Scandinavian Journal of Gastroenterology. 2014; 49: 1091–1095

ORIGINAL ARTICLE

National trends and inpatient outcomes of inflammatory bowel disease patients with concomitant chronic liver disease DOUGLAS L. NGUYEN1,3, MATTHEW L. BECHTOLD2 & MOHAMMAD MAZEN JAMAL1,3 1

Division of Gastroenterology and Hepatology, University of California-Irvine, Orange, CA, USA, 2Division of Gastoenterology and Hepatology, University of Missouri-Columbia, Missouri, USA, and 3Division of Gastroenterology and Hepatology, Veterans Administration Long Beach Health Care Systems, Long Beach, USA

Abstract Background. There is little information on the frequency of chronic liver disease among hospitalized patients with inflammatory bowel disease (IBD). In this study, we seek to define the common etiologies contributing to chronic liver disease among IBD patients and to identify potential risk factors predictive of increased mortality in this population. Methods. We analyzed the Nationwide Inpatient Sample from 1988 to 2006 to determine the frequency of chronic liver disease among patients with IBD and to determine their in-hospital outcomes. Results. From 1988 to 2006, the age-adjusted rate of chronic liver disease among hospitalized patients with IBD has increased from 4.35 per 100,000 persons in 1988–2001 to 7.45 per 100,000 persons in 2004–2006. The most common etiologies contributing to chronic liver disease among IBD patients were: primary sclerosing cholangitis, unspecified chronic hepatitis, chronic hepatitis C, and nonalcoholic fatty liver disease. Compared to IBD patients without liver disease, there was more than a twofold higher rate of inpatient morality among IBD patients with concomitant liver disease (2.7% vs. 1.3%, p < 0.01). The multivariate analysis showed that factors predictive of inpatient mortality include age >50, spontaneous bacterial peritonitis, ascites, hepatic encephalopathy, presence of cirrhosis, malnutrition, Clostridium difficile colitis, and hospital-acquired pneumonia. Conclusion. There is a higher rate of inpatient mortality among patients with concomitant IBD and chronic liver disease compared to IBD alone. Early recognition and management of complications related to portal hypertension among patients with IBD and chronic liver disease is particularly important in order to reduce inpatient mortality and morbidity.

Key Words: chronic liver disease, inflammatory bowel disease, primary sclerosing cholangitis

Introduction Liver diseases are common extra-intestinal manifestations of both Crohn’s disease (CD) and ulcerative colitis (UC). Current data suggests that primary sclerosing cholangitis (PSC) is the most common chronic liver disease among inflammatory bowel disease (IBD) patients, with up to 7.5% of IBD patients having a concomitant diagnosis of PSC [1]. However, autopsy series show that fatty liver disease can be present in up to 80% of IBD cases [2,3]. A small number of studies have examined the prevalence of liver dysfunction in different IBD

cohorts. Depending on the definition used, the reported prevalence of liver disease among IBD patients ranged from 3% to 50% cases [4]. Abnormal liver biochemistries are most commonly transient and have previously been linked to the activity of IBD, though having no impact on long-term prognosis [5,6]. The Mayo Clinic experience demonstrated that abnormal liver biochemistries were present in up to 30% of IBD patients, and among patients with a diagnosis of chronic liver disease, there appears to be a negative impact on long-term mortality [4]. Current studies are from single center experiences and the prevalence of chronic liver disease among

Correspondence: Douglas Nguyen, MD, Division of Gastroenterology and Hepatology, University of California-Irvine, 101 The City Drive, Orange, CA 92697, USA. Tel: +1 714 456 6745; E-mail: [email protected]

(Received 5 April 2014; revised 28 April 2014; accepted 30 April 2014) ISSN 0036-5521 print/ISSN 1502-7708 online  2014 Informa Healthcare DOI: 10.3109/00365521.2014.921326

