Efﬁcacy and Safety of Traditional Medical Therapies for ...Consti-pation was combined with the following terms: osmotic laxatives, irritant laxatives, stimulant laxatives, bulk lax-

American Journal of Gastroenterology ISSN 0002-9270C© 2005 by Am. Coll. of Gastroenterology doi: 10.1111/j.1572-0241.2005.40925.xPublished by Blackwell Publishing

Efficacy and Safety of Traditional Medical Therapies forChronic Constipation: Systematic ReviewDavendra Ramkumar, M.D., and Satish S.C. Rao, M.D., Ph.D.Division of Gastroenterology, University of Iowa Carver College of Medicine, Iowa City, Iowa

OBJECTIVES: Constipation is common, and its treatment is unsatisfactory. Although many agents have beentried, there are limited data to support their use. Our aim was to undertake a systematic review ofthe efficacy and safety of traditional medical therapies for chronic constipation and to makeevidence-based recommendations.

METHODS: We searched the English literature for drug trials evaluating treatment of constipation by usingMEDLINE and PUBMED databases from 1966 to 2003. Only studies that were randomized,conducted on adult subjects, and published as full manuscripts were included. Studies wereassigned a quality score based on published methodology. Standard forms were used to abstractdata regarding study design, duration, outcome measures, and adverse events. By using thecumulative evidence of published data for each agent, recommendations were made regardingtheir use following the United States Preventive Services Task Force guidelines.

RESULTS: Good evidence (Grade A) was found to support the use of polyethylene glycol (PEG) and tegaserod.Moderate evidence (Grade B) was found to support the use of psyllium, and lactulose. There was apaucity of quality data regarding many commonly used agents including milk of magnesia, senna,bisacodyl, and stool softeners.

CONCLUSIONS: There is good evidence to support the use of PEG, tegaserod, lactulose, and psyllium. Surprisingly,there is a paucity of trials for many commonly used agents. These aspects should be consideredwhen designing trials comparing new agents with traditional therapies because their use may notbe well validated.

(Am J Gastroenterol 2005;100:936–971)

INTRODUCTION

Constipation is a common problem, with an estimated preva-lence of 2–20% (1–4). It is one of the more common present-ing complaints to both general practitioners and gastroen-terologists, and carries a significant economic impact (4, 5).Constipation appears to be more prevalent in the elderly,women, nonwhites, and persons in lower socio-economic andeducation classes (4).

Although a common problem, the treatment of constipa-tion has been far from satisfactory. A recent metaanalysissuggested that there was little credible evidence to supportmany of the drugs that are commonly used in the treatmentof this disorder (6). However, this analysis lumped all agentsinto a single “laxative group,” which may have obscured anybenefits of individual medications.

It is generally recommended that lifestyle measures such asadequate hydration, nonstrenuous exercise, increased naturalfiber intake, and dedicated time to have a bowel movementbe attempted first before medical therapy is tried. It shouldbe noted that none of these measures has been validated in aproper controlled trial.

With regard to medical therapy, the following categoriesof drugs have been used to treat constipation:

a. Bulk or hydrophilic laxatives—psyllium (isphagula),methylcellulose, bran, celandine, plantain derivatives, andaloe vera

b. Surfactant or softening or wetting agents—docusate,poloxalkol

c. Osmotic laxatives—lactulose, sorbitol, milk of magne-sia (MOM) (magnesium hydroxide), polyethylene glycol(PEG) solutions

d. Peristaltic stimulants or sometimes referred to as irri-tant laxatives—senna, bisacodyl, danthron, cascara, ery-thromycin, misoprostol

e. Others (prokinetic, prosecretors)—colchicine, tegaserod.

The purpose of this systematic review is to assess the avail-able evidence in the English literature, particularly random-ized, controlled trials addressing the efficacy and safety ofvarious medical therapies in adult patients with chronic con-stipation.

936

Efficacy and Safety of Traditional Medical Therapies 937

Literature SearchMEDLINE and PUBMED databases for the period from1966 to 2004 were used to search the literature. Consti-pation was combined with the following terms: osmoticlaxatives, irritant laxatives, stimulant laxatives, bulk lax-atives, fecal softeners, lactulose, sorbitol, MOM, magne-sium sulfate, PEG, senna, bisacodyl, danthron, cascara,psyllium, methylcellulose, calcium polycarbophil, isphagula,bran, celandin, plantain, alovera, aloe vera, docusate, polox-alkol, mineral oil, glycerine, misoprostol, erythromycin, lox-iglumide, tegaserod, herbal remedies, traditional medicine,Chinese herbal, plantain, and colchicine. Exploded termswere reviewed, and where appropriate, the search was ex-panded to include them.

Abstracts of the English language articles were allscreened. Potentially relevant studies were then reviewed, andselection criteria applied. The bibliographies of the studiesfound by this method and in reviews were manually searched.

Selection CriteriaStudies were included if they were (i) randomized (open-labeled or placebo-controlled, parallel design or crossoverdesign) comparing the agent in question with placebo, orcomparing two separate agents for efficacy and safety in pa-tients with chronic constipation; (ii) conducted using adultsubjects; and (iii) published in full manuscript form.

Data Extraction and AnalysisThe articles were reviewed and the relevant data were ab-stracted to standard forms. Data extracted included (i) thetherapy studied; (ii) the control agent; (iii) study design; (iv)number of patients; (v) mean age or age range; (vi) analysisby sex if available; (vii) duration of the study or crossoverperiods and, where necessary, wash-out intervals; (viii) out-come measures including stool frequency and consistency,straining, use of rescue medications; (ix) results in the formof percentage improvement or other suitable variable measur-ing the degree of change in the outcome measure in individ-ual patients as well as between patients treated with differentmeasures; and (x) an assessment of adverse reactions andother aspects of the safety of the treatment measure. Meta-analysis was not performed.

Qualitative Assessment of Study MethodologyThe identified studies were carefully assessed using criteriapreviously established (7, 8), for methodology that minimizesbias and enhances validity of trials about therapy. The follow-ing were evaluated (i) how randomization was performed anddescribed; (ii) use of concealed allocation; (iii) blinding; and(iv) completeness of follow-up. A scoring system was thenused to rate the strength of the studies. A score of 1 or 2was given for randomization (2 for appropriate randomiza-tion technique and concealed allocation explicitly stated ordescribed, 1 for study simply described as “randomized”).Scores of 0–2 were given for blinding (2 when both subjectsand investigators were explicitly said to be blinded to the

treatment by use of identical placebo or other technique, 1when the study is described as “double-blind,”, and 0 whenthe study was not double-blind). A score of 0 or 1 was givenfor frequency of withdrawals (1 when the number of with-drawals and reason for withdrawals were stated and 0 whenno statement was made pertaining to withdrawals). Thus, thequality score ranged from 1 to 5 with 5 being the highest pos-sible score. The studies were all reviewed by both authors,and scored independently. When there were discrepancies,the papers were reviewed again, and the final scores weredecided by consensus.

Levels of Evidence and Grading of RecommendationsThe strength of evidence and grading of recommendationswas as utilized by the U.S. Preventive Services Task Force(9).

LEVELS OF EVIDENCE.

(i) Good evidence (Level I)—consistent results from well-designed, well-conducted studies.

(ii) Fair evidence (Level II)—results show benefit, butstrength is limited by the number, quality, or consistencyof the individual studies.

(iii) Poor evidence (Level III)—insufficient because of lim-ited number or power of studies, flaws in their design orconduct.

CLASSIFICATION OF RECOMMENDATIONS.

(i) Grade A—good evidence in support of the use of amodality in the treatment of constipation.

(ii) Grade B—moderate evidence in support of the use of amodality in the treatment of constipation.

(iii) Grade C—poor evidence to support a recommendationfor or against the use of the modality.

(iv) Grade D—moderate evidence against the use of themodality.

(v) Grade E—good evidence to support a recommendationagainst the use of a modality.

RESULTS

Effectiveness and Safety of PEG Solution in the Treatmentof ConstipationPEG is a nonabsorbable, nonmetabolized osmotic agent thatis most often used in lavage solutions for gut cleansing forcolonoscopy and surgery. Its use as a laxative has garneredmuch interest recently.

Eight studies were found that satisfied the selection criteria(10–17). These are summarized in Table 1. Five of these stud-ies evaluated the efficacy of PEG solutions versus placebo,while one compared PEG solution to lactulose in patients withchronic constipation. Another evaluated the efficacy and tol-erance of PEG solutions, lactulose, and placebo in relieving

938 Ramkumar and RaoTa

ble

1.S

umm

ary

ofth

eT

rial

sE

valu

atin

gth

eE

ffica

cyan

dS

afet

yof

PE

Gin

the

Tre

atm

ento

fC

onst

ipat

ion

Pati

ent

Out

com

esS

tudy

Mea

nA

geR

efer

ence

sS

core

Inte

rven

tion

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

105

PE

G(8

–16

ozs)

orpl

aceb

oD

oubl

e-bl

ind,

plac

ebo-

cont

roll

ed,

cros

sove

r

3262

28/9

Two

5-da

ype

riod

sw

ith

2-da

yw

asho

ut

SF,

SC

,SE

,and

UO

LS

F(P

EG

7.75

±4.

55v

spl

aceb

o4.

88±

2.62

,p

<0.

01)

SC

(PE

G2.

56±

1.17

vs

plac

ebo

1.91

±0.

94,

p<

0.05

)

Sid

eef

fect

sof

PE

G(c

ram

ping

,gas

,na

usea

,unp

leas

ant

tast

e,an

dlo

ose

stoo

ls)

wer

em

inim

alan

dto

lera

ble∗

115

PE

G(2

6g/

day)

orla

ctul

ose

(20

g/da

y)

Mul

tice

nter

,ra

ndom

ized

,co

mpa

rativ

e

99of

115

com

plet

ed30

to≤6

5>

43to

6594

/21

4+

8w

kop

enS

Fan

dst

rain

ing,

use

ofad

diti

onal

laxa

tives

,liq

uid

stoo

ls,fl

atus

,bl

oati

ngru

mbl

ing,

abdo

min

alpa

in

PE

Ggr

oup

had

mor

eda

ilyst

ools

(1.3

vs

0.9,

p=

0.00

5)an

dle

ssst

rain

ing

(p=

0.00

01)

Low

dose

PE

Gw

asm

ore

effe

ctiv

eth

anla

ctul

ose

and

bett

erto

lera

ted

(on

visu

alan

alog

scal

ep

=0.

001)

No

sign

ifica

ntad

vers

eev

ents

inei

ther

grou

p.S

igni

fica

ntly

mor

efl

atul

ence

,wit

ha

tend

ency

tom

ore

bloa

ting

,ab

dom

inal

pain

,an

dru

mbl

ing

wit

hla

ctul

ose

Ope

n-la

bele

dph

ase

wit

hP

EG

foll

owin

gth

est

udy—

nolo

ssof

effi

cacy

Les

sus

eof

addi

tion

alla

xativ

esin

the

PE

Ggr

oup

(p=

0.04

)12

3P

EG

(17

g/da

y)or

plac

ebo

Ran

dom

ized

doub

lecr

osso

ver

tria

lof

plac

ebo

vs17

gof

PE

Gda

ily

2347

.722

/17-

day

plac

ebo

cont

rolp

erio

dth

en14

days

SF,

SC

,EO

D,

cram

ping

,rec

tal

irri

tati

on,a

ndfl

atus

.Adv

erse

even

ts

PE

Gin

crea

sed

mea

nda

ilyS

Fto

1pe

rda

yby

the

last

7da

ysof

the

14-d

aytr

eatm

ent

peri

odvs

plac

ebo,

whi

chpr

ovid

edab

out

1bo

wel

mov

emen

tev

ery

2da

ysdu

ring

the

last

wee

kof

ther

apy

(p=

0.00

01)

No

sign

ifica

ntad

vers

eev

ents

.No

clin

ical

lysi

gnifi

cant

chan

ges

inbl

ood

chem

istr

y,co

mpl

ete

bloo

dco

unt(

CB

C),

orur

inal

ysis

The

rew

asst

atis

tica

llysi

gnifi

cant

impr

ovem

enti

nsu

bjec

tive

scor

esfo

rS

C,E

OD

,cra

mpi

ng,

and

rect

alir

rita

tion

,bu

tnot

for

pass

age

offl

atus

(con

tinu

ed)

Efficacy and Safety of Traditional Medical Therapies 939Ta

ble

1.C

onti

nued

Pati

ent

Out

com

esS

tudy

Mea

nA

geR

efer

ence

sS

core

Inte

rven

tion

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

135

PE

G(1

7g/

day)

orpl

aceb

oM

ulti

cent

er,

doub

le-b

lind

,pl

aceb

o-co

ntro

lled

para

llel

grou

ptr

ial

48of

55co

mpl

eted

25P

EG

,23

plac

ebo

48±

1537

/11

4w

kpl

aceb

oru

nin

,the

n8

wk

(tw

o4-

wk

peri

ods)

SF,

SC

,str

aini

ng,u

seof

laxa

tives

,and

tran

sitt

imes

PE

Gin

crea

sed

SF

at4

wk

and

atth

e8

wk

(PE

G:4

.8±

2.3

vs

plac

ebo:

2.8

±1.

