Gut lavage fluid protein concentrations: Objective measures of disease activity in inflammatory bowel disease

GASTROENTEROLOGY

1993;104:1064-1071

Gut Lavage Fluid Protein Concentrations: Objective Measures of Disease Activity in Inflammatory Bowel Disease CHINTAMANENI P. CHOUDARI, and ANNE FERGUSON Gastro-Intestinal

Unit, Western

General

SEAMUS Hospital

O’MAHONY,

and University

GORDON

of Edinburgh,

Background: Fluid obtained by whole gut lavage normally contains traces of immunoglobulin (lg) G, albumin, and a-1-antitrypsin; higher concentrations have been found in patients with inflammatory bowel disease (IBD). Methods: In a prospective study, 53 lavages were performed in 45 IBD patients (27 Crohn’s disease, 18 ulcerative colitis), in whom disease activity was simultaneously assessed by Crohn’s Disease Activity Index or Powell Tuck index. Concentration of IgG in lavage fluid was measured by enzyme-linked immunosorbant assay, and of albumin and a-l-antitrypsin by immunoturbidimetry. Results: For IgG, concentrations in lavage fluid correlated closely with activity indices: in Crohn’s disease, r = 0.723 (P < O.OOOl), in ulcerative colitis, r = 0.714 (P < 0.0001). Results for albumin and a-1-antitrypsin concentrations were similar to those for IgG, but they were less sensitive in detecting active disease. However, this method cannot be used as a diagnostic test for IBD; normal results were obtained for IgG in 6 (all inactive) of 42 lavages in patients who had unequivocal radiological or endoscopic abnormalities. Conclusions: Assay of protein concentrations in gut lavage fluid is a simple, objective means of grading disease activity in patients with IBD; its potential uses are likely to be in the evaluation of complex cases and in clinical trials.

Edinburgh,

provide Whole

T

Disease

Activity

Index (CDAI),

oped in the 197Os,’ has proved

a reliable

develand re-

markably reproducible measure of health status in patients with Crohn’s disease (CD). In theory, the CDAI is vulnerable to criticism because it is heavily weighted by subjective factors of general well-being and abdominal pain, and investigators have argued that the CDAI is not so much a measure of disease activity as of general disability. Many factors apart from intestinal inflammation will contribute to the index, including infection, obstruction, malnutrition, psychological dysfunction, and the side effects of drugs. Nevertheless, the CDAI and similar indices for ulcerative colitis (UC) activity, such as the Powell-Tuck index (PTI),’

gut lavage

tal surgery.3

appropriate

processing,

nity. 4s Normally,

provides

or colorec-

w h o le gut

rate groups

lavage

concentrations

higher

than

those

concentrations

of IgG

fluid

were

albumin

lavage

In the present performed

with

content,

plasma

leak-

mucosa.6

prospective

study, gut lavage has been

on 53 occasions

in 45 well-characterized

IgG and of two plasma-derived a- 1 -antitrypsin

high

a positive

IgG and albumin

with CD or UC, and WGLF

visit the CDAI

IBD we

significantly

We also found

that these tests are measuring

age across inflamed

(WGLF)

with active

that

of controls.5’6

between

for the immu-

of IgG, but in two sepa-

of WGLF

correlation

se-

and, after

humoral

of 10 and 44 patients

found

colonic

It contains

ideal material

of intestinal

only trace amounts

patients

as a bowel

(Ig) and antibodies

investigation

suggesting

used

endoscopy,

a gut perfusate.

immunoglobulins

contains

widely

enema,

in IBD.

The clear fluid passed after initial

is essentially

clinical

of any new ap-

of disease activity

is now

for barium

cleansing

MWANTEMBE,

for the assessment

to the measurement

preparation

creted

OBEDY

Scotland

standards

proach

blood he Crohn’s

BRYDON,

concentrations

proteins,

albumin

of and

(A 1 AT), were measured. At the same or PTI as appropriate was calculated,

was taken for measurements

of erythrocyte

sedi-

mentation rate (ESR), platelet count, and C-reactive protein (frequently used, single indices of activity), and the physician’s by using

a visual

We aimed

global analogue

to establish

assessment

was quantified

scale (VAS)

of O-120.

whether

assay of WGLF

pro-

teins could be used as a diagnostic test for IBD and to assess to what extent concentrations of proteins in WGLF parallel the global indices, CDAI and PTI. Do the results of this investigation make it possible to Abbreviations used in this paper: AlAT, alpha-1-anti-trypsin; ELISA, enzyme-linked immunosorbant assay; ESR, erythrocyte sedimentation rate; PRU, Protein Reference Unit; PTI, Powell-Tuck index; VAS, visual analogue scale; WGLF, whole gut lavage fluid. 0 1993 by the American Gastroenterological Association 00 16-5065/93/$3.00

