Comparing Current Screening Modalities for Colorectal...

Comparing Current Screening Modalities for Colorectal

Cancer and Precancerous Lesions: Is Colonoscopy

the Method of Choice?

Puneet K. Singh*

Saba University School of Medicine, Saba, Dutch Caribbean

*Corresponding author: Puneet K. Singh; [email protected]

Colorectal cancer (CRC) is the third most common form of cancer and the second leading cause of cancer death in the Western

world. Presently, screening tools such as colonoscopy, sigmoidoscopy, fecal occult blood test (FOBT) and computed tomographic

colonography (CTC) are available for CRC screening. The debate over which screening tool is most effective in detecting CRC and

precancerous lesions is ongoing. Many recent studies have identified colonoscopy as the most sensitive and specific screening

modality for CRC. However, a number of factors have prevented colonoscopy from being widely accepted. Less invasive techniques

such as sigmoidoscopy and CTC are growing in popularity among physicians and patients who are apprehensive about colonoscopy

screening; although many still are yet to experience the procedure first-hand. This literature review will attempt to validate the

growing theory that colonoscopy is superior to other modalities for the diagnosis and screening of CRC and reduces the risk of CRC

mortality. In order to do so, the paper will compare the risks and benefits of colonoscopy to sigmoidoscopy and CTC. It will further

look at the different aspects that encompass a patient’s decision to partake in screening, such as basic knowledge about CRC, history

of CRC in the family, advice from physicians and individual beliefs about what screening entails. Finally, this paper will propose ways

in which colonoscopy screening can be improved and thus surpass other screening modalities to universally become the first choice

for CRC screening.

Keywords: colonoscopy; colorectal neoplasms; sigmoidoscopy; CT colonography; mass screening.

INTRODUCTIONolorectal cancer (CRC) is the third most commonform of cancer and the second leading cause of

cancer death in the Western world, equally affectingboth men and women.1 In 2012, the United States hadan estimated 143,460 individuals diagnosed with CRCand 51,690 related deaths.2 The vast majority of CRCswithin North America are sporadic with fewer than5% directly related to chronic inflammatory diseases orhereditary causes of CRC, such as familial adenomatouspolyposis (FAP) and hereditary non-polyposis coloncancer (HNPCC).1 Sporadic CRC is due to mutationscausing histological changes within the luminal aspectof colonic mucosa which slowly progress to benignadenomatous polyps of varying types: tubular, tubulo-villous and villous.3 These precancerous lesions canincrease in size, become dysplastic and eventuallytransform into overt carcinomas. The slow progressionof these changes causes age to be one of the greatestrisk factors for CRC. It is estimated that 90% of allCRC cases occur after the age of 50 in both men andwomen.4 Along with family history and age, othersignificant risk factors for CRC include obesity, tobacco

and alcohol abuse, stress, inflammatory bowel diseases(e.g., ulcerative colitis) and diet.3 With its long list of riskfactors and worldwide prominence, it is imperative thathealth care providers and patients become moreknowledgeable about CRC and the ways in which todetect its precursor lesions at early and docile stages.

A number of different techniques are currently em-ployed to screen for polyps and CRC. Epidemiologicalstudies have shown a decline in the incidence andmortality of CRC over the years, which is primarilyattributed to increases in screening test use.5 Specificguidelines outlining which tests should be used andwhen they should be administered have been estab-lished by a number of prominent medical societies andorganizations. The United States Preventative ServicesTask Force (USPSTF) recommends three main screeningmethods: high-sensitivity fecal occult blood test (FOBT)annually, flexible sigmoidoscopy every 5 years withFOBT every 3 years or colonoscopy every 10 years.5 TheAmerican Cancer Society and The American Collegeof Physicians’ (ACP) recommendations mirror those ofUSPSTF. These bodies also agree that patients with one

Review Article

MSRJ # 2014 VOL: 04. Issue: Fall

epub September 2014; www.msrj.org

Medical Student Research Journal 034

http://www.msrj.org

or more first-degree relatives with CRC, or a heredi-tary syndrome that predisposes them to CRC, shouldreceive screening in the second or third decade of life,6

while in average-risk patients, screening should bedone between the ages of 50 and 75.7 There is a strongbelief that screening after the age of 75 may no longerbe beneficial for patients and may in fact cause harm.4,6

Since the 1990s, the dominant screening test for CRCin the United States has been colonoscopy.8 Colono-scopy allows for direct visualization of the entire colon,from the appendiceal orifice to the dentate line,and also facilitates biopsy sampling or polypectomy oflesions that may appear abnormal. However, there is stillan ongoing debate in the medical community overwhich screening test is superior in the prevention anddetection of CRC. Moreover, with the introductionof newer screening methods such as Computed Tomo-graphic Colonography (CTC) and fecal DNA testing,choosing the best screening method has become moredifficult for both physicians and patients.

This paper will review both the advantages and dis-advantages of colonoscopy, sigmoidoscopy and CTC.The paper will forgo discussion of FOBT as it attemptsto focus on invasive screening techniques that aremore procedurally similar to colonoscopy so thataspects of the patient experience during each techni-que can be appropriately compared. Other factorsaffecting a patient’s decision to engage in regular CRCscreening and the role of primary care providers ininforming their patients about each method will alsobe analyzed.

Through a contemporary literature review, this paperwill examine whether colonoscopy is the superiormethod for the diagnosis and screening of CRC, andthus whether it has a greater capacity to reduce therisk of death from CRC as compared to other screeningmodalities.

METHODSThe main database used to obtain scholarly articles

cited in this literature review was PubMed at www.pubmed.org. A number of different search strategieswere used to narrow down articles. One strategyincluded keywords such as: ((colon cancer) AND (colo-noscopy) AND (surveillance)). Another strategy used‘colonoscopy’, ‘epidemiology’ and ‘colorectal neoplasms’as MeSH terms with ‘mass screening OR screening.’Subsequent searches focused on other modalities ofCRC screening with the use of ‘sigmoidoscopy’ and‘CT colonography’ as MeSH terms and with the sub-heading ‘therapeutic use.’ The filters used in all searches

included: past 5 years (2008�2013), clinical trial, rando-mized control trials (RCTs), humans, English and full textavailable. A few articles were also attained from otherdatabases such as Medscape, EBSCOhost and GoogleScholar using variations of the search strategies, key-words and filters described above.

In all articles selected, the study population of in-terest was high- and low-risk patients, aged 50 orolder, living within North America and other developednations. Other inclusion criteria included choosingarticles that were published in prominent journals orby recognized and valued medical organizations.

Criteria used to exclude articles from this paperinclude factors such as a small study population andarticles categorized as ‘review articles’, although alimited number were consulted to obtain relevantbackground information on the pathophysiology, epi-demiology and diagnosis of CRC.

Articles that met these criteria were then compiledinto an ‘Evidence Table’ (Table 1) that outlines the keyfindings of each.

RESULTS

Comparing Colonoscopy to Sigmoidoscopy and CTC

ColonoscopyColonoscopy is a screening method that allows

inspection of the entire colon and enables biopsy ofneoplastic lesions through polypectomy. This methodis conducted under sedation and is currently theleading tool for CRC screening.8 Many of the presentlyknown benefits of colonoscopy stem from population-based cohort studies that analyze the effects ofcolonoscopy on incidence and mortality amongcommunities around the world. In Ontario, Canada,Rabeneck et al9 conducted a large prospective studybetween 1993 and 2006 where they found the ratesof complete colonoscopy screening increased in allregions of the province. Within the population thatunderwent screening, the incidence rates and mortalityrates of CRC were lower in the younger age group (50�69 years) and lower for women within all age groups.When mortality rate was adjusted for confoundingfactors associated with increased risk of CRC death,such as increased age, male gender, lower income andrural residence, greater colonoscopy use was overallassociated with decreased mortality from CRC. Further-more, the study identified that for every 1% increasein colonoscopy rates in the cohort’s individual regionof residence (each participant was assigned to 1 of13 regions based on their address in Ontario), there

Puneet K. Singh Is Colonoscopy the Method of Choice?




http://www.pubmed.org

http://www.pubmed.org

http://www.msrj.org

Ta

ble

1.

Evid

en

ceta

ble

ou

tlin

ing

arti

cle

san

alyz

ed

inth

isre

vie

w

Firs

tau

tho

rD

ate

of

pu

blic

atio

nSt

ud

yd

esi

gn

Leve

lo

fe

vid

en

ceSt

ud

yp

op

ula

tio

nT

he

rap

yo

re

xpo

sure

Ou

tco

me

/re

sult

s

Ad

ler,

An

dre

as2

01

3P

rosp

ect

ive

coh

ort

wit

ho

ut

con

tro

ls

42

1co

lon

osc

op

ists

we

rest

ud

ied

fro

mp

riva

tep

ract

ice

sin

Be

rlin

,b

etw

ee

nO

ct.

