Inflammatory Bowel Disease (IBD) consists of ulcerative colitis (UC) and Crohn’s disease (CD), which are characterized by chronic inflammation with relapsing and remitting course involving the colonic mucosa and entire gastrointestinal tract, respectively  . Both conditions result in decreased quality of life and increased morbidity   . Hospitalization due to IBD burdens the healthcare resource utilization system; the mean annual hospitalization cost for a CD patient was $2595 and $1895 for a UC patient between 2003 and 2004 .
Acute appendicitis is a common diagnosis with a lifetime incidence risk of about 8% among the general population . The treatment of choice is laparoscopic appendectomy owing to less wound complications and faster recovery times . Though it is one of the common emergent abdominal surgeries among hospitalized patients, not much is known about the hospital outcomes in IBD patients. Association of appendectomy with IBD has been studied previously, but there is paucity of research addressing the hospital outcomes of IBD patients undergoing appendectomy.
Thus we sought to compare hospital outcomes including in-hospital mortality, post-surgical complications, length of hospital stay and hospital costs between patients with and without IBD who underwent appendectomy for appendicitis.
2.1. Data Sources
All data were extracted from the Nationwide Inpatient Sample (NIS) between 2009 and 2013. NIS is the largest all-payer inpatient database in the United States. The database represents approximately a 20% sample of nonfederal, acute-care hospitals in the United States. The sampling frame includes community and general hospitals and academic medical centers comprising approximately 90% of all-hospital discharges in the United States. Each data entry includes a unique identifier, demographic variables (including age, gender and race/ethnicity), type of admission, source of admission, principal and secondary diagnoses, primary insurance payer, total hospital charges, and length of stay (LOS). Missing demographic information was summrized in Supplementary Table S1. Healthcare resource utilization included LOS and hospital costs. The comorbidity burden was stratified based on Elixhauser score <3 and ≥3 .
The “Data and Specimen Policy” and “Human Subjects Research Policy” at The Ohio State University does not require Institutional Review Board approval for population-based public data set .
2.2. Study Population and Definition of Variables
Our study consisted of all adult patients (≥18 years old) who had a principal discharge diagnosis of appendicitis and underwent appendectomy, identified by appropriate International Classification of Diseases, Ninth Revision, and Clinical Modification codes (ICD-9-CM) (Appendix 1). The patients were considered to have IBD if the secondary discharge diagnosis was either UC or CD. These criteria have been previously used and validated at a tertiary care academic center  . Elective admissions and patients with discharge diagnoses of both UC and CD were excluded.
Our outcomes of interest were In-hospital Mortality, LOS, total hospital cost and incidence of post-surgical complications.
Statistical analysis was performed using SAS 9.4 (SAS Institute, Cary, NC), employing appropriate survey estimation commands and strata weights. Continuous variables were summarized with means and confidence intervals. Categorical variables were summarized with frequencies and percentages. Differences between patients with and without IBD undergoing appendectomy were analyzed using X2 tests or Student’s t-tests, as appropriate. Multivariate logistic regression was used to calculate adjusted odds ratios and corresponding 95% confidence intervals for in-hospital mortality, incidence of post-surgical complications, and all other bivariate outcomes. Multivariate linear regression was used to analyze LOS and hospital cost. Each multivariate model was adjusted for patient demographics and hospital characteristics, sample size permitting. The models for LOS and hospital costs were additionally adjusted for type of appendicitis, surgery type and significant post-operative complications. The mortality and post-operative complications models were additionally adjusted for appendicitis type and surgery type.
3.1. Demographic Characteristics of Patients and Hospitals
A total of 849,312 patients were included during the study period, of which there were 1845 patients with UC (0.22%), 2416 patients with CD (0.28%), and 845,051 patients with no IBD (99.50%) after 381,592 patients were excluded. Univariate analysis (Table 1 and Supplementary Table S2) demonstrated that patients with UC were more likely to be olde0r and to have more comorbidities, including anemia, chronic blood loss, congestive heart failure, depression,
Table 1. Univariate analysis of non-inflammatory bowel disease (IBD), Crohn’s disease (CD), and ulcerative colitis (UC) patient and hospital characteristics.