GUT LAVAGE

April 1993

grade severity

within

“active”

disease or to detect min-

imal abnormalities in patients with CDAI of 5150 or PTI of 14 (i.e., classified as inactive)? Does this approach tests

offer any advantage (ESR,

physician’s

C-reactive global

the purpose

over conventional

protein,

assessment

platelet (recorded

the gastrointestinal

or a

Patients week preceding indices

Material and Methods

patients

Subjects

the CDAI

female).

The diagnosis

confirmed various

subgroups

current

macroscopic

patient

were studied times.

performed aration

or PTI was calculated

1.

Undergoing

and radiology.

were

Full de-

were classified

of previous

and

for colonoscopy

The study was approved of the Lothians

by the Medicine

Area Ethics

of Research

of 45 Patients

was being or barium as prepto empty

Parts of gastrointestinal tract previously resected Crohn’s disease None

Small bowel lleocecal f terminal leum

Segmental colonic resection

With IBD

Gut Lavage

Patients ethylene

drank

4 L of isotonic

glycol-based

No. of patients

oratories,

Inc.,

Braintree,

lo-15

minutes.

Usually

followed

by liquid

several

large-volume,

within

2 8

2

Small bowel lleocolonic Colonic Rectosigmoid Perianal Microscopic only Small bowel No macroscopic disease Small bowel lleocolonic lleocolonic Colonic

Total Ulcerative colitis Previously documented macroscopic disease Rectosigmoid

Left sided

27

mixing

3

18

No macroscopic disease Rectosigmoid No macroscopic disease Left sided No macroscopic disease Pancolitis

Kent,

phony1

fluoride

(1 mmol/L); 3 1 1 2 32

Aliquots

Sample

3 9 1 2 1 5 21

human

within

ward or

of col-

a GF/A

(What-

glass fibre filter.

in phosphate

The

in bracksaline

WGLF

acid

phenyl

(2 mmol/L);

calf serum

with

buffered

ethylenediaminetetraacetic

in 95% ethanol

methyl sodium

in sul-

azide

(5% volume).

were then stored at -70°C.

Methods

analysis

for IgG was by enzyme-linked

assay (ELISA),’ IgG

for

suite.

10 minutes

through

saline (15 mmol/L);

of processed

munosorbant

in-patient

(final concentrations

and newborn

Analytical

by

An ali-

was collected

investigation

England)

inhibitor

sodium

buffered

were

were added to 10 mL of the filtrate,

after each addition

phosphate

feces followed

of Whole Gut Lavage Fluid

ets): soya bean trypsin

anti-human 11

out-patient

Ltd.,

(80 pg/mL);

Lab-

by an experienced

in a general

were processed

reagents

solid

clear fluid “stools.”

20 mL, was filtered

Scientific

following

dard,

4

Total

man

poly-

Braintree

were

was supervised either

Processing Fluid,

which

clear specimen

The procedure

Specimens

2 hours

feces,

virtually

in the gastrointestinal

No. of lavages

nonabsorbable

MA) at a rate of 250 mL every

passed

lection. 15

Committee.

lavage fluid (Golytely;

nurse and performed

on 53 Occasions

Macroscopic disease at time of lavage

tests.

Subcommittee

Lavage Protocol

and 1 UC

and 1 man with UC was studied gut lavage

signs, and results

into

resections

Five CD patients

and in 5 cases it was performed

Whole

In

disease activ-

of the blood

in 4 cases lavage was performed

Details

of overall

and

fashion.

on a VAS, this being

analysis. Clinical

in the standard

assessment

quot of the first completely Table

Igs. All

by one of us (C.P.C.),

ity for the day of lavage was recorded

as bowel preparation

for surgery;

ESR, C-reac-

and serum

based on symptoms,

In 44 of the 53 studies,

enema examination;

assessed

the physician’s

count,

10

by the nature twice,

platelet

and albumin

27

1. The patients

disease.

haemoglobin,

total protein,

for the

and biochemical

with IBD were studied,

endoscopy,

in Table

a diary card of symptoms

and 18 with UC (8 male,

and disease extent in all patients

by histology,

tails are given

three

16 female)

completed

were clinically

addition,

A total of 45 patients

with an elemental

the lavage. Haematological

included

tive protein,

with CD (11 male,

1065

Clinical Assessments

on a VAS for

of this investigation)?

