Inflammatory bowel disease in the elderly

Ulcerative colitis and Crohn's disease both may occur in the elderly. In many populations, a second peak in the incidence of inflammatory bowel disease occurs near age 70. Clinical manifestations of inflammatory bowel disease in the elderly are generally similar to those seen in younger patients, although there is a tendency for both ulcerative colitis and Crohn's disease to involve more distal segments of the gut in older patients. Ischemic and infectious colitis, diverticulitis, and malignancy can all masquerade as inflammatory bowel disease in the elderly. Recent epidemiologic and clinical reports indicate that the outlook for older patients with inflammatory bowel disease is more favorable than previously suspected.     

Ileorectal anastomosis for inflammatory disease of the colon

Summary We have reported long-term results in the cases of 42 patients following total colectomy and ileorectal anastomosis for inflammatory bowel disease. In this group, 35 patients had Crohn's disease and seven had ulcerative colitis. Five of those seven patients with ulcerative colitis had carcinoma of the colon at the time of colectomy. A diverting loop ileostomy was constructed in 14 of the 35 patients who had Crohn's colitis at the time of operation, and none of these patients had any anastomotic leakage either before or after the ileostomy was closed. However, there were three patients with Crohn's colitis in whom anastomotic leaks developed postoperatively; all three patients died. In the group with ulcerative colitis, one patient had an anastomotic leak but there was no operative mortality. Of the 29 patients with Crohn's disease followed for one to 18 years, 12 (41 per cent) developed recurrences in the ileum and/or rectum, and seven of these patients had to have their anastomoses taken down. Read at the meeting of the American Society of Colon and Rectal Surgeons, New Orleans, Louisiana, May 2 to 6, 1976.     

The Cause of Death in Inflammatory Bowel Disease: A Comparison of Death Certificates and Hospital Charts in Rochester, New York

Objective: This study examines the causes of death from Crohn's disease and ulcerative colitis by comparing death certificates with hospital charts as part of an ongoing, community-based analysis in Rochester, NY. Methods: A registry of 1358 inflammatory bowel disease patients followed from January 1973 to December 1989 was analyzed for the cause of death by a study of death certificates as well as by a study of hospital records, including surgical pathology and autopsy records. A panel of physicians defined specific criteria for diagnosis, cause of death, and relation of death to inflammatory bowel disease. Results: One hundred and thirty patients (59 with ulcerative colitis and 71 with Crohn's disease) from the registry were found to have death certificates recorded by Monroe County during this period. There was an 80% concordance of the death certificate to the hospital record for the cause of death and its relationship to inflammatory bowel disease. Discordance was noted in cases of colon cancer and surgical complications. Conclusions: Sixty-eight percent of Crohn's disease and 78% of ulcerative colitis patients died from causes unrelated to their inflammatory bowel disease. Deaths caused by Crohn's disease decreased from 44% in the 1973–1980 period to 6% in the 1981–1989 period. Crohn's disease was it direct cause of death in 25% of the female patients, whereas only 6% of male patients died directly of Crohn's disease. Colorectal cancer caused 14% of the deaths in ulcerative colitis patients, three times more often than in Crohn's disease patients. Excluding cancer, there were only two deaths directly due to ulcerative colitis, both in the first 2 yr after diagnosis.     

Inflammatory bowel disease: An uncommon problem in Singapore

Fifty patients with inflammatory bowel disease (ulcerative colitis, 40; Crohn's disease, seven; indeterminate colitis, three) treated in one gastroenterology unit in Singapore over a 10 year period were reviewed. Clinical features were similar to those described in Western patients. Of the three main races of Singapore it was found that Indians are more susceptible to these diseases than Chinese or Malays. A survey of all gastroenterologists in Singapore indicated a possible prevalence of 8.6 per 100,000 people for ulcerative colitis and 1.3 per 100,000 people for Crohn's disease. These prevalence rates are much lower than those reported for Western populations.     

