The diagnostic accuracy of colonic ultrasound, to problem solve suspected colon cancer - PowerPoint Presentation

Slide1

The diagnostic accuracy of colonic ultrasound, to problem solve suspected colon cancer following diagnostic CT. Can we reduce the need for Colonoscopy?

Gangi A, Ball C, Beable R, Higginson, A. Queen Alexandra Hospital, Portsmouth

Background

~ 700,000 adults are investigated in England per year for suspected colorectal cancer.In symptomatic patients:Colonoscopy is the gold standard.CT colonography (CTC) is the gold standard when radiological imaging of the whole colon is required.

Alternative investigations are not in routine clinical use.

Background

CTC+/- colonoscopy at times can be limited byResidual faeces and luminal fluidSuboptimal luminal distensionSegmental colitis

Complex diverticular disease

Patient discomfort

At our centre we carry out transabdominal ultrasound

to:

problem

solve cases where CTC is indeterminate

avoid colonoscopy and proceed directly to surgery

select

patients for endoscopic mucosal resection (EMR)

Aims

To determine the diagnostic accuracy of focused ultrasound in characterisation of colonic pathology in patients following abdominal CT or CT colonography.

Method

Consecutive patients from 2010- 2017 were extracted from a database by patient number and included in this preliminary retrospective study. Inclusion criteria:

Patients

with suspected colonic

cancer

A CT abdomen/CTC followed by a colonic US carried out within 3

months

Histopathology

, colonoscopy or 2 year clinical follow up

results

Method

Referral pathway

CT colon/CT abdomen

Colonoscopy

Colonic US

CT colon/CT

abdomen

Colonoscopy

Colonic US

+/-

Surgery

Method

Ultrasound technique:2 GI radiologists (25 years total experience; >10,000 bowel US)Toshiba Aplio 500, Canon

Aplio

i800

Low (3-5MHz) and high (5-17MHz) frequency probes

Overall assessment

Assessment from caecum to

sigmoid

Bowel wall layer assessment

Loss of submucosal layer

Hypoechoic echotexture

Superb micro-vascular ultrasound (

SMI) -Initial experience

Results

44 patients selected from the database met the criteria and were reviewed (22 female; mean age 74)Reference standard: 29/44 patients had histopathological results15/44 had 2+

years clinical/radiological

follow up

Results

Ultrasound results:Overall, US correctly evaluated for colonic cancer in 42 of 44 (95%)25/44

patients had a final diagnosis of colonic

cancer. US correctly identified 25/25 (100%).

3/44

had tubulovillous adenoma (TVA).

US correctly identified

1/3 (33.3%).

Technically inadequate in 1

case due to body habitus

Both

US and CT missed a

TVA

with adenocarcinoma transformation in a complex diverticular disease patient.

Results

Ultrasound results:

Sensitivity of 93%, specificity 100%, PPV 100%, NPV 89%,

Results

CT results:In 10/44 (23%) US had a greater diagnostic accuracy than CT7 indeterminate2 false positive

1 false negative

Indeterminate:

3 x technical factors (untagged faeces, sub-diagnostic)

4 x focal or eccentric thickening

2 of which had adenocarcinoma, 1 TVA

Transverse colon adenocarcinoma

Figure

1

(a

)–(b):

CTC: T2 circumferential mid distal transverse colon tumour.

US: T3b transverse colon tumour with anterior abdominal glide.

Histopathology confirmed a pT3 adenocarcinoma.

b)

a)

TVA

Figure

2

(a

)–(b): CTC

: 2cm

C-shaped lesion within the proximal sigmoid colon. US: a focal lesion with appearances in keeping with a TVA which was confirmed at histopathology.

a)

b)

Untagged faeces

Figure

3

(a

)–(b): US vs CTC. CTC: untagged faecal debris in caecum giving impression of a rolled edged mucosal lesion. US: normal bowel wall layers.

Clinical follow up and CTC 4 years later confirmed this.

a)

b)

Diverticulosis

Figure 4

(a

)–(b): CTC: Thickening of the sigmoid colon ? Diverticular or cancer. US: Resolving inflammatory change around the sigmoid colon.

Confirmed with colonoscopy and

biopsy to be sigmoid diverticulosis.

a)

b)

Diverticulitis

Figure 5

(a

)–(b): CTC:

Ascending colon

colitis. US: ulcerating tumour within the diverticular segment.

Confirmed with

histopathology to be adenocarcinoma.

a)

b)

Results

Colonoscopy results:17/44 (39%) patients had a colonoscopy as part of the investigation work up. In 8, colonoscopy pre USIn 9, colonoscopy post US

Of these 11/17 (65%) were non diagnostic :

Not

safe to proceed x1

Insufflation problems x1

Did not tolerate

x3

Residual

stool/poor prep x3 (

3x for 1 patient)

Unable to pass through inflammatory stricture x3

Discussion

Financial considerationsUS <20 mins = £30Colonoscopy = £440Only 39% of patients had colonoscopyCost if all had colonoscopy = £19,360Cost for 17 colonoscopy + 27 US = £8,260

Which is a saving of £11,100

From 2010- 2017 there are projected to be a total of 250 patients who meet the inclusion criteria

Projected saving of £62,525

Key findings

Colonic ultrasound has a high sensitivity of 93% and specificity 100%.

Higher diagnostic accuracy

than CTC

in complex patients.

There is a financial advantage to carrying out a problem solving

colonic US

rather than colonoscopy.

Conclusion

Colonic ultrasound in the hands of a GI specialist radiologist has high diagnostic accuracy.In suspected colon cancer, we have demonstrated that colonic US

has a role in

problem

solving

where colonoscopy has failed or

in complex/technically

inadequate

CTC examinations.

At our hospital,

CTC + coloni

c US

is used to confirm colon cancer thereby allowing patients to go

directly to surgery

.

A larger retrospective cohort or prospective study is needed to confirm these findings.

Thank you.

Anmol.Gangi@porthosp.nhs.uk