treatment

IN IBD

diet.

clinical

count)

tract before

FLUIDS

(North

reference

East serum

using

Laboratories) from

im-

affinity-purified and,

the Protein

goat as stan-

Reference

Unit (PRU), Royal Hallamshire Hospital, Sheffield. Albumin and Al AT were assayed by immunoturbidimetric procedures. Scottish

Anti-human

Antibody

was obtained 6000 reagent

Production

albumin

was supplied

Unit, and anti-human

by the Al AT

from the PRU. WGLF was diluted in PEG in tris buffer pH7 with and without antibody

and incubated for 15 minutes. Turbidity was measured at 340 nmol/L in a flow-through spectrophotometer (PU8610, Phillips). Standard plots of albumin and Al AT

1066

CHOUDARI

ET AL.

GASTROENTEROLOGY

values over the range O-200 pg/mL were prepared from Human Reference Serum (PRU), diluted in 0.9% NaCl containing 48 g/L PEG 3350.

Statistical

were normal

and

borderline

CDAI

Correlation coefficients were calculated by using the Spearman Rank Correlation test.

hemicolectomy of arthritis with

Results

CDAI

ease distribution treatment. scribed

and

were heterogeneous

in their

In CD, the CDAI

24 lavages tients

studied

were in patients

below with

as having

a CDAI

in patients in patients

and

current

IgG

concentration

were

high

present

in 23 patients

with

(de-

abdominal

and 8 in pa-

WGLF

IgG concentration

disease)

(in remission).

In patients

from 0 to 13; 13 lavages were

with active disease (PTI of >4), and 8 were in remission with a PTI of 14.

next worse;

6 months

of whom

pain

had

his growth

were all taking

a normal

subjects: 20 men, age age range 24-88, who

diet without

being

given

mL, and for AlAT

~19

yg/mL.’

Values for WGLF protein concentrations tients with CD and UC are shown in Figure plotted

against

CDAI

in pa1 and are

or PTI as appropriate.

Diagnostic Potential Fluid Tests

of Whole Gut Lavage

The possibility that WGLF protein concentrations are always high when there is overt intestinal mucosal inflammation (on radiological or endoscopic visualization) would provide a means of screening patients for the presence or absence of IBD. This issue has been considered by analysis of the results for were WGLF IgG. High values for IgG concentration present in only 24 of the 27 lavages performed in patients who had unequivocal macroscopic small bowel or colonic CD and in 13 of the 15 lavages in patients with UC who had positive radiograph or endoscopic findings. There was also one patient with microscopic evidence of colonic CD and no macroscopic disease; however the WGLF IgG concentration was high.

Whole Gut Lavage Fluid Proteins and Disease Activity IgG, Crohn’s disease. In 32 lavages tients

with

CD, values

for WGLF

from paIgG concentration

22 had active CD and unexof 119 and

of 20 pg/mL.

During

stopped

and pain

have resolved

a

the

became

with systemic

IgG, ulcerative colitis. In 21 lavages with

UC, values

for WGLF

in 8 patients

Both discrepant equivocally

ste-

(of whom

disease

roid treatment

active

two weeks before

and were

or oral corticoste-

Albumin. In 32 lavages from patients for 15 patients

albumin

concentration

(8 of these patients

in the others

min concentration of whom

disease).

who had had un-

respectively.

values for WGLF and CDAI

pa-

7 were in remis-

(12 with

cases were patients

active

from

IgG concentration

in the early stages of Sulphasalazine

im-

munosuppressive or anti-inflammatory drugs. Normal values are for IgG 110 pg/mL, for albumin 526 l.tg/

IgG

a CDAI

sion) and high in 13 patients

63 immunologically normal range 15-80, and 43 women,

of WGLF

boy with perianal

both problems

were normal

Reference ranges for WGLF IgG, albumin, and A 1 AT have previously been established by studies of

and

roid therapy. tients

Whole Gut Lavage Fluid Protein Concentrations

was one patient

of 10 I_tg/mL

concentrations

plained

of >150

after right

of 185.

Abnormally

disease. A 13-year-old

CDAI

a

to reflect activity

of CD. There

range was from 24 to 321;

active

5150

with UC, the PTI ranged

previous

in dis-

than

in

with

spondyli-

and high ESR shortly

for CD were thought

WGLF

man

of 168. He had ankylosing

rather

No. 4

of these were in patients

1 was in an 18-year-old

tis, and his anorexia

Analysis

The patients

in 9. Seven

remission,

Vol. 104,

ranged

with CD,

were normal

were in remission

from

168-283).