1092

D. L. Nguyen et al.

IBD patients has not been fully defined. Further, the trends in the prevalence of chronic liver disease among IBD patients has not been well documented. Using the Nationwide Inpatient Sample (NIS), we seek to define common etiologies contributing to chronic liver disease among hospitalized IBD patients in a large nationwide cohort and to identify potential risk factors predictive of increased in-hospital mortality in this population. Methods Data sources The NIS is a component of the Healthcare Cost and Utilization Project and is the largest all-payer inpatient care database in the USA. It is composed of all-payer information on hospital inpatient admissions from states participating in the Healthcare Cost and Utilization Project from 1988 to 2006. The NIS compiles data from more than 8 million inpatient stays from approximately 1000 hospitals each year. It is designed to approximate a 20% stratified sample of patients from all community hospitals in the USA and has been used by researchers and policymakers to identify, track, and analyze national trends in healthcare use, access, charges, quality, and outcomes. The NIS contains primary and secondary diagnoses, primary and secondary procedures, admission and discharge status, and patient demographics. Data collection We extracted and analyzed all hospital discharges between 1988 and 2006 among patients who were >17 years of age with a primary or secondary diagnosis of UC or CD as identified by an International Classification of Diseases, ninth revision, Clinical Modification (ICD-9-CM) codes of 555.0 to 556.9. Among the IBD patients identified, patients with underlying chronic liver disease were identified. Etiologies contributing to chronic liver disease were found using the following ICD-9-CM codes: chronic hepatitis C (070.41, 070.44, 070.54, 070.7), chronic hepatitis B (070.21–070.33), unspecified chronic hepatitis (571.49, 571.9), autoimmune hepatitis (571.42), PSC (576.1), primary biliary cirrhosis (571.6), and nonalcoholic fatty liver disease (571.8). We excluded alcoholic liver disease from our study because of its unique pathogenesis from other chronic liver diseases. Demographic information including age, sex, hospital size, hospital location, and ethnicity of these patients was obtained. Using ICD-9-CM codes, in-hospital morbidity including Clostridium difficile colitis (008.45), hospital-acquired pneumonia

(480–486), hepatic encephalopathy (572.2), spontaneous bacterial peritonitis (567.23), malnutrition (263.0, 263.8, 263.9), presence of cirrhosis (571.5), presence of ascites (789.59), and variceal hemorrhage (456.0) were identified. Statistical analysis For each year of the study, the average annual age-adjusted incidence rate for IBD patients who had underlying chronic liver disease were calculated by multiplying the age-specific rates by age-specific weights. The weights used in the age-adjustment of the data were the proportion of the year 2000 standard US population within each age group. The ageadjusted rates were averaged across 3- to 4-year intervals for the purpose of trend analysis. Medians with ranges were reported in the case of continuous variables with non-normal distribution, and means with standard deviation in the case of continuous variables with normal distribution. Multivariate regression models using SPSS General Log-linear software was used for identifying factors predictive of in-hospital mortality in our study population. Results Baseline characteristics The baseline characteristics of IBD patients with concurrent chronic liver disease are summarized in Table I. The mean age in the study cohort was 48.10 ± 17.45 years in the UC group and 49.38 ± 15.29 years in the CD group. In the CD group, 60.5% were males and in the UC group 48.9% were males. IBD patients with underlying chronic liver disease were most frequently hospitalized in a large hospital. The hospital geographic location was evenly distributed throughout the country. National trend in prevalence of chronic liver disease among IBD patients From 1988 to 2006, the age-adjusted rate of chronic liver disease among hospitalized IBD patients has nearly doubled from 4.35 per 100,000 persons in 1988–2001 to 7.45 per 100,000 persons in 2004– 2006 (Figure 1). The incidence of chronic liver disease during 2004–2006 among UC patients is 4.61 per 100,000 persons and for CD patients the incidence of chronic liver disease is 2.84 per 100,000. As illustrated in Figure 1, the frequencies of chronic liver patients appear to be higher among UC patients compared to CD patients across all time points.

Outcomes of IBD patients with chronic liver disease Table I. Baseline characteristics of IBD patients with underlying chronic liver disease.