Favorable response to parenteral nutrition and medical therapy in Crohn's colitis. A report of 38 patients comparing severe Crohn's and ulcerative colitis

The courses of 38 patients with severe, uncomplicated acute colitis (16 with Crohn's colitis and 22 with ulcerative colitis) were analyzed retrospectively. The patients were placed on total parenteral nutrition and treated concomitantly with corticosteroids, antibiotics (often metronidazole), sulfasalazine, and/or azathioprine. Fifteen of the 16 Crohn's colitis patients were initially managed without surgery. Four patients subsequently relapsed, two responded to reinstituted medical therapy, and two underwent colon resection 2 and 4 years later. Of 22 ulcerative colitis patients, 16 required surgery during the initial hospitalization, one patient subsequently had surgery, and one died after refusing surgery. Three of the other four continue in remission on medical therapy. Thus, there were significant differences in this series between the clinical courses of severe ulcerative colitis and severe Crohn's colitis. While most of the ulcerative colitis patients with severe disease underwent colectomy, most of the patients with severe but uncomplicated Crohn's colitis responded to aggressive medical therapy, of which total parenteral nutrition and perhaps bowel rest seemed to be an important part. Afterwards, the majority remained in remission on long-term medical therapy.     

Inflammatory bowel diseases in Indo-Canadians with and without antineutrophil cytoplasmic autoantibodies.

A sequentially evaluated cohort of Indo-Canadians with either ulcerative colitis or Crohn's disease were prospectively examined for antineutrophil cytoplasmic autoantibodies (ANCA). Of 84 patients, 62 had ulcerative colitis and 22 had Crohn's disease. About one-third were born in Canada, and two-thirds were migrants from India or other countries, particularly East African nations. There was a disease-based and geographically based male predominance. The mean age of Canadian-born patients was significantly less than that of those born in other countries. Moreover, for migrants, the mean duration of residence in Canada before developing disease was 8.9 years for Crohn's disease patients and 13.5 years for ulcerative colitis patients. Moderate to severe disease was present; virtually all those with Crohn's disease had colonic involvement, and most of those with ulcerative colitis had extensive colonic disease. Overall, 40 of 84 (48%) were seropositive for ANCA, including a majority of those with ulcerative colitis but not Crohn's disease. In addition, eight had cytoplasmic ANCA, a reported seromarker for extensive colitis. Seropositive and seronegative patients were similar in age, sex, birth or duration of residence in Canada, site and severity of disease, familial history and complications, including pouchitis. This study supports the view that these diseases arise in individuals with a genetic predisposition following exposure to some, as yet unknown, environmental factor.     

Indeterminate colitis: the real story

PURPOSE: Up to one in five patients undergoing surgery for ulcerative colitis will have ambiguous histology, with features of both ulcerative colitis and Crohn's disease, and are categorized as having indeterminate colitis. We hypothesized that functional outcomes in indeterminate colitis patients undergoing ileal pouch-anal anastomosis are comparable with those of ulcerative colitis patients undergoing ileal pouch-anal anastomosis. METHODS: Physician-conducted interviews of 120 consecutive ileal pouch-anal anastomosis patients with a preoperative diagnosis of ulcerative colitis were reviewed, with a mean follow-up of 54 months. All colectomy specimens were reviewed by a single pathologist. Any changes in histologic diagnosis from ulcerative colitis to indeterminate colitis or Crohn's disease, frequency of postoperative complications, pouch function, and long-term postoperative medication usage were recorded. RESULTS: Although postoperative fistulas were more common in indeterminate colitis than ulcerative colitis (26 vs. 10 percent; P = 0.02, chi-squared), no indeterminate colitis patient required a permanent ileostomy as compared with six ulcerative colitis patients. Long-term functional results were similar. Overall, two-thirds of patients developed pouchitis. Ulcerative colitis and Crohn's disease patients were more likely to have had >3 episodes of pouchitis (58 and 72 percent) compared with indeterminate colitis patients (29 percent; P = 0.006, chi-squared). A greater number of Crohn's disease patients required maintenance oral antibiotic therapy (64 percent) to achieve satisfactory functional results compared with both indeterminate colitis and ulcerative colitis patients (20 and 28 percent; P = 0.014, chi-squared). CONCLUSIONS: Although ileal pouch-anal anastomosis patients with indeterminate colitis have more postoperative fistulas, long-term function is equal to that of ulcerative colitis patients and better than Crohn's disease patients. Ileal pouch-anal anastomosis should be offered to patients with indeterminate colitis and those with severe colitis in whom clear differentiation between indeterminate colitis and ulcerative colitis cannot be made.     

