The Association Between Corticosteroid Use and Development of Fractures Among IBD Patients in a Population-Based Database

Transcription

1 THE AMERICAN JOURNAL OF GASTROENTEROLOGY Vol. 98, No. 8, by Am. Coll. of Gastroenterology ISSN /03/$30.00 Published by Elsevier Inc. doi: /s (03) The Association Between Corticosteroid Use and Development of Fractures Among IBD Patients in a Population-Based Database Charles N. Bernstein, M.D., James F. Blanchard, M.D., Ph.D., Colleen Metge, Ph.D., and Marina Yogendran, M.Sc. Department of Internal Medicine, University of Manitoba; University of Manitoba Inflammatory Bowel Disease Clinical and Research Centre; Department of Community Health Sciences, University of Manitoba; Faculty of Pharmacy, University of Manitoba; and Manitoba Centre for Health Policy, Winnipeg, Manitoba, Canada OBJECTIVE: Because the rate of fracture among patients with inflammatory bowel disease (IBD) is only slightly higher than that in the general population, it is important to define high-risk groups worthy of diagnostic evaluation or prophylactic interventions. Corticosteroid use has been considered in other diseases to be a risk for fracture, although not all studies in IBD are concordant on this point. We aimed to determine whether patients with IBD drawn from a population-based database who sustain fractures are more likely to have been using corticosteroids than a matched group of IBD patients who did not fracture. METHODS: We extracted from our population-based University of Manitoba Inflammatory Bowel Disease Epidemiology Database the number of patients with a new diagnosis of fracture between the years From within our Inflammatory Bowel Disease Epidemiology Database, we extracted a control group of IBD patients who did not develop fractures matched to the case group who did by age, gender, diagnosis, year of diagnosis, and geographic area of residence. We linked our cohorts with Manitoba Health s Drug Program Information Network to study corticosteroid use within 2 yr before fracture diagnosis. The Drug Program Information Network is a population-based database, established in 1995, which records all prescription drugs. RESULTS: Fractures were identified in 13 patients with Crohn s disease and in 28 patients with ulcerative colitis. The control group included 103 Crohn s disease and 173 ulcerative colitis patients who did not fracture. In Crohn s disease, for the group who fractured compared with the controls who did not fracture, corticosteroid use before fracture was evident in seven (54%) compared with 21 (22%) who did not fracture ( , df 1, p 0.035). In ulcerative colitis, for the group who fractured compared with the controls who did not fracture, corticosteroid use before fracture was evident in five (18%) compared with 37 (21%) who did not fracture ( , df 1, p 0.861). Fracture cases were more likely to be exposed to oral corticosteroids (OR 1.75; 95% CI ), but this result is not significant. Regarding corticosteroid dosing among the 12 patients with IBD who fractured and used corticosteroids, the mean total days supply was 314 days 236 days compared with 258 days 278 days in those who did not fracture (p 0.16). The prescribed daily dose among corticosteroid users was comparable for those who fractured versus those who did not fracture (18 mg/day vs 21 mg/day, p 0.90). CONCLUSIONS: Patients who require corticosteroids in Crohn s disease should be considered at risk for fracture. Further research is required to delineate after how much corticosteroid use are subjects at risk and/or after what duration of active disease. (Am J Gastroenterol 2003;98: by Am. Coll. of Gastroenterology) INTRODUCTION The prevalence of osteopenia among patients with inflammatory bowel disease (IBD) ranges from 40% to 50%, and frank osteoporosis is seen in approximately 15% of patients (1 5). These estimates are based on the measurement of bone mineral density (BMD), where osteoporosis is defined as a bone mass T score 2.5 (6). Recently, we have reported on a population basis the incidence rate of fractures among patients with IBD and the relative risk compared with an age-, gender- and geographically matched control group (7). We reported that patients with IBD have a 40% increased risk of fracturing compared with the general population (7). The two other population-based reports of fracture risk in IBD have reported either no increased fracture risk (8) or an increased risk for patients with Crohn s disease only (9). Thus, there is a discrepancy between the degree of osteoporosis and hence fracture risk suggested by BMD studies on subjects presenting at specialized centers compared with population-based fracture studies. Nonetheless, some subjects with IBD are at increased risk, and it is