20

06

and

Mar

,2

00

8.

Ove

rall,

ato

talo

f1

2,1

34

colo

no

sco

pie

sw

ere

exa

min

ed

.

Co

lon

osc

op

yT

hre

ep

atie

nt

fact

ors

corr

ela

ted

wit

had

en

om

ad

ete

ctio

nra

te(A

DR

):ag

e,

sex

and

low

qu

alit

yo

fb

ow

el

pre

par

atio

n.

Fact

ors

that

acco

un

ted

for

41

.4%

of

inte

r-p

hys

icia

nva

riab

ility

inA

DR

incl

ud

ed

:th

en

um

be

ro

fC

ME

me

eti

ng

sat

ten

de

dan

dd

iffe

ren

ces

ine

qu

ipm

en

tu

sed

du

rin

gsc

ree

nin

g.

Atk

in,

We

nd

y2

01

3R

CT

11

,58

0p

atie

nts

,5

5ye

ars

or

old

er

(me

dia

nag

ew

as6

8ye

ars;

ran

ge

61�7

5),

wit

hsy

mp

tom

ssu

gg

est

ive

of

CR

C(c

han

ge

inb

ow

el

hab

its,

rect

alb

lee

din

go

rab

do

min

alp

ain

)fr

om

21

UK

ho

spit

als.

1,0

47

we

rera

nd

om

lyas

sig

ne

dto

colo

no

sco

py

and

53

3as

sig

ne

dto

CT

C.

86

4(5

5%

)o

fth

eco

ho

rtw

ere

wo

me

n.

Pat

ien

tsw

ere

recr

uit

ed

be

twe

en

Mar

.2

00

4an

dD

ec.

20

07

.

CT

Can

dco

lon

osc

op

y1

60

(30

%)

pat

ien

tsin

the

CT

Cg

rou

ph

adad

dit

ion

alco

lon

icin

vest

igat

ion

com

par

ed

wit

h8

6(8

.2%

)in

the

colo

no

sco

py

gro

up

.A

lmo

sth

alf

the

refe

rral

saf

ter

CT

Cw

ere

for

smal

l(B

10

mm

)p

oly

ps

or

clin

ical

un

cert

ain

ty.

De

tect

ion

rate

so

fC

RC

or

larg

ep

oly

ps

we

re1

1%

for

bo

thp

roce

du

res.

CT

Cm

isse

d1

of

29

CR

Can

dco

lon

osc

op

ym

isse

dn

on

e(o

f5

5)

Bax

ter,

Nan

cyN

.2

00

9C

ase

con

tro

l3

10

,29

2ca

sep

atie

nts

and

51

,46

0co

ntr

ols

we

rep

art

of

this

stu

dy

po

pu

lati

on

.Fi

veco

ntr

ols

mat

che

db

yag

e,

sex,

ge

og

rap

hic

loca

tio

nan

dso

cio

eco

no

mic

stat

us

we

rera

nd

om

lyse

lect

ed

for

eac

hca

sep

atie

nt.

Par

tici

pan

tsw

ere

be

twe

en

the

age

so

f5

2an

d9

0w

ho

rece

ive

da

CR

Cd

iag

no

sis

fro

mJa

n.

19

96

toD

ec.

20

01

and

die

do

fC

RC

by

De

c.2

00

3.

Co

lon

osc

op

yC

om

ple

teco

lon

osc

op

yw

asst

ron

gly

asso

ciat

ed

wit

hfe

we

rd

eat

hs

fro

mle

ft-s

ide

dC

RC

bu

tn

ot

fro

mri

gh

t-si

de

dC

RC

Bax

ter,

Nan

cyN

.2

01

2C

ase

con

tro

l3

Ato

tal

of

9,4

58

case

sw

ere

ide

nti

fie

d.

Pat

ien

tsw

ere

age

d7

0�8

9an

dd

iag

no

sed

wit

hC

RC

be

twe

en

Jan

.1

99

8an

dD

ec.

20

02

.T

hre

eco

ntr

ols

(pat

ien

tsw

ith

ou

tC

RC

)w

ere

mat

che

dto

eac

hca

se.

Th

ep

rim

ary

exp

osu

rein

the

case

sw

asco

lon

osc

op

yp

refo

rme

dfr

om

1Ja

n.

19

91

to6

mo

nth

sb

efo

reth

ere

fere

nce

dat

e.

All

case

and

con

tro

ld

ata

we

reco

llect

ed

fro

mth

eSE

ER-M

ed

icar

ed

ata.

Co

lon

osc

op

yC

olo

no

sco

py

isas

soci

ate

dw

ith

are

du

ced

risk

of

de

ath

fro

mC

RC

,w

ith

ast

ron

ge

rd

ete

ctio

nra

tefo

rd

ista

lvs

.p

roxi

mal

CR

C.

Th

eo

vera

llas

soci

atio

no

fre

du

ced

risk

of

de

ath

was

stro

ng

est

ifco

lon

osc

op

yw

asp

erf

orm

ed

by

ag

astr

oe

nte

rolo

gis

t,as

op

po

sed

tosu

rge

on

so

rp

rim

ary

care

pro

vid

ers

.

Is Colonoscopy the Method of Choice? Puneet K. Singh

036 Medical Student Research Journal MSRJ # 2014 VOL: 04. Issue: Fall


http://www.msrj.org

Ta

ble

1(C

on

tin

ued

)

Firs

tau

tho

rD

ate

of

pu

blic

atio

nSt

ud

yd

esi

gn

Leve

lo

fe

vid

en

ceSt

ud

yp

op

ula

tio

nT

he

rap

yo

re

xpo

sure

Ou

tco

me

/re

sult

s

Bre

tag

ne

,Je

an-F

ran

cois

20

10

Re

tro

spe

ctiv

eco

ho

rt3

Eig

hte

en

en

do

sco

pis

tsw

ere

use

din

this

stu

dy;

14

wo

rke

din

pri

vate

pra

ctic

ean

d4

wo

rke

din

pu

blic

ho

spit

als.

Th

esc

ree

nin

gg

rou

pin

clu

de

das

ymp

tom

atic

me

nan

dw

om

en

age

db

etw

ee

n5

0an

d7

4w

ith

no

oth

er

risk

fact

or

for

CR

C.

Th

est

ud

yw

asco

nd

uct

ed

be

twe

en

20

03

and

20

07

inIll

ee

t-V

ilan

e,

Bri

ttan

y.T

he

firs

tro

un

do

fsc

ree

nin

gw

asin

20

03

and

20

04

wit

h2

13

,63

5p

arti

cip

ants

and

the

seco

nd

rou

nd

in2

00

5an

d2

00

6w

ith

22

4,5

04

par

tici

pan

ts.

Co

lon

osc

op

yIn

tere

nd

osc

op

icva

riab

ility

had

no

eff

ect

on

can

cer

de

tect

ion

ina

scre

en

ing

pro

gra

mw

ith

ah

igh

com

plia

nce

rate

wit

hco

lon

osc

op

yaf

ter

FOB

T;

ho

we

ver,

itd

idin

flu

en

ceth

eid

en

tifi

cati

on

of

ade

no

mas

.

Co

urt

ne

y,R

yan

J.2

01

3C

ase

seri

es

wit

ho

ut

con

tro

ls

41

,59

2at

-ris

kin

div

idu

als

(56�8

8ye

ars

of

age

)w

ere

ran

do

mly

sele

cte

dfr

om

the

Hu

nte

rC

om

mu

nit

ySt

ud

y(a

lon

git

ud

inal

com

mu

nit

yco

ho

rtin

Au

stra

lia)

be

twe

en

De

c.2

00

4an

dD

ec.

20

07

tota

kep

art

ina

qu

est

ion

nai

reb

ym

ail.

1,1

17

par

tici

pan

tsre

turn

ed

the

qu

est

ion

nai

res;

76

0re

spo

nd

en

tsw

ere

elig

ible

for

scre

en

ing

and

anal

ysis

.

Co

lon

osc

op

y6

3%

of

par

tici

pan

tsre

ceiv

ed

CR

Csc

ree

nin

g,

wit

hth

em

ajo

rity

be

ing

‘po

ten

tial

lyh

igh

risk

’p

arti

cip

ants

(84

%).