Long-term prognosis for patients with ulcerative proctosigmoiditis (ulcerative colitis confirmed to the rectum and sigmoid colon).

Despite its frequency, ulcerative proctosigmoiditis is poorly recognized and its prognosis is not well understood. This study reviews experience at the Cleveland Clinic from 1960 through 1972, with a 97% follow-up. A total of 359 patients with a diagnosis of ulcerative proctosigmoiditis were followed for a mean of 11 years (range 3.2--17.7 years). There were 219 men and 140 women. Mean age at diagnosis was 40.6 years. All had rectal bleeding, typical proctosigmoidoscopic findings of diffuse granularity, mucosal ulcerations and friability, and normal findings above the rectosigmoid on barium enema. The major complaint in all patients was rectal bleeding, and there were few other symptoms. Perianal disease was absent. On the whole, the patients were older than the usual patients with ulcerative colitis; 37 were older than age 61 at diagnosis. The disease remained localized and the prognosis accordingly was good for 90% of the patients. Extension to the proximal colon developed in 10% (36) of patients, and operation was required in 8% (29), two for toxic megacolon. Operation was carried out a mean of 5.6 years after diagnosis: 2.2% (8) of the patients died during the period of follow-up, six after operation. One patient had carcinoma of the sigmoid colon 12 years after diagnosis. Only one patient was subsequently found to have Crohn's disease. Thirty-nine patients died from causes unrelated to colitis. This study suggests that ulcerative proctosigmoiditis as defined by typical sigmoidoscopic appearance and a normal barium enema is relatively common, that proctosigmoidoscopic findings are identical with those of universal ulcerative colitis, but that the outlook is much better. Ulcerative proctosigmoiditis is rarely confused with Crohn's disease, and carcinoma is rare. Recurrences do occur, however, and proximal extension develops in 10% of patients, usually within 5 years after the diagnosis is first made.     

Specific gastroduodenoscopic findings in Crohn''s disease: Comparison with findings in patients with ulcerative colitis and gastroesophageal reflux disease

BACKGROUND: Crohn's disease patients often carry gastroduodenal lesions. However, few reports have addressed specific gastroduodenoscopic findings in Crohn's disease patients. METHODS: The gastroduodenoscopic findings of 63 Crohn's disease patients were examined. Those of 62 ulcerative colitis and 63 age- and gender-matched gastroesophageal reflux disease patients were also reviewed as controls. Findings of bamboo-joint-like appearance, gastric antral erosions, and duodenal lesions were the specific findings that were highlighted. RESULTS: Of 63 Crohn's disease patients, 47 (75%) had at least one of the specific gastroduodenoscopic findings, and the prevalence of these findings was significantly higher in Crohn's disease patients than in ulcerative colitis and gastroesophageal reflux disease patients (ulcerative colitis, 24/62, 39%; gastroesophageal reflux disease, 15/63, 24%, P < 0.0001). In particular, bamboo-joint-like appearance was almost unique to Crohn's disease patients (Crohn's disease, 28/63, 44%; ulcerative colitis, 3/62, 5%; gastroesophageal reflux disease, 0/63, 0%, P < 0.0001). Analysis of the relationship between the Crohn's disease patient's background and gastrodunodenoscopic findings revealed that both patients with disease affecting the ileum and those with previous gut operations were more likely to exhibit the specific gastroduodenoscopic findings (P = 0.030 and P = 0.043, respectively). CONCLUSION: Specific gastroduodenoscopic findings were observed in Crohn's disease patients. In particular, bamboo-joint-like appearance could be a unique marker of Crohn's disease.     