Mean age (years) Gender Male Female Geographic location East Central South West Hospital size Small Medium Large Health insurance Medicare Medicaid Private insurance Other Chronic hepatitis B Chronic hepatitis C Primary Sclerosing Cholangitis Primary biliary cirrhosis Nonalcoholic fatty liver disease Autoimmune hepatitis Unspecified chronic hepatitis

Ulcerative colitis n = 8434

Crohn’s disease n = 6850

48.10 ± 17.45

49.38 ± 15.29

5101 (60.5%) 3333 (39.5%)

3349 (48.9%) 3501 (51.1%)

22.9% 24.1% 29.3% 23.7%

21.9% 22.3% 35.9% 19.9%

793 (9.4%) 1973 (23.4%) 5668 (67.2%)

739 (10.8%) 1761 (25.7%) 4350 (63.5%)

2159 (25.6%) 751 (8.9%) 4976 (59.0%) 548 (6.5%) 109 (1.3%) 928 (11.0%) 4268 (50.6%) 354 (4.2%) 1071 (12.7%) 219 (2.6%) 1485 (17.6%)

2295 (33.5%) 870 (12.7%) 3096 (45.2%) 589 (8.6%) 212 (3.1%) 1499 (21.9%) 1815 (26.5%) 219 (3.2%) 1391 (20.3%) 185 (2.7%) 1529 (22.3%)

contributing to chronic liver disease in CD patients were: PSC (26.5%), unspecified chronic hepatitis (22.3%), chronic hepatitis C (21.9%), nonalcoholic fatty liver disease (20.3%), primary biliary cirrhosis (3.2%), chronic hepatitis B (3.1%), and autoimmune hepatitis (2.7%).

Factors predictive of in-hospital mortality

The most common etiologies contributing to chronic liver disease among UC patients were: PSC (50.6%), unspecified chronic hepatitis (17.6%), nonalcoholic fatty liver disease (12.7%), chronic hepatitis C (11.0%), primary biliary cirrhosis (4.2%), autoimmune hepatitis (2.6%), and chronic hepatitis B (1.3%). In contrast, the most common etiologies

The trend of in-hospital mortality among IBD patients with and without concomitant chronic liver disease is illustrated in Figure 2. The crude mortality rate among IBD patients without concomitant chronic liver disease has remained stable throughout the indexed period. However, the mortality rate of IBD patients with chronic liver disease has steadily been decreasing since 1988–1991. Despite a trend toward decreased mortality in the IBD with chronic liver disease group, there was more than a twofold overall higher incidence of mortality among the IBD group with chronic liver disease compared to the group with IBD alone in our study period (2.7% vs. 1.3%, respectively). The multivariate analysis (Table II) showed that factors predictive of inpatient mortality among IBD patients with liver disease include: age >50 (odds ratio [OR] = 1.83; 95% confidence interval [CI]: 1.44, 2.34), presence of ascites (OR = 1.92; 95% CI: 1.23, 2.73), hepatic encephalopathy (OR = 3.77, 2.75, 5.16), spontaneous bacterial peritonitis (OR = 2.93; 95% CI: 2.24, 3.83), presence of cirrhosis (OR = 1.93; 95% CI: 1.51, 2.48), Clostridium difficile colitis (OR = 2.46; 95% CI: 1.29, 4.32), hospital-acquired pneumonia (OR = 2.44; 95% CI:

8.00 7.00 Age-adjusted rate per 100,000

1093

6.00 5.00 Crohn's disease 4.00

Ulcerative colitis Combined

3.00 2.00 1.00 0.00 1988–1991

1992–1995

1996–1999 Year

2000–2003

2004–2006

Figure 1. Illustration of national estimated age-adjusted frequencies of IBD patients with concurrent chronic liver disease.

D. L. Nguyen et al. Crude in-hospital mortality rate (%)

1094

4.00 3.50 3.00 2.50 IBD with liver disease

2.00

IBD alone

1.50 1.00 0.50 0.00 1988–1991 1992–1995 1996–1999 2000–2003 2004–2006 Year

Figure 2. Illustration of trends of in-hospital mortality rate among IBD patients with and without concomitant chronic liver disease.

