to optimize the management of UC with goal to decrease colectomy rate and avoid postoperative complications following colectomy.

Contributions of authors Ing Shian Soon IS participated in conceiving the study concept and design, acquisition of data, statistical analysis and interpretation of data, and drafting of the manuscript. Jennifer CC deBruyn JD participated in conceiving the study concept and design, interpretation of the data, and critical revision of the manuscript for important intellectual content. James Hubbard JH participated in the statistical analysis and interpretation of data. Iwona Wrobel IW participated in conceiving the study concept and design, interpretation of the data, and critical revision of the manuscript for important intellectual content. Reg Sauve RJ participated in conceiving the study concept and design, interpretation of the data, and critical revision of the manuscript for important intellectual content. David L Sigalet DS participated in critical revision of the manuscript for important intellectual content. Gilaad G. Kaplan GK participated in conceiving the study concept and design, interpretation of the data, and critical revision of the manuscript for important intellectual content.

Conflict of interest Gilaad Kaplan has served as an advisory board member for Jansen, Abbvie, and Schering-Plough. He has served as a speaker for Jansen, Schering-Plough, and Abbvie. Research funding from Shire and Abbvie. Jennifer deBruyn has participated in advisory board meetings for Merck and Janssen Inc. and has received research grant support and travel support from Janssen Inc. The other authors do not have financial disclosures.

Acknowledgments Dr. Soon was supported by the Alberta Children's Hospital Research Institute for Child and Maternal Health and the Canadian Institute of Health Research (CIHR) Training Program in Genetics, Child Development and Health research fellowship. Dr. Kaplan is supported through a New Investigator Award from the CIHR and a Population Health Investigator Award from the Alberta Heritage Foundation for Medical Research (AHFMR).

Appendix A. Postoperative in-hospital complications based on the International Classification of Diseases 9th Revision, Clinical Modification (ICD-9-CM) codes

Postoperative complications Mechanical wound Delayed wound healing Postoperative hematoma Postoperative non-infected seroma Dehiscence of operative wound Persistent postoperative fistula Infectious Postoperative infection Postoperative infected seroma Postoperative septic wound complication Postoperative skin infection Postoperative intra-abdominal abscess Postoperative subdiaphragmatic abscess Bacteremia or sepsis Urinary Postoperative urinary retention Postoperative urinary tract infection Postoperative acute renal failure Pulmonary Postoperative atelectasis Postoperative pneumonia (aspiration) Acute respiratory failure Postoperative pneumothorax Postoperative pulmonary edema Gastrointestinal Postoperative small bowel obstruction Postoperative ileus Postoperative nausea and vomiting Postoperative pancreatitis Complication of anastomosis of gastrointestinal tract Hepatic failure following surgery Cardiovascular Deep vein thrombosis

ICD-9-CM 998.83 998.12 998.13 998.31 or 998.32 998.6 998.5 998.51 998.59 998.59 998.59 998.59 790.7, 038.0–038.9 997.5 997.5 997.5 997.3 997.3 518.5, 518.4, 518.81, 518.82 512.2 518.4 997.4 997.4, 560.1 997.4 997.4 997.4 997.4 453.40–453.42, 453.8, 453.9 415.11 997.02

Postoperative pulmonary embolism Postoperative cerebrovascular infarction or hemorrhage Cardiac arrest or heart failure resulting 997.1 from a procedure Acute myocardial infarction 410.0–410.9 Postoperative shock 998.0 Cardiac arrest 427.5 Complication during the surgical procedure Accidental puncture or laceration 998.2 Foreign body accidentally left during 998.4 procedure Hemorrhage complicating procedure 998.11 54.12, 54.61 Reopening of a recent laparotomy sitea a

Procedural ICD-9-CM codes.

Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

Colectomy in children with ulcerative colitis

References 1. Molodecky NA, Soon IS, Rabi DM, Ghali WA, Ferris M, Chernoff G, et al. Increasing incidence and prevalence of the inflammatory bowel diseases with time, based on systematic review. Gastroenterology Jan 2012;142(1):46–54 [e42; quiz e30]. 2. Frolkis AD, Dykeman J, Negrón ME, Debruyn J, Jette N, Fiest KM, et al. Risk of surgery for inflammatory bowel diseases has decreased over time: a systematic review and meta-analysis of population-based studies. Gastroenterology Nov 2013;145(5): 996–1006. 3. Soon IS, Wrobel I, deBruyn JC, Sauve R, Sigalet DL, Kaplan BS, et al. Postoperative complications following colectomy for ulcerative colitis in children. J Pediatr Gastroenterol Nutr Jun 2012;54(6):763–8. 4. Kaplan GG, Seow CH, Ghosh S, Molodecky N, Rezaie A, Moran GW, et al. Decreasing colectomy rates for ulcerative colitis: a population-based time trend study. Am J Gastroenterol Dec 2012;107(12):1879–87. 5. El-Baba M, Lin CH, Klein M, Tolia V. Outcome after surgical intervention in children with chronic inflammatory bowel disease. Am Surg Dec 1996;62(12):1014–7. 6. Rintala RJ, Lindahl HG. Proctocolectomy and J-pouch ileo-anal anastomosis in children. J Pediatr Surg Jan 2002;37(1):66–70. 7. Patton D, Gupta N, Wojcicki JM, Garnett EA, Nobuhara K, Heyman MB. Postoperative outcome of colectomy for pediatric patients with ulcerative colitis. J Pediatr Gastroenterol Nutr Aug 2010;51(2):151–4. 8. Sako M, Kimura H, Arai K, Koganei K, Kito F, Sugita A, et al. Restorative proctocolectomy for pediatric patients with ulcerative colitis. Surg 2006;36(2):162–5. 9. Pakarinen MP, Natunen J, Ashorn M, Koivusalo A, Turunen P, Rintala RJ, et al. Long-term outcomes of restorative proctocolectomy in children with ulcerative colitis. Pediatrics May 2009;123(5):1377–82. 10. de Silva S, Ma C, Proulx MC, Crespin M, Kaplan BS, Hubbard J, et al. Postoperative complications and mortality following colectomy for ulcerative colitis. Clin Gastroenterol Hepatol 2011;9(11):972–80. 11. Agency for Healthcare Quality, Research. KID database documentation. Available at: http://www.hcup-us.ahrq.gov/kidoverview. jsp [Accessed Jan 8, 2012]. 12. Heaton P, Tundia N, Schmidt N, Wigle P, Kelton C. National burden of pediatric hospitalizations for inflammatory bowel disease: results from the 2006 Kids' Inpatient Database. J Pediatr Gastroenterol Nutr Apr 2012;54(4):477–85, http://dx. doi.org/10.1097/MPG.1090b1013e318239bc318279. 13. Debruyn J, Soon I, Hubbard J, Wrobel I, Panaccione R, Kaplan G. Nationwide temporal trends in incidence of hospitalization and surgical intestinal resection in pediatric inflammatory bowel diseases in the United States from 1997 to 2009. Inflamm Bowel Dis Oct 2013;19(11):2423–32, http://dx.doi.org/10.1097/MIB. 2420b2013e3182a56148. 14. Ma C, Crespin M, Proulx MC, DeSilva S, Hubbard J, Prusinkiewicz M, et al. Postoperative complications following colectomy for ulcerative colitis: a validation study. BMC Gastroenterol 2012;12:39. 15. U.S. Bureau of the Census. Available at: http://www.census. gov/popest/index.html [Accessed December 11, 2011]. 16. United States Department of Labor. Bureau of Labor Statistics. Consumer price index. Available at: http://www.bls.gov/cpi/ [Accessed December 11, 2011]. 17. Stata Statistical Software. College Station (TX): Stata Corporation; 2009. 18. Cosnes J, Gower-Rousseau C, Seksik P, Cortot A. Epidemiology and natural history of inflammatory bowel diseases. Gastroenterology 2011;140(6):1785–94 [e1784].