6;p

<0.

002)

PE

Gde

crea

sed

stra

inin

gat

defe

cati

on(p

<0.

01)

The

rew

asno

diff

eren

cebe

twee

nco

ntro

lsan

dP

EG

-tre

ated

pati

ents

inab

dom

inal

sym

ptom

san

dsi

deef

fect

s

PE

Gim

prov

edS

C(p

<0.

02)

Oro

anal

,lef

tcol

on,

and

rect

altr

ansi

tti

mes

wer

esi

gnifi

cant

lysh

orte

ned

byP

EG

trea

tmen

tP

EG

decr

ease

dus

eof

laxa

tives

(p<

0.03

)14

5P

EG

(17

g/da

y)or

plac

ebo

Dou

ble-

blin

d,pl

aceb

o-co

ntro

lled

,pa

rall

elgr

oup

stud

y

70of

78M

ean

43(1

8–73

)58

/12

4w

kru

nin

wit

hP

EG

foll

owed

by rand

omiz

atio

nto

PE

Gor

plac

ebo

for

20w

kin

resp

onde

rs

SF

and

mod

alit

yof

evac

uati

on,U

OL

,an

dre

leva

ntsy

mpt

oms

PE

Ggr

oup

had

sign

ifica

ntly

high

erS

F(w

eek

12:7

.4vs

4.3

BM

/wk,

and

wee

k24

:7.4

vs5.

4B

M/w

k)

No

sign

ifica

ntdi

ffer

ence

sin

adve

rse

even

tsbe

twee

nth

etw

ogr

oups

for

sym

ptom

s(i

nclu

ding

naus

ea,

vom

itin

g,an

ddi

scom

fort

),la

bora

tory

valu

es,h

eart

rate

,and

bloo

dpr

essu

re

Ate

nd,7

7%of

the

PE

Ggr

oup

and

20%

ofth

epl

aceb

ogr

oup

wer

eno

tcon

stip

ated

(p<

0.01

)P

EG

repo

rted

less

hard

/pel

lety

stoo

lsat

defe

cati

onL

ess

UO

Lin

PE

Ggr

oup

and

redu

ced

mea

nnu

mbe

rof

PE

Gsa

chet

sus

edP

EG

grou

pha

dle

ssdr

opou

tsfo

rth

erap

yfa

ilur

e(1

6vs

3;p

<0.

005)

.

940 Ramkumar and Rao

154

PE

G(1

7g/

day)

orpl

aceb

oR

ando

miz

ed,

plac

ebo-

cont

roll

ed,

blin

ded,

mul

tice

nter

para

llel

tria

l

151

45.2

131

F20

M2

wk

SF,

SC

,EO

D,c

ram

ps,

and

flat

us.

Inve

stig

ator

and

pati

ents

ubje

ctiv

eas

sess

men

tof

perc

epti

onof

trea

tmen

tef

fect

iven

ess

SF

impr

oved

wit

hP

EG

wee

k2

4.5

BM

/wk

com

pare

dto

plac

ebo

2.7

BM

/wk

(p<

0.00

1)

No

sign

ifica

ntch

ange

sin

adve

rse

sym

ptom

s,C

BC

,bl

ood

chem

istr

ies,

and

urin

alys

isw

ere

note

dbe

twee

nth

etw

ogr

oups

Pati

ente

valu

atio

nsof

SC

show

edsi

gnifi

cant

impr

ovem

enti

nth

eac

tive

trea

tmen

tgr

oup

(p<

0.00

1)Pa

tien

teva

luat

ions

ofE

OD

show

edsi

gnifi

cant

impr

ovem

enti

nth

eac

tive

trea

tmen

tgr

oup

(p<

0.00

1)In

vest

igat

or(p

<0.

005)

and

pati

ent

(p<

0.00

1)su

bjec

tive

asse

ssm

ento

fpe

rcep

tion

oftr

eatm

ent

effe

ctiv

enes

sw

ere

bett

erw

ith

PE

G16

3P

EG

(8oz

s/da

y),

plac

ebo,

lact

ulos

e(3

0m

l/da

y)

Ran

dom

ized

,tr

iple

cros

sove

raf

ter

cont

rol

run-

in

57—

allo

nop

iate

s18

–50

Not st

ated

One

wee

kru

nin

cont

rol(

notr

eatm

ent)

foll

owed

by3

trea

tmen

tph

ases

of2

wk

each

SF,

SC

,UO

L,E

OD

,an

dS

EP

EG

solu

tion

and

lact

ulos

epr

oduc

edm

ore

“non

hard

”st

ools

than

the

plac

ebo

(p<

0.01

)an

dco

ntro

l(p

<0.

003)

.PE

Gpr

oduc

edth

elo

oses

tst

ool(

p<

0.00

01)

com

pare

dw

ith

the

cont

rol

No

chan

gein

elec

trol

ytes

note

dbe

twee

nth

egr

oups

.L

actu

lose

had

mor

ead

vers

eef

fect

s∗(p

resu

mab

lyab

dom

inal

pain

,bl

oati

ng,a

ndna

usea

,bas

edon

the

desc

ript

ion

ofth

em

etho

d)T

here

wer

eno

sign

ifica

ntdi

ffer

ence

sin

stoo

lco

nsis

tenc

yin

eith

erex

peri

men

talg

roup

,bu

tbot

hw

ere

bett

erth

anha

ving

noth

ing

orju

stth

epl

aceb

oU

OL

incr

ease

don

lyin

grou

ptr

eate

din

the

run-

inco

ntro

lper

iod


Tabl

e1.

Con

tinu

ed

Pati

ent

Out

com

esS

tudy

Mea

nA

geR

efer

ence

sS

core

Inte

rven

tion

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

174

Hyp

o-os

mot

icP

EG

4000

(For

lax)

10or

20g

and

iso-

osm

otic

PE

G33

50(T

rans

ipeg

)5.

9or

11.8

g

Pro

spec

tive,

rand

omiz

ed,

doub

le-b

lind

,pa

rall

el-

grou

p

263

of26

652

±18

.585

% wer

eF

Ran

dom

ized

to1

of4

trea

tmen

tgr

oups

:Hig

hor

low

dose

PE

G40

00,o

rhi

ghor

low

dose

PE

G33

50fo

r4

wk

SF

seco

ndar

yef

fica

cypa

ram

eter

sin

clud

edS

C,d

ate

ofoc

curr

ence

offi

rstm

otio

n,S

t,re

ctal

evac

uati

on,

abdo

min

alpa

in,

and

dist

ensi

on

SF

was

sign

ifica

ntly

incr

ease

dco

mpa

red

wit

hba

seli

nein

all

trea

tmen

tgro

ups

(p=

0.00

01)

wit

hno

diff

eren

cebe

twee

ngr

oups

All

med

icat

ions

wer

ew

ell

tole

rate

d.T

hehi

gher

dose

grou

psha

dm

ore

repo

rts

ofdi

arrh

ea.

Dis

tens

ion,

flat

ulen

ce,a

ndab

dom

inal

pain

occu

rred

even

lyin

allf

our

grou

ps∗ .

SC

sign

ifica

ntly

impr

oved

com

pare

dw

ith

base

line

inal

ltr

eatm

entg

roup

s(p

=0.

0001

).T

hepe

rcen

tage

ofpa

tien

tsw

ith

norm

alS

Cw

assi

gnifi

cant

lyhi

gher

for

stan

dard

-dos

eP

EG

3350

vsbo

thm

axim

um-d

ose

trea

tmen

ts(p

<0.

01).

≥67.

3%ha

dth

eir

firs

tst

oolw

ithi

n1

day

ofst

arti

ngtr

eatm

ent

Rec

tale

vacu

atio

n,st

rain

ing,

bloa

ting

,an

dpa

inw

ere

also

sign

ifica

ntly

impr

oved

com

pare

dw

ith

base

line

inal

ltr

eatm

entg

roup

s(p

=0.

0001

)

SF,

stoo

lfre

quen

cy;S

C,s

tool

cons

iste

ncy;

St,

stra

inin

g;U

OL

,use

ofad

diti

onal

laxa

tives

;EO

D,e

ase

ofde

feca

tion

;SE

,sid

eef

fect

s;,n

otot

herw

ise

spec

ified

.


Table 2. Methodologic Quality of Trials with PEG

Statement onReference Randomization Blinding Withdrawals Total Score

10 2 2 1 511 2 2 1 512 1 2 0 314 2 2 1 513 2 2 1 515 1 2 1 416 1 2 0 317 1 2 1 4

A score of 1 or 2 was given for randomization (2 for appropriate randomizationtechnique and concealed allocation explicitly stated or described, 1 for study simplydescribed as “randomized”). Scores of 0–2 were given for blinding (2 when bothsubjects and investigators were blinded to the treatment by use of identical placeboor other technique, 1 when the study is described as “double-blind,” and 0 when thestudy was not double-blind). Score of 0 or 1 was given for frequency of withdrawals (1when the number of withdrawals and reason for withdrawals were stated and 0 whenno statement was made pertaining to withdrawals).

opiate-induced constipation (16). The last study comparedtwo doses of two commercially available PEG formulations,an iso-osmotic preparation, and a hypo-osmotic solution (17).Constipation was variously defined, with only two studiesutilizing the Rome criteria to identify suitable patients (13,14). Six of the studies evaluated the short-term efficacy andsafety of PEG solution. The longest duration in this categorywas 8 wk. One trial looked at the same parameters over a 6-month period (14) and therefore provides longer-term data.The qualitative assessment of these studies and the extracteddata are presented in Table 2.

PEG is an effective form of treatment with few side effectsand it is modestly more effective than lactulose. Decisionanalysis modeling suggests that PEG, despite its higher cost,is ultimately more cost-effective than lactulose, at least fromthe perspective of the National Health Service of the UnitedKingdom. (18).

PEG: Level I Evidence, Grade A Recommendation.

Efficacy and Safety of Lactulose in the Treatmentof ConstipationLactulose is a nonabsorbable synthetic disaccharide whichfunctions as an osmotic laxative and which appears to havebeen accorded the role of the standard against which neweragents are compared for efficacy and safety. Three studiescompared lactulose with placebo (19–21). The others wereall comparisons with other agents in which the efficacy oflactulose was determined by the improvement from baselinein the parameters that assess constipation (11, 16, 22–27).The characteristics and results are summarized in Table 3.The qualitative assessments of these studies are summarizedin Table 4.

Lactulose appears to be an effective and safe agent foruse in idiopathic constipation, with the most common sideeffects (bloating, flatulence, and loose stools) being an ex-tension of its mechanism of action. Compared to PEG solu-tions, lactulose was less efficacious and had more side effects

(11, 16). An open-labeled, randomized, parallel study whichcompared lactulose, psyllium, and placebo suggested that thetwo treatment agents were equally effective in the treatmentof constipation (26). A single trial that compared the efficacyand safety of lactulose and sorbitol suggested that the twoagents were similarly effective, but lactulose had a greaterpropensity to cause nausea (23).

Lactulose: Level II Evidence, Grade BRecommendation.

Sorbitol: Level III Evidence, Grade CRecommendation.

Efficacy and Safety of MOM in the Treatment ofConstipationThere is one trial in the English literature evaluating the ef-ficacy of MOM in the treatment of constipation (28). MOMwas compared to a bulk laxative and was found to causemore frequent bowel movements than bulk laxative, and addi-tional laxative was not needed as often as with bulk laxative.Stool consistency was more normal during the magnesiumhydroxide treatment. In two patients, serum magnesium wasover 1.25 mmol/L after the magnesium hydroxide treatmentbut there were no clinical signs of hypermagnesemia. Thisis more likely to be a problem in patients with renal insuf-ficiency. Hypermagnesemia, paradoxically, causes paralyticileus that in turn leads to obstipation. This study suggeststhat MOM is effective, but there is a risk of hypermagne-semia with frequent use.

There are no trials that have evaluated the utility of mag-nesium sulfate (Epsom salts) in the treatment of constipationafter 1966.

MOM: Grade III Evidence, Grade CRecommendation.