Lower gastrointestinal bleeding and ischemic colitis.

OBJECTIVE: To compare the incidence and clinical characteristics of lower gastrointestinal (LGI) bleeding due to ischemic colitis with those with LGI bleeding of other causes. METHODS: A chart review was performed of patients admitted with LGI bleeding to Sunnybrook and Women's College Health Sciences Centre, Toronto, Ontario, between July 1, 1997 and January 31, 2000. RESULTS: Of 124 patients with LGI bleeding, 24 cases were due to ischemic colitis, 62 to diverticulosis, 11 to inflammatory bowel disease (IBD) and 27 to all other causes ('others'). The average ages of patients in each group were 66.5, 76.5, 40.5 and 77.5 years, respectively. Patients with ischemic colitis were statistically younger than those with diverticular bleeding and 'others'. Patients with IBD were younger than those in the other three groups. The only statistical difference for vascular disease risks was hypertension, because of its absence from the IBD group. Three patients with ischemic colitis underwent blood transfusions, while 23 with diverticulosis, 15 'others' and none with IBD received blood. Three patients with ischemic colitis and one patient from the 'others' group died. More women (75) than men (49) had LGI bleeding - in total and within each subgroup. Of women with LGI bleeding, many more with ischemic colitis (44.4%) than with diverticulosis (3.0%), IBD (0%) or 'others' (5.6%) were taking estrogen. CONCLUSIONS: Ischemic colitis was the second most common cause of LGI bleeding. LGI bleeding from all causes was more common in women than in men. Many more women in the ischemic colitis group than in the other groups were using estrogen therapy.     

Fusobacterium varium localized in the colonic mucosa of patients with ulcerative colitis stimulates species-specific antibody

BACKGROUND: Microbial agents are a possible cause of ulcerative colitis. We have previously reported evidence of bacteria invading the colonic mucosa of patients with ulcerative colitis. We have isolated bacteria from inflamed colonic mucosa, examined the localization of the species in the mucosa, and assayed for serum antibodies to the bacteria. METHODS: Cohorts of 31 per group were enrolled from patients with active ulcerative colitis, Crohn's disease, ischemic colitis, and colon adenomas. A group of 31 healthy controls were also studied. The presence of bacteria in biopsies of patients with ulcerative colitis was analyzed by both isolation and immunohistochemistry. Sera from patients were tested for bacterial antibodies using both Western blots and enzyme-linked immunosorbent assay (ELISA). RESULTS: Only sera from patients with ulcerative colitis gave specific reactions with Fusobacterium varium in Western blot assays. The detection rate of specific bands was higher for patients with ulcerative colitis (61%) than for subjects with either Crohn's disease (13%) or healthy controls (29%) (P < 0.001 and P = 0.021, respectively). The ELISA showed that the mean optical densities with extracts of F. varium as antigen were significantly higher for ulcerative colitis patients than for subjects with either Crohn's disease or healthy controls (P < 0.001). Immunohistochemical detection of F. varium in colonic mucosa was significantly higher in patients with ulcerative colitis (84%) than for subjects with either Crohn's disease (16%) or other controls (3-13%) (P < 0.001). CONCLUSIONS: Fusobacterium varium bacteria were present in a significant number of patients with active ulcerative colitis, and should be tested in therapeutic trials in order to confirm the causal relationship between F. varium and ulcerative colitis.     

Incidence and clinical significance of lactose malabsorption in ulcerative colitis and Crohn's disease