Th

ose

sig

nif

ican

tly

mo

relik

ely

toh

ave

rece

ive

dte

stin

gw

ere

age

db

etw

ee

n6

5an

d7

4an

dw

ho

had

atso

me

po

int

rece

ive

dsc

ree

nin

gad

vice

fro

mth

eir

fam

ilyp

hys

icia

n,

or

had

dis

cuss

ed

fam

ilyh

isto

ryo

fC

RC

wit

hth

eir

do

cto

r.2

1%

of

tho

sein

the

‘at

or

slig

htl

yab

ove

ave

rag

e-r

isk’

gro

up

had

rece

ive

dsc

ree

nin

gas

pe

ro

ffic

ial

scre

en

ing

gu

ide

line

s.G

uid

elin

es

we

resi

gn

ific

antl

ym

ore

like

lyto

be

follo

we

din

the

‘mo

de

rate

lyin

cre

ase

dri

sko

rp

ote

nti

ally

hig

hri

sk’

gro

up

s(4

5%

).C

olo

no

sco

py

wit

hin

5ye

ars

was

the

mo

stco

mm

on

lyre

com

me

nd

ed

gu

ide

line

and

was

sig

nif

ican

tly

mo

relik

ely

tob

ere

com

me

nd

ed

toin

div

idu

als

wit

hp

riva

tein

sura

nce

or

wh

oh

add

iscu

sse

dfa

mily

his

tory

of

CR

Cw

ith

ad

oct

or.

de

Wijk

ersl

oo

th,

Th

om

asR

.2

01

2R

CT

1T

his

stu

dy

was

of

8,8

44

Du

tch

citi

zen

sag

ed

50�7

4,

invi

ted

by

mai

lfo

rp

op

ula

tio

n-b

ase

dC

RC

scre

en

ing

inA

mst

erd

aman

dR

ott

erd

am,

be

twe

en

Jun

.2

00

9an

dA

ug

.2

01

0.

Invi

tati

on

sw

ere

ran

do

mly

allo

cate

d2

:1to

colo

no

sco

py

(5,9

24

)o

rC

TC

(2,9

20

)

Co

lon

osc

op

yan

dC

TC

Be

fore

scre

en

ing

,in

div

idu

als

exp

ect

ed

colo

no

sco

py

and

bo

we

lp

rep

arat

ion

tob

em

ore

bu

rde

nso

me

than

CT

Csc

ree

nin

g.

Ho

we

ver,

wh

en

ind

ivid

ual

sp

arti

cip

ate

din

scre

en

ing

,C

TC

was

sco

red

asm

ore

bu

rde

nso

me

than

colo

no

sco

py.





http://www.msrj.org

Ta

ble

1(C

on

tin

ued

)

Firs

tau

tho

rD

ate

of

pu

blic

atio

nSt

ud

yd

esi

gn

Leve

lo

fe

vid

en

ceSt

ud

yp

op

ula

tio

nT

he

rap

yo

re

xpo

sure

Ou

tco

me

/re

sult

s

Allo

cati

on

was

stra

tifi

ed

for

age

,se

x,SE

Sst

atu

sb

ase

do

nd

ata

of

Stat

isti

csN

eth

erl

and

sd

eW

ijker

slo

oth

,T

ho

mas

R.

20

12

RC

T1

Be

twe

en

Jun

.2

00

9an

dA

ug

.2

01

0,

8,8

44

ind

ivid

ual

sag

ed

50�7

5,

ne

ver

be

fore

scre

en

ed

for

CR

C,

we

rera

nd

om

lyal

loca

ted

tou

nd

erg

oe

ith

er

colo

no

sco

py

scre

en

ing

(5,9

24

)o

rC

TC

(2,9

20

).A

lloca

tio

nw

asb

ase

do

nag

e,

sex

and

soci

oe

con

om

icst

atu

s;d

ata

fro

mSt

atis

tics

Ne

the

rlan

ds.

Pat

ien

tsw

ere

fro

mth

eA

mst

erd

aman

dR

ott

erd

amre

gio

n.

Co

lon

osc

op

yan

dC

TC

Th

em

ost

fre

qu

en

tly

cite

dre

aso

ns

toac

cep

tsc

ree

nin

gw

ere

ear

lyd

ete

ctio

no

fp

recu

rso

rle

sio

ns

and

CR

C,

and

con

trib

uti

on

tosc

ien

ce.

Th

em

ost

fre

qu

en

tly

cite

dre

aso

ns

tod

ecl

ine

we

reth

eu

np

leas

antn

ess

of

the

exa

min

atio

n,

the

inco

nve

nie

nce

of

the

pre

par

atio

n,

ala

cko

fsy

mp

tom

san

d‘n

oti

me

/to

om

uch

eff

ort

’.El

de

rly

ind

ivid

ual

sci

ted

the

abse

nce

of

sym

pto

ms

asa

mo

resi

gn

ific

ant

reas

on

ton

ot

un

de

rgo

colo

no

sco

pie

sas

com

par

ed

toth

eo

vera

llst

ud

yp

op

ula

tio

n.

Th

em

ost

fre

qu

en

tly

cite

dre

aso

nfo

rd

ecl

inin

gco

lon

osc

op

yw

asth

eu

np

leas

antn

ess

of

the

exa

min

atio

n.

Fen

ton

,Jo

shu

aJ.

20

11

Pro

spe

ctiv

eco

ho

rt3

Th

est

ud

yp

op

ula

tio

nin

clu

de

dp

atie

nts

atte

nd

ing

app

oin

tme

nts

at2

acad

em

icp

rim

ary

care

clin

ics

�U

niv

ers

ity

of

Cal

ifo

rnia

,D

avis

,an

dM

ed

ical

Ce

nte

rin

Sacr

ame

nto

,C

A.,

be

twe

en

Sep

t.2

00

8an

dD

ec.

20

09

.5

0p

atie

nts

age

db

etw

ee

n5

0an

d7

5w

ere

elig

ible

for

CR

Cb

ase

do

nth

ecr

ite

ria

that

the

yh

adn

eve

rh

ad:

FOB

Td

uri

ng

the

pas

tye

ar,f

lexi

ble

sig

mo

ido

sco

py

du

rin

gth

ep

ast

5ye

ars,

colo

no

sco

py

in1

0ye

ars.

Inad

dit

ion

,2

0fa

mily

care

ph

ysic

ian

sw

ere

sele

cte

dfr

om

the

2cl

inic

s.

Co

lon

osc

op

yA

sco

mp

are

dto

pat

ien

tvi

sits

wit

ho

ut

CR

Csc

ree

nin

gd

iscu

ssio

n,

visi

tsw

ith

dis

cuss

ion

we

reas

soci

ate

dw

ith

incr

eas

ed

pe

rce

ive

dan

dsc

ree

nin

gin

ten

tio

naf

ter

the

visi

tb

ut

no

sig

nif

ican

tch

ang

ein

pe

rce

ive

db

en

efi

ts,

bar

rie

rso

rse

lf-e

ffic

acy.

Wit

hin

6m

on

ths,

17

of

38

pat

ien

ts(4

5%

)w

ho

dis

cuss

ed

scre

en

ing

com

ple

ted

scre

en

ing

com

par

ed

wit

h0

of

12

pat

ien

tsw

ho

did

no

td

iscu

sssc

ree

nin

g.

Gra

ser,

A.

20

09

Pro

spe

ctiv

eco

ho

rtw

ith

ou

tco

ntr

ols

43

11

asym

pto

mat

ic,

ave

rag

e-r

isk

adu

lts

(17

1m

en

and

14

0w

om

en

),ag

ed

be

twe

en

50

and

81

un

de

rwe

nt

sam

ed

aysc

ree

nin

go

f5

dif

fere

nt

scre

en

ing

inte

rve

nti

on

s.O

nly

30

7p

atie

nts

com

ple

ted

itin

full,

as4

sub

ject

sw

ith

dre

w.

Sig

mo

ido

sco

py

CT

C,

FIT

,FO

BT

,C

olo

no

sco

py

Sen

siti

viti

es

of

OC

,CT

C,F

S,FI

Tan

dFO

BT

for

adva

nce

dco

lon

icn

eo

pla

sia

we

re1

00

,96

.7.

83

.3,

32

and

20

%,

resp

ect

ive

ly.

