NON INVASIVE MONITORING OF MUCOSAL HEALING IN IBD. THE ROLE OF BOWEL ULTRASOUND Fabrizio Parente Gastrointestinal Unit, A.Manzoni Hospital, Lecco & L.Sacco School of Medicine,University of Milan - Italy
Mucosal imaging in IBD ITEMS Endoscopic visualization of ileo-colonic mucosa has a prognostic role during severe flare up of disease both in CD and UC. Mucosal healing after short-term medical therapies, as assessed by endoscopy, seems to be associated with a better prognosis in CD and, especially, in UC. Can Bowel Ultrasound be used as a surrogate of endoscopy in assessing mucosal response to medical therapies in UC?
Endoscopy and severe Ulcerative colitis Colonoscopy is feasible and safe in over 90% of severe attacks or flare-up of ulcerative colitis Severe endoscopic activity is based on: Deep ulcers Extensive loss of mucosal layer with or without residual mucosal areas Well-like ulcers Large mucosal erosions Colectomy was performed in 43/46 patients with severe endoscopic colitis as compared with only 9/39 with moderate endoscopic colitis (OR 41, 95%CI 10,5-164) Carbonnel F, et al. Dig Dis Sci 1994;39:1550-7
Endoscopy and severe ulcerative colitis Deep ulcers and severe endoscopic rectal inflammation at admission are significantly more common in patients who subsequently will require colectomy. Severity of endoscopical picture constitutes one of the most important predictors of refractoriness to intravenous steroids. Travis S, et al. Gut 1996;38:905-10
Endoscopy and severe ulcerative colitis Probability of colectomy-free survival according to the endoscopic severity of ulcerative colitis at presentation. Statistical significant difference (p=0,0001) Hazard Ratio 18,1 (95%CI 2,7-21,2) Daperno M, et al. Dig Liver Dis 2004;36(1):21-8
Endoscopic criteria of severity (ECS) according to anatomical severity (ACS) of colonic Crohn s disease ACS+ group ACS- group No. of patients 68 10 ECS 1 32 0 ECS 2 65 2 ECS 3 34 0 At least one ECS 95% 20% Nahon, Am J Gastroenterol 2002
Mucosal healing with biological therapies in CD By courtesy of Prof. Rutgeerts D Haens Van Assche, Leuven, Belgium
Endoscopic healing in CD reduces the risk of surgery and hospitalisation ACCENT I patients who demonstrated healing at one or both control visits had a lower incidence of hospitalisation or surgery *only intra-abdominal surgeries analyzed Rutgeerts Gastrointest. Endoscopy 2006; 63: 433-42
Mucosal healing in ulcerative colitis
Cumulative relapse risk based on mucosal healing at endoscopy after 5-ASA treatment Meucci G, et al. DDW 2006
Clinical remission of UC after 30 weeks according to mucosal healing at 8 weeks with Infliximab Rutgeerts P, et al. N Engl J Med 2005;353:2462-76
Long-term risk of colectomy according to mucosal healing after short-term treatment with Infliximab Kohn A, et al. Aliment Pharmacol Ther 2007; 26: 747-756
MUCOSAL HEALING IN ULCERATIVE COLITIS The key question is whether the mucosal healing itself is able to improve the outcome of the disease, and therefore if the clinical evaluation alone is not sufficient to monitor UC patients in clinical practice. Clear clinical guidelines based on endoscopic followup in UC are not yet available and, moreover, patients are reluctant to be re-endoscoped during follow-up because of the invasiveness of colonoscopy. Can bowel US be employed as surrogate of colonoscopy in evaluating mucosal healing in ulcerative colitis?
Bowel wall thickening seen at US in the cross section and longitudinal scans (ileal CD)
Current applications of Bowel ultrasound in IBD Primary imaging technique in patient with clinically suspected IBD (especially Crohn disease where conventional bowel US has sensitivity and specificity of 78-94% and 79-100%). Evaluation of anatomical distribution and extension of lesions along the bowel in an already known IBD both at primary diagnosis and during follow-up. Detection of abominal complications of IBD (especially strictures, fistulas and abscesses in CD, where sensitivities are 86%, 71% and 90%,respectively). Evaluation of local disease activity (favourable results with the aids of Power Doppler function and iv contrast agents).