The incidence of lactose malabsorption was investigated in 85 patients with ulcerative colitis and 71 patients with Crohn's disease by means of lactose tolerance tests and disaccharidase determinations in small intestinal mucosa. Eight patients with ulcerative colitis (9%) and four with Crohn's disease (6%) had lactose malabsorption. A control group displayed a similar incidence. It is concluded that lactose malabsorption is not particularly common in ulcerative colitis and Crohn's disease. If it is present, its aetiology seems to be unrelated to the intestinal disease.Transitional lactose malabsorption was detected in two cases during a relapse of ulcerative colitis.Institution of a lactose-free (or lactose-poor) diet was an important supporting measure in seven patients who were unaware of their milk intolerance, in particular in two with ileostomy. Therefore, it is recommended that a lactose tolerance test should always be performed in patients with ulcerative colitis or Crohn's disease.Twenty-one patients with ulcerative colitis and nine with Crohn's disease, none of whom had lactose malabsorption, were placed on milk-free diets. A beneficial effect was noticed in five of the patients with ulcerative colitis, and in three of those with Crohn's disease. The mechanism is unknown.Evidence is presented that milk allergy is not responsible for the beneficial effect of a lactosefree diet in patients with associated lactose malabsorption.     

Compared azathioprine efficacy in ulcerative colitis and in Crohn's disease

AIM: To compare the 6-month efficacy and tolerance of azathioprine in 68 patients with steroid-resistant or steroid-dependent chronic ulcerative colitis (n=30) or Crohn's disease (n=38).METHODS: Clinical remission was defined as a Crohn's Disease Activity Index<150 for Crohn's disease and number of non-bloody stools<=3/day for ulcerative colitis, associated with prednisone requirement<=10 mg/day.RESULTS: Seventy-three per cent of patients with ulcerative colitis had distal or left-sided colitis and 84% of patients with Crohn's disease had pancolitis. Azathioprine was discontinued early for side-effect in 8 (26.7%) patients with ulcerative colitis and in 8 (21.1%) patients with Crohn's disease (NS). In patients treated at least 6 months by azathioprine, clinical remission rates were 77.3% and 70% for chronic ulcerative colitis and Crohn's disease (NS). Complete corticosteroids weaning was obtained significantly more often in ulcerative colitis patients than in Crohn's disease patients (59.1% vs 30%; P<0.05).CONCLUSION: Azathioprine seems to be at least as effective and equally tolerated in steroid-resistant or steroid-dependent chronic ulcerative colitis or Crohn's disease patients.     

The natural history of corticosteroid therapy for inflammatory bowel disease: a population-based study

BACKGROUND & AIMS: The aim of this study was to determine the 1-year outcome after the first course of corticosteroids in an inception cohort of patients with inflammatory bowel disease. METHODS: All patients in Olmsted County, Minnesota, diagnosed with Crohn's disease (n = 173) or ulcerative colitis (n = 185) from 1970 to 1993 who were treated with systemic corticosteroids were identified (4 denied research authorization). Immediate outcome (30 days) and 1-year outcome after the first course of corticosteroids were determined. RESULTS: Seventy-four (43%) patients with Crohn's disease and 63 (34%) with ulcerative colitis were treated with corticosteroids. Immediate outcomes for Crohn's disease were complete remission in 43 (58%), partial remission in 19 (26%), and no response in 12 (16%). Immediate outcomes for ulcerative colitis were complete remission in 34 (54%), partial remission in 19 (30%), and no response in 10 (16%). One-year outcomes for Crohn's disease were prolonged response in 24 (32%), corticosteroid dependence in 21 (28%), operation in 28 (38%), and lost to follow-up in 1 (1%). One-year outcomes for ulcerative colitis were prolonged response in 31 (49%), corticosteroid dependence in 14 (22%), and operation in 18 (29%). CONCLUSIONS: Most patients with Crohn's disease and ulcerative colitis initially respond to corticosteroids. At 1 year, 32% of patients with Crohn's disease and 48% with ulcerative colitis are corticosteroid free without operation.     

Controlled trial of bowel rest in the treatment of severe acute colitis.

In a prospective, randomised clinical trial, 47 patients with severe, acute, non-infective colitis treated with 60 mg intravenous prednisolone daily, received either bowel rest with parenteral nutrition or oral diet. Although those who received 'bowel rest' experienced a reduction in daily stool weight, there were no differences in the operation or mortality rates between the groups. Fourteen of the 27 patients with ulcerative colitis, but none of the 16 patients with Crohn's disease required urgent surgery. Bowel rest did not affect the outcome in severe ulcerative colitis treated with intravenous prednisolone. Ulcerative colitis and Crohn's colitis behaved differently in the acute attack.     