Co

mb

inat

ion

of

FSw

ith

FIT

or

FOB

Td

idn

ot

incr

eas

ese

nsi

tivi

tyfo

rad

van

ced

CR

C.




http://www.msrj.org

Ta

ble

1(C

on

tin

ued

)

Firs

tau

tho

rD

ate

of

pu

blic

atio

nSt

ud

yd

esi

gn

Leve

lo

fe

vid

en

ceSt

ud

yp

op

ula

tio

nT

he

rap

yo

re

xpo

sure

Ou

tco

me

/re

sult

s

Excl

usi

on

scr

ite

ria

incl

ud

ed

the

abse

nce

of

spe

cifi

csy

mp

tom

so

fco

lon

icd

ise

ase

(me

len

a,h

em

ato

che

zia,

dia

rrh

ea,

rele

van

tch

ang

es

inst

oo

lfr

eq

ue

ncy

or

abd

om

inal

pai

n).

Sub

ject

sw

ere

also

exc

lud

ed

ifth

ey

had

pri

or

OC

wit

hin

the

last

5ye

ars,

po

siti

vefa

mily

his

tory

for

CR

C,a

his

tory

of

or

pre

sen

tIB

D,

he

red

itar

yC

RC

syn

dro

me

s,a

bo

dy

we

igh

tg

reat

er

than

15

0kg

or

seve

reca

rdio

vasc

ula

ro

rp

ulm

on

ary

dis

eas

e.

CT

Ch

adlo

wse

nsi

tivi

tyfo

rle

sio

nsB

6m

min

size

,b

ut

de

tect

ion

of

larg

ean

dad

van

ced

lesi

on

s,se

nsi

tivi

tyw

asco

mp

arab

leto

OC

.4

6%

of

sub

ject

sp

refe

rre

dC

TC

,w

hile

37

%p

refe

rre

dO

Cfo

rfu

ture

scre

en

ing

.

Ho

ff,

Ge

ir2

00

9R

CT

1T

he

stu

dy

incl

ud

ed

13

,6

53

sub

ject

sin

the

scre

en

ing

gro

up

(of

wh

ich

on

ly8

,84

6at

ten

de

dsc

ree

nin

g),

and

41

,09

2su

bje

cts

we

rein

the

con

tro

lg

rou

p.

Gro

up

sco

nsi

ste

do

fan

eq

ual

nu

mb

er

of

me

nan

dw

om

en

,ag

ed

55�6

4,

wh

ow

ere

ran

do

mly

sele

cte

dfr

om

2ar

eas

inN

orw

ay;

city

of

Osl

oan

dT

ele

mar

kco

un

ty(u

rban

and

mix

ed

urb

anan

dru

ral

po

pu

lati

on

s).

Th

est

ud

yw

asco

nd

uct

ed

fro

mJa

n.

19

99

toD

ec.

20

00

.

Sig

mo

ido

sco

py

Re

du

ctio

nin

inci

de

nce

of

CR

Cw

ith

fle

xib

lesi

gm

oid

osc

op

ysc

ree

nin

gw

asn

ot

see

naf

ter

7-y

ear

follo

w-u

p.

Mo

rtal

ity

fro

mC

RC

was

no

tsi

gn

ific

antl

yre

du

ced

inth

esc

ree

nin

gg

rou

p.

Joh

nso

n,

C.

Dan

iel

20

08

Cas

ese

rie

sw

ith

ou

tco

ntr

ols

4T

he

stu

dy

incl

ud

ed

2,5

31

asym

pto

mat

icp

atie

nts

,5

0ye

ars

or

old

er

wh

ou

nd

erw

en

tC

TC

follo

we

db

ysa

me

day

colo

no

sco

py

be

twe

en

Feb

.2

00

5an

dD

ec.

20

06

.P

atie

nts

we

ree

xclu

de

dif

the

yp

rese

nte

dan

ysy

mp

tom

so

fC

RC

(an

em

ia,

me

len

a,h

em

ato

che

zia)

mo

reth

ano

nce

inth

ep

ast

6m

on

ths,

ifth

ey

had

IBD

,h

ere

dit

ary

colo

rect

ald

ise

ase

,if

the

yh

adre

ceiv

ed

colo

no

sco

py

inth

ep

ast

5ye

ars

or

had

ap

osi

tive

FOB

Tre

sult

.

CT

Can

dco

lon

osc

op

yC

TC

ide

nti

fie

d9

0%

of

the

ade

no

mas

or

can

cers

me

asu

rin

g1

0m

mo

rm

ore

insu

bje

cts.

Ko

,C

ynth

iaW

.R

etr

osp

ect

ive

cro

ss-s

ect

ion

alst

ud

y

4A

tota

lo

f3

28

,16

7co

lon

osc

op

ies

we

reco

nd

uct

ed

by

12

,91

0p

rovi

de

rs(e

ith

er

gas

tro

en

tero

log

ists

,su

rge

on

s(g

en

era

lo

rco

lore

ctal

)o

rp

rim

ary

care

ph

ysic

ian

Co

lon

osc

op

yG

astr

oe

nte

rolo

gis

tsw

ere

mo

stlik

ely

and

ge

ne

ral

surg

eo

ns

we

rele

ast

like

lyto

de

tect

po

lyp

sd

uri

ng

colo

no

sco

py.

Dia

gn

ost

icb

iop

syra

tes

we

reh

igh

est

for

fam

ilyp

hys

icia

ns

and

low

est

for

colo

rect

alsu

rge

on

s.





http://www.msrj.org

Ta

ble

1(C

on

tin

ued

)

Firs

tau

tho

rD

ate

of

pu

blic

atio

nSt

ud

yd

esi

gn

Leve

lo

fe

vid

en

ceSt

ud

yp

op

ula

tio

nT

he

rap

yo

re

xpo

sure

Ou

tco

me

/re

sult

s

Snar

ep

oly

pe

cto

my

and

po

lyp

rem

ova

lra

tes

we

reh

igh

est

for

gas

tro

en

tero

log

ists

and

low

est

for

fam

ilyp

hys

icia

ns.

Leib

erm

an,

Dav

idA

.2

00

0C

ase

con

tro

l3

Asy

mp

tom

atic

adu

lts

we

rese

lect

ed

fro

m1

3V

ete

ran

sA

ffai

rsm

ed

ical

cen

ters

inth

eU

nit

ed

Stat

es,

be

twe

en

Feb

.1

99

4an

dJa

n.

19

97

.1

7,7

32

pat

ien

tsw

ere

scre

en

ed

for

en

rolm

en

t,3

,19

6w

ere

en

rolle

d;3

,12

1o

fth

ee

nro

lled

pat

ien

ts(9

7.7

%)

un

de

rwe

nt

com

ple

teco

lon

ice

xam

inat

ion

.T

he

me

anag

eo

fth

ep

atie

nts

was

62

.9ye

ars,

and

96

.8%

we

rem

en

.

Co

lon

osc

op

yC

olo

no

sco

py

can

de

tect

adva

nce

dco

lon

icn

eo

pla

sms

inas

ymp

tom

atic

adu

lts

wh

ich

can

no

tb

ed

ete

cte

dw

ith

sig

mo

ido

sco

py.

Man

ser,

Ch

rist

ine

N.

Feb

ruar

y2

01

2N

on

-ra

nd

om

ized

pro

spe

ctiv

eco

ho

rt

21

91

2sc

ree

ne

dan

d2

0,7

74

con

tro

lp

arti

cip

ants

fro

ma

rura

lar

ea

of

Swit

zerl

and

Co

lon

osc

op

yC

olo

no

sco

py

wit

hp

oly

pe

cto

my

sig

nif

ican

tly

red

uce

dC

RC

inci

de

nce

and

can

cer-

rela

ted

mo

rtal

ity

inth

eg

en

era

lp

op

ula

tio

n.

Rab

en

eck

,Li

nd

a2

01

0P

op

ula

tio

n-

bas

ed

pro

spe

ctiv

eco

ho

rt

3T

his

stu

dy

incl

ud

ed

2,4

12

,07

7p

arti

cip

ants

age

db

etw

ee

n5

0an

d9

0(m

ean

age

of

64

)in

On

tari

o,

Can

ada.

Of

the

coh

ort

,8

4.2

%liv

ed

inan

urb

anar

ea,

53

.7%

we

rew

om

en

and

86

.5%

had

aco

-mo

rbid

ity

sco

reo

f0

.Ea

chm

em

be

ro

fth

eco

ho

rtw

asas

sig

ne

dto

1o

f1

3re

gio

ns

bas

ed

on

the

irad

dre

ssin

the

pro

vin

ce.

Pat

ien

tsw

ere

exc

lud

ed

ifth

ey

had

ap

revi

ou

sd

iag

no

sis

of

CR

C,

ulc

era

tive

colit

is,

Cro

hn

’sd

ise

ase

or

had

the

irre

sid

en

cein

the

Sou

thEa

stLo

cal

He

alth

Inte

gra

tio

nN

etw

ork

du

rin

gth

eti

me

of

the

stu

dy.