1Patients with codes for both open and laparoscopic procedures were classified as open.
hypertension, liver disease, fluid and electrolyte disorders, valvular disease and weight loss when compared with patients without IBD. Patients with CD were more likely to have anemia, rheumatoid arthritis or collagen vascular disease, coagulopathy, fluid and electrolyte disorders, psychoses, and weight loss compared to patients without IBD.
Multivariate analysis adjusting for age, gender, race, insurance, income, Elixhauser score, and hospital size, type and region revealed that, UC patients were more likely to have non-perforated appendicitis (aOR 1.36, 95% CI (1.04, 1.79)) when compared with non-IBD patients (Table 2). CD patients were more likely to have perforated appendicitis (aOR 1.37, 95% CI (1.67, 1.11)) and require open appendectomies (aOR 1.56, 95% CI (1.96, 1.27)) (Table 3).
In-hospital mortality was not different between CD and non-IBD patients after appendectomy and we were unable to compare the mortality between UC and non-IBD patients, as there were no deaths among the UC patients.
3.2.2. Hospital Resource Utilization (Length of Stay and Hospital Cost)
After adjusting for patient and hospital characteristics, appendicitis type (non-perforated versus perforated), surgery type (laparotomy versus open) and significant post-surgical complications, both UC and CD patients after appendectomy had longer LOS (adjusted coefficient, 0.45 days, 95% CI (0.11, 0.79); adjusted coefficient, 0.91 days, 95% CI (0.69, 1.14), respectively) when compared
Table 2. Univariate and multivariate analyses of length of stay (LOS), hospital costs, post-operative deep vein thrombosis, type of appendicitis and type of surgery between patients with UC and no IBD.
1Adjusted for age, gender, race, insurance, income, Elixhauser score, hospital size, type, and region, appendicitis type, surgery type and deep vein thrombosis. 2Adjusted for age, race, insurance, income, and Elixhauser score. 3Adjusted for age, gender, race, insurance, income, Elixhauser score, hospital size, type, and region. 4Patients with codes for both open and laparoscopic procedures were classified as open.
Table 3. Univariate and multivariate analyses of mortality, LOS, hospital costs, post-operative pulmonary embolism and anemia, type of appendicitis and type of surgery between patients with CD and no IBD.
1Adjusted for age, gender, race, insurance, Elixhauser score, hospital type, and region, appendicitis type, and surgery type. 2Adjusted for age, gender, race, insurance, income, Elixhauser score, hospital size, type, and region, appendicitis type, surgery type, GI complications, hemorrhage, hematoma, seroma, pulmonary embolism and post-operative anemia. 3Adjusted for age, gender, race, insurance, Elixhauser score, hospital region, appendicitis type, and surgery type. 4Adjusted for, age, gender, race, insurance, income, Elixhauser score, hospital size, type, and region. 5Patients with codes for both open and laparoscopic procedures were classified as open.
with non-IBD patients. Similarly, UC and CD patients had higher hospital costs (adjusted coefficient, $902, 95% CI (10, 1794); adjusted coefficient, $1952, 95% CI (1343, 2562), respectively) when compared with non-IBD patients (Table 2 and Table 3).
3.2.3. Post-Surgical Deep Vein Thrombosis, Pulmonary Embolism, and Post-Operative Anemia
After adjusting for patient and hospital characteristics, appendicitis type, and surgery type, CD patients after appendectomy were more likely to develop pulmonary embolism (PE) (aOR, 7.06, 95% CI (2.19, 22.79)) and post-operative anemia (aOR, 2.23, 95% CI (1.21, 4.10)) compared to non-IBD patients, whereas UC patients were more likely to develop deep vein thrombosis (DVT) only (aOR, 9.79, 95% CI (2.41, 39.75)) (Table 2 and Table 3).