2 1798 Bernstein et al. AJG Vol. 98, No. 8, 2003 important to determine what risk factors are relevant specifically in IBD. There has been much written on potential risk factors for osteoporosis in IBD. Although many factors may be relevant for IBD, one factor of proven risk is age (7 9). Another risk factor that has been a subject of some conflicting data is the use of corticosteroids. Systemic inflammation may be a risk factor, because proinflammatory cytokines may affect bone mass, and recent studies have established the link between osteoclastogenesis and T-cell function through the receptor for activated nuclear factor -B (RANK) and its ligand RANKL (10, 11). Although some studies have reported an association between corticosteroid use and osteoporosis in IBD (10), no study has reported this with adequate control for active inflammatory disease. In IBD, patients receiving corticosteroids are invariably ones with active inflammation. Furthermore, it has been shown that many IBD patients may have diminished bone mass in the absence of ever having used corticosteroids (12, 13). Even bone mass data from corticosteroid users with a variety of chronic conditions (i.e., asthma, polymyalgia rheumatica, rheumatoid arthritis) may not be directly extrapolated to IBD because among these conditions there are wide ranges as to the effects of corticosteroids (14, 15). Finally, no study has determined whether corticosteroid use correlated with fracture risk in IBD, a much more meaningful end point than BMD measurements. We aimed to test the hypothesis that corticosteroid use is associated with an increased risk of bone fracture among persons with IBD. We conducted a nested case control analysis. We extracted subjects within our University of Manitoba Inflammatory Bowel Disease Epidemiology Database who had sustained fractures in the years and matched them by age, gender, duration of disease, and geographic residence with IBD subjects within the database who had not sustained a fracture in those years. We then linked the matched cohort with Manitoba Health s Drug Program Information Network (DPIN). We extracted from the DPIN corticosteroid usage. Because DPIN was established only in 1995, we aimed to determine if a relationship existed between corticosteroid use within the 2 yr before sustaining a fracture. MATERIALS AND METHODS The University of Manitoba Inflammatory Bowel Disease Epidemiology Database In 1995, we established the University of Manitoba Inflammatory Bowel Disease Epidemiology Database (16). This is a population-based database. In establishing it, we created an administrative definition of IBD, which has a sensitivity and specificity of approximately 90% for diagnosing a case as IBD as validated against patient self-report and chart review. We identify cases as being truly IBD if they have five contacts within the administrative database of health claims of Manitoba Health (the provincial body that provides universal health insurance to all Manitoba residents) or at least three claims if they have been residents within the province for 2 yr or less. Patients are identified within the Manitoba Health database by a personal health identification number (PHIN), a unique identifier that allows for longitudinal tracking of health history and records all outpatient physician and hospital-based contacts. The University of Manitoba Inflammatory Bowel Disease Epidemiology Database contains only non-nominal data, and subjects case histories are tracked by a scrambled PHIN method to ensure confidentiality. We extracted from our University of Manitoba Inflammatory Bowel Disease Epidemiology Database the number of patients with a new diagnosis of fracture between the years We used International Classification of Diseases, 9th Revision, codes 820.xx, 805.xx, 807.0x x, and 813.xx to identify fractures at the hip, spine, ribs, and wrists, respectively. From within our University of Manitoba Inflammatory Bowel Disease Epidemiology Database, we extracted a control group of IBD patients who did not fracture matched to the case group who developed fractures. We selected five controls for each case, matching on age, gender, diagnosis (Crohn s disease vs ulcerative colitis), year of IBD diagnosis, and geographic area of residence. When fewer than five potential controls were available, we analyzed only those available controls and maintained the matching process. Manitoba Health s DPIN We have recently linked our University of Manitoba IBD Epidemiology Database with DPIN to study pharmaceutical usage by IBD patients (17). The DPIN database is a population-based database, which