Co

lon

osc

op

yIn

cre

ase

dco

lon

osc

op

yu

sew

asas

soci

ate

dw

ith

mo

rtal

ity

red

uct

ion

fro

mC

RC

atth

ep

op

ula

tio

nle

vel.

Sin

gh

,H

arm

ind

er

20

10

Po

pu

lati

on

-b

ase

dp

rosp

ect

ive

coh

ort

33

2,3

06

ind

ivid

ual

sfr

om

Man

ito

ba,

On

tari

o,

wh

oh

adn

eg

ativ

ere

sult

so

nin

itia

lco

lon

osc

op

ysc

ree

nin

gb

etw

ee

nA

pr.

1,

19

87

and

Sep

t.3

0,

20

07

.St

ud

yp

op

ula

tio

nin

clu

de

dp

atie

nts

age

db

etw

ee

n5

0an

d8

0.

Ind

ivid

ual

so

uts

ide

this

age

ran

ge

,th

ose

wit

hp

rio

rsi

gm

oid

osc

op

y,IB

D,r

ese

ctiv

eco

lore

ctal

surg

ery

or

CR

Cw

ere

exc

lud

ed

.

Co

lon

osc

op

yT

he

rew

asa

29

%re

du

ctio

nin

ove

rall

CR

Cm

ort

alit

yan

da

47

%re

du

ctio

nin

mo

rtal

ity

fro

md

ista

lC

RC

(SM

R,

bu

tn

ore

du

ctio

nin

mo

rtal

ity

fro

mp

roxi

mal

CR

C)

Th

ere

du

ctio

nin

mo

rtal

ity

fro

md

ista

lC

RC

rem

ain

ed

sig

nif

ican

tfo

rg

reat

er

than

10

year

s




http://www.msrj.org

Ta

ble

1(C

on

tin

ued

)

Firs

tau

tho

rD

ate

of

pu

blic

atio

nSt

ud

yd

esi

gn

Leve

lo

fe

vid

en

ceSt

ud

yp

op

ula

tio

nT

he

rap

yo

re

xpo

sure

Ou

tco

me

/re

sult

s

Sch

oe

n,

Ro

be

rtE.

20

12

RC

T1

Ato

tal

of

15

4,9

00

me

nan

dw

om

en

age

d5

5�7

4p

arti

cip

ate

din

this

stu

dy

fro

m1

99

3th

rou

gh

20

01

.In

div

idu

als

we

rera

nd

om

lyas

sig

ne

dto

eit

he

ran

inte

rve

nti

on

gro

up

(scr

ee

nin

gw

ith

fle

xib

lesi

gm

oid

osc

op

yw

ith

rep

eat

scre

en

ing

at3�5

year

s)o

rto

the

usu

alca

reg

rou

p.

Sig

mo

ido

sco

py

Inci

de

nce

of

CR

Caf

ter

am

ed

ian

follo

w-u

po

f1

1.9

year

sw

as1

1.9

case

sp

er

10

,00

0p

ers

on

-ye

ars

inth

ein

terv

en

tio

ng

rou

p,

asco

mp

are

dw

ith

15

.2ca

ses

pe

r1

0,0

00

pe

rso

n-y

ear

sin

the

usu

al-c

are

gro

up

.R

ed

uct

ion

sw

ere

see

nin

inci

de

nce

of

CR

Cin

bo

thp

roxi

mal

and

dis

tal

colo

n.

Mo

rtal

ity

was

red

uce

db

y5

0%

on

lyin

dis

tal

CR

Cw

ith

no

chan

ge

inp

roxi

mal

.vo

nW

agn

er,

Ch

rist

ian

20

12

RC

T1

54

7p

atie

nts

wit

hsy

mp

tom

ssu

gg

est

ive

of

CR

Cw

ho

we

rera

nd

om

lyas

sig

ne

dat

ara

tio

of

2:1

tou

nd

erg

oe

ith

er

colo

no

sco

py

(36

2)

or

CT

C(1

85

)O

fth

ese

,3

88

resp

on

de

dto

ap

ost

-te

stq

ue

stio

nn

aire

;21

2w

om

en

(55�8

7ye

ars

old

)an

d1

76

me

n(5

5�9

6ye

ars

old

).3

37

pat

ien

tsre

spo

nd

ed

toth

efo

llow

-up

qu

est

ion

nai

re;1

99

wo

me

n(5

5�8

7ye

ars

old

)an

d1

38

me

n(5

5�9

2ye

ars

old

).

Co

lon

osc

op

yan

dC

TC

Inth

esh

ort

-te

rman

alys

is,p

atie

nts

are

mo

relik

ely

toac

cep

tC

TC

asa

be

tte

rm

eth

od

of

surv

eill

ance

than

colo

no

sco

pie

s.H

ow

eve

r,af

ter

lon

g-t

erm

follo

w-u

p,p

atie

nts

no

ted

that

colo

no

sco

py

off

ere

dso

me

be

ne

fits

Zal

is,

Mic

hae

lE.

20

12

Pro

spe

ctiv

eco

ho

rtw

ith

ou

tco

ntr

ols

46

05

asym

pto

mat

icm

en

and

wo

me

n,

age

d5

0�8

5,

wit

hm

od

era

teto

ave

rag

eri

sko

fC

RC

we

rere

cru

ite

dfr

om

fou

rin

stit

uti

on

s:M

assa

chu

sett

sG

en

era

lH

osp

ital

;B

rig

ham

and

Wo

me

n’s

Ho

spit

al;

Un

ive

rsit

yo

fC

alif

orn

ia,

San

Fran

cisc

oV

ete

ran

sA

ffai

rsM

ed

ical

Ce

nte

r;an

dN

ort

hSh

ore

Me

dic

alC

en

ter,

be

twe

en

Jun

.2

00

5an

dO

ct.

20

10

No

n-c

ath

arti

cC

TC

and

colo

no

sco

py

No

n-c

ath

arti

cC

TC

was

able

toac

cura

tely

de

tect

ade

no

mas

10

mm

or

gre

ate

rb

ut

less

accu

rate

for

smal

ler

lesi

on

sw

he

nco

mp

are

dto

colo

no

sco

py

scre

en

ing

of

sam

eco

ho

rt

RC

T�

ran

do

miz

ed

con

tro

ltr

ials

;C

TC

�co

mp

ute

dto

mo

gra

ph

icco

lon

osc

op

y;O

C�

op

tica

lco

lon

osc

op

y;FS

�fl

exi

ble

sig

mo

ido

sco

py;

FIT

�fe

cal

imm

un

och

em

ical

sto

ol

test

ing

;

FOB

T�

feca

lo

ccu

ltb

loo

dte

stin

g;

CR

C�

colo

rect

alca

nce

r;IB

D�

infl

amm

ato

ryb

ow

el

dis

eas

e.





http://www.msrj.org

was a statistically significant decrease in the hazardof death by 3%.9

Similar results showing significantly decreased CRCincidence and mortality in groups undergoing colono-scopy screening were found in two other population-based prospective studies conducted by Manseret al10, in Switzerland, and by Singh et al11 in Manitoba,Canada. Singh et al11 analyzed a cohort of individualswho had previously undergone CRC screening withonly colonoscopy between April 1984 and September2007 and had received negative results (no polyps/CRC). The overall reduction in CRC mortality within thescreened population of this study was 29%, with thelargest reduction in mortality rates (39%) seen during a5�10 year follow-up, as compared to the generalpopulation.11 Importantly, the study also found therewere differences in the morality rates associated withspecific locations in the colon. There was a statis-tically significant 47% reduction in distal CRC deaths,but no reduction in deaths from proximal CRC.11 Thereduction in mortality due to distal CRC remainedsignificant for up to 10 years following the study’sconclusion.11 A case-control study carried out by Baxteret al12 presented mirroring results, finding that colono-scopy screening not only decreased CRC mortalityin cases vs. controls but also that this screening wasassociated with fewer deaths from left-sided CRC ascompared to right-sided.12

Many recent studies have discovered that discrepan-cies during colonoscopy-specific detection of CRC andprecancerous lesions may be operator dependent.Bretagne et al13 identified that differences in theperformance of 18 endoscopists analyzed in their studyresulted in large ranges of adenoma detection rates(ADR). However, when assessing the detection rate ofactual CRC, these operant-dependent factors did notindependently influence the varied range of rates, aspatient age and sex also played a role.13 Another studyby Adler et al14 went on to identify what it believedwere the specific factors that defined the efficacy andquality of screening by colonoscopists. The moststatistically significant associations, with 41.4% of theinter-physician variability in ADR, were the number ofContinuing Medical Education (CME) meetings eachcolonoscopist attended and the characteristics of theirindividual instruments.14

