A prospective study of the course of Crohn's disease

Vitamin B12 deficiency in inflammatory bowel disease: prevalence, risk factors, evaluation, and management

BACKGROUND

Management of vitamin B(12) (cobalamin, Cbl) deficiency in inflammatory bowel disease (IBD) is often not evidenced-based because of uncertainty on whether it causes enough malabsorption to result in clinical disease. This systematic review examines whether IBD predisposes to Cbl deficiency. We provide an approach to the management of abnormal Cbl values in IBD based on current literature and consensus-based guidelines.

METHODS

An extensive search of the Ovid MEDLINE and EMBASE databases by independent reviewers identified 42 articles with a total of 3732 patients evaluating Cbl deficiency in IBD.

RESULTS

Crohn's disease without ileal resection, regardless of disease location in the ileum, did not increase the risk for Cbl deficiency. Ileal resections greater than 30 cm were associated with Cbl deficiency in Crohn's disease, whereas those less than 20 cm were not. The effects of 20 to 30 cm resections were inconsistent. Ulcerative colitis did not predispose to deficiency. All studies failed to use confirmatory biomarker testing as stipulated by diagnostic guidelines for Cbl deficiency.

CONCLUSIONS

This literature does not support an association of Crohn's disease in general, regardless of ileal involvement, with Cbl deficiency. Only ileal resections greater than 20 cm in Crohn's disease predispose to deficiency and warrant treatment. Based on these findings, we suggest a diagnostic and therapeutic algorithm. All findings and recommendations require verification in further studies using confirmatory biomarkers as per diagnostic guidelines for Cbl deficiency. Serum Cbl levels alone are likely insufficient to diagnose deficiency in asymptomatic patients.

HM-PAO-labeled leukocyte scintigraphy in pediatric patients with inflammatory bowel disease

Leukocyte scintigraphy (LS) was performed in 20 pediatric patients with inflammatory bowel disease (IBD: 10 with ulcerative colitis, 2 with indeterminate colitis, and 8 with Crohn disease) in different stages of clinical activity. Leukocytes were separated from 15 to 60 ml venous blood and were labeled in vitro with [99mTc]HM-PAO. The segmental extent (small intestine; ascending, transverse, and descending colon; and recto-sigmoideum) of the process was determined by LS. The uptake of each bowel segment was scored in relation to the bone marrow uptake. The scintigraphic activity, calculated by summing the segment scores, was compared with laboratory parameters. The mean labeling efficacy was 76% (60-86%). The segmental extent of the process determined by LS was compared with the results of barium enema or colonoscopy with regard to 32 bowel segments. The sensitivity, specificity, and accuracy of LS were 93, 88, and 91%, respectively. Two extraintestinal manifestations (abdominal abscess and joint involvement) were also detected by LS. These lesions were verified by computed tomography (CT) (abscess) and on the basis of the clinical outcome (arthritis). The scintigraphic activity correlated with the C-reactive protein (CRP) level (r = 0.82, p < 0.001), the alpha 2-globulin level (r = 0.63, p < 0.02), the sedimentation rate (r = 0.51, p < 0.05), and the fS iron level (r = -0.66, p < 0.005). LS is applicable in pediatric patients. The method is an excellent technique for assessment of the extent of IBD in children. Extraintestinal manifestations of IBD can also be investigated by LS. The scintigraphic activity is a useful parameter for determination of the activity of IBD in children.

Quantification of disease activity in Crohn's disease by computer analysis of Tc-99m hexamethyl propylene amine oxime (HMPAO) labelled leucocyte images