9 19. Falcone Jr RA, Lewis LG, Warner BW. Predicting the need for colectomy in pediatric patients with ulcerative colitis. J Gastrointest Surg Mar-Apr 2000;4(2):201–6. 20. Gower-Rousseau C, Dauchet L, Vernier-Massouille G, Tilloy E, Brazier F, Merle V, et al. The natural history of pediatric ulcerative colitis: a population-based cohort study. Am J Gastroenterol Aug 2009;104(8):2080–8. 21. Van Limbergen J, Russell RK, Drummond HE, Aldhous MC, Round NK, Nimmo ER, et al. Definition of phenotypic characteristics of childhood-onset inflammatory bowel disease. Gastroenterology Oct 2008;135(4):1114–22. 22. Sandborn WJ, Rutgeerts P, Feagan BG, Reinisch W, Olson A, Johanns J, et al. Colectomy rate comparison after treatment of ulcerative colitis with placebo or infliximab. Gastroenterology Oct 2009;137(4):1250–60 [quiz 1520]. 23. Williet N, Pillot C, Oussalah A, Billioud V, Chevaux JB, Bresler L, et al. Incidence of and impact of medications on colectomy in newly diagnosed ulcerative colitis in the era of biologics. Inflamm Bowel Dis Sep 2012;18(9):1641–6. 24. Kaplan GG, McCarthy EP, Ayanian JZ, Korzenik J, Hodin R, Sands BE. Impact of hospital volume on postoperative morbidity and mortality following a colectomy for ulcerative colitis. Gastroenterology Mar 2008;134(3):680–7. 25. Kaplan GG, Hubbard J, Panaccione R, Shaheen AA, Quan H, Nguyen GC, Dixon E, et al. Risk of comorbidities on postoperative outcomes in patients with inflammatory bowel disease. Arch Surg Aug 2011;146(8):959–64. 26. Randall J, Singh B, Warren BF, Travis SPL, Mortensen NJ, George BD. Delayed surgery for acute severe colitis is associated with increased risk of postoperative complications. Br J Surg Mar 2010;97(3):404–9. 27. Nelson R, Liao C, Fichera A, Rubin DT, Pekow J. Rescue therapy with cyclosporine or infliximab is not associated with an increased risk for postoperative complications in patients hospitalized for severe steroid-refractory ulcerative colitis. Inflamm Bowel Dis Jan 2014;20(1):14–20, http://dx.doi.org/10.1097/1001.MIB. 0000437497.0000407181.0000437405. 28. Kennedy R, Potter DD, Moir C, Zarroug AE, Faubion W, Tung J. Pediatric chronic ulcerative colitis: does infliximab increase post-ileal pouch anal anastomosis complications? J Pediatr Surg Jan 2012;47(1):199–203. 29. Yang Z, Wu Q, Wang F, Wu K, Fan D. Meta-analysis: effect of preoperative infliximab use on early postoperative complications in patients with ulcerative colitis undergoing abdominal surgery. Aliment Pharmacol Ther Nov 2012;36(10):922–8, http://dx.doi.org/10.1111/apt.12060 [Epub 12012 Sep 12024]. 30. Adamiak T, Walkiewicz-Jedrzejczak D, Fish D, Brown C, Tung J, Khan K, et al. Incidence, clinical characteristics, and natural history of pediatric IBD in Wisconsin: a population-based epidemiological study. Inflamm Bowel Dis 2013;19(6):1218–23, http://dx.doi.org/10.1097/MIB.1210b1013e318280b318213e. 31. Kappelman M, Moore K, Allen J, Cook S. Recent trends in the prevalence of Crohn's disease and ulcerative colitis in a commercially insured US population. Dig Dis Sci 2013;58(2): 519–25 [2013/02/01]. 32. Molodecky NA, Panaccione R, Ghosh S, Barkema HW, Kaplan GG. Challenges associated with identifying the environmental determinants of the inflammatory bowel diseases. Inflamm Bowel Dis Aug 2011;17(8):1792–9. 33. Guller U, Jain N, Hervey S, Purves H, Pietrobon R. Laparoscopic vs open colectomy: outcomes comparison based on large nationwide databases. Arch Surg Nov 2003;138(11): 1179–86. 34. Guller U, Hervey S, Purves H, Muhlbaier LH, Peterson ED, Eubanks S, et al. Laparoscopic versus open appendectomy: outcomes comparison based on a large administrative database. Ann Surg Jan 2004;239(1):43–52.

Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

10 35. Khan NA, Quan H, Bugar JM, Lemaire JB, Brant R, Ghali WA. Association of postoperative complications with hospital costs and length of stay in a tertiary care center. J Gen Intern Med Feb 2006;21(2):177–80.

I.S. Soon et al. 36. Lawthers AG, McCarthy EP, Davis RB, Peterson LE, Palmer RH, Iezzoni LI. Identification of in-hospital complications from claims data. Is it valid? Med Care Aug 2000;38(8): 785–95.

Please cite this article as: Soon IS, et al, Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States, J Crohns Colitis (2014), http://dx.doi.org/10.1016/j.crohns.2014.05.002

Rising post-colectomy complications in children with ulcerative colitis despite stable colectomy rates in United States.

In children with ulcerative colitis, data on temporal colectomy trends and in-hospital post-colectomy complications are limited. Thus, we evaluated ti...
494KB Sizes 2 Downloads 2 Views