Bowel US in active ulcerative colitis COLONIC WALL FEATURES Thickening: 5-8 mm Echopattern: stratified or hypoechoic (severe activity) Vascularity: often hypervascularization in flare-up Haustration: often absent SITE AND EXTENSION Rectum always involved (difficult to see at US) No skip areas Colite Ulcerosa EXTRAINTESTINAL ALTERATIONS Fibrofatty mesenterial proliferation: rare Lymph adenopathy: rare Worlicek et al. 87; Arienti et al. 99, Parente et al. 02
Significance of predominantly hypoechoic echopattern: correlation with histology and vascularity as detected by Doppler-US (resected colonic specimens) Parente, Alim Pharmacol & Ther 2005
Bowel vascularity, as assesssed by power Doppler US, in determing local inflammatory disease activity
Thickened bowel segment with greatly increased intramural blood flow as depicted by high definition Doppler US (e-flow technique)
Thickening of colonic walls with stratified echopattern and disappearance of haustra coli sigmoid and descending colon. Cross section Longitudinal US section
Complete disappearance of colonic haustration with thickening of colonic wall in a patient with UC flare-up (descending colon)
Parente F, Aliment Pharmacol Ther. 2003;18:1009-16
Thickening of rectal walls (visualized throughout the bladder window) in a patient with severe ulcerative proctitis
Relationship between US activity index, MRI activity index and endoscopic activity index (Baron) in UC Pascu M, Inflamm Bowel Dis. 2004;10:373-82
Endoscopical activity index of ulcerative colitis (Baron score) and US activity index of disease based on degree of colonic bowel thickening and presence of intramural blood flow at power Doppler US. Endoscopical activity index US activity index
US-activity index before and after 8 weeks of high dose steroids in pts with severe UC flare up Parente et al,uegw Berlin 2006
Study design Population: 83 patients with severe UC (all had moderate to severe disease graded 2-3 according to the Baron score) requiring high-dose steroids were recruited. Time 0 3 month 9 month 15 months UC pts Endoscopy Endoscopy Endoscopy Endoscopy Bowel US Bowel US Bowel US Bowel US Lab Lab Lab Lab Visit Visit Visit Visit
Severe: Truelove e Witts classification of ulcerative colitis Diarrhoea: 6 or more motions per day, with blood Fever: mean evening temperature > 37.5, or any time of day over 37.7 on at least 2 out of 4 days Tachycardia: mean pulse rate over 90 per minute Anemia: Hemoglobin of 75% or less, allowing for recent transfusion ESR: more than 30 mm in 1 hour Mild: Mild diarrhoea: less than 4 motions per day, with only small amounts of blood No fever No tachycardia Mild anemia ESR < 30 mm in 1 hour Moderately Severe Intermediate between mild and severe
Baron score for ulcerative colitis Grade 0 Lack of lesions. Normal vascular pattern Grade 1 Hyperemic mucosa. Mucosa granularity. Mucosal oedema. Indistinct vascular pattern. Grade 2 Mucosa friability. Contact bleeding. Superficial ulcerations. Free luminal blood or pus. Grade 3 Deep or coalescing ulcers. Spontaneous bleeding.
Material and methods Endoscopic severity of UC was graded 0-3 according to Baron score, US severity was also graded 0-3 according to the maximum colonic wall thickening and vascularity (grade 0: CTW < 4mm; grade 1: CTW 4-6mm; grade 2: CTW 6-8 mm,grade 3: CTW > 8 mm) Patients clinically responsive to steroids (slowly tapered within 4 mths) and then maintained on mesalazine 1.6-2.4 g/day were followed-up for 15 mths with repeated colonoscopy and bowel US at 3, 9 and 15 mths from entry. Statistical analysis Concordance between clinical, endoscopic and US scores at 0, 3, 9 and 15 mths was determined by kappa statistics. Unconditioned multiple regression analysis was used to assess predictivity of Truelove, Baron and US scores measured at 3 and 9 mths on the development of a UC relapse (Baron score 2-3) at 15 mths.