In this study of the nationwide inpatient database, patients with IBD undergoing appendectomy for appendicitis, although not incurring attributable mortality, was associated with increasing healthcare resource utilization with longer duration of hospitalization, and higher hospital costs. Further, IBD patients also incurred increased risk of post-surgical complications, specifically, DVT in UC patients and PE and post-operative anemia in patients with CD. Furthermore, patients with CD were more likely to have perforated appendicitis and open appendectomy compared to patients with no IBD. To our knowledge, this is the first population-based study evaluating hospital outcomes of IBD patients admitted with appendicitis.
While adult patients with non-perforated appendicitis can be safely discharged on the same day after laparoscopic appendectomy without high rates of complications , IBD patients in general have higher complexity, increased hospital costs and health resource utilization . The study findings are similar to published data on IBD patients undergoing elective colectomy for colorectal cancer or diverticular disease where they contribute to higher healthcare resource utilization including prolonged hospitalization and higher hospital costs when compared with non-IBD patients . Despite controlling for type of appendicitis and appendectomy, patients with UC and CD were associated with longer LOS and higher hospital costs. Compared to patients with UC, those with CD were more likely to have perforated appendicitis, which may be partially attributed to transmural inflammation associated with CD. Consequently, CD patients required more open appendectomies.
IBD patients undergoing appendectomy for appendicitis had increased risk of DVT or PE, which is in agreement with previous studies  . Although chronic inflammation may play a role, DVT or PE could be an extra-intestinal manifestation of IBD as these are not associated with other inflammatory conditions like rheumatoid arthritis and celiac disease . IBD patients have 2 - 3 fold-increased risk of venous thromboembolism  , particularly after surgery (19). Hence, pharmacologic DVT prophylaxis is recommended in all hospitalized patients with IBD unless contraindicated .
Patients with CD, but not UC, were more likely to have post-operative anemia compared to patients without IBD. Anemia is more prevalent in IBD patients due to iron deficiency, anemia of chronic disease, or vitamin B12 deficiency  with CD patients having higher incidence than UC patients . In addition, perioperative blood transfusion has been shown to be associated with post-operative infection and worse hospital outcomes in CD . Therefore, pre-operative optimization of anemia, for instance intravenous iron therapy, may improve outcomes of CD patients undergoing surgery .
We note that this study has several limitations. Accuracy of ICD-9-CM coding could not be verified by chart review, owing to privacy safeguards. However, the Agency for Healthcare Research and Quality has reviewed the NIS database and found good reliability . In addition, any errors in administrative data should be distributed non-differentially across all groups . Furthermore, administrative discharge codes have been previously used and validated for outcomes research in a variety of diseases  . Due to the large number of patients included in the study, there is a risk of identifying statistically significant findings which are of no clinical significance. We reported ORs and 95% CIs to determine the strength of the correlations and improve interpretation. There are no laboratory values or medication data in the NIS (i.e., unable to determine use of immuno-modulators, or post-operative DVT prophylaxis), and no details about disease duration or distribution. This study demonstrated only an association between IBD and worse hospital outcomes in patients with appendicitis undergoing appendectomy. However, determination of causality would require a prospective study.
Despite the above limitations, our study also has several strengths. The NIS database provides a large number of patients with a discharge diagnosis of IBD who underwent appendectomy for appendicitis, which may not be possible from single center or multicenter studies. The benefit of using the NIS database is the results also represent the current national in-patient health-care utilization of IBD patients with appendicitis who underwent appendectomy.
IBD patients with appendicitis undergoing appendectomy had worse hospital outcomes when compared to non-IBD patients. Efforts should be made to give DVT prophylaxis for all IBD patients undergoing appendectomy as the risk of venous thromboembolism (DVT and PE) is very high. Pre-operative anemia in CD patients should be treated adequately with intravenous iron infusions as correcting anemia would improve surgical outcomes. Though laparoscopic approach is ideal for appendectomy, it is reasonable to have low threshold to convert to open appendectomy in CD patients, owing to increased association with perforated appendicitis.
Table A1. ICD-9 codes.
Table S1. Missing data.
Table S2. Univariate analysis of Elixhauser comorbidities of non-inflammatory bowel disease (IBD), Crohn’s disease (CD), and ulcerative colitis (UC).