Inflammatory Bowel Disease and Thrombosis: A National Inpatient Sample (NIS) Study

Background: Inflammatory bowel disease (IBD) is a chronic inflammatory disease of the gastrointestinal mucosa, including two disorders, regional enteritis and ulcerative colitis. These disorders are characterized by defective immune regulation, in which T cells in the gastrointestinal mucosa secrete inflammatory cytokines and an inflammatory response that increases venous thromboembolism (VTE) risk, 2-3 times greater than in the general population. The mechanism by which thrombosis occurs and how to identify those at risk remain unclear.

Methods: We conducted a cross-sectional analysis of discharge data from the National Inpatient Sample (NIS) between January 1, 2009 and December 31, 2014. The NIS is an all-payer database that approximates a 20% stratified sample of discharges from U.S. hospitals participating in the Healthcare Cost and Utilization Project. Discharges among individuals with and without IBD were evaluated by age, race, gender, primary and secondary diagnoses, demographics, and length of stay. De-identified discharge data were classified by the International Classification of Diseases, Ninth Revision (ICD-9) diagnostic codes for thrombosis, including deep venous thrombosis (DVT), pulmonary embolism (PE); arterial thrombosis and arterial embolism; intra-abdominal thromboses, including portal vein thrombosis, splenic/other vein thrombosis, and mesenteric vein thrombosis; and other thrombosis, including cerebral artery embolism/ thrombosis, intracranial venous sinus thrombosis, retinal vein occlusion; and hepatic vein thrombosis. Discharge-level weights from the NIS database were used to produce estimates representative of all community hospitals in the U.S. Thrombosis prevalence was compared between patients with and without IBD, as were demographics (age, race, gender), length of stay, inpatient mortality, and risk factors for thrombosis (dehydration, malnutrition, ports/catheters, chronic steroid use, malignancy, surgery, trauma, and immobilization). Medical conditions i.e. hypertension, renal failure. Continuous data (e.g. age) were compared by weighted simple linear regression. Categorical data (e.g. gender and risk factors) were compared across groups by the Rao-Scott Χ² test. Weighted multivariable logistic regression models for thrombosis were fitted to estimate odds ratios and 95% confidence intervals.

Results: During the 5-year period 2009-2014, the prevalence of thrombosis in patients with IBD (unweighted N= 374,315) was 7.5%, significantly higher than in non-IBD patients (unweighted N=44,748,771), 4.5%, p<0.001. Individuals with IBD and thrombosis were older than IBD without thrombosis, 57.5 years vs. 50.4 years, p<0.001. The most common types of thrombosis were mesenteric vein thrombosis, 7.4%, PE, 6.5%, DVT 6.2%, and portal vein thrombosis, 2.5%. The most common risk factors for thrombosis included surgery, 53.7%, ports/catheters, 13.2%, malignancy, 13.1%, dehydration, 12.4%, malnutrition, 8.9%, and chronic steroid use, 8.2%, which, except dehydration, were all more common than in IBD without thrombosis, p<0.001. Hypertension, 45.2%, hyperlipidemia, 21.5%, cardiac disease, 16.6%, diabetes, 14.9% and renal failure, 13.6%, each p<0.001, were also more common in those with thrombosis. Severity of illness was higher, 1.42 vs. 0.96, length of stay longer, 7.7 vs.5.5 days, and inpatient mortality higher, 3.8% vs. 1.5%, each p<0.001. After adjusting for age and comorbidity index, risk factors remaining significant for thrombosis included port/catheters, OR=1.73 (CI 1.66,1.81), chronic steroids, OR=1.61 (1.54,1.69), malnutrition, OR=1.34 (1.28,1.41), and malignancy, OR=1.13 (1.08,1.18), each p<0.001, while dehydration, OR=0.86 (0.83,0.90), and Asian race, OR=0.61 (0.52,0.72), each p<0,001, appeared protective against thrombosis.

Conclusion: This NIS analysis indicates the prevalence of thrombosis in IBD is 1.7-fold greater than in those without IBD. Even after adjusting for age and severity of illness, the odds of thrombosis in IBD are 73% higher with ports, 61% higher with chronic steroids, 34% higher with malnutrition, and 13% higher with malignancy. These findings suggest the importance of anticoagulation, particularly in IBD patients admitted with ports/catheters, steroids, malnutrition, and/or malignancy and suggest potential benefit if continued long-term in the outpatient setting.