records all prescription drugs prescribed for each resident of Manitoba registered with the provincial health care system and has been in effect since The linkage was performed using a scrambled PHIN method. Thus, for all database usage, no nominal patient information is used. Previously, we have used this linkage to produce unique and important data regarding prescription drug use by patients with IBD and have assessed trends in the use of certain IBD-specific drugs (17). We determined whether IBD patients with bone fracture had increased exposure to corticosteroid use before fracture. We analyzed the DPIN database for use of oral corticosteroids within 2 yr before the onset of the bone fracture in the index case. Exposure to corticosteroids was defined as oral prednisone use at doses greater than 5 mg/day for more than 2 wk or other forms of corticosteroid (such as oral budesonide), continuously for at least 2 months. Conditional logistic regression was used to estimate the OR and 95% CI for the association between corticosteroid exposure (Y/N) and fracture. We performed matched conditional logistic regression using the SAS procedure PHREG (SAS Institute, Cary, NC). We also compared the mean total days supply and the mean daily dose of oral prednisone used by cases versus

3 AJG August, 2003 Corticosteroid Use and Fractures in IBD Patients 1799 Table 1. Characteristics of and (n 13) Crohn s Disease (n 103) (n 28) Ulcerative Colitis (n 173) Age, mean (SD) 59.4 (23.5) 52.5 (19.1) 70.5 (16.5) 64.7 (16.5) Female, n. (%) 9 (69.2) 53 (51.5) 18 (64.3) 110 (63.6) Geographic region, urban (n, %) 8 (61.5) 59 (57.3) 18 (64.3) 111 (64.2) Decade of diagnosis, 1980s (%) 7 (53.8) 56 (54.4) 12 (42.9) 102 (59.0) controls. A nonparametric Wilcoxon rank sum test was used to test for differences between those on oral prednisone who fractured and those on oral prednisone who did not fracture. A Kruskal Wallis test was used to determine whether the differences in days supply (days on prednisone) and daily dose (in mg) found between prednisone-use fracturers and nonfracturers were significant. Analysis was performed using SAS statistical software, Version 8.2 (SAS Institute, Cary, NC). This study has been approved by the University of Manitoba Health Research Ethics Board, Manitoba Health s Health Information Privacy Committee, and the study review committee of the Manitoba Center for Health Policy. RESULTS We identified 41 cases of IBD with fracture. These were matched with 276 IBD patients who did not sustain fracture. The mean age of the sample was 61.0 (SD 18.7), and 61% were women (Table 1). Approximately two thirds of the sample was urban based, which reflects the Manitoba urban/ rural resident proportions. The majority of IBD cases were originally diagnosed in the decade of the 1980s (Table 1). Among the fracture group, 12 (30%) had used corticosteroids compared with 60 (22%) (p 0.32) who did not fracture. Thirteen patients with Crohn s disease and 28 patients with ulcerative colitis were identified who fractured. These subjects were matched with 103 subjects with Crohn s disease and 173 subjects with ulcerative colitis who did not fracture. Corticosteroid use within the 2 yr before fracture was evident in seven (54%) of Crohn s disease subjects who fractured compared with 23 (22%) of subjects in the Crohn s disease control group who did not fracture ( , df 1, p 0.035). Corticosteroid use within the 2 yr before fracture was evident in five (18%) of ulcerative colitis subjects who fractured compared with 37 (21%) of subjects in the ulcerative colitis control group who did not fracture ( , df 1, p 0.861). There were too few subjects with fractures to meaningfully assess women versus men in each of the disease groups. For all cases of IBD with fracture exposed to corticosteroids (n 12), the mean total days supply was 314 days 236 days (median 330 days), and there was a statistical trend for difference compared with the total days supply of those who did not fracture (mean 257 days 278 days, median 152 days, p 0.16). The prescribed daily dose among corticosteroid users was comparable for those who fractured versus those who did not fracture (18 mg/day vs 21 mg/day, p 0.90). Compared with controls who did not fracture, patients seemed to be about 75% more likely to have been exposed to corticosteroids (OR 1.75, 95% CI ). However, the result is not significant (Table 2). None of the patients who fractured were using budesonide, and of patients without fractures, only two used budesonide. Thus, it was decided to exclude all users of budesonide from the analysis (because a direct conversion to equivalent doses of prednisone is difficult). DISCUSSION Our