The reliability and clinical applications of computerised image analysis measurement of bowel uptake of Tc-99m HMPAO labelled leucocytes has been examined as a measure of disease activity in Crohn's disease. In 54 studies carried out on 33 patients with established Crohn's disease, the mean 'scan score', a quantitative assessment of image intensity, was 82.1 SEM (13.6), in patients with clinically active disease compared to 24.7 (7.0) in those with quiescent disease, p < or = 0.0005. A significant correlation was found between the scan score and Crohn's Disease Activity Index (rs = 0.52, p < 0.0001), and Harvey and Bradshaw Simple Index (rs = 0.4, p < 0.004). A low scan score correctly identified seven patients whose raised Crohn's Disease Activity Index incorrectly indicated active disease because symptoms used in calculation of the index were not caused by active inflammation. Of the laboratory measurements, the scan score correlated with the haemoglobin (rs = 0.66, p < 0.0001), albumin level (rs = -0.6, p < 0.0001), C-reactive protein (rs = 0.7, p < 0.0001), alpha-acid glycoprotein (rs = 0.57, p < 0.001), and platelet count (rs = 0.47, p < or = 0.006), but not with the erythrocyte sedimentation rate (rs = 0.2, p < or = 0.25). The scan score was raised in all patients who had clinically active disease but normal laboratory tests. The results of this study indicate that the scan score provides an objective indicator of disease activity in Crohn's disease which may be superior to clinical indices, and also to laboratory tests which although objective are often normal in the presence of active disease.

Prediction of surgery for obstruction in Crohn's ileitis. A study of 64 patients

The charts of 64 patients with Crohn's ileitis were reviewed to determine what characteristics at the time of first observation at a specialized center were predictive of surgery for irreversible obstruction, using Cox's proportional hazard model. Individual variables which proved significant (P less than 0.01) included age at presentation to the clinic, serum albumin, sedimentation rate, and the presence of radiologic strictures. Patients eventually requiring surgery were older than the rest because of a later date of symptom onset; furthermore, patients with strictures on initial x-rays had not been ill for a longer period of time than those who did not have strictures. These data suggest that fibrosis and stenosis are characteristic of a subgroup of Crohn's patients, rather than being inevitable phases of disease evolution. Multivariate analysis revealed that patients with low serum albumin, high sedimentation rate, and roentgenologic stenosis at first observation were ten times as likely to require surgery for obstruction at 10 years than those without any of these three unfavorable parameters.

Specificity of indium-111 granulocyte scanning and fecal excretion measurement in inflammatory bowel disease--an autoradiographic study

The validity of 111In granulocyte scanning and fecal excretion measurement, as a reflection of loss of cells into the gastrointestinal tract, was studied using an autoradiographic technique in 11 patients in whom 111In granulocyte scan and colonoscopy were carried out simultaneously. 111In granulocytes were injected 1.5-4 hr prior to colonoscopy, and intraluminal fluid, mucosal brushings, and colonic biopsies were collected during the colonoscopy. In two patients with no histological evidence of inflammatory bowel disease, and four patients with clinically and histologically inactive inflammatory bowel disease, no 111Indium was detected in fluid, brushing, or biopsies. In five patients with active disease, 85% of the 111In activity in colonic fluid was precipitated by low-speed centrifugation. Autoradiography confirmed that the label remained attached to whole granulocytes in colonic fluid and mucosal brushings. Studies on biopsies, at intervals up to 4 1/2 hr following labeled granulocyte injection, demonstrated labeled polymorphonuclear neutrophils (PMNs) on the inflamed epithelial surface, with occasional cells in crypt abscesses by 110 min. We conclude that the techniques of 111In granulocyte scanning and fecal counting in patients with IBD are specifically measuring cell loss; labeled PMNs are capable of migrating through the gastrointestinal mucosa, in active disease, within 2 hr of administration.

Relationship between clinical and laboratory parameters and length of lesion in Crohn's disease of small bowel

The correlation of 22 commonly used clinical and laboratory abnormalities with linear extent of the lesion was studied in 70 patients with Crohn's ileitis, 16 of whom had inactive disease and 54 active disease. Extent was measured radiologically using a well-validated double-contrast technique. In the patients with active disease, lesion length was significantly correlated with weight loss, serum albumin, total protein, and serum iron. In the group without active inflammation, pain and abdominal mass were significantly correlated with lesion extent. No correlation was found between linear extent of lesion and the following: an index of inflammatory activity (New Crohn's Disease Activity Index), several acute-phase reactants, and the components of the complete blood count. Only total protein and serum iron had a significant regression coefficient following a procedure of stepwise regression. No mathematical model was found capable of satisfactorily predicting the length of lesion.