2.30, 2.59), and malnutrition (OR = 3.24; 95% CI: 3.15, 4.34) Geographic location, hospital size, and variceal hemorrhage were not predictive of increased mortality. Discussion Our study is the first large-scale study attempting to define the frequency of chronic liver disease among hospitalized IBD patients, which brings new insights into our understanding of liver diseases in this population. In this national cohort of hospitalized patients, the estimated age-adjusted frequency of chronic liver disease among IBD patients have nearly doubled from 4.35 per 100,000 persons in 1988– Table II. Factors predictive of increased risk for in-hospital mortality.

Age >50 years Gender Male Hospital bed size Small Medium Large Region of hospital Northeast Central South West Presence of ascites Variceal hemorrhage Hepatic encephalopathy Spontaneous bacterial peritonitis Presence of cirrhosis Clostridium difficile colitis Hospital-acquired pneumonia Malnutrition

Odds ratio

95% Confidence interval

1.83

1.44, 2.34

0.99

0.79, 1.24

1.10 0.83 1.00

0.73, 1.66 0.64, 1.07

0.70 1.03 1.01 1.00 1.92 1.62 3.77 2.93 1.93 2.46 2.44 3.24

0.50, 1.06 0.72, 1.46 0.73, 1.38 1.23, 0.58, 2.75, 2.24, 1.51, 1.29, 2.30, 3.15,

2.73 4.54 5.16 3.83 2.48 4.32 2.59 4.34

2001 to 7.45 per 100,000 persons in 2004–2006. Across all time points, the age-adjusted rate of chronic liver disease appears to be higher in the UC group compared to the CD group. This is likely secondary to a higher incidence of PSC in the UC group. Our finding is consistent with prior reports that the presence of IBD, particularly with UC, is between 40% and 90% of patients with a diagnosis of PSC [7,8]. In this study, we demonstrated that the most common etiologies contributing to chronic liver disease among IBD patients include PSC, nonalcoholic fatty liver disease, unspecified chronic cirrhosis, and chronic hepatitis C infection. In a recent report from the Cleveland Clinic, the estimated prevalence of nonalcoholic fatty liver disease among IBD patients was only 8.2% compared to 33.6% of the general population [9]. Our study also supports that though nonalcoholic fatty liver disease is not a major contributor as an etiology for the development of clinically significant chronic liver disease among IBD patients, there has been over a 60% increase in the development of nonalcoholic fatty liver disease among IBD patients from 0.78 per 100,000 in 1988–1999 to 1.27 per 100,000 in 2004–2006 (data not shown). This relative increase of 60% is by far the largest increase among the studied etiologies for the development of chronic liver disease. In our cohort, nonalcoholic fatty liver disease appears to be the largest growing etiology for the development of chronic liver disease among IBD patients, similar to the trends we are observing nationally [9–11]. This may be related to the increasing incidence of obesity among IBD patients and the use of corticosteroids [11]. Further, among the patients with reported unspecified chronic hepatitis in this study, it is unclear which portion of those patients may also have underlying nonalcoholic fatty liver disease as an etiology for the development of chronic liver disease.

Outcomes of IBD patients with chronic liver disease Mendes et al. in a small cohort study from the Mayo Clinic has shown that IBD patients with underlying chronic liver disease had a poorer prognosis compared to patients with IBD alone [4]. Though our study demonstrated that since 1988 the crude rate for in-hospital mortality has steadily been declining across all time points, the rate of mortality among IBD patients with chronic liver disease was higher compared to IBD patients without concomitant chronic liver disease. The improved mortality rate is likely related to improvements in medical care for patients with chronic liver disease. Strong factors predictive of in-hospital mortality were related to portal hypertension and cirrhosis, including presence of ascites, spontaneous bacterial peritonitis, hepatic encephalopathy, and protein calorie malnutrition. Consistent with our prior report [12], variceal hemorrhage was not predictive of increased in-hospital mortality. This is likely secondary to a combination of early recognition of variceal hemorrhage, aggressive resuscitation strategies, and improved endoscopic techniques for the management of variceal bleeding. Further, in our cohort, the presence of Clostridium difficile colitis resulted in a twofold higher risk of in-hospital mortality. The analysis of administrative datasets has several limitations. First, the study identifies patients using ICD-9-CM coding for IBD cases and chronic liver disease, which we are unable to validate by chart review. Second, we are unable to identify methods (i.e., biochemical testing, liver biopsy, abdominal imaging, clinical presentation) which accurately diagnose chronic liver disease in IBD patients. Thirdly, the NIS dataset does not allow us to determine the current therapies that the patients are on for the treatment of their IBD (i.e., immunomodulators and biologics) and whether or not being on these medications aggravated their underlying liver disease. Further, the trends presented in this study only reflect the inpatient population and it is unclear if the same trend exists in the outpatient setting. Future studies will need to better define the prevalence of chronic liver disease among IBD patients in the outpatient setting. Additionally, it would be important to understand the relationship between severity of IBD and its effect on the underlying liver disease. It would also be interesting to define if treatment for IBD affects the natural history of the liver disease and which classes of IBD medications has the greatest effect. In conclusion, it is clear in our study and prior studies that IBD patients with underlying chronic