Albu-

levels were high in 17 patients,

had active

all

disease.

In 21 lavages from patients with UC, values for WGLF albumin concentration were normal for 12 patients

(of whom

mally

high

8 were in remission)

in 9 patients,

and were abnor-

all of whom

had active

dis-

ease.

AlAT.

In 32 lavages

from

patients

with

CD,

values for WGLF Al AT concentration were normal for 17 patients (of whom 7 were in remission) and were high in 15 patients

(14 with active disease).

In 21

lavages from patients with UC, values for WGLF Al AT concentration were normal in 14 patients (of whom 7 were in remission) with active disease).

and high in 7 patients

(6

Correlation of Lavage Proteins Concentrations With Activity Indices in Active and Inactive Disease Correlation coefficients for WGLF protein concentrations against activity indices are detailed in Table 2. For both CD and UC and for all three proteins studied, there was a highly significant positive correla-

GUT LAVAGE

April 1993

a00

FLUIDS

r200

IN IBD

.

1067

.

.

150 ?? *

/1 100 WGLF IgG concentration 75 @g/ml) I

150ti1 IOOWGLF IgG concentration @g/ml) 75.

??e

.

50.

. .

251, 0

100

50

150

200

250

O< 0

>300

. 2

4

6

CDAI

i b

12

.

14

.

8 .

0’

. WGLF albumin concentration 75 @g/ml) 1

. .e

.

50-

0

0.0

??

.

??

25

25

0 0

50

100

150

200

250

0 -0

300

2

4

6

.

CDAI

1

8

10

12

14

PTI

.

60 WGLF AlAT concentration @g/ml)

10

.

lOOWGLF albumin ;;gy;;tration 75.

80

8

z-200 1

.

100

;

PTI

2001 150

??

??

WGLF AIAT concentration kg/ml)

40

.

. .

.

40 .

20

20

0 0

50

100

150

200

250

0 7300

-0

2

4

CDAI

Figure 1. Concentrations

tion

between

centration, Because,

activity

of IgG, albumin,

index

with r values as shown

clearly

and

calculated

10

12

14

Al AT in WGLF taken from patients with CD and UC and plotted against CDAI and PTI, respectively.

protein

con-

from 0.439 to 0.723.

in Figure

1, the relation-

ship between activity index and lavage protein centration is not linear but biphasic, correlations separately

8 PTI

and WGLF

ranging

6

conwere

for cases with active and inactive

PTI. In inactive trations

CD and UC, WGLF

were independent

of activity

Conventional Laboratory Disease Activity For the purposes

protein

concen-

indices.

Indices of

of this analysis,

patients

with

disease (as defined by CDAI and PTI). Calculations were also performed for the subsets of CD patients

CD who had a CDAI of > 150 and those with UC who had a PTI of >4 have been designated as having active

with active colonic and active small bowel disease. The results (Table 2) show that there are significant correlations only in active disease; that the highest T values were found for IgG vs. CDAI in active CD (r =

disease. The numbers of cases with abnormal and normal values for ESR, C-reactive protein, and platelet count in the active and remission groups are given in Table 3. This table also includes, for comparison, num-

0.821), particularly in the subset with active small bowel CD (r = 0.970); and that in active UC, although there was good correlation between WGLF IgG concentration and the PTI (T = 0.740), the other two proteins did not show significant correlation with the

bers with normal and abnormal WGLF IgG, albumin and Al AT values, and the physician’s overall assessment based on a VAS. The three blood indices were normal in most patients with inactive disease. However, they were also

CHOUDARI ET AL.

1068

Table 2. Concentrations

GASTROENTEROLOGY Vol. 104. No. 4

of Proteins

in WGLF From Patients

With Active and Inactive

IgG concentration vs. activity index No. of lavages Crohn’s disease All Active Inactive Active colonic Active small bowel Ulcerative colitis All Active Inactive

r 0.723 0.82 1 0.217 0.778 0.970

32 24 8 8 10

With CDAI and PTI

Albumin concentration vs. activity index P

r


0.588 0.547 -0.352 0.507 0.547


0.714 0.740 _-0.318

21 13 8

IBD; Correlations


P

r

P


0.439 0.293

0.160 0.468 0.359

co.01 NS NS NS NS

0.527 0.600 -0.04 1

<0.02 co.05 NS

co.01 NS NS NS


0.701 0.610 -0.378


A 1AT concentration vs. activity index

co.05 NS

NOTE. P > 0.05. NS, not significant.