study has proved that subjects with Crohn s disease, but not ulcerative colitis, who sustain a fracture compared with subjects with these diseases who have not fractured are more likely to have used corticosteroids within 2 yr and more likely to have used greater duration of corticosteroids. Because this was an epidemiological study without access to patient charts, we cannot determine the disease activity status of the subjects in either the fracture or no-fracture groups. It is plausible to assume, however, that those using corticosteroids had more active disease because the era of the study was before availability of infliximab in Manitoba (another agent considered for treatment of active disease), and corticosteroids are widely prescribed for active IBD in Manitoba (17). Hence, this study does not establish irrefutably that corticosteroids are a key risk factor for fracture because it may have been the inflammatory disease that put those patients who fractured at principle risk. It should be noted that two studies have suggested that disease activity has no effect on BMD (12, 18), so the debate as to whether active disease or the treatment thereof with corticosteroids can lead to fractures rages on. However, this study does identify that the group of patients who warrant and use corticosteroids (and this in most cases will include those with active inflammatory disease) are at risk for fracture. Table 2. Association Between Fracture and Corticosteroid Use No. (%) Exposed (n 40) (n 276) Univariate OR (95% CI) Oral corticosteroids 12 (30.0) 60 (21.7) 1.75 ( )

4 1800 Bernstein et al. AJG Vol. 98, No. 8, 2003 Although there are advocates of screening all IBD subjects with dual energy x-ray absorptiometry (DXA) who use corticosteroids, total dose has been inversely associated with BMD in several studies of IBD (19 23) but not in others (24, 25). Current corticosteroid use has been reported to show both no effect on BMD (24) and diminished BMD (26, 27). One study showed that current corticosteroid use was associated with lower BMD in Crohn s disease, but not ulcerative colitis (28). Although some consider Crohn s disease to be an increased risk for fracture compared with ulcerative colitis, not all fracture data support this (7). Our current study actually found more cases with ulcerative colitis who fractured compared with Crohn s disease, despite a slightly higher prevalence of Crohn s disease in the population (16). It is noteworthy that we found corticosteroid use to be associated with fractures in Crohn s disease, but not ulcerative colitis, suggesting potentially different mechanisms of bone disease in each disease. Establishing specific risk factors for fracture in IBD is of paramount importance to help risk stratify patients as to who needs either investigations of BMD (with DXA) and/or who might need therapeutic interventions. Risk factors for osteoporosis in IBD have been assumed to be the same risk factors as in other disease states. It is imperative that studies be conducted to help risk stratify subjects with IBD because, although they have an increased risk for fracture, it is at most a mildly increased risk. The strength of our study is that it is population based, thereby eliminating the bias of only recruiting IBD subjects who present to specialized IBD clinics. Furthermore, the DPIN database is comprehensive and population based, eliminating the likelihood that prescriptions for corticosteroids would be missed. The weakness of our study is that, in the absence of access to full clinical data, we cannot control for other potential confounding variables, such as smoking, gonadal state, and nutritional status and that we cannot distinguish whether it is the corticosteroids or the active disease that is the real risk for fractures. Nonetheless, we have established a risk factor for fracture to consider in patients with Crohn s disease other than patients age, which has already been established as a fracture risk factor. Patients who require corticosteroids in Crohn s disease should be considered at risk for fracture. Further research is required to delineate after how much corticosteroid use are subjects at risk and/or after what duration of active disease. ACKNOWLEDGMENTS We thank Andre Wajda, M.Sc., for help with creating the matched cohorts. Charles N. Bernstein, M.D., is supported in part by a Canadian Institutes of Health Research Investigator Award and by the Crohn s and Colitis Foundation of Canada Research Scientist Award. We are indebted to Health Information Services, Manitoba Health, for providing data on health care use. The results and conclusions are those of the authors, and no official endorsement