1095

liver disease have a poorer long-term prognosis. Therefore, early recognition and management of complications related to portal hypertension among patients with IBD and chronic liver disease is particularly important in order to reduce inpatient mortality and morbidity.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

References [1] Broome U, Bergquist A. Primary sclerosing cholangitis, inflammatory bowel disease, and colon cancer. Semin Liver Dis 2006;26:31–41. [2] Perett AD, Higgins G, Johnston HH, Massarella GR, Truelove SC, Wrigth R. The liver in Crohn’s disease. Q J Med 1971;40:187–209. [3] Perrett AD, Higgins G, Johnston HH, Massarella GR, Truelove SC, Wrigth R. The liver in ulcerative colitis. Q J Med 1971;40:211–38. [4] Mendes FD, Levy C, Enders FB, Loftus EV Jr, Angulo P, Lindor KD. Abnormal hepatic biochemistries in patients with inflammatory bowel disease. Am J Gastroenterol 2007;102:344–50. [5] Loftus EV Jr, Sandborn WJ, Lindor KD, Larusso NF. Interactions between chronic liver disease and inflammatory bowel disease. Inflamm Bowel Dis 1997;3:288–302. [6] Broomé U, Glaumann H, Hellers G, Nilsson B, Sörstad J, Hultcrantz R. Liver disease in ulcerative colitis: an epidemiological and follow up study in the country of Stockholm. Gut 1994;35:84–9. [7] Escorsell A, Parés A, Rodés J, Solís-Herruzo JA, Miras M, dela Morena E. Epidemiology of primary sclerosing cholangitis in Spain. Spanish Association for the Study of the Liver. J Hepatol 1994;21:787–91. [8] Tung BY, Brentnall TA, Kowdley KV, Emond M, Kimmey MB, Stevens AC, et al. Diagnosis and prevalence of ulcerative colitis in patients with sclerosing cholangitis (abstract). Hepatology 1996;24:169A. [9] Sourianarayanane A, Garg G, Smith TH, Butt MI, McCullough AJ, Shen B. Risk factors of non-alcoholic fatty liver disease in patients with inflammatory bowel disease. J Crohns Colitis 2013;7:e279–85. [10] Kemmer N, Neff GW, Franco E, Osman-Mohammed H, Leone J, Parkinson E, et al. Nonalcoholic fatty liver disease epidemic and its implications for liver transplantation. Transplantation 2013;96:860–2. [11] McGowan CE, Jones P, Long MD, Barritt AS. Changing shape of disease: nonalcoholic fatty liver disease in Crohn’s disease-a case series and review of the literature. Inflamm Bowel Dis 2012;18:49–54. [12] Jamal MM, Samarasena JB, Hashemzadeh M. Decreasing in-hospital mortality for oesophageal variceal hemorrhage in the USA. Eur J Gastroenterol Hepatol 2008;20:947–55.

Copyright of Scandinavian Journal of Gastroenterology is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

National trends and inpatient outcomes of inflammatory bowel disease patients with concomitant chronic liver disease.

There is little information on the frequency of chronic liver disease among hospitalized patients with inflammatory bowel disease (IBD). In this study...
213KB Sizes 1 Downloads 3 Views