normal disease,

in many patients with and this was particularly

assessments C-reactive count

in active protein

UC; ESR was normal was

was normal

considerable

unequivocally active so in the group of 13

normal

in

platelet

In an initial

in 8.

to WGLF,

not directly fected. tion

Factors

control

of illness

related

in a patient

to the length “disease

with

of bowel

such as the severity

and nonspecific

IBD is

grossly

af-

There

practice

have been many

cal and/or

laboratory

criteria,

and claims

gut lavage

lavage

fluid

and

(WGLF),

have

reported

that

appropriately

stantial

amounts

a few clini-

whole

of IgG (<5 pg/mL) there

with

in fluid from

were

in WGLF

significantly

from

was confirmed active

of albumin

as applied of IgA but

10 patients

in a separate

CD who also had sub-

in WGLF,

indicating

because

are widely

that the in-

fecal Al AT concentraas indices of GI

accepted

protein loss. With a view to the introduction of this new test into clinical practice, we also carried out a retrospective review of results for 140 lavages in symptomatic patients’ to establish whether concentrations of these proteins in WGLF are high in conditions other than active IBD. Assays of WGLF obtained on 102 occa-

been

gut

processed,

of ELISA

concentrations

whereas

in the protocol

tion and clearance

made for the significance of single laboratory determinations in some situations. 14-‘* We have been exploiting the relatively new method of bowel cleansing by whole

for clini-

and inflamma-

these tests may be detecting plasma leakage into gut.’ In further work, assay of AlAT has been

a single

have

appraisal high

CD.5 This

of patients

cluded

to develop

index of illness in IBD’,2,8-‘3 by combining

active

cohort

are also rele-

and research.

attempts

immunity

IgG concentrations

with

of local inflamma-

activity”

subjects,

higher

vant. There is an urgent need for simple and objective indices of these separate components of overall disability for use in clinical

technical

we found

only trace amounts

Discussion The degree

as a source of material

of intestinal

tion5

in 8 cases,

10, and

potential

cal evaluation

has

Table 3. Presence of Normal or Abnormal Values for Conventional Laboratory Indices of Disease Activity, Global Assessment, WGLF IgG, Albumin, and Al AT Concentrations in Patients With CD and UC Arbitrarily Subdivided Into Active or Inactive IBD on the Basis of CDAI or PTI, Respectively ESR

C-reacwe

VAS

protein

~

Platelet count

No. of

Normal

High

Normal

kwages

(X20)

(>20)

(<1.5)

24

5

19

8

8

5

3

8

High (>

I .5)

~

~

WGLF IgG

WGLF albumln

WGLF AIAT

COnCentratlOn

c0ncentrat10n

concentration

____

Normal

Normal

High

I6

6

I8

I

0

7

I

7

(O-60)

High (61-120)

Normal

~ High

~

Normal

Mgh

Normal

High

(526)

(>26)

(S 19)

(>19)

(
@IO)

23

2

22

7

I7

IO

14

I

7

I

8

0

7

I

I

12

4

9

7

6

2

7

I

8

0

7

I

Crohn’s disease Active (CDAI > 150) Inactive (CDAI 5 150) Ulceratw colibs Actwe (PTI > 4) lnactwe (PTI c 4)

13

8

5

IO

3

8

5

2

8

8

0

8

0

7

I

6

II

GUT LAVAGE

April 1993

sions

from

revealed albumin

95 patients

WGLF

IgG concentrations

with

lymphangiectasia,

other

conditions

of varying

were present

activity

and in only

findings

are relatively

We now report vages in patients investigated indices

with

for various

this completely

objective

These

results

and Al AT sup-

protein

concentra-

of active

study based on 53 laof disease

and composite

procedure

disease

activity

as the reference

activity

for

in CD and a similar

measure

index

IBD suggests

of

for UC, the

PTI.

from

who have extensive

chronic

merely

tissues

men

as a result

when

abundant

some

study

shows

that WGLF

protein

centrations,

particularly

IgG,

tween

and inactive

IBD and, within

active

discriminate

well

conbe-

the subsets

bumin,

patients

with

CDAI

porting

the

present

I 150 and PTI 14, clinical

trials

strongly

practice

such values are taken as successful end points, ing that remission has been achieved. However screening

WGLF

protein

or diagnostic

assays cannot

tests for IBD.

sup-

whereby indicat-

Although

high

However,

in active

of the three proteins The

differ subthree

or the epithelial

We have also considered could

explain

tive and

and

in relation

inactive

differ,

disease.

