Use and Misuse of Endoscopy in Ulcerative Colitis Gary R Lichtenstein, MD Director, Center for IBD University of Pennsylvania School of Medicine Hospital.

Use and Misuse of Endoscopy in Ulcerative Colitis

Gary R Lichtenstein, MD

Director, Center for IBD

University of Pennsylvania School of Medicine

Hospital of the University of PA

Philadelphia, PA

Gary Lichtenstein, MDDisclosures

Research, Advisory, and/or Honorarium

Uses of Endoscopy in IBD

• Diagnosis• Disease extent• Prognostication• Assessment of Activity/Healing• Stricture evaluation and dilation• Dysplasia Surveillance• Diagnose/Control Bleeding• Pouch Evaluation• Endoscopic Ultrasound• Video Capsule Endoscopy

I. Diagnosis

• The Gold Standard for Diagnosis of UC (also Crohn’s colitis and ileocolitis)

A. Periappendiceal Red Patch

Rubin D, et al. Dig Dis Sci (2010) 55:3495–3501

• PARP is PARP is NOTNOT Crohn’s Disease Crohn’s Disease

• Present in 29/367 (7.9%) patientsPresent in 29/367 (7.9%) patients

•

•

A. Periappendiceal Red Patch

Rubin D, et al. Dig Dis Sci (2010) 55:3495–3501PARP- Periappendiceal Red Patch

I. Diagnosis

• Mimickers of Crohn’s ileitis and colitis

? Aphthous Ulcer ? Aphthous Ulcer

Yes Yes Diagnosis: Early Crohn’s DiseaseDiagnosis: Early Crohn’s Disease

• Exudate washes offExudate washes off

? Aphthous Ulcer ? Aphthous Ulcer

Yes Yes Diagnosis: Early Crohn’s DiseaseDiagnosis: Early Crohn’s Disease

• ExcavationExcavation

? Aphthous Ulcers ? Aphthous Ulcers

No – Lymphoid Aggregates No – Lymphoid Aggregates Diagnosis: Normal TIDiagnosis: Normal TI

• Exudate does not Exudate does not wash offwash off

? Aphthous Ulcers in the Colon ? Aphthous Ulcers in the Colon

No – Pseudomembranous ColitisNo – Pseudomembranous Colitis

• Exudate washes off• No underlying

excavation

I. Diagnosis

• Lymphoid aggregates mimic aphthous ulcerations

• Pseudmembranes in colon mimic aphthous ulcers– Lead to erroneous suggestion of the presence

of Crohn’s Disease in patients with Ulcerative Colitis

II. Need to BiopsyII. Need to Biopsy

Patient Presentation• Male, age 59 yrs. old College

Professor at a University in Louisianawell until 1 year ago

• Symptoms– Pain– Diarrhea (6-8 loose stools/day)– 5-lb weight loss

• Physical examination– Tender LLQ– No mass

• Laboratory values– WBC: 4,500 cells/µL– Hgb: 10.5 g/dL– CRP: 5.5 mg/dL

• Laboratory values– Albumin: 3.5 g/dL– Stool – C & S, C diff, O & P-

negative

• Colonoscopy– Erythematous, granular friable

mucosa from the anal verge to 35 cm

– No biopsies

• Rx– Prednisone 40 with taper 5mg

every week with symptom resolution

– At 20 mg a day diarrhea recurs.

CRP = C-reactive protein; LLQ = right lower quadrant.

Repeat Colonoscopy

• Colonoscopy: Loss of vascular pattern, granularity, pinpoint erosions, touch friability to 35 cm (left colon) with superficial ulcerations. Random biopsies obtained.

Need to BiopsyNeed to Biopsy

Qu Z, et. al . Human Pathology . 2009; 40, 572–577

Strongyloides Colitis Strongyloides Colitis

Endemic areas : - Appalachian region and Louisiana in the United States- Regions with large influx of tourists and emigrants from these endemic areas, southeastern Asia, and southern, eastern, and central Europe also have high incidence and prevalence of the disease .

Who Gets Disease: - The infection may remain clinically indolent.- When the host is immune-compromised, hyperinfection syndrome (i.e., larvae overload in

the lung and involvement of the rest of the gastrointestinal system) and

disseminated strongyloidiasis (i.e., involvement of other organs) occur with a mortality rate near 90% Qu Z, et. al . Human Pathology . 2009; 40, 572–577

Strongyloides Colitis Strongyloides Colitis

Qu Z, et. al . Human Pathology . 2009; 40, 572–577

Infectious Colitis that Mimics UC Infectious Colitis that Mimics UC

Rameshshanker R., et. al . World J Gastrointest Endosc 2012 June 16; 4(6): 201-211

No Randomized Trials

Smith GCS, Pell JP. Br Med J. 2003;327:1549.

Parachute use to prevent death and major trauma related to gravitational challenge: Systematic review of randomised controlled trials

CONCLUSION:Parachutes reduce the risk of injury after gravitational challenge, but their effectiveness has not been proved with randomised controlled trials

III. Disease Extent

• Assess extent of disease– when having active symptoms

• Mucosa may have complete endoscopic healing or remission with when in remission.