Some researchers investigated if the specific special-ties of those carrying out colonoscopies played anyrole in the variability of ADR and CRC detection. Baxteret al15 found that although colonoscopy screeningreduced the risk of CRC mortality (regardless of the

specialty of the endoscopist), there was a stronger asso-ciation if a gastroenterologist performed the colono-scopy as opposed to a non-gastroenterologist (e.g., asurgeon or primary care provider). Conclusively, gastro-enterologists provided significantly more protectionfrom CRC death than other providers.15 A study byKo et al16 further identified variability in frequencyof procedures performed by each specific specialty(Fig. 1). Overall, multivariate analysis determined thatnon-gastroenterologists were least likely to detect andremove polyps, and likelihood of diagnostic bio-psy was significantly lower for all surgeons (general/colorectal).16

SigmoidoscopyUnlike colonoscopy, flexible sigmoidoscopy is perfor-

med without sedation, has limited bowel preparationand is thus more often provided by general practi-tioners or non-physicians.1 The use of flexible sigmoi-doscopy CRC screening was analyzed in a Germanobservational study by Graser et al17 and two RCTs: thePLCO trial conducted by Schoen et al18 and the firstof the three Norwegian Colorectal Cancer Prevention(NORCCAP) trials carried out by Hoff et al.19

The PLCO trial mirrored findings presented in manyolder observational trials that showed flexible sigmoi-doscopy conferring protection against CRC mortalityand incidence.8 In this study, a 21% reduction in CRCincidence was observed in the intervention group ascompared to the usual care group, and CRC incidencein specific locations of the colon also showed signifi-cant reductions: 29% in the distal colon and 19% in theproximal.18 Overall, CRC mortality was reduced by 26%in the intervention group as compared to the usual-care group. However, when observing location-specificmortality rates in distal and proximal parts of the colon,the PLCO trial found that distal CRC mortality wasreduced by 50%, but no significant change in mortalitywas observed for proximal CRC (143 and 147 deaths;relative risk, 0.97; 95% CI, 0.77�1.22; P�0.81).18

Compared to the PLCO trial, the NORCAPP trial obser-ved a larger reduction in mortality rates (59%) amongsubjects who took part in sigmoidoscopy screening.19

Nevertheless, like the PLCO trial, some findings ofNORCAPP also substantiated discrepancies in cancermortality rates among discrete locations of the colonwhen sigmoidoscopy was performed. Among the inter-vention group, a greater reduction in both incidenceand mortality (76%) of rectosigmoidal cancer was foundas opposed to CRC.19 Thus, benefits of sigmoidoscopy




http://www.msrj.org

were once again shown to be limited to areas of thedistal colon.

The last study analyzing sigmoidoscopy screeningwas a prospective study carried out by Graser et al.17

The sensitivities of five different screening methods:sigmoidoscopy, CTC, colonoscopy, fecal immunochemi-cal stool testing (FIT) and FOBT were all tested inparallel among asymptomatic subjects. Flexible sigmoi-doscopy was 83.3% sensitive for advanced colonicneoplasia (CRC) and only 68% sensitive to adenomas]10 mm. Combining sigmoidoscopy with FOBT or FITenabled an increased detection of large adenomas(76.2 and 71.4%, respectively) as compared to sigmoi-doscopy alone (68%). However, when these tests werecombined for the detection of advanced CRC, noincrease in sensitivity was observed. Although flexiblesigmoidoscopy showed to be a superior test to FOBTand FIT, it was unable to surpass the advanced sen-sitivity of colonoscopy and CTC in detection of CRC andadenomas of all sizes.17

Computed Tomographic ColonographyCTC is a minimally invasive screening tool that is

currently undergoing testing in a number of trials. Likecolonoscopy, CTC provides examination of the entirecolon and rectum; however, it allows for computeri-zed 3D and advanced 2D imaging not available withcolonoscopy.1 In order to compare the efficiency ofCTC to colonoscopy in CRC screening and detection,three observational studies and one UK-based multi-center RCT were analyzed.17,20,21

All three observational studies focused on compar-ing the sensitivity and specificity of CTC in detectingadenomas of various sizes and neoplastic lesions tothat of colonoscopy, with additional comparison toother screening modalities (sigmoidoscopy, FIT andFOBT) completed by Graser et al.17 The study popula-tions assessed in all three studies were comparable andincluded average risk, asymptomatic patients (eachstudy using similar exclusion criteria) who were aged50 or older.17,20,21 All studies presented similar results(Table 2).

Although similarities between the sensitivity andspecificity of CTC and colonoscopy for the detectionof large neoplastic lesions were found, discrepanciesbecame evident in all studies when detecting adeno-mas of smaller sizes, specifically between 5 and 6 mmin diameter (Table 2). All three studies concludedthat CTC was significantly less sensitive for smallerlesions than colonoscopy. Measurements of specificityshowed similar trends.17,20,21 In two of the studies,the median sizes of missed lesions were 7 mm21 and6 mm.20 Graser et al17 found that CTC only missed oneadenoma with advanced histology in the B10 mm sizegroup.

The UK-based RCT carried out by Atkin et al22 pre-sented similar findings to those seen in the observa-tional studies. The sensitivity of CTC to CRC was 85% inthis RCT as compared to 93% with colonoscopy. Still,the most significant discrepancy in CTC screeningpresented by this study was its discovery that a greaternumber of patients assigned to the CTC screening

Figure 1. Variability in rate of polyp detection, biopsy and polyp removal among provider specialty. Gastroenterologists have thehighest rate of polyp detection, polypectomy and polyp removal. General surgeons are least likely to detect polyps, whilecolorectal surgeons have the lowest diagnostic biopsy rate. Family physicians have the highest rate of biopsy, but lowest rate ofpolyp removal. (Modified from Ko et al.16)





http://www.msrj.org

group needed to undergo additional colonic inves-tigations (after initial screening) as compared to thecolonoscopy group (30.0% vs. 8.2%).22 Within the colo-noscopy group, the major reason for additional screen-ing was incomplete colonoscopy (did not reach thececum) as seen in 11.3% of patients. In contrast, themajor causes for additional CTC investigations werelow predictive value for CRC or polyps ]10 mm(15.6%) and failure to confirm the presence of small(B10 mm) polyps (9.2%). In both cases, the additionalinvestigation was a new or repeat colonoscopy; a moreinvasive procedure than CTC. Finally, this RCT was theonly study to identify a statistically significant differ-ence in men and women with regard to the need foradditional investigations after screening. Men were sixtimes more likely to need further investigation afterCTC compared to colonoscopy, while women wereonly two times more likely.22

Patient Experience, Education and CompliancePatient experiences, perspectives on CRC screening

and compliance to screening guidelines were alsoanalyzed. Research conducted by von Wagner et al23

found that individuals undergoing colonoscopy weresignificantly less satisfied, more worried, experiencedmore physical discomfort and reported more adverseeffects such as ‘feeling faint or dizzy’ than those takingpart in CTC screening. This study further noted thatpatients had a better experience with CTC screeningthan with colonoscopy.17,20�22 However, this initialdissatisfaction with colonoscopy was not absolute, asvon Wagner et al23 identified that patients undergoingCTC had a greater number of post-procedure referralrates as compared to those who took part in colono-scopy screening (33% vs. 7%). Thus, the study con-cluded that after 3 months, patients reported greatersatisfaction with the long-term outcomes of their colo-noscopy screening compared to CTC23; a result alsofound by Atkin et al.22

It is likely that because the overall benefits of colo-

noscopy are not known by patients initially, the nega-

tive connotations surrounding CRC screening are factors

that deter patients from actually fulfilling screening

guidelines. A RCT conducted by de Wijkerslooth et al24

examined the reasons for participation and non-

participation in CRC screening among a study popula-

tion who had never undergone screening in two

regions of the Netherlands. This study found that the

most significant reason to participate in CRC screening

(either colonoscopy or CTC) was ‘it allows early detec-

tion of precursor lesions’ (the most decisive reason in

both screening modalities; 72% for colonoscopy vs.