Efficacy and Safety of Stimulant Laxatives in theTreatment of ConstipationFor the period reviewed, no placebo-controlled trials werefound. The studies reviewed all compared a laxative contain-ing only an irritant/stimulant agent, or agents containing anirritant/stimulant as one of its active ingredients with otheragents (22, 24, 25, 27, 29–34). Tables 5 and 6 summarizethe study characteristics and methodologic assessment. Threestudies that compared a preparation containing psyllium andsenna with lactulose suggested that the fiber/stimulant com-bination was more efficacious and may even be more cost-effective (22, 24, 25). Another study compared lactulose withirritant laxatives containing senna, anthraquinone derivatives,or bisacodyl (31). Lactulose was found to be more effectivethan each of these three irritant laxatives. A comparison be-tween bisacodyl and bisoxatin acetate, both irritant laxatives,showed similar efficacy (32). The latter agent is no longermarketed. Preparations with and without senna were evalu-ated (30); laxation appears to be similar, with the combina-tion appearing to cause more side effects compared with just


Tabl

e3.

Sum

mar

yof

the

Tri

als

Eva

luat

ing

the

Effi

cacy

and

Saf

ety

ofL

actu

lose

inth

eT

reat

men

tof

Con

stip

atio

n

Pati

ent

Out

com

esS

tudy

Mea

nA

geR

efer

ence

sS

core

Inte

rven

tion

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

193

Lac

tulo

se(6

0m

l/da

y)or

plac

ebo

Con

stip

ated

subj

ects

doub

le-b

lind

para

llel

2410

took

lact

ulos

e14

took

plac

ebo

28.2

22/2

1w

kba

seli

ne—

allt

ook

plac

ebo

(sin

gle-

blin

d),

the

seco

ndw

eek

was

ado

uble

-bli

ndtr

eatm

ent

peri

od

SF,

stoo

lwei

ght,

volu

me,

SC

,and

wat

erco

nten

t

SF

—la

ctul

ose

syru

ppr

oduc

edcl

inic

ally

and

stat

isti

cally

sign

ifica

ntin

crea

ses

(4.5

vs1.

6st

ools

per

wee

k,p

<0.

05)

Adv

erse

effe

cts

wer

eal

lex

tens

ions

(flat

ulen

ce)

ofth

eph

arm

acol

ogic

effe

cts

ofth

edr

ugan

dw

ere

inge

nera

lwel

lto

lera

ted

SC

—la

ctul

ose

prod

uced

stoo

lsof

soft

erco

nsis

tenc

yco

mpa

red

toba

seli

neva

lues

,as

wel

las

toa

sucr

ose-

trea

ted

cont

rolg

roup

inbo

thno

rmal

and

cons

tipa

ted

subj

ects

Lac

tulo

sepr

oduc

edst

ools

ofgr

eate

rw

eigh

t,vo

lum

e,an

dw

ater

cont

ent

214

Lac

tulo

se(1

5–30

ml/

day)

orpl

aceb

odo

sees

cala

tion

allo

wed

Mul

tice

nter

,ra

ndom

ized

,do

uble

-bli

nd,

plac

ebo-

cont

roll

ed,

para

llel

stud

y

103

>60

Not st

at-e

d2-

wk

run-

in.

3-w

ktr

eatm

ent,

then

2-w

kle

adou

t

SF,

UO

L,p

ossi

ble

SE

The

succ

ess

rate

for

lact

ulos

ew

as86

vs60

%pl

aceb

o(p

<0.

02).

Tru

lyco

nsti

pate

dpa

tien

tsw

ere

dete

rmin

edba

sed

onth

eir

laxa

tive

requ

irem

ent

post

trea

tmen

t.T

hesu

cces

sra

teof

lact

ulos

ew

asin

crea

sed

to80

%in

thes

epa

tien

tsan

dpl

aceb

o33

%(p

<0.

01)

Non

em

enti

oned


204

Lac

tulo

se(3

0m

lof

50%

solu

tion

)or

plac

ebo

(glu

cose

)

Ran

dom

ized

,do

uble

-bli

nd,

plac

ebo-

cont

roll

ed

47of 55

—nu

rsin

gho

me

pati

ents

19la

ctul

ose,

23gl

ucos

e

84.7

39F

6M

2-w

kru

n-in

,12

-wk

trea

tmen

t,th

en1-

wk

obse

rvat

ion

SF,

sym

ptom

sof

cons

tipa

tion

,ep

isod

esof

stoo

lim

pact

ion,

UO

L

SF

—la

culo

sew

assu

peri

orto

gluc

ose

wit

hm

ean

BM

per

day

of0.

7vs

0.5

(p<

0.02

)an

din

the

perc

enta

geof

days

inw

hich

atle

asto

nebo

wel

mov

emen

tocc

urre

d(p

<0.

05)

No

adve

rse

clin

ical

ofla

bora

tory

effe

cts

Red

ucti

onin

the

seve

rity

ofea

chof

cram

ping

,gri

ping

,fl

atul

ence

,te

nesm

us,b

loat

ing

was

grea

ter

wit

hla

ctul

ose.

For

reli

efof

allfi

vesy

mpt

oms,

lact

ulos

ew

asm

ore

effe

ctiv

eth

angl

ucos

e(p

<0.

04)

The

redu

ctio

nin

the

num

ber

offe

cal

impa

ctio

ns(6

inth

ela

ctul

ose

pati

ents

vs66

inth

eco

ntro

ls)

was

high

lysi

gnifi

cant

(p<

0.01

5)T

hela

ctul

ose

pati

ents

need

edfe

wer

enem

asth

andi

dth

eco

ntro

ls11

5P

EG

(26

g/da

y)or

lact

ulos

e(2

0g/

day)

Mul

tice

nter

,ra

ndom

ized

,co

mpa

rativ

e

115–

91co

mpl

eted

30to

≤65

>43

to65

94/2

1Tw

o5-

day

peri

ods

wit

h2-

day

was

hout

SF

and

St

PE

Ggr

oup

had

mor

eda

ilyst

ools

and

less

stra

inin

g

No

sign

ifica

ntad

vers

eev

ents

inei

ther

grou

p.M

ore

flat

ulen

cew

ith

lact

ulos

eL

owdo

seP

EG

was

mor

eef

fect

ive

than

lact

ulos

ean

dbe

tter

tole

rate

dO

pen-

labe

led

phas

ew

ith

PE

Gfo

llow

ing

the

stud

y—no

loss

ofef

fica

cy

(con

tinu

ed)


ble

3.C

onti

nued

Pati

ent

Out

com

esS

tudy

Mea

nA

geR

efer

ence

sS

core

Inte

rven

tion

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

163

PE

G(8

ozs/

day)

,pl

aceb

o,la

ctul

ose

(30

ml/

day)

Ran

dom

ized

,tr

iple

cros

sove

raf

ter

cont

rol

run-

in

57—

allo

nop

iate

s18

–50

Not st

ated

1-w

kru

nin

cont

rol(

notr

eatm

ent)

foll

owed

byth

ree

trea

tmen

tph

ases

of2

wk

each

SF,

SC

,and

EO

DP

EG

solu

tion

and

lact

ulos

epr

oduc

edm

ore

“non

hard

”st

ools

than

the

plac

ebo

(p<

0.01

)an

dco

ntro

l(p

<0.

003)

.PE

Gpr

oduc

edth

elo

oses

tst

ool(

p<

0.00

01)

com

pare

dw

ith

the

cont

rol

Lac

tulo

seha

dm

ore

adve

rse

effe

cts

The

rew

ere

nosi

gnifi

cant

diff

eren

ces

inS

Cin

eith

erex

peri

men

tal

grou

p,bu

tbot

hw

ere

bett

erth

anha

ving

noth

ing

orju

stth

epl

aceb

o22

2L

actu

lose

30cc

ora

prep

arat

ion

cont

aini

ngis

pagh

ula

(psy

lliu

m)

and

senn

a(A

giol

ax)

Ope

n,ra

ndom

ized

,an

dco

ntro

lled

cros

sove

r

30lo

ng-s

tay

geri

atri

cpa

tien

ts

81.8

25/5

1-w

kru

nin

foll

owed

two

5-w

ktr

eatm

ent

peri

ods

sepa

rate

dby

1w

k

SF,

SC

,UO

L(b

isac

odyl

),an

dS

E

The

Agi

olax

prod

uced

4.5

BM

/wk

(in

both

peri

ods)

com

pare

dw

ith

2.2

and

1.9

per

wk

inpe

riod

son

ean

dtw

o,re

spec

tivel

y,fo

rla

ctul

ose

No

sign

ifica

ntad

vers

eef

fect

sno

ted

The

freq

uenc

yof

loos

est

ools

was

grea

ter

wit

hA

giol

axth

anw

ith

lact

ulos

e(p

<0.

05)

245

Lac

tulo

se(3

0–60

ml/

day)

and

com

bina

tion

ofps

ylli

uman

dse

nna

pod

(Agi

olax

10–2

0m

l/da

y)

Mul

tice

nter

,do

uble

-bli

ndcr

osso

ver

77of

85lo

ng-s

tay

geri

atri

cpa

tien

ts

82.9

57/2

8S

enna

-fibr

eco

mbi

nati

onor

lact

ulos

ew

ith

mat

chin

gpl

aceb

ofo

rtw

o14

-day

peri

ods,

wit

h3–

5da

ysbe

fore

and

betw

een

trea

tmen

ts

SF,

SC

,EO

D;

devi

atio

nfr

omre

com

men

ded

dose

;dai

lydo

sean

dco

stpe

rst

ool;

adve

rse

effe

cts

SF

was

grea

ter

wit

hth

ese

nna-

fibr

eco

mbi

nati

on(0

.8pe

rda

yth

anla

ctul

ose

(0.6

per

day,

p≤

0.00

1)

No

diff

eren

cein

adve

rse

effe

cts

Sco

res

for

SC

and

EO

Dw

ere

sign

ifica

ntly

high

erfo

rth

ese

nna-

fibr

eco

mbi

nati

onth

anfo

rla

ctul

ose


The

reco

mm

ende

ddo

sew

asex

ceed

edm

ore

freq

uent

lyw

ith

lact

ulos

eth

anth

ese

nna-

fibr

eco

mbi

nati

on(χ

2=

8.38

,p

≤0.

01)

The

cost

per

stoo

lwas

appr

oxim

atel

yfo

urti

mes

high

erfo

rla

ctul

ose

than

for

the

senn

a-fi

ber

com

bina

tion

262

Lac

tulo

se(3

0m

l/da

y)or

isph

agul

a(p

syll

ium

)7

g/da

y

Ope

n,ra

ndom

ized

,pa

rall

elgr

oup

stud

y

112

50.6

Not st

ated

4w

kS

F,S

C,S

t,gl

obal

impr

ovem

ent,

med

icat

ion

acce

ptab

ilit

y

Bot

htr

eatm

ents

resu

lted

inst

atis

tica

llysi

gnifi

cant

(p<

0.00

01)

incr

ease

sin

stoo

lfre

quen

cyov

erba

seli

nebu

tnot

betw

een

the

trea

tmen

tgro

ups

(bas

elin

e2

per

wee

kvs

6.5

per

wee

kfo

rla

ctul

ose

and

7.5

per

wk

for

ispa

ghul

a)

No

seri

ous

adve

rse

effe

cts

Bot

htr

eatm

ents

caus

edim

prov

edS

C(p

=0.

027)

,but

ther

ew

ere

nodi

ffer

ence

sbe

twee

nth

egr

oups

The

rew

asno

sign

ifica

ntdi

ffer

ence

inS

tT

here

was

nosi

gnifi

cant

clin

ical

diff

eren

cein

glob

alim

prov

emen

tM

ore

pati

ents

foun

dis

pagh

ula

unpa

lata

ble

at28

days

(15.

7vs

4.2%

)p

=0.

063

(con

tinu

ed)


Tabl

e3.

Con

tinu

ed

Pati

ent

Out

com

esS

tudy

Mea

nA

geR

efer

ence

sS

core

Inte

rven

tion

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

253

Lac

tulo

se15

ml

orA

giol

ax10

ml(

mix

ture

ofis

pagh

ula

and

senn

a)

Ran

dom

ized

,do

uble

-bli

nd,

cros

sove

r

77lo

ng-s

tay

geri

atri

cpa

tien

ts

82.9

57/2

0Tw

o14

-day

trea

tmen

tpe

riod

sw

ith

3–5

dla

xativ

efr

eepe

riod

befo

rean

dbe

twee

ntr

eatm

ents

SF,

SC

,EO

D,

adve

rse

effe

cts

Agi

olax

resu

lted

inst

atis

tica

llysi

gnifi

cant

incr

ease

sin

SF

(0.8

per

day

vs

0.6

per

day

p<

0.00

1)

No

diff

eren

ces

inad

vers

eef

fect

sbe

twee

nth

etr

eatm

entg

roup

s

Agi

olax

resu

lted

inst

atis

tica

llysi

gnifi

cant

impr

ovem

enti

nS

C(p

<0.