Although

the

for the

proteins been

bacterial

proteases

during

gut (l-2

hours).

Ex vivo

leaking the transit

techniques

three

ra-

diological or endoscopic features of IBD such as cobblestone ulcers and long, stringlike strictures. Nevertheless in our clinical practice we are finding that assay of WGLF IgG is a valuable aid to diagnosis in some clinical situations, e.g., in patients with microscopic CD, when IBD coexists with severe diverticular disease, and in separating the effects of intestinal and joint disease in IBD patients with ankylosing spondylitis. There is abundant evidence to implicate IgG in the pathological lesions of IBD, particularly CD: immunohistochemistry reveals an excess of IgG-containing cells deep in the mucosa 19**O; isolated intestinal mononuclear cells from IBD patients spontaneously secrete high amounts of IgG*‘,*‘; and there are differences in

whereas

at

min-

into the

to pancreatic

and

of fluid along the

experiments

show

in 1 hour),

are unequivocal

assays

within

inactive

there

ac-

between-run,

proximally

exposed

pro-

between

fluids are processed

will have

factors

for the three

of variation,

are similar

5%.’ Although

utes of collection,

analytical

results

to discrimination

co-efficients

within-run,

10%-l

whether

the different

37°C degradation of albumin by unprocessed is significant (loss of lo%-40% of measured

even when

proteins

basal lam-

values are found in many patients, the data in this paper clearly show that values are normal in clinically disease,

IBD,

are pres-

concentrations

in plasma.

are

the gut lu-

ina.

gut lumen

be used as

tests

entering

weight (IgG, 150,000 daltons; alAl AT, 45,000 daltons), and daltons;

69,000

used

in most

in IgG cells).

sensitive

proteins

relative

those

CD patients

of the intestine

we suggest that there is a selective increase in mucosato-lumen permeability in active IBD operating either

these

normal

these

of bleeding. their

concentrations

to be replete

plasma

from

teins studied

tests were completely

active

in molecular

of patients with active disease, closely parallel disease activity as defined by CDAI and PTI. Furthermore, sensitive

of

albumin

with

inactive

ulceration

that

amounts

ent in WGLF, differ

known

be argued

measuring

stantially

patients

all three protein

were low in WGLF

It could

from

the findings

proteins,

that at least some of the IgG is plasma-

at the level of the capillary

The present

be-

the abundant

However,

fluid

Furthermore,

(diseased

and

can substitute

and Al AT, in lavage derived.

IBD.

1069

distribution

from

of two plasma

assay of WGLF proteins test for IBD and whether

index, the CDAI,

derived

mucosa.

or complement currently used clinical tests and indices of disease activity. We have used the best currently available

ered that it was probably high concentrations

at the time of the lavage. The aims of this study

were to determine whether can be used as a diagnostic

subclass

IgG cells in diseased

distribution single

IgG

with

UC or CD fully characterized

to anatomical

cell

non-

markers

a prospective

plasma

IN IBD

tween controls, UC and CD patients.23 Thus, when we first noted the presence of IgG in WGLF, we consid-

with

WGLF

specific

intestinal

1 of 32 patients

for albumin

the view that high

regard

in a patient

(a value of 11 yg/mL).

and comparable ported tions

IBD

in 1 of 5 patients

IBD forms of colitis,

with

with

high concentrations of IgG in 65 (64%), of in 53 (52%), and of AlAT in 37 (36%). High

FLUIDS

that

at

WGLF albumin

IgG and A 1 AT are relatively

resis-

tant to such proteolysis. This could explain the differences in results for IgG and albumin in active disease but not the relative centrations. Various

scientific

insensitivity approaches

of disease activity in viewed.24,25 Laboratory

of WGLF

Al AT con-

to the measurement

IBD have recently been retests such as ESR, platelet

count, acute phase proteins, are useful. However they may be normal in active IBD, particularly in UC and in small bowel CD (e.g., Table 3) and will be positive in situations of active inflammation or infection without gut involvement. Other techniques, such as labeled leucocyte studies26*27 and measures of GI protein loss by Al AT clearance,28-30 have advantages and disad-

1070

CHOUDARI

vantages,

ET AL.

which

GASTROENTEROLOGY

have

been

where.6924725 In general, applied tion

without

against

being

fully

discussed

else-

they have been developed subjected

the carefully

to prospective

developed

indices

such as CDAI.