68% for CTC).24 The most significant reason for non-

participation with respect to colonoscopy was ‘the

examination strikes me as unpleasant’ (66%) while for

CTC the reasons were both lack of time and absence of

symptoms.24 A second RCT looked at the ‘expected’

burden of screening before colonoscopy or CTC and

compared it to the actual (‘perceived’) burden experi-

enced during either procedure.25 This research discov-

ered that although participants expected colonoscopy

to be more burdensome than CTC, in reality they

experienced significantly more overall burden with CTC

(79% with colonoscopy vs. 82% with CTC).25

Many of the reasons mentioned for and against

screening participation stem from a lack of patient

knowledge about CRC and its prevention, and most

importantly from a lack of doctor�patient communica-

tion about specific guidelines for screening. A case

series by Courtney et al26 identified that within their

study population, only 63% of the cohort had ever

received any sort of CRC screening (FOBT/ sigmoido-

scopy or colonoscopy), with the majority of this subset

being ‘potentially high risk’ participants (84%). Overall,

individuals significantly more likely to have received

testing were those who were either between the ages

of 65 and 74, had at some point received screening

Table 2. Differences in sensitivity and specificity of CTC and colonoscopy in detecting adenomatous lesions of various sizes

CTC (large lesions; �10 mm)CTC vs. Colonoscopy

(small lesions; 5�6 mm)

Study Sensitivity (%) Specificity Sensitivity (%)

Graser et al17 96.7 n/a 59.2 94.6Johnson et al20 90 86% 78 100Zalis et al20 91 85% 59 76

Colonoscopy and CTC have similar efficacy in detecting large lesions; however, colonoscopy is significantly more sensitive than CTC for smaller

lesions.




http://www.msrj.org

advice from their family physician or had discussedfamily history of CRC with their doctor.26

Similar results permeated from a prospective studyby Fenton et al27 which took a deeper look at what agroup of physicians actually discussed with patientsduring visits. In 76% of the interactions, physiciansdiscussed CRC screening for a median time of 2.5 min.Physicians described one or two modalities for screen-ing, with colonoscopy always being mentioned. Doctorsdiscussed benefits of CRC screening in more than halfof the encounters, but less often commented on risks/susceptibility to CRC, barriers to screening, or self-efficacy of screening. After all visits were complete, itwas found that patients in the discussion group had asignificantly increased knowledge of risk/susceptibilityto CRC and had an increased intention to undergoscreening. Unfortunately, there was no significant changein perceived benefits of screening, barriers or self-efficacy in this discussion group compared to whenthey initially began the visit. Finally, a 6-month follow-up revealed that 45% of patients within the group thatdiscussed CRC screening actually underwent screening,as compared to the non-discussion group in which nopatient was screened.27

DISCUSSIONScreening for CRC is an integral part of cancer

prevention and has the capacity to positively impactCRC mortality rates. Colonoscopy has proven to be aleading method of CRC screening and its use hasincreased in many regions across North America.9,11

However, detection of CRC via colonoscopy does notoccur uniformly within the colon, and specific ‘burdens’of screening are discouraging individuals from partici-pating in colonoscopy. Thus, prematurely acceptingcolonoscopy to be the superior screening modality iserroneous. Many factors must be considered whendefining a tool as superior including patient prefer-ences, user-dependent skills, success of CRC detectionand cost-effectiveness. Until research can address thesefactors and clearly define superiority, the debate overwhich method to choose for CRC screening remains.

Unlike research on sigmoidoscopy and CTC, currentscholarly literature analyzed in this study has notproduced RCTs studying colonoscopy as a methodof CRC screening. Many of the colonoscopy-focusedstudies analyzed in this review were observational(level 4) studies � a major limitation of this paper.Some studies lacked control groups, and their cohortswere often too small. In addition, each study focused

on only one or two screening modalities at a time, thuspreventing grouped analysis of common variables. Theinitiation of RCTs with large cohorts comparing eachspecific modality in parallel and with more universaldata analysis techniques is necessary. In addition, moreprospective studies looking at the long-term benefitsof colonoscopy are needed as current research showsmany of the benefits of colonoscopy are observedseveral years following the initial procedure.11 However,to accurately determine long-term benefits, studiesmust also focus on populations closer to the age of50 as loss to follow-up due to death can negativelyimpact results.

Other limitations of this review included restrictingsearch strategies with the filter ‘full text available’ dur-ing data collection and also focusing on only a selectgroup of screening modalities. Studies on FOBT orFIT could have expanded the scope of this paper andenabled a more comprehensive comparison of allscreening tools recommended by current guidelines.

Nevertheless, this paper addresses several importantaspects of colonoscopy screening. First, a number ofproblems still remain in the actual effectiveness ofcolonoscopy screening. Many articles determined thatcolonoscopy was more beneficial in detecting distalCRC as opposed to proximal. Two RCTs identified thatsigmoidoscopy also presented with similar caveats.18,19

Although both screening tools were limited in thelocation they could optimally perform, sigmoidoscopyproved to cause a greater reduction in distal CRCmortality as compared to colonoscopy. The PLCOtrial found that mortality was reduced by 50% in thedistal colon using sigmoidoscopy compared to colono-scopy,18 while the NORCAPP trial also identified thatsigmoidoscopy’s greatest reduction of mortality (76%)was seen for rectosigmoidal cancer (specific to thedistal colon).19 Both values were higher than the 47%reduction in distal CRC mortality found via colono-scopy.11 Identifying ways to optimize screening of boththe proximal and distal colon is therefore necessaryto enable colonoscopy to surpass the strengths ofsigmoidoscopy.

Another major issue associated with colonoscopywas the variations in results due to the level of exper-tise of each colonoscopist. Gastroenterologists provedto be the most efficient when compared to surgeonsand primary care physicians (Table 2). These perfor-mance differences can greatly impact the accuratedetection of CRC and precancerous lesions. Further-more, these differences in expertise may prevent the





http://www.msrj.org

significantly at-risk populations from being screenedby the most skilled provider. Individuals from higher-income households (�$70,000 compared to 5$39,999)26

are more likely to take part in CRC screening, as aspecialist appointment is more costly than a primarycare visit. This is an unfortunate fact considering lowincome is a significant risk factor for CRC mortality.9 It isimperative that all health care professionals performingcolonoscopies attain standardized training and con-tinually strive to advance their skills so every patientcan receive screening from an equally qualified colo-noscopist. One way in doing so may be to increaseattendance at CME meetings, as this had a positiveassociation with ADR in primary care providers.14

In spite of the disadvantages of colonoscopy use,this screening modality is still the most sensitive andspecific test for detecting CRC and precancerouslesions of all sizes. Studies comparing colonoscopy toCTC and sigmoidoscopy identified that the rank fromhighest to lowest for specificity and sensitivity wascolonoscopy �CTC �sigmoidoscopy. Although CTCwas the closest to colonoscopy in sensitivity and speci-ficity for CRC, it was 20�30% less sensitive for lesionsB6 mm in size than colonoscopy. Thus, exclusive useof CTC over colonoscopy risks missing small lesionsthat can present similar threats of cancerous growthas large ones. As many articles have noted, this failureto detect small lesions forces patients to endure addi-tional investigations via colonoscopy in order toidentify all those that are missed. Patients thus becomeburdened with extra tests leading to unnecessarystress and worry.

Currently only 65.1% of the US population is up-to-date on screening for CRC as recommended bystandard guidelines.7 Among studied populations, themajor reason for participation in both colonoscopy andCTC was to identify precancerous lesions, while themain reason to not participate was the thought thatthe colonoscopy procedure would be unpleasant andthe belief that a lack of symptoms did not warrantundergoing CTC. Furthermore, when looking at rea-sons to choose specific screening modalities patientsalso assumed colonoscopy screening to be more bur-densome than CTC due to its preparation and un-pleasantness. However, patients admitted that in thelong run colonoscopy was less burdensome,25 suggest-ing that patient expectations or beliefs may often bedue to a lack of knowledge and guidance. Under-standing the factors that shape a patient’s views onCRC screening is essential in learning how to present

screening in a positive light and how to createeducational material that may further empower pa-tients to comply with guidelines.

Much research has shown that the most significantfactor in promoting screening is the interaction be-tween a physician and patient. In one study, discus-sions about the risks and benefits of CRC, options forscreening and family history of CRC occurred in only76% of patient�doctor meetings.27 Nevertheless, ofthe group of patients whose physicians did make anattempt to provide educational information, 45% wenton to take part in colonoscopy screening. This studyhighlighted the fact that a simple conversation canenable individuals to take action. If physicians takeadequate time to have detailed discussions with all at-risk patients and eliminate any myths of the procedure,it is likely that participation in colonoscopy screeningwill significantly increase.