005)

Agi

olax

resu

lted

inst

atis

tica

llysi

gnifi

cant

impr

ovem

enti

nE

OD

(p=

0.02

)23

5S

orbi

tola

ndla

ctul

ose—

upto

60m

l/da

yof

eith

er

Ran

dom

ized

,do

uble

-bli

nd,

cros

sove

rtr

ial

3065

–86

All

men

Lac

tulo

sean

d70

%so

rbit

olw

ere

each

give

nfo

r4

wk

prec

eded

bya

2-w

kw

asho

utpe

riod

SF,

pref

eren

ceof

laxa

tive,

adve

rse

effe

cts

The

aver

age

num

ber

ofbo

wel

mov

emen

tspe

rw

eek

was

6.71

wit

hso

rbit

olan

d7.

02w

ith

lact

ulos

e,an

dth

eav

erag

enu

mbe

rof

days

per

wee

kw

ith

bow

elm

ovem

ents

was

5.23

wit

hso

rbit

olan

d5.

31w

ith

lact

ulos

e

The

rew

ere

nosi

gnifi

cant

diff

eren

ces

betw

een

sorb

itol

and

lact

ulos

ein

any

outc

ome

mea

sure

dex

cept

naus

ea,w

hich

was

incr

ease

dw

ith

lact

ulos

e(p

≤0.

05).

11pa

tien

tsst

ated

apr

efer

ence

for

sorb

itol

,12

for

lact

ulos

e,an

d7

had

nopr

efer

ence


On

avi

sual

anal

ogsc

ale

mea

suri

ngse

veri

tyof

cons

tipa

tion

(0–1

00m

m),

the

aver

age

scor

efo

rso

rbit

olw

as35

.6vs

37.1

mm

for

lact

ulos

eT

heso

rbit

olan

dla

ctul

ose

trea

tmen

tpe

riod

sw

ere

also

sim

ilar

inpe

rcen

tof

bow

elm

ovem

ents

reco

rded

as“n

orm

al”

271

Lac

tulo

se(2

0m

l/da

y)or

Dor

bane

x10

ml/

day—

mix

ture

ofpo

loxa

lkol

and

anan

thra

quin

one—

orco

ntro

l(e

nem

asif

noB

Min

5da

ys)

Ran

dom

ized

,op

entr

ial

wit

hco

ntro

lgr

oup

37lo

ng-s

tay

geri

atri

cpa

tien

ts

82N

ot stat

edR

ando

mly

assi

gned

toon

eof

the

trea

tmen

tgr

oups

for

27da

ys

80%

tran

sitt

ime

(TT

—ti

me

for

80%

ofin

gest

edm

arke

rsto

bede

feca

ted)

All

pati

ents

had

80%

TT

prio

rto

trea

tmen

t.10

/14,

12/1

5,an

d6/

6ha

s80

%T

Taf

ter

6d

ays

foll

owin

g27

days

oftr

eatm

ent

wit

hla

ctul

ose,

Dor

bane

x,an

dco

ntro

l,re

spec

tivel

y

No

seri

ous

adve

rse

effe

cts.

19la

ctul

ose,

23gl

ucos

e


Table 4. Methodologic Scores for the Studies Evaluating Lactulose


11 2 2 1 519 1 2 0 322 1 0 1 223 2 2 1 524 2 2 1 520 1 2 1 421 2 2 0 426 1 0 1 225 1 2 0 327 1 0 0 116 1 2 0 3

A score of 1 or 2 was given for randomization (2 for appropriate randomizationtechnique and concealed allocation explicitly stated or described, 1 for studysimply described as “randomized”). Scores of 0–2 were given for blinding (2when both subjects and investigators were blinded to the treatment by use ofidentical placebo or other technique, 1 when the study is described as “double-blind,” and 0 when the study was not double-blind). Score o f 0 or 1 was givenfor frequency of withdrawals (1 when the number of withdrawals and reason forwithdrawals were stated, and 0 when no statement was made pertaining to withdrawals).

psyllium. Sodium picosulfate, a stimulant laxative similar tobisacodyl, has been compared to senna. The agents appearto be similar in their beneficial effects on stool frequency,although sodium picosulfate seems to cause more side ef-fects. An agent comprised of the combination of a surfactant(poloxalkol) and an anthraquinone performed well comparedwith placebo in postpartum constipation (33), and similar tolactulose in another study (27). This combined agent wassimilar in efficacy to the stimulant sodium picosulfate (34).

While few studies suggest that combining stimulant lax-atives with fiber or surfactant agents may provide some re-lief of symptoms in patients with constipation, there are noplacebo-controlled trials.

Stimulant Laxatives: Level III Evidence,Grade C Recommendation.

Efficacy and Safety of Bulk Laxatives in the Treatmentof ConstipationA study which compared increased dietary fiber with regu-lar diet in posthysterectomy patients suggested that increaseddietary fiber is beneficial in improving stool frequency, stoolconsistency, time to defecate, and other symptoms of difficultdefecation (35). The individual agents that can be used to sup-plement the diet with fiber are discussed below. The charac-teristics of the trials evaluating bulking agents and qualitativeassessment of the methodology are summarized in Tables 7and 8.

Efficacy and Safety of Methylcellulose in the Treatmentof ConstipationThere are no placebo-controlled trials. One study (36) com-paring three doses of methylcellulose against psyllium satis-fied the screening criteria and is summarized in Table 5. Thelack of an appropriate control group and its low methodologicscore argues against accepting the results.

Methylcellulose: Level III Evidence, Grade CRecommendation.

Efficacy and Safety of Bran in the Treatmentof ConstipationThe studies evaluating the efficacy of bran and increased di-etary fiber on constipation all suggest benefits (37–41). Ina trial of the addition of wheat bran and corn bran supple-ments compared with no fiber supplementation, there wasimprovement in stool frequency and consistency, with noappreciable side effects (37). When bran was compared tojust a regular diet in elderly patients, there was a decreasein laxative requirements, but, paradoxically, these patientsseemed to require more assistance with actual defecationas evidenced by increased use of enemas and suppositories(41). A smaller study compared wheat bran with regular dietin elderly patients (42). There was significant improvementin stool frequency and consistency (42). No significant dif-ference was found in laxative or suppository requirements.Another study also showed the beneficial effects on stoolfrequency, and also suggested that oroanal transit times im-proved, but only in patients with slow colonic transit and notin those with slow rectal transit times (38). Comparison ofcorn bran and wheat bran showed that both products werebeneficial from the standpoint of improvement in stool fre-quency and intestinal transit times, but corn bran was rated bypatients as being better at relieving the symptoms of consti-pation (39). A comparison of bran with senna suggested thatthere was no significant difference on frequency and consis-tency of stools, but bran decreased the incidence of “large”stools (40).

Level III Evidence, Grade C Recommendation.

Efficacy and Safety of Psyllium in the Treatmentof ConstipationPsyllium (ispagula), a derivative of the husk of Plantagoovata, has been evaluated in several trials (30, 36, 43–47).Compared with placebo, psyllium seems to clearly improvestool frequency (43, 44). One study suggested that total guttransit time improved (44), while another suggested that therewas no change in colon transit (43). Similarly, the effect onstool consistency in these placebo-controlled trials was con-troversial with one study suggesting no change (44) and an-other suggesting significant improvement (43). A third largertrial with a single-blind design comparing psyllium withplacebo showed statistically significant improvement in bothstool frequency and consistency with both the investigator,and patient noting significant improvement in the constipa-tion (48). In an open trial, psyllium was noted to be superiorto three different stimulant/irritant laxatives, lactulose, andmagnesium sulfate in the treatment of constipation as well asbeing more palatable and acceptable to patients (46). A com-parison of a preparation of psyllium with senna, and psylliumalone showed the combination to be more effective than


Tabl

e5.

Sum

mar

yof

the

Tri

als

Eva

luat

ing

the

Effi

cacy

and

Saf

ety

ofIr

rita

ntan

dS

tim

ulan

tLax

ativ

esin

the

Tre

atm

ento

fC

onst

ipat

ion

Pati

ent

Out

com

es

Ref

eren

ces

Sco

reIn

terv

enti

onS

tudy

Des

ign

NM

ean

Age

Wom

enD

urat

ion

Out

com

eM

easu

reR

esul

tsS

afet

yA

naly

sis

312

Lac

tulo

se15

ml

b.i.d

.or

“irr

itan

t”la

xativ

es(o

fpa

tien

tsch

oice

)co

ntai

ning

senn

a,an

thra

quin

one

deri

vativ

esor

bisa

cody

l

Ope

n,ra

ndom

ized

,cr

osso

ver

194

of22

7N

otst

ated

Not

stat

edTw

otr

eatm

ent

peri

ods

of1

wk

sepa

rate

dby

1-w

kw

asho

ut

Sto

olco

nsis

tenc

y,si

deef

fect

sB

yda

y7,

58%

ofth

ela

ctul

ose-

trea

ted

grou

pw

ere

pass

ing

ano

rmal

stoo

lw

here

ason

ly42

%(p

<0.

001)

ofth

epa

tien

tsre

ceiv

ing

an“r

rita

nt”

laxa

tive

Lac

tulo

sepr

epar

atio

nha

da

pers

iste

ntca

rry-

over

effe

ctth

anth

eir

rita

ntla

xativ

es

No

diff

eren

cew

asno

ted

inth

ere

cord

ing

ofsi

deef

fect

sdu

ring

trea

tmen

tand

nont

reat

men

tpe

riod

s

322

Bis

acod

yl10

mg/

day

orbi

soxa

tin

acet

ate

60m

g/da

y

Ran

dom

ized

,do

uble

-bli

nd,

com

para

tive

51of

6121

–69

All

mal

epr

ison

ers

Two

cons

ecut

ive

4-w

kpe

riod

sO

nset

offi

rstB

M,

stoo

lfre

quen

cy,

cons

iste

ncy,

pati

ent

sati

sfac

tion

,ad

vers

ere

acti

ons

The

rew

ere

nodi

ffer

ence

sbe

twee

nbi

soxa

ntin

and

bisa

cody

lin

the

mea

nti

me

tofi

rst

BM

(7vs

5h)

Bis

acod

ylap

pear

edto

caus

em

ore

abdo

min

alpa

in

No

diff

eren

cein

stoo

lfr

eque

ncy

(1.7

per

day

vs2.

1pe

rda

y)N

odi

ffer

ence

inst

ool

cons

iste

ncy

The

rew

ere

less

sati

sfac

tion

wit

hbi

saco

dyl(

73%

vs82

%)

222

Lac

tulo

se30

ccor

apr

epar

atio

nco

ntai

ning

ispa

ghul

a(b

ulki

ngag

ent)

and

senn

a(A

giol

ax)

Ope

n,ra

ndom

ized

,an

dco

ntro

lled

cros

sove

r

30lo

ng-s

tay

geri

atri

cpa

tien

ts

81.8

25F

5M

1-w

kru

nin

foll

owed

two

5-w

ktr

eatm

ent

peri

ods

sepa

rate

dby

1w

k

Sto

olfr

eque

ncy,

cons

iste

ncy,

use

ofad

diti

onal

laxa

tive

(bis

acod

yl)

and

side

effe

cts

The

Agi

olax

prod

uced

4.5

stoo

lsa

wee

k(i

nbo

thpe

riod

s)co

mpa

red

wit

h2.

2an

d1.

9pe

rw

eek

inpe

riod

son

ean

dtw

o,re

spec

tivel

y,fo

rla

ctul

ose

The

freq

uenc

yof

loos

est

ools

was

grea

ter

wit

hA

giol

axth

anw

ith

lact

ulos

e(p

<0.

05)

No

sign

ifica

ntad

vers

eef

fect

sno

ted

(con

tinu

ed)


ble

5.C

onti

nued

Pati

ent

Out

com

es

Ref

eren

ces

Sco

reIn

terv

enti

onS

tudy

Des

ign

NM

ean

Age

Wom

enD

urat

ion

Out

com

eM

easu

reR

esul

tsS

afet

yA

naly

sis

302

Psy

lliu

m(P

)an

dps

ylli

umw

ith

senn

a(P

S)

Ope

n,ra

ndom

ized

,si

ngle

-bli

ndco

ntro

lled

40of

4226

.135

F5

M1

wk

plac

ebo

and

1w

ktr

eatm

ent

SF,

SC

sym

ptom

sw

ith

BM

sB

oth

laxa

tives

incr

ease

dS

F[P

3.6

BM

/wk

vsP

S6.

8B

M/w

k(p

<0.