Furthermore,

studies

are expensive,

involve

standard labeled

exposure

and

evaluaactivity leucocyte

to radioactiv-

ity, and depend on complete fecal collection. The findings with these tests relate to events in macroscopically affected

gut and thus differ

the gut lavage technique, more diffuse phenomenon. There

may be problems

this

could

from

either

specific

treatment. tion

of whole

changes

fluid

WGLF

before

endoscopy,

for laboratory

the course of a standard collected

as described

influenced

in

without

in WGLF between

assayed

by ELISA

active and inactive

the degree direct

deterioration.

of activity.

acceptable

approach

nity will prove

to clinical useful

Concentration discriminates that

investigation

in the analysis

and immunomodulatory

17.

18.

very well grades

19.

this new and of gut immu-

of illness

response to treatment in some complex and has considerable potential in clinical inflammatory

of IgG

IBD and accurately

We suggest

16.

Lavage fluid,

is aesthetically

to laboratory staff, resembling urine rather than feces, and specimens can be stored at -70°C for several months

15.

for bowel

and surgery,

work-up.

14.

by this

can often be obtained

clinical above,

test

and of

20.

IBD patients trials of antiregimens.

21.

References 1. Best WR, Becktel JM, Singleton JW, Kern F. Development of a Crohn’s disease activity index: National Cooperative Crohn’s Disease Study. Gastroenterology 1976;70:439-444. 2. Powell-Tuck J, Bown RL, Lennard-Jones JE. A comparison of oral prednisolone given as a single or multiple daily doses for active proctocolitis. Stand J Gastroenterol 1978; 13:833-837. 3. DiPalma JA, Brady CE, Stewart DL, Karlin DA, McKinney MK, Clement DJ, Coleman TW, Pierson WP. Comparison of colon cleansing methods in preparation for Colonoscopy. Gastroenterology 1984;86:856-860. 4. Gaspari MM, drennan PT, Soloman SM, Elson CO. A method of obtaining, processing, and analysing human intestinal secretions for antibody content. J lmmunol Methods 1988; 110:85-g 1. 5. O’Mahony S, Barton JR, Crichton S, Ferguson A. Appraisal of gut lavage in the study of intestinal humoral immunity. Gut 1990;31:1341-1344. 6. O’Mahony S, Choudari CP, Barton JR, Walker S, Ferguson A. Gut

No. 4

lavage fluid proteins as markers of activity in inflammatory bowel disease. Stand J Gastroenterol 199 1;26:940-944. 7. Bryndon WG, Choudari CP, Ferguson A. Relative specificity for active inflammatory bowel disease of plasma-derived proteins in gut lavage fluid. Eur J Gastroenterol Hepatol (in press). 8. Truelove SC, Witts LJ. Cortisone in ulcerative colitis: Final report on a therapeutic trial. Br Med J 1955;2: 104 1 - 1048. 9. Harvey RF, Bradshaw JM. A simple index of Crohn’sdisease activity. Lancet 1980;1:514. 10. Van Hees PAM, Van Elteran PH, Van Lier HJJ, Van Tongeren JHM. An index of inflammatory activity in patients with Crohn’s disease. Gut 1980;2 1:279-286. 11. De Dombal FT, Burton IL, Clamp SE, Goligher JC. Short term course and prognosis of Crohn’s disease. Gut 1974; 15:435443. 12. Myren J, Bouchier IAD, Watkinson G, Softley A, Clamp SE, De Dombal FT. The OMGE multinational inflammatory bowel disease survey 1976-1982: a further report on 2,657 cases. Stand J Gastroenterol 1984; lS(Suppl 95):1-27. 13. Sandler RS, Jordan MC, Kupper LL. Development of a Crohn’s index for survey research. J Clin Eprdemiol 1988;41:45 l-458.

for collec-

treatment,

radiology,

analysis

resulting of the

protocol is not

tran-

function;

changes

method should lend itself to clinical trials. Because gut lavage is now widely used preparation

used,

intestinal

or pharmacological

a

of isotope-

effects

the standard

gut lavage

with

to measure

defecatory

or mimic

anti-inflammatory

dietary

obtained

if the treatment

influences

mask

Because

concurrent

studies

diet therapy,

sit rate or otherwise

those appears

in interpretation

based or Al AT clearance such as elemental

with

which

Vol. 104,

22.

23.

24.

25.

26.