CONCLUSIONIn conclusion, colonoscopy has proven to be the

most sensitive and specific tool for CRC screening andhas allowed for significant reductions in CRC incidenceand mortality. In order to ascertain that colonoscopy issuperior to sigmoidoscopy and CTC in all aspects ofCRC such as effectiveness in detecting both distal andproximal CRC, convenience of screening, efficiency ofscreening and patient preference, a number of factorsmust be addressed. First, large-scale RCTs looking atall three screening modalities together must beinitiated in order to understand the exclusive benefitsof colonoscopy and to move away from observationalstudies that are clouding current research. Next,eliminating the weakness of colonoscopy in detectingproximal CRC is imperative in order to ensure that itprovides the greatest advantage possible. In addition,in order to enable patients to understand the lifesaving benefits of colonoscopy, misconceptions andnarrowed views about this modality must be thor-oughly addressed. Empowerment can start within thedoctor�patient relationship. Once patients become moreaware of their health and more knowledgeable aboutall of the preventative procedures available to maintaintheir wellbeing, they may be more inclined to takeaction. Finally, performing colonoscopies must becomea more standardized procedure. All colonoscopistsshould ideally learn the same techniques and haveaccess to the same quality of screening tools to ensurethat operant-dependent differences do not confoundthe results of colonoscopy screening.




http://www.msrj.org

Conflict of interest and funding: The author declares noconflict of interest.

REFERENCES1. Holt PR, Kozuch P, Mewar S. Colon cancer and the elderly:

from screening to treatment in management of GI disease

in the elderly. Best Pract Res Clin Gastroenterol 2009; 23:

889�907. doi: 10.1016/j.bpg.2009.10.0102. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012.

CA Cancer J Clin 2012; 62: 10�29. doi: 10.3322/caac.201383. Manne U, Shanmugam C, Katkoori VR, Bumpers HL, Grizzle

WE. Development and progression of colorectal neoplasia.

Cancer Biomark 2010; 9: 235�65. doi: 10.3233/CBM-2011-01604. Qaseem A, Denberg TD, Hopkins RH, Jr., et al. Screening

for colorectal cancer: a guidance statement from the

American College of Physicians. Ann Intern Med 2012; 156:

378�86. doi: 10.7326/0003-4819-156-5-201203060-000105. Centers for Disease Control and Prevention. Vital signs: colo-

rectal cancer screening, incidence, and mortality�United States,

2002�2010. MMWR Morb Mortal Wkly Rep 2011; 60: 884�9.6. Lieberman D. Colorectal cancer screening: practice guide-

lines. Dig Dis 2012; 30(Suppl 2): 34�8. doi: 10.1159/0003418917. Centers for Disease Control and Prevention. Vital signs:

colorectal cancer screening test use � United States, 2012.

MMWR Morb Mortal Wkly Rep 2013; 62: 881�8.8. Kahi CJ, Anderson JC, Rex DK. Screening and surveillance

for colorectal cancer: state of the art. Gastrointest Endosc

2013; 77: 335�50. doi: 10.1016/j.gie.2013.01.0029. Rabeneck L, Paszat LF, Saskin R, Stukel TA. Association between

colonoscopy rates and colorectal cancer mortality. Am J

Gastroenterol 2010; 105: 1627�32. doi: 10.1038/ajg.2010.8310. Manser CN, Bachmann LM, Brunner J, Hunold F, Bauerfeind P,

Marbet UA. Colonoscopy screening markedly reduces the

occurrence of colon carcinomas and carcinoma-related death: a

closed cohort study. Gastrointest Endosc 2012; 76: 110�17.

doi: 10.1016/j.gie.2012.02.04011. Singh H, Nugent Z, Demers AA, Kliewer EV, Mahmud SM,

Bernstein CN. The reduction in colorectal cancer mortality

after colonoscopy varies by site of the cancer. Gastroenter-

ology 2010; 139: 1128�37. doi: 10.1053/j.gastro.2010.06.05212. Baxter NN, Goldwasser MA, Paszat LF, Saskin R, Urbach

DR, Rabeneck L. Association of colonoscopy and death from

colorectal cancer. Ann Intern Med 2009; 1501�813. Bretagne JF, Hamonic S, Piette C, et al. Variations between

endoscopists in rates of detection of colorectal neoplasia and

their impact on a regional screening program based on colono-

scopy after fecal occult blood testing. Gastrointest Endosc 2010;

71: 335�41. doi: 10.1016/j.gie.2009.08.03214. Adler A, Wegscheider K, Lieberman D, et al. Factors deter-

mining the quality of screening colonoscopy: a prospective

study on adenoma detection rates, from 12,134 examinations

(Berlin colonoscopy project 3, BECOP-3). Gut 2013; 62: 236�41.

doi: 10.1136/gutjnl-2011-30016715. Baxter NN, Warren JL, Barrett MJ, Stukel TA, Doria-Rose

VP. Association between colonoscopy and colorectal cancer

mortality in a US cohort according to site of cancer andcolonoscopist specialty. J Clin Oncol 2012; 30: 2664�9.doi: 10.1200/JCO.2011.40.477216. Ko CW, Dominitz JA, Green P, Kreuter W, Baldwin LM.Specialty differences in polyp detection, removal, and biopsyduring colonoscopy. Am J Med 2010; 123: 528�35.doi: 10.1016/j.amjmed.2010.01.01617. Graser A, Stieber P, Nagel D, et al. Comparison of CTcolonography, colonoscopy, sigmoidoscopy and faecal occultblood tests for the detection of advanced adenoma in anaverage risk population. Gut 2009; 58: 241�8. doi: 10.1136/gut.2008.15644818. Schoen RE, Pinsky PF, Weissfeld JL, et al. Colorectal-cancerincidence and mortality with screening flexible sigmoido-scopy. N Engl J Med 2012; 366: 2345�57. doi: 10.1056/NEJMoa111463519. Hoff G, Grotmol T, Skovlund E, Bretthauer M, NorwegianColorectal Cancer Prevention Study G. Risk of colorectal cancerseven years after flexible sigmoidoscopy screening: randomisedcontrolled trial. BMJ 2009; 338: b1846. doi: 10.1136/bmj.b184620. Johnson CD, Chen MH, Toledano AY, et al. Accuracy of CTcolonography for detection of large adenomas and cancers. NEngl J Med 2008; 359: 1207�17. doi: 10.1056/NEJMoa080099621. Zalis ME, Blake MA, Cai W, et al. Diagnostic accuracyof laxative-free computed tomographic colonography fordetection of adenomatous polyps in asymptomatic adults:a prospective evaluation. Ann Intern Med 2012; 156: 692�702.doi: 10.7326/0003-4819-156-10-201205150-0000522. Atkin W, Dadswell E, Wooldrage K, et al. Computedtomographic colonography versus colonoscopy for investi-gation of patients with symptoms suggestive of colorectalcancer (SIGGAR): a multicentre randomised trial. Lancet 2013;381: 1194�202. doi: 10.1016/S0140-6736(12)62186-223. von Wagner C, Ghanouni A, Halligan S, et al. Patientacceptability and psychologic consequences of CT colonographycompared with those of colonoscopy: results from a multicenterrandomized controlled trial of symptomatic patients. Radiology2012; 263: 723�31. doi: 10.1148/radiol.1211152324. de Wijkerslooth TR, de Haan MC, Stoop EM, et al. Reasons forparticipation and nonparticipation in colorectal cancer screening:a randomized trial of colonoscopy and CT colonography. Am JGastroenterol 2012; 107: 1777�83. doi: 10.1038/ajg.201225. de Wijkerslooth TR, de Haan MC, Stoop EM, et al. Burdenof colonoscopy compared to non-cathartic CT-colonographyin a colorectal cancer screening programme: randomisedcontrolled trial. Gut 2012; 61: 1552�9. doi: 10.1038/ajg.201226. Courtney RJ, Paul CL, Sanson-Fisher RW, et al. Individual- andprovider-level factors associated with colorectal cancer screeningin accordance with guideline recommendation:a community-level perspective across varying levels of risk. BMCPublic Health 2013; 13: 248. doi: 10.1186/1471-2458-13-24827. Fenton JJ, Jerant AF, von Friederichs-Fitzwater MM,Tancredi DJ, Franks P Physician counseling for colorectalcancer screening: impact on patient attitudes, beliefs, andbehavior. J Am Board Fam Med 2011; 24: 673�81. doi:10.3122/jabfm.2011.06.110001





http://www.msrj.org

Comparing Current Screening Modalities for Colorectal...

Documents

Transcript of Comparing Current Screening Modalities for Colorectal...