001)

]B

oth

Pan

dP

Sin

crea

sed

wet

and

dry

stoo

lwei

ghts

wit

hth

ead

ded

effe

ctof

the

senn

acl

earl

yev

iden

t.O

nly

the

psyl

lium

wit

hse

nna

incr

ease

dst

ool

moi

stur

e

Pgr

oup

3/22

cram

ping

and

gas

PS

grou

p7/2

2ha

dcr

amps

,un

com

fort

able

diar

rhea

,bl

oati

ngga

s,an

dna

usea

Bot

hla

xativ

esim

prov

edS

C(s

imil

ar)

Bot

hla

xativ

espr

ovid

eda

high

degr

eeof

subj

ectiv

ere

lief

245

Lac

tulo

se(L

)30

ml/

day

and

com

bina

tion

offi

bre—

ispa

ghul

a(p

syll

ium

)an

dse

nna

pod

(PS

)—M

anev

ac10

–20

ml/

day

Mul

tice

nter

,do

uble

-bli

nd,

cros

sove

r

77of

85lo

ng-s

tay

geri

atri

cpa

tien

ts

82.9

57F

Sen

na-fi

bre

com

bina

tion

orla

ctul

ose

wit

hm

atch

ing

plac

ebo

for

two

14-d

aype

riod

s,w

ith

3–5

days

befo

rean

dbe

twee

ntr

eatm

ents

SF,

SC

,EO

D;

devi

atio

nfr

omre

com

men

ded

dose

;dai

lydo

se,

and

cost

per

stoo

l;ad

vers

eef

fect

s

SF

was

grea

ter

wit

hth

ese

nna-

fibr

eco

mbi

nati

on(0

.8pe

rda

yth

anla

ctul

ose

(0.6

per

day,

p≤

0.00

1)S

core

sfo

rS

Can

dE

OD

wer

esi

gnifi

cant

lyhi

gher

for

the

senn

a-fi

bre

com

bina

tion

than

for

lact

ulos

eT

here

com

men

ded

dose

was

exce

eded

mor

efr

eque

ntly

wit

hla

ctul

ose

than

the

senn

a-fi

bre

com

bina

tion

(χ2=

8.38

,p

≤0.

01)

The

cost

per

stoo

lwas

appr

oxim

atel

yfo

urti

mes

high

erfo

rla

ctul

ose

than

for

the

senn

a-fi

ber

com

bina

tion

No

diff

eren

cein

adve

rse

effe

cts


253

Lac

tulo

se15

ml

orA

giol

ax10

ml(

mix

ture

ofIs

pagh

ula

(psy

lliu

m)

and

senn

a)

Ran

dom

ized

,do

uble

-bli

nd,

cros

sove

r

77lo

ng-s

tay

geri

atri

cpa

tien

ts

82.9

57F

20M

Two

14-d

aytr

eatm

ent

peri

ods

wit

h3–

5da

ysla

xativ

efr

eepe

riod

befo

rean

din

betw

een

trea

tmen

ts

SF,

SC

,EO

D,

adve

rse

effe

cts

Agi

olax

resu

lted

inst

atis

tica

llysi

gnifi

cant

incr

ease

inS

F(0

.8pe

rda

yvs

0.6

per

day,

p<

0.00

1)A

giol

axre

sult

edin

stat

isti

cally

sign

ifica

ntim

prov

emen

tin

SC

(p<

0.00

5)A

giol

axre

sult

edin

stat

isti

cally

sign

ifica

ntin

crea

ses

in(p

=0.

02)

EO

D

No

diff

eren

ces

inad

vers

eef

fect

sbe

twee

nth

etr

eatm

entg

roup

s

292

Sod

ium

pico

sulf

ate

SP

10m

g/da

yor

stan

dard

ized

senn

a(S

)2

tabs

/day

Ope

n,ra

ndom

ized

para

llel

50lo

ng-s

tay

geri

atri

cpa

tien

ts

78.5

36F

14M

2w

kS

tool

freq

uenc

y,ti

min

g,co

nsis

tenc

y,an

dsi

deef

fect

s

The

num

ber

ofB

Ms

per

wee

kw

ere

sim

ilar

for

S(4

.41)

and

SP

(4.9

7)S

Pca

used

mor

elo

ose

orun

form

edst

ools

than

senn

a

The

rew

ere

few

min

orsi

deef

fect

sal

lrel

ated

toth

eac

tion

ofth

edr

ugs

onth

ebo

wel

333

Dor

bane

x(m

ixtu

reof

polo

xalk

olan

ddi

hydr

o-an

thro

quin

olon

e)(P

D)

orpl

aceb

o

Ran

dom

ized

,do

uble

-bli

nd,

plac

ebo-

cont

roll

ed

200

post

par-

tum

pati

ents

wit

hco

n-st

ipat

ion

Chi

ld-

bear

ing

age

rang

e(c

anno

tin

fer

from

data

)

All

wom

en6

days

Sto

olou

tput

,re

quir

emen

tfor

enem

asan

dsu

ppos

itor

ies,

By

the

seco

ndda

yof

trea

tmen

t50%

inth

etr

eatm

entg

roup

had

aB

Mco

mpa

red

wit

hle

ssth

anon

e-ei

ghth

inth

epl

aceb

ogr

oup

(p<

0.00

1).

Dec

reas

edre

quir

emen

tfo

ren

emas

and

supp

osit

orie

s(s

tati

stic

ally

sign

ifica

nt)

No

adve

rse

effe

cts

obse

rved

(con

tinu

ed)


Tabl

e5.

Con

tinu

ed

Pati

ent

Out

com

es

Ref

eren

ces

Sco

reIn

terv

enti

onS

tudy

Des

ign

NM

ean

Age

Wom

enD

urat

ion

Out

com

eM

easu

reR

esul

tsS

afet

yA

naly

sis

272

Lac

tulo

se20

ml

b.i.d

.or

Dor

bane

x(m

ixtu

reof

polo

xalk

olan

ddi

hydr

o-an

thro

quin

olon

e)(P

D)—

10m

l/da

y,or

enem

aif

noB

Min

5da

ys(c

ontr

olgr

oup)

Ope

n,ra

ndom

ized

,co

ntro

lled

35of

37lo

ng-s

tay

elde

rly

pati

ents

8130

F7

M27

days

Impr

ovem

enti

ntr

ansi

ttim

esby

Hin

ton

met

hod

The

rew

ere

stat

isti

cally

sign

ifica

ntim

prov

emen

tsin

the

tran

sitt

imes

inth

ela

ctul

ose

and

PD

grou

psbo

thw

ithi

nth

egr

oup

com

pare

dw

ith

base

line

,and

whe

nco

mpa

red

wit

hth

eco

ntro

lgro

up

No

sign

ifica

ntad

vers

eef

fect

sno

ted

342

Sod

ium

pico

sulf

ate

(SP

—10

mg/

day)

orD

orba

nex

(mix

ture

ofpo

loxa

lkol

and

dihy

dro-

anth

roqu

inol

one

(PD

—10

ml/

day)

Ope

n,ra

ndom

ized

,cr

osso

ver

38of

40lo

ng-s

tay

pati

ents

75.6

31F

7M

2-w

kru

nin

foll

owed

bytw

o2-

wk

trea

tmen

tpe

riod

sw

ith

a3-

day

was

hout

inbe

twee

n

Sto

olfr

eque

ncy,

size

,sid

eef

fect

sT

heav

erag

enu

mbe

rof

BM

per

day

incr

ease

dfr

om0.

56in

the

pret

rial

peri

od,

to0.

96fo

rS

Pan

d0.

86fo

rP

DB

oth

SP

and

PD

resu

lted

inno

rmal

izat

ion

inst

ools

ize

inth

em

ajor

ity

ofpa

tien

ts

Hig

hin

cide

nce

ofin

cont

inen

cein

the

grou

pth

atre

ceiv

edS

Pas

thei

rfi

rst

trea

tmen

t.O

ther

effe

cts

wer

em

inim

al


Table 6. Methodologic Scores of the Studies Evaluating the Stimu-lant and Irritative Laxatives


22 1 0 1 224 2 2 1 525 1 2 0 330 1 0 1 231 1 0 1 232 1 2 1 429 1 0 1 227 1 0 1 233 1 2 0 334 1 0 1 2

A score of 1 or 2 were given for randomization (2 for appropriate randomizationtechnique and concealed allocation explicitly stated or described, 1 for studysimply described as “randomized”). Scores of 0–2 were given for blinding (2when both subjects and investigators were blinded to the treatment by use ofidentical placebo or other technique, 1 when the study is described as “double-blind,” and 0 when the study was not double-blind). Score o f 0 or 1 was givenfor frequency of withdrawals (1 when the number of withdrawals and reason forwithdrawals were stated, and 0 when no statement was made pertaining to withdrawals).

psyllium alone. However, the combination appeared to causemore side effects. A study evaluating the efficacy of psylliumcompared with docusate revealed that psyllium was superiorin its effect on stool frequency, stool water content, total stooloutput, and the combination of several objective measuresof constipation (45). The combination of psyllium, celandin,and aloe vera was superior to placebo in the treatment of con-stipation (47). Three studies that compared a combination ofpsyllium and senna with lactulose are discussed in the sec-tion on lactulose and stimulant/irritant laxatives (22, 24, 25).The comparison of two preparations of psyllium and senna,one with a higher dose of senna, revealed that the preparationwith the higher senna dose increased stool frequency morethan the other (49). In elderly, bed-ridden, nursing home pa-tients, psyllium and calcium polycarbophil have been notedto be similar in their effect in improvement in stool frequency,stool consistency, and ease of defecation (50). Psyllium andmethylcellulose were similarly effective in constipated sub-jects in another study (36).

Level II Evidence, Grade B Recommendation.

The Efficacy and Safety of Calcium Polycarbophil in theTreatment of ConstipationApart from the study mentioned in the discussion of psyllium,no other trials were found in the English literature evaluatingthis drug. Additional clinical trials are required to furtherevaluate the utility of this agent.

Level III Evidence, Grade C Recommendation.

Efficacy and Safety of Cisapride in the Treatmentof ConstipationThree randomized studies are presented (51–53). These aresummarized in Tables 9 and 10. Two of these studies includedplacebo (51, 53); in the third (52) study, comparative results

of efficacy were inferred from periods when study patientswere not receiving any therapy. The results demonstratingthe efficacy of cisapride in the treatment of idiopathic con-stipation were seen in a pilot study as well (54). Open trialshave evaluated the efficacy of cisapride in the constipationof Parkinson’s disease (55, 56); over the long-term, the effi-cacy may wane (56). The drug may also be useful for treatingconstipation patients with spinal cord injury (57) and sys-temic sclerosis (58), although this remains to be validated ina controlled manner.

Though it appears that cisapride may be a useful agentin the treatment of idiopathic constipation, it is no longermarketed in the United States.

Cisapride: No Recommendation.

Efficacy and Safety of Colchicine in the Treatmentof ConstipationNo randomized studies in otherwise healthy patients with id-iopathic constipation have been reported, apart from a singlestudy that is in abstract only. One randomized trial performedon developmentally disabled patients is presented (59). Thisis summarized in Tables 11 and 12. In an open-labeled trialin seven patients with chronic constipation, the mean num-ber of spontaneous bowel movements significantly increased(p < 0.05) from 1.7 ± 0.5 noted during routine therapy ofconstipation with laxatives and enemas to 6.4 ± 0.7 per week;mean colonic transit time significantly (p < 0.05) decreasedfrom 58.1 ± 2.5 to 47.1 ± 5.0 h; and symptoms of abdominalpain, nausea, and bloating significantly (p < 0.05) improvedduring therapy with colchicine (60). This pilot study wasfollowed by a double-blind, placebo-controlled, randomized,crossover study in 16 chronically constipated subjects. Asalluded to above, this has been published in abstract formonly to date (61). Colchicine 0.6 mg t.i.d. was the dose andschedule utilized in a 4-wk treatment interval. Colchicine re-sulted in reduced transit time and increased number of bowelmovements per week (9.9 vs 3.8) compared with placebo.A significant placebo effect was noted. No significant sideeffects were observed. Observations in patients with Parkin-son’s disease (62) and persistent constipation after total ab-dominal colectomy with ileorectostomy for colonic inertia(63) suggest that colchicine may be useful in these settings aswell.

The utility of colchicine in the treatment of chronic idio-pathic constipation remains to be confirmed.

Colchicine: Level III Evidence, Grade CRecommendation.

Effectiveness and Safety of Misoprostol in the Treatmentof ConstipationOnly one suitable randomized trial was found (64). The num-ber of patients was small (n = 8). The qualitative assessmentof this study and the extracted data are presented in Tables 13and 14.


Tabl

e7.

Sum

mar

yof

the

Tri

als

Eva

luat

ing

the

Effi

cacy

and

Saf

ety

ofth

eB

ulk

Lax

ativ

es

Pati

ent

Out

com

esM

ean

Age

Ref

eren

ces

Sco

reIn

terv

enti

onS

tudy

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

371

10g

fibe

rfr

omco

rn-b

ased

bisc

uits

(gp

A),

whe

atbr

an(g

pB

),or

noin

terv

enti

on(g

pC

)

Ran

dom

ized

open

inpa

tien

tsin

the

thir

dtr

imes

ter

ofpr

egna

ncy

4028

2-w

kba

seli

nefo

llow

edby

2-w

ktr

eatm

ent

SF,

SC

,pre

senc

eof

pain

,str

aini

ng,

and

bloo

d

Mea

nda

ilydi

etar

yfi

bre

inta

kein

allg

roup

s,in

the

firs

t2w

k20

.4g

was

sim

ilar

toth

atin

the

gene

ral

popu

lati

on.