by Manitoba Health is intended or should be inferred. Reprint requests and correspondence: Charles N. Bernstein, M.D., John Buhler Research Centre, 804F-715 McDermot Avenue, Winnipeg, Manitoba, Canada R3E 3P4. Received Dec. 12, 2002; accepted Apr. 30, REFERENCES 1. Bernstein CN. Calcium and bone issues in inflammatory bowel disease. Gastroenterol Int 1997;10: Compston JE. Review article: Osteoporosis, corticosteroids and inflammatory bowel disease. Aliment Pharmacol Ther 1995;9: Robinson RJ, Iqbal SJ, Abrams K, et al. Increased bone resorption in patients with Crohn s disease. Aliment Pharmacol Ther 1998;12: Fries W, Dinca M, Luisetto G, et al. Calcaneal ultrasound bone densitometry in inflammatory bowel disease A comparison with double energy x-ray absorptiometry. Am J Gastroenterol 1998;93: Silvennoinen JA, Karttunene TJ, Niemela SE, et al. A controlled study of bone mineral density in patients with inflammatory bowel disease. Gut 1995;37: Kanis JA, Melton LJ, III, Christiansen C, et al. The diagnosis of osteoporosis. J Bone Min Res 1994;9: Bernstein CN, Blanchard JF, Leslie W, et al. The incidence of fractures among patients with IBD: A population-based study. Ann Intern Med 2000;133: Loftus EV Jr, Crowson CS, Sandborn WJ, et al. Long-term fracture risk in patients with Crohn s disease: A populationbased study in Olmsted County, Minnesota. Gastroenterology 2002;123: Vestergaard P, Mosekilde L. Fracture risk in patients with celiac disease, Crohn s disease, and ulcerative colitis: A nationwide follow-up study of 16,416 patients in Denmark. Am J Epidemiol 2002;156: Bernstein CN. Risk factors and prevalence of bone disease in inflammatory bowel disease. In: Williams CN, Bursey RF, Gall DG, Martin F, McLeod RS, Sutherland LR, Wallace JL, eds. Trends in inflammatory bowel disease therapy Dordrecht: Kluwer Academic Publishers, 1999: Kong YY, Penninger JM. Molecular control of bone remodeling and osteoporosis. Exp Gerontol 2000;35: Ghosh S, Cowen S, Hannan WJ, Ferguson A. Low bone mineral density in Crohn s disease, but not in ulcerative colitis, at diagnosis. Gastroenterology 1994;107: Schoon EJ, Blok BM, Geerling BJ, et al. Bone mineral density in patients with recently diagnosed inflammatory bowel disease. Gastroenterology 2000;119: Genarri C, Civitelli R. Glucocorticoid induced osteoporosis. Clin Rheum Dis 1986;12: Reid DM, Kennedy NSJ, Smith MA, et al. Total body calcium in rheumatoid arthritis; effects of disease activity and corticosteroid treatment. BMJ 1982;285: Bernstein CN, Blanchard JF, Rawsthorne P, Wajda A. The epidemiology of Crohn s disease and ulcerative colitis in a central Canadian province: A population-based study. Am J Epidemiol 1999;149: Metge CJ, Blanchard JF, Peterson S, Bernstein CN. Use of pharmaceuticals by IBD patients: A population-based study. Am J Gastroenterol 2001;96: Ardizzone S, Bollani S, Bettica P, et al. Altered bone metabolism in inflammatory bowel disease: There is a difference

5 AJG August, 2003 Corticosteroid Use and Fractures in IBD Patients 1801 between Crohn s disease and ulcerative colitis. J Intern Med 2000;247: Compston JE, Judd D, Crawley EO, et al. Osteoporosis in patients with inflammatory bowel disease. Gut 1987;28: Abitbol V, Roux C, Chaussade S, et al. Metabolic bone assessment in patients with inflammatory bowel disease. Gastroenterology 1995;108: Silvennoinen JA, Karttunen TJ, Niemela SE, et al. A controlled study of bone mineral density in patients with inflammatory bowel disease. Gut 1995;37: Schulte C, Dignass AU, Mann K, Goebell H. Reduced bone mineral density and unbalanced bone metabolism in patients with inflammatory bowel disease. Inflamm Bowel Dis 1998; 4: Fries W, Dinca M, Luisetto G, et al. Calcaneal ultrasound bone densitometry in inflammatory bowel disease A comparison with double X-ray densitometry of the lumbar spine. Am J Gastroenterol 1998;93: Bjarnason I, Macpherson A, Mackintosh C, et al. Reduced bone density in patients with inflammatory bowel disease. Gut 1997;40: Staun M, Tjellesen L, Thale M, et al. Bone mineral content in patients with Crohn s disease. A longitudinal study in patients with bowel resections. Scand J Gastroenterol 1997;32: Bernstein CN, Seeger LL, Sayre JW, et al. Decreased bone density in inflammatory bowel disease is related to corticosteroid use and not disease diagnosis. J Bone Miner Res 1995; 10: Robinson RJ, Al Azzawi F, Iqbal SJ, et al. Osteoporosis and determinants of bone density in patients with Crohn s disease. Dig Dis Sci 1998;43: Jahnsen J, Falch JA, Aadland E, Mowinckel P. Bone mineral density is reduced in patients with Crohn s disease but not in patients with ulcerative colitis: A population based study. Gut 1997;40:313 9.