Distribution of 74 UC patients at baseline according to various selected variables
Concordance between Baron score (gold standard) and Truelove score at 3 months Baron score at 3 mths Truelove score at 3 months Frequency mild moderately severe severe Total 0-I 48 3 0 51 II 7 6 0 13 III 4 6 0 10 Totale 59 15 0 74 Statistics Value ASE 95% Confidence Intervals Kappa 0.3486 0.0911 0.1700 0.5272 Weighted Kappa 0.3795 0.0844 0.2140 0.5450
Concordance between Baron score (gold standard) and Truelove score at 15 months Baron score at 15 mths Truelove score at 15 mths Frequency mild moderately severe severe Total 0-I 48 1 0 49 II 6 6 1 13 III 1 5 6 12 Total 55 12 7 74 Statistics Value ASE 95% confidence intervals Kappa 0.5923 0.0858 0.4242 0.7604 Weighted Kappa 0.6868 0.0734 0.5430 0.8306
Concordance between Baron score (gold standard) and Ultrasound score at 3 months Baron score at 3 mths US score at 3 mths Frequency 0-1 2 3 Total 0-I 49 2 0 51 II 2 6 5 13 III 0 3 7 10 Total 51 11 12 74 Statistics Value ASE 95% Confidence intervals Kappa 0.6600 0.0774 0.5084 0.8117 Weighted Kappa 0.7622 0.0578 0.6488 0.8755
Concordance between Baron score (gold standard) and US score at 15 months Baron score at 15 mths US score at 15 mths Frequency 0-1 2 3 Total 0-I 48 1 0 49 II 0 12 1 13 III 0 3 9 12 Total 48 16 10 74 Statistics Value ASE 95% confidence intervals Kappa 0.8677 0.0542 0.7614 0.9740 Weighted Kappa 0.9042 0.0398 0.8263 0.9821
Ulcerative colitis Follow-up - Activity SEVERE ACTIVE DISEASE QUIESCENT DISEASE Increased wall thickness (> 6 mm) Hypoechoic echopattern of colonic walls Hypervascularization (increased Doppler sig.) Disappearance or reduction of haustration Normal wall thickness Stratified echopattern of colonic wall
Distribution of 74 patients with UC by Baron score at 15 months according to age, sex and various selected covariates. ^Estimated by unconditional multiple logistic regression models after allowance for smoking status (backward selection).
Distribution of 74 patients with UC, by Baron score at 15 months according to Baron and US score after 3 and 9 months. ^ Estimated by unconditional multiple logistic regression models after allowance for smoking status and appendicectomy (backward selection).
Conclusion I We confirm that not achieving complete mucosal healing after short-term treatment with high-dose steroids is predictive of negative outcome at 15 months. Therefore, the most important objective of any pharmacological therapy for active UC should be achieving endoscopic remission. Severe US score after short-term medical therapy is almost as accurate as endoscopic score in predicting negative disease outcome at 15 months.
Conclusion II Therefore, bowel US may constitute, in expert hands, an accurate surrogate of colonoscopy in order to assess response to treatment and may therefore be preferred to endoscopy in clinical practice, with the exclusion of diseases confined to the rectum only.
Thank you Dr Roberto Bianco Dr Laura Norsa Consultants Radiologist Dr Gianluca Sanpietro Consultant Surgeon Dr Pietro Zerbi Lecturer in Hystopathology
Conclusion Response to high-dose steroids based on clinical evaluation only at 3 mths from starting therapy fails to detect those steroid-dependent UC pts who still have endoscopic activity and are therefore at high risk of recurrence at 15 mths. Only bowel US at 3 mths, other than colonoscopy, may identify this subset of pts who require a more aggressive medical approach.
Oral-contrast enhanced bowel US
RESULTS Considering endoscopy as the reference test for good response to steroids, in the three visits we showed inconsistent concordance between 0-I Baron scores and Truelove score (weighted K ranged between 0.38 and 0.94), but high and consistent concordance between 0-I Baron scores and US scores (weighted K between 0.76 and 0.90).
RESULTS On logistic regression analysis, patients with high Baron score (2-3) at 3 months, regardless of their Truelove score, had a high risk of severe endoscopic activity at the 15 th months (OR 5.2; 95% CI: 1.6-17.6). Similarly, a significant direct association, was also shown between severe US scores (2-3) at the 3 rd and at the 15 th month (OR 9.1; 95% CI: 2.5-33.5);