The

rew

asa

tend

ency

for

the

fibe

rsu

pple

men

ted

grou

psto

expe

rien

cele

sspa

inan

dst

rain

ing,

but

this

did

nota

chie

vest

atis

tica

lsi

gnifi

canc

e

Inth

efi

nal2

-wk

chan

ges

infi

ber

inta

kes

wer

eG

pA

,m

ean

incr

ease

7.2

g/da

y(p

<0.

001)

;Gp

B,m

ean

incr

ease

9.1

g/da

y(p

<0.

001)

;Gp

Cm

ean

decr

ease

3.50

g/da

y(p

<0.

005)

.S

tool

freq

uenc

yim

prov

edin

Gps

Aan

dB

wit

hno

chan

ges

inG

pC

.S

tool

freq

uenc

yan

dco

nsis

tenc

yim

prov

edin

Gps

Aan

dB

,wit

hno

chan

ges

inG

pC

.38

4B

ran

6.6

gt.i

.d.

orpl

aceb

oR

ando

miz

ed,

doub

le-b

lind

,pl

aceb

o-co

ntro

lled

,cr

osso

ver

24of

2937

(20–

65)

26/3

3w

eek

base

line

,th

entw

oco

nsec

utiv

e4-

wk

trea

tmen

tpe

riod

s

Seg

men

talt

rans

itti

mes

,SF,

stoo

lw

eigh

t,an

dot

her

sym

ptom

s

SF

for

plac

ebo

was

1.2

±1.

3pe

rw

kvs

3.5

±1.

8fo

rbr

an(p

<0.

01)

No

sign

ifica

ntad

vers

eef

fect

s

The

rew

asno

diff

eren

cein

stoo

lwei

ght

Dur

ing

bran

trea

tmen

tor

oana

ltra

nsit

tim

eno

rmal

ized

only

inpa

tien

tsw

ith

slow

colo

nic

tran

sita

ndno

tin

thos

ew

ith

slow

rect

altr

ansi

t


The

occu

rren

cean

dse

veri

tyof

the

mos

tfr

eque

ntac

com

pany

ing

sym

ptom

sof

chro

nic

cons

tipa

tion

did

not

diff

erw

ith

plac

ebo

and

bran

trea

tmen

ts39

110

gb.

i.d.o

fco

rnbr

anor

whe

atbr

an

Ope

n,ra

ndom

ized

,co

ntro

lled

1026

.310

/01-

wk

adju

stm

ent,

2-w

kco

ntro

l,an

d2-

wk

trea

tmen

tpe

riod

Feca

lwei

ght,

feca

lm

oist

ure

cont

ent,

SF,

inte

stin

altr

ansi

ttim

e,an

dsy

mpt

oms

The

adm

inis

trat

ion

ofbo

thbr

anpr

oduc

tsw

asas

soci

ated

wit

han

incr

ease

infe

cal

wei

ght(

157%

,p

<0.

05),

and

whe

reas

Non

ere

port

ed

Bra

nin

crea

sed

SF

(2.3

±0.

3to

3.5

±0.

5B

M/w

k;55

%,

p<

0.05

)B

ran

decr

ease

din

test

inal

tran

sitt

ime

(50%

)Pe

rcen

tage

feca

lm

oist

ure

incr

ease

don

lyw

ith

whe

atbr

an(6

7.4–

72.1

%)

Cor

nbr

anw

assi

gnifi

cant

lybe

tter

than

whe

atbr

anin

reli

evin

gsy

mpt

oms

ofco

nsti

pati

on(p

<0.

05)

492

Agi

olax

and

Lun

elax

Ope

n,ra

ndom

ized

,co

ntro

lled

,cr

osso

ver

19of

2083

16/4

1-w

kru

n-in

foll

owed

bytw

oco

nsec

utiv

etr

eatm

ent

peri

ods

SF,

SE

,tas

teof

the

prep

arat

ion,

use

ofen

emas

The

SF

tend

edto

behi

gher

wit

hL

unel

ax(w

hich

cont

ains

ahi

gher

dose

ofse

nna)

than

wit

hA

giol

ax

No

side

effe

cts

seen

The

rew

ere

nodi

ffer

ence

inth

e“e

ase

ofsw

allo

win

g”T

here

wer

eno

diff

eren

cein

“eas

eof

adm

inis

trat

ion”

and

The

rew

ere

nodi

ffer

ence

sin

“tas

te”

(con

tinu

ed)


Tabl

e7.

Con

tinu

ed

Pati

ent

Out

com

esM

ean

Age

Ref

eren

ces

Sco

reIn

terv

enti

onS

tudy

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

401

10g

unre

fine

dbr

an,1

0g

ofbr

anas

bran

bisc

uits

or10

mlo

fse

noko

t

Ope

n,ra

ndom

ized

23ps

ycho

-ge

riat

ric

pati

ents

50.3

15/8

Eac

hre

ceiv

edth

eth

ree

agen

tsin

thre

eco

nsec

utiv

epe

riod

sof

3w

k,th

ese

quen

cera

ndom

lyde

term

ined

SF,

SC

,siz

eof

bow

elm

ovem

ents

,use

ofen

emas

The

rew

asno

sign

ifica

ntdi

ffer

ence

SF.

The

rew

asno

sign

ifica

ntdi

ffer

ence

inus

eof

enem

as.

The

rew

asno

sign

ifica

ntdi

ffer

ence

inth

eco

nsis

tenc

yof

the

mot

ions

.T

here

was

litt

ledi

ffer

ence

inth

epr

opor

tion

of“l

arge

”st

ools

inth

eth

ree

trea

tmen

tper

iods

,but

asi

gnifi

cant

lyhi

gher

prop

orti

onof

smal

ler

stoo

lsw

ere

repo

rted

wit

hbr

an.

Non

em

enti

oned

483

Ispa

ghul

a(p

syll

ium

)3.

6g

t.i.d

.or

plac

ebo

Mul

tice

nter

,ra

ndom

ized

,pl

aceb

o-co

ntro

lled

,si

ngle

-bli

nd,

para

llel

201

4915

1/50

14da

ysS

F,S

C,a

bdom

inal

disc

omfo

rt,#

ofsa

chet

sta

ken

SF

—sh

owed

stat

isti

cally

sign

ifica

ntin

crea

ses

from

2.3

per

wee

kba

seli

ne,t

o7

per

wee

kw

ith

ispa

ghul

aan

d4.

5pe

rw

eek

for

plac

ebo.

No

sign

ifica

ntad

vers

eef

fect

s

SC

—th

enu

mbe

rof

hard

-or

pell

et-l

ike

stoo

lsw

ere

also

sign

ifica

ntly

low

er,

and

loos

eor

wat

ery

stoo

lsin

crea

sed

inth

eis

pagh

ula

grou

pS

tati

stic

ally

sign

ifica

ntde

crea

ses

inab

dom

inal

disc

omfo

rt,

and

stra

inin

gin

the

isph

agul

agr

oup

Bot

hin

vest

igat

ors

and

pati

ents

note

dst

atis

tica

llysi

gnifi

cant

impr

ovem

enti

nth

eco

nsti

pati

on


422

Whe

atbr

an1.

5–4.

5g

qdor

regu

lar

diet

Ope

n,ra

ndom

ized

,pa

rall

el

12of

14lo

ng-s

tay

geri

atri

cpa

tien

ts

8014

/06

wk,

then

plac

ebo

grou

pw

ere

trea

ted

wit

hbr

anas

wel

l

SF,

stoo

lsiz

e,S

C,

EO

D,U

OL

SF

—th

enu

mbe

rof

spon

tane

ousl

ypa

ssed

BM

sin

the

bran

grou

pw

asst

atis

tica

llysi

gnifi

cant

com

pare

dto

the

plac

ebo

grou

p

One

pati

ent

repo

rted

diffi

cult

ysw

allo

win

gbr

an

All

pati

ents

inth

ebr

angr

oup,

and

the

pati

ents

inth

epl

aceb

ogr

oup

foll

owin

gco

mm

ence

men

tof

bran

supp

lem

enta

tion

show

edim

prov

emen

tin

the

quan

tity

and

cons

iste

ncy

ofst

ools

pass

ed41

1B

ran

(0.5

–1.5

g)or

regu

lar

diet

Ope

n,ra

ndom

ized

,co

ntro

lled

37of

50>

60yr

—ex

tend

edca

reho

spit

al

—13

wk

Req

uire

men

tfor

laxa

tive

use

The

rew

asan

over

all

decr

ease

inla

xativ

eus

ein

the

bran

grou

p.H

owev

er,t

here

was

also

anin

crea

sein

the

use

ofen

emas

and

supp

osit

orie

s

No

sign

ifica

ntad

vers

eef

fect

s

353

Hig

hfi

ber

(22.

9g

daily

)di

etor

regu

lar

diet

Ope

n,ra

ndom

ized

,co

ntro

lled

35of

49pa

tien

tsun

derg

oing

radi

calh

ys-

tere

ctom

y17

cont

rol

(C)

18tr

eatm

ent

(T)

3549

/07

mon

ths

SF,

SC

,EO

Dga

s,cr

ampi

ngS

F—

Tgr

oup

had

asi

gnifi

cant

incr

ease

inth

efr

eque

ncy

ofB

M(p

=0.

0096

);th

isw

asno

tsee

nin

the

Cgr

oup.

The

Tgr

oup

took

less

tim

eto

defe

cate

(p<

0.00

1)bu

thad

mor

eB

Ms

acco

mpa

nied

byga

s(p

<0.

001)

.T

heC

grou

pha

dsi

gnifi

cant

lym

ore

BM

wit

hcr

amps

(p<

0.00

1),s

trai

ning

(p<

0.00

1),a

ndre

tent

ion

(p<

0.00

1)

No

sign

ifica

ntad

vers

eef

fect

s

SC

—th

eC

grou

pha

dsi

gnifi

cant

lym

ore

BM

s,w

hich

wer

eha

rd(p

<0.

001)

(con

tinu

ed)


ble

7.C

onti

nued

Pati

ent

Out

com

esM

ean

Age

Ref

eren

ces

Sco

reIn

terv

enti

onS

tudy

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

462

Ispa

ghul

a(I

S)

(psy

lliu

m)

3.5

g,or

anot

her

laxa

tive

(lac

tulo

se,

bisa

cody

l,do

cusa

te,

senn

a,an

dm

agne

sium

sulf

ate)

Ope

n,m

ulti

cent

er,

rand

omiz

ed,

cont

roll

ed

381

of39

422

4(5

6.9%

)is

pagh

ula

170

(43.

1%)

othe

rla

xativ

e

30–5

925

0/13

94

wk

SF

/SC

ofso

ilin

g,ac

cept

abil

ity,

pala

tabi

lity

,tr

eatm

ent-

rela

ted

upse

ts

ISw

asas

sess

edby

the

GP

sto

besu

peri

orto

the

othe

rtr

eatm

ents

inim

prov

ing

bow

elfu

ncti

onan

din

over

all

effe

ctiv

enes

s.IS

was

mor

epa

lata

ble

and

acce

ptab

leto

pati

ents

.IS

prod

uced

ahi

gher

perc

enta

geof

norm

al,

wel

l-fo

rmed

stoo

lsan

dfe

wer

hard

stoo

lsth

anot

her

laxa

tives

.In

cide

nces

ofso

ilin

g,di

arrh

ea,a

ndab

dom

inal

pain

wer

elo

wer

inth

eIS

grou

p.

No

sign

ifica

ntad

vers

eef

fect

s

433

Psy

lliu

m24

g/da

yor

plac

ebo

Sin

gle-

blin

d,ra

ndom

ized

,pl

aceb

o-co

ntro

lled

fibe

rin

terv

enti

onw

ith

cros

sove

r

10>

605/

54

wk

inea

char

mS

F,S

C,a

ndw

eigh

tsda

ilyFi

ber

decr

ease

dto

talg

uttr

ansi

ttim

efr

om53

.9h

(pla

cebo

cond

itio

n)to

30.0

h(p

<.0

5).

Sto

olw

eigh

tsan

dco

nsis

tenc

yw

ere

not

sign

ifica

ntly

impr

oved

byfi

ber.