Sachar DB, Smith H, Chan S, Cohen LB, Lichtiger S, Messer J. Erythrocyte sedimentation rate as a measure of clinical activity in inflammatory bowel disease. J Clin Gastroenterol 1986;8:647650. Fegan EA, Dyck RF, Maton PN, Hodgson HJF, Chadwick VS, Petrie A, Pepys MB. Serum levels of C-reactive protein in Crohn’s disease and ulcerative colitis. Eur J Clin Invest 1982;12:351359. Jansen KB, Jarnum S, Koudahl G, Kristensen M. Serum orosomucoid in ulcerative colitis: its relation to clinical activity, protein loss and turnover of albumin and IgG. Stand J Gastroenterol 1976;11:177-183. Meyers S, Wolke A, Field SP, Feuer EJ, Johnson JW, Janowitz HD. Faecal alpha,-antitrypsin measurement: an indicator of Crohn’s disease activity. Gastroenterology 1985;89:13-18. Harries AD, Fitzsimons E, Fifield R, Dew MJ, Rhodes J. Platelet count: a simple measure of activity in Crohn’s disease. Br Med J 1983;286: 1476. Baklien K, Brandtzaeg P. lmmunohistochemical characterisation of local immunoglobulin formation in Crohn’s disease of the ileum. Stand J Gastroenterol 1976; 11:447-457. Rosekrans PCM, Meijer CJLM, Van Der Wal AM, Cornelisse CJ, Lindeman J. lmmunoglobulin containing cells in inflammatory bowel disease of the colon: a morphometric and immunohistochemical study. Gut 1980;2 1:94 J-947. MacDermott RP, Nash GS, Bertovich MJ, Seiden MV, Bragdon MJ, Beale MG. Alterations of IgM, IgG, and IgA synthesis and secretion by peripheral blood and intestinal mononuclear cells from patients with ulcerative colitis and Crohn’s disease. Gastroenterology 198 1;8 1:844-852. Verspaget HW, Pena AS, Weterman IT, Lamers CBHW. Disordered regulation of the in vitro immunoglobulin synthesis by intestinal mononuclear cells in Crohn’s disease. Gut 1988; 29:503-510. Kett K, Rognum TO, Brandtzaeg P. Mucosal subclass distribution of immunoglobulin G-producing cells is different in ulcerative colitis and Crohn’s disease of the colon. Gastroenterology 1987;93:9 19-924. Singleton JW. Clinical activity assessment in inflammatory bowel disease. Digestive Diseases and Sciences 1987;32(Suppl): 42S-45s. Camilleri M, Proano M. Advances in the assessment of disease activity in mflammatory bowel disease. Mayo Clin Proc 1989;64:800-807. Saverymuttu SH, Cammileri M, Rees H, Lavender JP, Hodgson HJF, Chadwick VS. lndium 11 1-Granulocyte scanning in the

April 1993

27.

28.

assessment of disease extent and disease activity in inflammatory bowel disease. Gastroenterology 1986;90: 1 12 I- 1128. Crama-Bohbouth GE, Arndt JW, Pena AS, Verspaget HW, Tjon A, Tham RT, Weterman IT, Pauwels EK, Lamers LB. Value of indium1 1 1 granulocyte scintigraphy in the assessment of Crohn’s disease of the small intestine: prospective investigation. Digestion 1988;40:227-236. Florent C, L’Hirondel C, Desmazures C, Aymes C, Bernier JJ. Intestinal clearance of Alpha,-antitrypsin: A sensitive method for the detection of protein-losing enteropathy. Gastroenterology

1981;81:777-780. 29. Boirivant M, Pallone F, Ciaco A, Leoni M, Fais S, Torsoli A. Usefulness offaecal Alpha,-antitrypsin clearance and faecal concentration as early indicator of postoperative asymptomatic recurrence in Crohn’s Disease. Dig Dis Sci 1991;36:347-352. 30. Strygler B, Nicar MJ, Santangelo WC, Porter JL, Fordtran JS. Al-

GUT LAVAGE FLUIDS IN IBD

1071

pha,-antitrypsin excretion in stool in normal subjects and in patients with gastro-intestinal disorders. Gastroenterology 1990:99: 1380- 1387.

Received January 6, 1992. Accepted November 24, 1992. Address requests for reprints to C. P. Choudari, M.D., Gastro-lntestinal Unit, Western General Hospital, Edinburgh EH4 2XU, Scotland, U.K. Supported by grants from the Scottish Hospitals Endowment Research Trust, the Nestle Foundation, and the Edinburgh Intestinal Immunology Research Fund. The authors thank the nursing staff of the Gastro-Intestinal Unit for their work in specimen collection, Jackie Johnstone and Norman Anderson for technical assistance, and our consultant colleagues for permission to study their patients.