Disease Progressionin Ulcerative Colitis

Adapted from Miner PM. In Kirsner JB, ed. Inflammatory Bowel Disease, 5th edition. Philadelphia, Pa: WB Saunders Company; 2000.

1. Farmer RG, et al. Dig Dis Sci. 1993;38:1137-1146. 2. Stonnington CM, et al. Gut. 1987;28:1261-1266.3. Leijonmarck CE, et al. Gut. 1990;31:329-333. 4. Langholz E, et al. Gastroenterology. 1994;107:3-11.

5. Sinclair TS, et al. Gastroenterology. 1983;85:1-11.

Farmer1 Stonnington2 Leijonmarck3 Langholz4 Sinclair5

Number of patients

1116 182 1586 1161 537

Period of study 1960-83 1935-79 1955-84 1962-87 1967-76

Type of practice Referral Community Community Community Community

Initial extent < pancolitis

63% ~67% 63% 80% 89%

Extension From proctitis From left-sided colitis

46%70%

29% Not given Not given30% at 10

years

IV. Prognostication

• Assess histology to predict future probability of flare

• Mucosal healing predicts– lower colectomy rate– less steroid use in the future

• Treat individuals with potential for aggressive behavior with appropriately aggressive therapy.

Histologic Inflammation Predicts Relapse in UC Independent of Symptoms

Reproduced from Gut 32(2):174-178. Riley SA et al, 1991, with permission from BMJ Publishing Group Ltd.

Acute inflammatory cell infiltrate (N=27)

Crypt abscesses (N=27)

Mucin depletion (N=27)

Breaches in surface epithelium (N=27)

52%

25%

0

100

Infiltrate No infiltrate

56%

26%

Depletion Presence

78%

27%

Presence Absence

75%

31%

Presence Absence

Pat

ien

ts w

ith

re

lap

se

(%)

Pat

ien

ts w

ith

re

lap

se

(%)

P<0.02

P=0.02

P=0.1

P<0.005

Pat

ien

ts w

ith

re

lap

se

(%)

Pat

ien

ts w

ith

re

lap

se

(%)

80

60

40

20

0

100

80

60

40

20

0

100

80

60

40

20

0

100

80

60

40

20

Reprinted from Gastroenterology 120, Bitton A et al, Clinical, biological, and histologic parameters as predictors of relapse in ulcerative colitis, 13-20. Copyright

(2001), with permission from the American Gastroenterological Association.

Histologic Findings of Basal Plasmacystosis Predict Shorter Time to

Relapse in UCP

rop

ort

ion

of

Pat

ien

ts

in R

emis

sio

n

0

0.25

0.5

0.75

1

0 2 4 6 8 10 12Months of Study

Basal PlasmacytosisAbsencePresence

• Population-based cohort of IBD patients followed from 1990 to 1994 in Norway1

• Patients were treated with conventional therapies not including biologics1

• Among 495 pts available for analysis, mucosal healing observed at 1 year in 50% (UC) and 38% (CD)1

• In UC, mucosal healing was significantly associated with:

– less inflammation

– less corticosteroid treatment 5 years after diagnosis1

– fewer surgeries by 5 years1

• When f/u extended to 10 years:

– significantly fewer surgeries in patients with mucosal healing at 1 year2

Mucosal Healing Can Impact the Need for Surgery(IBSEN Study – Frøslie et al 2007, Solberg et al 2008)

25RGU11005

1. Frøslie KF, et al. Gastroenterology. 2007;133:412-422.2. Solberg IC, et al. Gut. 2008;57(Suppl II):A15. Abstract OP070.

Time in Years After 1-Year Visit

Pro

port

ion

of

UC

Pati

en

tsN

ot

Cole

cto

miz

ed

Patients with compromised mucosa 1 year after diagnosis showed a trend toward more surgeries. Frøslie KF, et al. Gastroenterology. 2007;133:412-422.

Patients without endoscopic activity at 1-year visit

Patients with endoscopic activity at 1-year visit

100

0 1 4 5 6 72 3 8

60

90

80

70

50

0

10

40

20

30

Impact of Mucosal Damage on Subsequent Colectomy in Ulcerative Colitis

P<0.05

Week 8 Mayo Endoscopy Subscore Predicts Corticosteroid-Free Symptomatic Remission at Week 30 During Anti-TNF Antibody Therapy- Results from ACT I and ACT II

Week 8 Mayo endoscopy

Subscore

Corticosteroid-free symptomatic

Remission, n/n (%)P value

0 30/65 (46) <.0001

1 35/102 (34)

2 8/71 (11)

3 2/31 (6.5)

Colombel JF, et al. Gastroenterology. 2011;141:1194-1201.

Mucosal Healing Correlates to Rate of Colectomy: Results from ACT 1 (Infliximab)

1 = MILD 2 = MODERATE 3 = SEVERE0 = NORMAL

Colombel JF, et al. Gastroenterology. 2011;141:1194-1201.