No

sign

ifica

ntad

vers

eef

fect

s

The

rew

asa

tren

dto

war

dan

incr

ease

inS

F(1

.3vs

0.8B

M/d

ay)

inth

efi

ber

grou

p44

4P

syll

ium

(10

g/da

y)an

dpl

aceb

o

Dou

ble-

blin

d,pl

aceb

o-co

ntro

lled

2251

14/8

8w

kw

ith

4-w

kru

n-in

onpl

aceb

oan

d4

wk

was

hout

SF,

SC

,EO

Dan

dw

eekl

yst

ool

wei

ght.

Col

ontr

ansi

tand

AR

M

SF

incr

ease

dsi

gnifi

cant

lyaf

ter

8w

kof

psyl

lium

trea

tmen

t(3.

8vs

2.9

BM

/wk,

p<

0.05

)

No

sign

ifica

ntad

vers

eef

fect

s

Sub

ject

sre

port

edim

prov

emen

tin

SC

(sto

olco

nsis

tenc

ysc

ore:

3.2

±0.

2vs

3.8

±0.

2,p

<0.

05)

onps

ylli

umE

OD

impr

oved

onps

ylli

um(p

ain

scor

e:2.

0±

0.4

vs2.

6±

0.5,

p<

0.05

)C

olon

tran

sita

ndA

RM

unch

ange

d


302

Psy

lliu

m(P

)7.

2g/

day

and

psyl

lium

wit

hse

nna

(PS

)(6

.5+

1.5

g/da

y)

Ope

n,ra

ndom

ized

,si

ngle

-bli

ndco

ntro

lled

40of

4226

.135

F5

M1-

wk

plac

ebo

and

1-w

ktr

eatm

ent

SF,

SC

,sym

ptom

sw

ith

BM

sB

oth

laxa

tives

incr

ease

dS

F[P

3.6

vsP

S6.

8B

M/w

k(p

<0.

001)

]

Pgr

oup

3/22

cram

ping

and

gas

PS

grou

p7/

22ha

dcr

amps

,un-

com

fort

able

diar

rhea

,bl

oati

ngga

s,an

dna

usea

Bot

hla

xativ

esim

prov

edS

C(s

imil

ar)

Bot

hP

and

PS

incr

ease

dw

etan

ddr

yst

ool

wei

ghts

wit

hth

ead

ded

effe

ctof

the

senn

acl

earl

yev

iden

t.O

nly

the

psyl

lium

wit

hse

nna

incr

ease

dst

oolm

oist

ure

Bot

hla

xativ

espr

ovid

eda

high

degr

eeof

subj

ectiv

ere

lief

455

Psy

lliu

m5.

1g

b.i.d

.and

docu

sate

100

mg

b.i.d

.

Mul

tice

nter

,ra

ndom

ized

,do

uble

-bli

nd,

para

llel

desi

gn

170

of18

737

.215

6/14

1-w

kw

asho

utfo

llow

edby

1-w

kba

seli

ne(p

lace

bo)

foll

owed

by2-

wk

trea

tmen

t

Sto

olw

ater

cont

ent

(SW

C),

stoo

lw

ater

wei

ght

(SW

W),

tota

lst

oolo

utpu

t,S

F

Com

pare

dto

base

line

,ps

ylli

umin

crea

sed

SW

Cvs

docu

sate

(psy

lliu

m2.

33%

vsdo

cusa

te0.

01%

,p

=0.

007

and)

Non

est

ated

Psy

lliu

mal

soin

crea

sed

SW

W(8

4.0

g/B

M)

vsdo

cusa

te(7

1.4

g/B

M);

p=

0.04

Tota

lsto

olou

tput

high

erw

ith

psyl

lium

(359

.9g/

wk)

vsdo

cusa

te(2

71.9

g/w

k);p

=0.

005

Psy

lliu

mha

da

high

erO

’Bri

enra

nk-t

ype

scor

eco

mbi

ning

obje

ctiv

em

easu

res

ofco

nsti

pati

on(p

syll

ium

475.

1;do

cusa

te40

3.9;

p=

0.00

2)S

Fw

assi

gnifi

cant

lygr

eate

rfo

rps

ylli

um(3

.5B

M/w

k)vs

docu

sate

(2.9

BM

/wk)

intr

eatm

entw

eek

2(p

=0.

02)

(con

tinu

ed)


ble

7.C

onti

nued

Pati

ent

Out

com

esM

ean

Age

Ref

eren

ces

Sco

reIn

terv

enti

onS

tudy

Des

ign

N(y

ear)

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

473

Psy

lliu

m,

cela

ndin

,and

aloe

vera

com

bina

tion

(CA

P)

1–3

caps

ules

per

day

orpl

aceb

o

Dou

ble-

blin

dra

ndom

ized

,pl

aceb

o-co

ntro

lled

32of

3551

.522

/10

28da

ysto

-2w

kpr

etri

alba

seli

ne

SF,

SC

,and

othe

rsy

mpt

oms

Inth

e(C

AP

)gr

oup,

bow

elm

ovem

ents

beca

me

mor

efr

eque

nt(4

.6vs

7.9

BM

/wk)

.(p

<0.

002)

No

side

effe

cts

dete

cted

∗

SC

—st

ools

soft

erin

CA

Pgr

oup

com

pare

dw

ith

plac

ebo.

(p<

0.00

2)L

ess

laxa

tive

depe

nden

cein

CA

Pgr

oup

(p<

0.01

)A

bdom

inal

pain

was

not

redu

ced

inei

ther

grou

p50

1C

alci

umpo

lyca

rbop

hil

(C)

2ta

bs/d

ayor

psyl

lium

(P)

two

teas

poon

s/da

y

Ope

n,ra

ndom

ized

cont

roll

ed,

cros

sove

r

32be

d-ri

dden

nurs

ing

hom

ere

side

nts

77±

10.4

26/6

Two

cons

ecut

ive

3-w

ktr

eatm

ent

peri

ods

SF,

SC

,EO

DT

here

wer

eno

stat

isti

cally

sign

ifica

ntch

ange

sin

SF

(for

C7.

2±

2.4

vsfo

rP

7.22

±2.

2)

Non

em

enti

oned

No

diff

eren

cein

EO

DN

odi

ffer

ence

inS

CM

ore

pati

ents

seem

edto

favo

rC

362

1,2,

or4

gof

met

hylc

ellu

-lo

se(M

)or

3.4

gmof

psyl

lium

(P)

Two

phas

e50

heal

thy

subj

ects

,59

chro

nica

llyco

nsti

pate

d

27(1

8–70

)28

(18–

70)

56/3

1-w

kru

n-in

(tak

ing

plac

ebo)

,the

n10

-day

trea

tmen

tpe

riod

wit

hon

eof

the

regi

men

sde

scri

bed

SF,

stoo

lwei

ght,

stoo

lwat

eran

dso

lids

.Adv

erse

even

ts

Hea

lthy

subj

ects

:m

ethy

lcel

lulo

sein

daily

dose

sof

4g

dem

onst

rate

da

stat

isti

cally

sign

ifica

ntin

crea

sein

feca

lfr

eque

ncy,

feca

lwat

er,

and

feca

lsol

ids

No

diff

eren

cein

the

inci

denc

eof

abdo

min

alcr

amps

,fl

atul

ence

,or

abdo

min

alpa

inbe

twee

nth

etr

eatm

ent

and

plac

ebo

peri

ods

Chr

onic

ally

cons

tipa

ted

indi

vidu

als—

alld

oses

ofM

and

Pst

atis

tica

llysi

gnifi

cant

incr

ease

sin

SF

Chr

onic

ally

cons

tipa

ted

indi

vidu

als—

alld

oses

ofM

and

Pst

atis

tica

llysi

gnifi

cant

incr

ease

dw

ater

cont

ent,

and

feca

lso

lids

The

rew

asno

incr

ease

inin

divi

dual

stoo

lwei

ght

from

any

ofth

ela

xativ

edo

ses

∗Bec

ause

this

stud

yis

doub

le-b

lind

and

plac

ebo-

cont

roll

ed,t

heto

tals

core

here

ishi

ghly

base

don

the

crit

eria

seto

ut;t

hest

udy

desc

ript

ion

reve

als

very

litt

lein

form

atio

n,an

das

such

the

over

allq

uali

tyis

nota

sgo

odas

the

scor

esu

gges

ts.


Table 8. Methodologic Scores of the Studies Evaluating the BulkLaxatives


38 1 2 1 437 1 0 0 139 1 0 0 149 1 0 1 240 1 0 0 148 1 1 1 342 1 0 1 241 1 0 0 135 2 0 1 346 1 0 1 230 1 0 1 244 1 2 1 445 2 2 1 543 1 1 1 347 1 2 1 450 1 0 0 136 1 1 0 2

A score of 1 or 2 was given for randomization (2 for appropriate randomizationtechnique and concealed allocation explicitly stated or described, 1 for studysimply described as “randomized”). Scores of 0–2 were given for blinding (2when both subjects and investigators were blinded to the treatment by use ofidentical placebo or other technique, 1 when the study is described as “double-blind,” and 0 when the study was not double-blind). Score o f 0 or 1 was givenfor frequency of withdrawals (1 when the number of withdrawals and reasonfor withdrawals were stated, and 0 when no statement was made pertaining towithdrawals).

Misoprostol: Level III Evidence, Grade C Recommen-dation.

Efficacy and Safety of Stool Softeners in the Treatment ofConstipationDocusate sodium and docusate calcium are the major drugs inthis category. Poloxalkol is another stool softener; no studieswere found in which this agent was used by itself in a treat-ment arm. The studies reviewed are summarized in Tables15 and 16. Three studies where it is combined with a stim-ulant/irritant laxative are summarized in the section dealingwith this latter class of drugs. Four studies are presented thatevaluate the utility of docusate in the treatment of consti-pation (45, 65–67). One of these compares it with psyllium(45). Psyllium appears to be superior to docusate sodium inthe doses utilized. The other studies either compare docusatesodium and docusate calcium or these drugs versus placebo.One study suggests that docusate calcium may be more effec-tive than docusate sodium (65). The efficacy of docusate inthe treatment of constipation is modest at best, with one study,albeit small, suggesting no significant benefit compared withplacebo (66). There are studies that have evaluated docusateprior to 1966, and these have been included in previous re-views (68).

Docusate: Level III Evidence, Grade CRecommendation.

Efficacy and Safety of Other Agents on the Treatmentof ConstipationA recent randomized, double-blind placebo-controlled trialevaluated the efficacy of tegaserod in the management ofchronic constipation (69). This drug is a selective agonist ofthe serotonin subtype 4 (5-HT4). It has been shown to enhancegastrointestinal motility in animals and healthy volunteers,and it has also been shown to be effective for symptom re-lief in patients with constipation predominant irritable bowelsyndrome. This was a large well-designed trial with a qual-ity score of 5. The doses used were tegaserod 2 mg b.i.d.,tegaserod 6 mg b.i.d., or placebo. A total of 1,116 patientscompleted the treatment phase of the trial. The mean age was47 yr and 90% of the subjects were women. Treatment du-ration was 12 wk. Responders were patients treated for ≥7days with an increase of ≥1 complete spontaneous BM perweek versus baseline during weeks 1–4 (primary variable)and weeks 1–12 (secondary variable). Other secondary vari-ables included SF, stool form, abdominal bloating/distention,straining, and abdominal pain/discomfort, and global assess-ment of constipation and bowel habits. Responder rates forcomplete spontaneous bowel movement during weeks 1–4were significantly greater (p < 0.0001) in the tegaserod 2mg twice daily (41.4%) and 6 mg twice daily groups (43.2%)versus placebo (25.1%). This effect was maintained over 12wk. Statistically significant improvements over placebo wereobserved across the majority of secondary variables for bothtegaserod doses. No rebound effect was observed after treat-ment withdrawal. As such, there appears to be clear, statisti-cally significant benefit of the two doses used versus placebo,with no clear benefit of the higher dose compared with thelower dose. Overall, tegaserod was well tolerated; headacheand nasopharyngitis, the most frequent adverse events, weremore common in the placebo group than in either tegaserodgroup.

Tegaserod: Level I Evidence, Grade A Recommen-dation.

A pilot study was undertaken to evaluate the efficacy oforal erythromycin in the treatment of idiopathic constipation(70). Eleven male patients were treated for 1 month with ery-thromycin (1 g/day for 2 wk then 500 mg/day for 2 wk) in anopen, nonrandomized trial. Colon transit time improved asdid stool frequency (2.3–6.7 per week). Two of the patientscomplained of borborygmi, otherwise there were no signif-icant side effects. These results suggest that erythromycinwarrants trial in a controlled manner to see if this dramaticimprovement could be reproduced.