Frequency and clinical evolution of indeterminate colitis: a retrospective multi-centre study in northern Italy. GSMII (Gruppo di Studio per le Malattie Infiammatorie Intestinali).

OBJECTIVE: To evaluate the prevalence and the clinical evolution of patients with an initial diagnosis of indeterminate colitis. DESIGN: Retrospective, observational study. SETTING: Fifteen gastrointestinal units in northern Italy. PARTICIPANTS: Patients with an initial diagnosis of indeterminate colitis seen between 1988 and 1993. INTERVENTIONS: Patients were traced through a common database and centres were requested to update their clinical follow-up. MAIN OUTCOME MEASURES: Frequency of patients with an initial diagnosis of indeterminate colitis among those with IBD; rate of patients who subsequently had a definite diagnosis of either Crohn's disease or ulcerative colitis. RESULTS: Fifty out of 1113 IBD patients (4.6%) had been diagnosed as having indeterminate colitis. During follow-up, 37 patients (72.5%) had a definite diagnosis of either Crohn's disease or ulcerative colitis. The cumulative probability of having a definite diagnosis of either ulcerative colitis or Crohn's disease was 80% 8 years after the first one (i.e. the first diagnosis). The probability of having a diagnosis of Crohn's disease was increased in patients with fever at onset, segmental endoscopic lesions or extra-intestinal complications and in current smokers. The probability of having a diagnosis of ulcerative colitis was increased in patients who had not undergone appendectomy before diagnosis. CONCLUSIONS: In our area, indeterminate colitis accounts for about 5% of initial diagnoses of IBD. In about 80% of patients, a diagnosis of either ulcerative colitis or Crohn's disease is made within 8 years. Several clinical and demographic features can help in identifying those patients more likely to have a subsequent diagnosis of Crohn's disease and those more likely to have a subsequent diagnosis of ulcerative colitis.     

Anergy to dinitrochlorobenzene and depression of T-lymphocytes in Crohn's disease and ulcerative colitis.

Skin reactivity to dinitrochlorobenzene (DNCB) and levels of circulating T-lymphocytes were measured in 15 patients with ulcerative colitis, 15 patients with Crohn's disease, and 12 normal control subjects. Diminished reactivity to DNCB was demonstrated in 87% of patients with Crohn's disease (P less than 0-001) and in 53% with ulcerative colitis (P less than 0-02), as compared with only 8-5% of controls; anergy was more frequent in Crohn's disease than in ulcerative colitis (P less than 0-05). Levels of circulating T-lymphoctes were also depressed in both Crohn's disease and ulcerative colitis (P less than 0-001) as compared with controls, with the values lower in Crohn's disease than in ulcerative colitis (P less than 0-02). There were no correlations of DNCB response with extent, duration, or severity of disease nor with T-cell levels within any patient group. These data provide further support for the concept that there is impairment of cell-mediated immunity among many patients with chronic inflammatory bowel disease, including both Crohn's disease and ulcerative colitis.     

DNA aneuploidy in Crohn''s disease and ulcerative colitis: results of a comparative flow cytometric study.

DNA ploidy and S-phase fractions were assessed by flow cytometry in colonic biopsy specimens from 28 patients with ulcerative colitis and 51 with Crohn's disease. Whereas only diploid DNA histograms were found in Crohn's disease and control subjects, three patients with ulcerative colitis exhibited DNA aneuploidy. In one case, aneuploidy was associated with low grade dysplasia. S-phase fractions were higher in ulcerative colitis (mean (SD) 17.8 (7.7)%) than in Crohn's disease (13.1 (4.6)%) or control subjects (14.2 (4.6)%), but did not correlate with either disease activity or duration in any group. In this study, aneuploidy was associated exclusively with ulcerative colitis, even in the absence of dysplasia. In view of the epidemiological differences in malignant colonic transformation between ulcerative colitis and Crohn's disease, this study suggests that flow cytometry may help to identify individuals with an increased cancer risk in ulcerative colitis.     