1.00

0.75

0.500 10 20 30 40 50

Time to colectomy or commercial infliximab use (weeks)

Pro

po

rtio

n

wit

ho

ut

cole

cto

my

or

com

mer

cial

in

flix

imab

use

Risk of Colectomy in Severe UC Patients with Severe Ulcerations

• Retrospective cohort• Prior to anti-TNF era • 85 consecutive patients with active UC• Severe endoscopic

Lesions (SELs):• Deep Ulcers• Well-like Ulcers• Large Mucosal Erosions• Extensive Loss of Mucosal

Layer with or withoutResidual Mucosal Areas

Carbonnel F, et al. Dig Dis Sci. 1994 Jul;39(7):1550-7.

93%

26%

No SEL

SEL

Colectomy 43/46 ptsColectomy 43/46 pts

Colectomy 9/39 pts

OR= 41, (95% CI 10.5-164)OR= 41, (95% CI 10.5-164)

Evolving Approach to Treating UC

Current: Modern Approach• Assessment of prognosis• “Optimization” of azathioprine/6-MP (dose or metabolites)• Earlier adoption of biologic therapy• Appreciation for the implications of a healed mucosa

Near Future Approach• Newer therapies with favorable safety and side-effect profiles • Individualized therapy based on genetics and physiology• Treatment to hard endpoints like mucosal healing or surrogates

of it• Disease monitoring to prevent relapse

V. Dysplasia Surveillance

Current ACG Surveillance Guidelines 2010Current ACG Surveillance Guidelines 2010(Secondary Prevention)(Secondary Prevention)

• Who:Who: left-sided or pan-UC more than 8-10 years (exception: PSC and UC- start immediately)

• Technique:Technique: random biopsies every 10 cm of mucosa; at least 33 biopsies; extra focus on nodules, masses, strictures

• How often: How often: q 6 months-2 years

• Outcome (reviewed by second pathologist): Outcome (reviewed by second pathologist): – High-grade dysplasia: colectomy

– Low-grade dysplasia: consider colectomy

– Indefinite dysplasia: increase surveillance?

– Atypia or indeterminate: treatment of active disease, repeat colonoscopy and biopsies

Kornbluth and Sachar, Ulcerative colitis practice guidelines (update). Am J Gastroenterol, 2010.

Myth

Most Dysplasia is Flat !

Random Biopsies Sample a Very Small Surface Area of the

Colorectum• Surface area of colorectum: 1578.1 + 301.0 cm2

• Surface area of biopsy forceps: 2.2-5 mm2

• Recommended “at least 33 biopsies”

• Percent surface area with this approach: 0.05%-0.1%

Sadahiro S. et al. Cancer, 1991.Rubin CE, et al. Gastroenterol, 1992.Kornbluth and Sachar. Am J Gastroenterol, 2004.

Biopsy Numbers Required in Dysplasia and Cancer Detection

Confidence Dysplasia Cancer

90% 33 35

95% 56 64

The World is not FlatDysplasia is Often Not “Invisible”!

• “Invisible”: indistinguishable from surrounding inflamed or quiescent mucosa

• “Visible”– Polypoid “adenoma-like” lesion– Irregular borders “spreading” lesion, not

endoscopically resectable (DALM)– Mass– Stricture

• Optical colonoscopy sensitivity (retrospective studies1,2): – Per lesion sensitivity: 61.6%-77.3%– Per patient sensitivity: 78.3%-89.3%

1Rutter MD, et al. Gastrointestinal Endoscopy, 2004 ;60:334–3392Rubin DT, et al. Gastrointestinal Endoscopy, 2007; 65:998-1004.3 Blonski W et al. Scand J. Gastronterol 2008;43(6):698-703.

Will New Technology Increase Detection of Neoplasia in IBD?

• High Definition Colonoscopes

• Chromoendoscopy– Dye spraying (Indigo Carmine, Methylene Blue)

– Narrow band imaging

• Magnifying endoscopy

• Fluorescence endoscopy

• Optical coherence tomography

• Confocal laser endomicroscopy

• Fecal DNA?

• Molecular assessment of biopsies?

Inflammatory polyps

Dysplasia Associated Lesion or Mass

Conventional Polyps: Endoscopic Features Suggesting Malignancy

Central Umbilication Firm (or hard) consistency when the head is pushed with a snare or forceps Satellite Lesions Irregular surface contour Focal ulceration Broadening of the stalk

Chromoendsocopy

Improves ability to detect lesions Improves ability to detect full extent of lesions Ability to differentiate neoplastic from non neoplastic lesions

Chromoendoscopy for Dysplasia in UC

Kiesslich R et al. Gastroenterol Clin N Am. 2012;41: 291–302

Type III, IV and V : are considered to be features of neoplastic lesionsKudo S, et al Gastrointest Endosc. 1996;44:95–96.

Modified Kudo Criteria

Chromoendoscopy : Indigo Carmine

Before (left) and after (right) application of Indigo Carmine.

Pit Patterns with Chromoendoscopy

Conclusion

Conclusion

Most dysplasia is raised- not flat Chromoendoscopy

Improves ability to detect lesions Improves ability to detect full extent of lesions Ability to differentiate neoplastic from non neoplastic lesions