Split ileostomy and ileocolostomy for Crohn''s disease of the colon and ulcerative colitis: a 20 year survey

The clinical course of 140 patients who have had a split ileostomy for ulcerative colitis or colonic Crohn's disease over a 20 year period is reported. In 37 patients with ulcerative colitis there was no sustained improvement. In the 102 patients with Crohn's disease there was an immediate clinical improvement in 95, which was sustained in 65. Thirty patients have subsequently required a proctocolectomy for persistent inflammation, and 28 are still defunctioned. Bowel continuity was restored after 61 split ileostomies and in 44 patients intestinal continuity remains intact at the present time (mean follow up since closure = 62.5 months, range 0-231 months). It is concluded that a split ileostomy is a safe conservative operation producing at least temporary improvement in severely ill and malnourished patients with Crohn's colitis, and that if a subsequent resection becomes necessary it may be less extensive than was thought applicable at the initial operation. In 27 patients a resection has not been required.     

Direct hospital costs for patients with inflammatory bowel disease in a Canadian tertiary care university hospital

OBJECTIVE: We set out to determine the direct costs of hospitalizations of patients with Crohn's disease and ulcerative colitis admitted to a university-affiliated tertiary care hospital and to contrast the costs of medical versus surgical inpatient care, Crohn's disease versus ulcerative colitis, and to identify dominant components of inpatient costs. METHODS: We used a patient-specific case costing system at Saint Boniface General Hospital, Winnipeg, Manitoba, for fiscal years 1994 and 1995. We extracted all inpatients whose hospital discharge abstracts included ICD-9-CM codes 555 (Crohn's disease) and 556 (ulcerative colitis) among the top eight discharge diagnoses, and performed a chart review on all cases to ensure that the hospitalization was for inflammatory bowel disease and the diagnoses were accurate. We analyzed cases based on their disease diagnosis, primary mode of therapy associated with the hospitalization (medical vs surgical), and their major diagnosis-related group (DRG). This study evaluated direct patient care costs only and costs are expressed in Canadian dollars. RESULTS: Of 362 hospital admissions, 325 were eligible and of these admissions 275 belonged to the digestive system DRGs. Seventy-one (37%) were admitted more than once during the 2 yr of the study, accounting for 202 (62%) of the total number of admissions. The mean cost per admission of all cases of Crohn's disease was $3,149 (95% confidence interval [CI], $2,665-$3,634) and for ulcerative colitis was $3,726 (95% CI $3,008-$4,445). Surgical therapy cases accounted for 49.8% of all admissions, 57.8% of all hospital days, and 60.5% of all costs. Patients treated surgically had more costly hospitalizations than those treated medically, particularly when analyzing only nontotal parenteral nutrition (TPN) cases. Surgical treatment admissions were significantly more costly for ulcerative colitis digestive DRG admissions than Crohn's disease. The nondigestive DRG admissions were more costly than the digestive DRGs in all categories although this was only statistically different among medically treated Crohn's disease. Patients treated medically were similarly costly whether they had Crohn's disease or ulcerative colitis. There was no significant difference in cost per admission among cases admitted multiple times, compared with those admitted only once. TPN cases accounted for 9.5% of cases but 27.1% of costs. TPN-associated hospitalizations were more costly than non-TPN-use hospitalizations but these costs were primarily driven by duration of stay rather than TPN use itself. For all cases, the top five cost categories in descending order were nursing unit bed-days, drugs and pharmacy, diagnostic lab tests, operating room, and diagnostic imaging and endoscopy. CONCLUSIONS: Using our system we could determine direct costs for inpatients with inflammatory bowel disease and the factors that determined increased costs. Medical therapy admissions were similarly costly between Crohn's disease and ulcerative colitis; however, surgical therapy admissions were costlier among ulcerative colitis patients. Admissions for nondigestive DRGs were more costly than those for digestive DRGs. TPN use identified a sicker group of patients who remained in the hospital longer than nonusers and, not surprisingly, these were the costliest patients.