Loxiglumide, a cholecystokinin antagonist, was evaluatedfor its efficacy in a prospective, randomized, double-blindcontrolled trial (71) in 21 nursing home patients with a meanage of 83 yr. There were 13 male and 8 female patients.They were randomized to receive loxiglumide 800 mg t.i.d.or identical placebo.


ble

9.S

umm

ary

ofth

eS

tudi

esE

valu

atin

gth

eE

ffica

cyan

dS

afet

yof

Cis

apri

dein

the

Tre

atm

ento

fC

onst

ipat

ion

Pati

ent

Out

com

es

Ref

eren

ces

Sco

reIn

terv

enti

onS

tudy

Des

ign

NM

ean

Age

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

515

Cis

apri

de20

mg

b.i.d

.or

plac

ebo

Ran

dom

ized

,do

uble

-bli

nd,

plac

ebo-

cont

roll

ed

126

(64

cisa

prid

e,62 pl

aceb

o)

50.5

75/5

14-

wk

base

line

phas

e,8-

wk

trea

tmen

tph

ase

and

4-w

kru

nou

t(p

lace

bo)

phas

e

SF,

SC

,EO

D,U

OL

,an

dov

eral

lsta

tew

ith

resp

ectt

oco

nsti

pati

on(v

isua

lana

log

scal

e)

Cis

apri

dean

dpl

aceb

oin

crea

sed

spon

tane

ous

SF

from

1.1

to3.

0B

M/w

k(p

≤0.

001)

and

from

1.2

to1.

5B

M/w

k(p

>0.

05),

resp

ectiv

ely

No

data

Lax

ativ

eco

nsum

ptio

nw

asde

crea

sed

from

3.6

to1.

8do

ses

per

wee

kby

cisa

prid

e(p

≤0.

001)

and

from

3.3

to2.

8by

plac

ebo

(p<

0.05

).B

oth

drug

sim

prov

edco

nsti

pati

onas

asse

ssed

byth

epa

tien

tby

mea

nsof

avi

sual

anal

ogsc

ale,

but

cisa

prid

edi

dso

toa

larg

erex

tent

than

plac

ebo

525

Dif

fere

ntdo

ses

ofci

sapr

ide

orno

trea

tmen

t

Ran

dom

ized

,do

uble

-bli

nd.

Two

grou

ps:

Aan

dB

119

Gp

A—

12w

kof

cisa

prid

e20

mg/

day,

then

12w

kof

free

dom

tota

keup

to20

mg/

day

in5

mg

incr

emen

ts.

Gp

B—

20m

g/da

yfo

r6

wk,

then

10m

g/da

yfo

r6

wk,

then

notr

eatm

entf

or12

wk

SF

and

UO

LS

Fw

asin

crea

sed

inbo

thgr

oups

duri

ngac

tive

trea

tmen

tand

was

not

redu

ced

whe

nth

edo

sew

asde

crea

sed

from

20to

10m

gtw

ice

daily

ingr

oup

Bbu

twas

mai

ntai

ned

ingr

oup

AL

axat

ive

inta

kefe

llby

50%

inbo

thgr

oups

,bu

tthi

sef

fect

was

mai

ntai

ned

duri

ngfo

llow

-up

ingr

oup

Aon

lyG

roup

Apa

tien

tsto

okne

arly

the

max

imum

dosa

geof

cisa

prid

eta

blet

sal

low

eddu

ring

foll

ow-u

p(3

.3ta

blet

spe

rda

y±

0.2

SE

M)

No

data


534

One

oftw

oci

sapr

ide

dose

s(5

or10

mg)

orpl

aceb

o

Ran

dom

ized

,do

uble

-bli

nd,

para

llel

grou

pde

sign

46of

48M

edia

nof

50yr

34F

12M

Thr

ee phas

es—

3-w

kba

seli

ne,

foll

owed

by12

-wk

doub

le-b

lind

peri

od,a

nd4-

wk

foll

ow-u

p

SF,

SC

,EO

D,U

OL

,S

EB

oth

cisa

prid

e5

and

10m

gin

crea

sed

stoo

lfr

eque

ncy

byab

out

70%

from

base

line

afte

r8–

12w

k(p

<0.

002)

,but

plac

ebo

also

incr

ease

dS

Fby

43%

(p<

0.07

).T

hedi

ffer

ence

betw

een

the

trea

tmen

tgro

ups

did

notm

eets

tati

stic

alsi

gnifi

canc

e

Pati

ents

inth

eci

sapr

ide

repo

rted

abdo

min

alcr

amps

(4),

dysp

epsi

a,ga

stri

cpr

oble

ms,

epig

astr

icpa

in,

flat

ulen

cedi

arrh

ea,a

nddi

zzin

ess

(1ea

ch)

Cis

apri

dede

crea

sed

SC

and

resu

lted

ingr

eate

rE

OD

com

pare

dw

ith

plac

ebo,

both

ina

stat

isti

cally

sign

ifica

ntm

anne

r(p

<0.

002

and

p<

0.00

2,re

spec

tivel

y)C

isap

ride

ther

apy

decr

ease

dla

xativ

eco

nsum

ptio

n,w

hile

the

plac

ebo

grou

pdi

dno

tR

elap

seaf

ter

wit

hdra

wal

ofth

erap

yoc

curr

edsl

owly

.No

sign

ifica

ntdi

ffer

ence

was

foun

dbe

twee

nth

etw

odi

ffer

entd

oses

ofci

sapr

ide


Table 10. Methodologic Scores of the Trials Evaluating Cisapride


51 2 2 1 552 2 2 1 553 1 2 1 4

A score of 1 or 2 was given for randomization (2 for appropriate randomizationtechnique and concealed allocation explicitly stated or described, 1 for studysimply described as “randomized”). Scores of 0–2 were given for blinding (2when both subjects and investigators were blinded to the treatment by use ofidentical placebo or other technique, 1 when the study is described as “double-blind,” and 0 when the study was not double-blind). Score o f 0 or 1 was givenfor frequency of withdrawals (1 when the number of withdrawals and reasonfor withdrawals were stated, and 0 when no statement was made pertaining towithdrawals).

It was noted that stool frequency improved from 3.9 perweek in the placebo group to 4.8 per week in the treatmentgroup (p < 0.006). Colon transit time was also significantlyimproved. No serious side effects or exocrine pancreatic in-sufficiency was encountered. These results suggest a largertrial involving more patients is indicated to further define therole of this agent in the treatment of constipation, and such atrial is ongoing.

No studies were found in the English literature that sup-ported the use of herbal remedies in the treatment of con-stipation, although there may be studies in the Chinese andJapanese literature.

DISCUSSION

The tables summarize individual studies where subjects wererandomized to receive either the active drug or placebo. A sur-prising observation was that apart from PEG and Tegaserod,there was paucity of placebo-controlled studies of high qual-ity. Without placebo-controlled trials, it is impossible to judgean agent’s efficacy. Also, in general, sample sizes were small.Consequently, two drugs, PEG and tegaserod, were accordeda Grade A recommendation and two drugs, lactulose and psyl-lium were given a Grade B recommendation. One of our keyobservations was that the definition of constipation was quitevaried among the various studies. While many of the studiesdefined constipation as the presence of less than 2 or 3 stoolsper week, and a few confirmed this in an observation periodprior to randomization, very few utilized criteria developedin consensus meetings, such as the Rome criteria. This pre-vented effective comparisons of trials that were otherwisesimilar. Similar difficulties were encountered when assessingoutcome measures for different trials. Stool frequency andmeasures of stool consistency were the most common param-eters that were assessed. Other measures that were reportedincluded ease of defecation use of additional laxatives, stoolweight, stool water content, and transit times. These problemsnotwithstanding, based on the results, the following recom-mendations can be made regarding the currently availableagents for the treatment of chronic constipation. Ta

ble

11.

Sum

mar

yof

the

Stu

dyE

valu

atin

gth

eE

ffica

cyan

dS

afet

yof

Col

chic

ine

inth

eT

reat

men

tof

Con

stip

atio

n

Pati

ent

Out

com

es

Stu

dyM

ean

Age

Out

com

eS

afet

yR

efer

ence

sS

core

Inte

rven

tion

Des

ign

N(y

ear)

F/M

Dur

atio

nM

easu

reR

esul

tsA

naly

sis

595

Col

chic

ine

(1.2

or1.

8m

g/da

y)or

plac

ebo

inde

velo

pmen

tally

disa

bled

pati

ents

Dou

ble-

blin

d,cr

osso

ver

11of

12co

mpl

eted

24–6

07/

54-

wk

base

line

,fo

llow

edby

two

8-w

ktr

eatm

ent

peri

ods

wit

ha

3-w

kw

ash-

outi

nbe

twee

n

SF,

and

need

for

addi

tion

alla

xativ

es

8of

11pa

tien

tsex

peri

ence

dan

impr

oved

bow

elpa

tter

nw

hile

onco

lchi

cine

com

pare

dw

ith

plac

ebo,

asde

fine

dby

anin

crea

sein

SF

ora

decr

ease

into

taln

umbe

rof

rect

alla

xativ

esus

ed.

7of

8pa

tien

tsw

hoha

dan

incr

ease

inS

Fre

quir

eda

decr

ease

inre

ctal

laxa

tive

use.

No

sign

ifica

ntcl

inic

alor

lab

com

plic

atio

nsw

ere

reco

gniz

ed


Table 12. Methodologic Score of the Study Evaluating Colchicinein the Treatment of Constipation

Statement on TotalReference Randomization Blinding on Withdrawals Score

59 2 2 1 5


Osmotic Laxatives(i) Lactulose (Kristalose)—three placebo-controlled trials

(19–21) with quality scores of 3, 4, 4. Evidence fair,Grade B recommendation.

(ii) PEG (Miralax)—five placebo-controlled trials (10, 12–15) (9, 11, 12, 13, 14) with quality scores of 5, 3, 5,5, 4; two lactulose-controlled trials (10, 15) with qualityscores of 5, 3. Evidence good, Grade A recommendation.PEG superior to lactulose.

Bulking Agents(i) Psyllium (Metamucil)—three placebo-controlled trials

(43, 44, 48) with quality scores of 3, ˙, 3; two lactulose-controlled trial (26, 46) with quality scores of 2, 2. Evi-dence fair, Grade B recommendation.

Table 13. Summary of the Study Evaluating the Efficacy and Safety of Misoprostol in the Treatment of Constipation

Patient Outcomes

Study Mean Age Outcome SafetyReferences Score Intervention Design N (year) F/M Duration Measure Results Analysis

64 4 Misoprostol(1,200µg/day) orplacebo

Randomized,double-blind,crossover

8 of 9 18 and older.Nototherwisequalified

9/0 Two 1-wktreatmentperiods(placebo ormisoprostol)separated by1-wkwashout

Colon transittime, SF,and stoolweight

Colonic transit timewas significantlyand consistentlydecreased bymisoprostolcompared toplacebo [66 ± 10.2vs 109.4 ± 8.1 h(p = 0.0005)]

No differencesin theincidences ofabdominalpain

Misoprostolsignificantlyincreased the totalstool weight perweek [976.5 ±288.8 vs 434.6 ±190.5 g(p = 0.001)]

Misoprostolsignificantlyincreased thenumber of stoolsper week comparedto placebo[6.5 ± 1.3 vs2.5 ± 0.11(p = 0.01)]

Table 14. Methodologic Score of the Study Evaluating Misoprostolin the Treatment of Constipation


64 1 2 1 4


(ii) Calcium polycarbophil (Perdiem fiber therapy)—onetrial against psyllium (50) with quality score of 1. Evi-dence poor, Grade C recommendation.

(iii) Bran—one placebo-controlled trial (38) with qualityscore of 4; one trial against “no treatment” (37) witha quality score of 1; one trial of wheat bran versus cornbran versus baseline (39) with a quality score of 1. Evi-dence poor, Grade C recommendation.

(iv) Methylcellulose—one nonplacebo-controlled trial witha quality score of 2 (36). Evidence poor, Grade C rec-ommendation.

Wetting AgentsA. Dioctyl sulfosuccinate—one trial of dioctyl calcium (not

available in the United States) and dioctyl sodium (DSS,Colace) versus placebo (65) with a quality score of 3; two


Tabl

e15

.S

umm

ary

ofth

eS

tudi

esE

valu

atin

gth

eE

ffica

cyan

dS

afet

yof

Sto

olS

ofte

ners

inth

eT

reat

men

tof

Con

stip

atio

n

Pati

ent

Out

com

es

Ref

eren

ces

Sco

reIn

terv

enti

onS

tudy

Des

ign

NM

ean

Age

F/M

Dur

atio

nO

utco

me

Mea

sure

Res

ults

Saf

ety

Ana

lysi

s

653

Dio

ctyl

sodi

umsu

l-fo

succ

inat

e(C

olas

e)(D

SS

)10

0m

gqd

or10

0m

gb.

i.d.o

rdi

octy

lca

lciu

msu

l-fo

succ

inat

e(S

urfa

k)D

CS

,240

mg

qd

Ran

dom

ized

,si

ngle

-bli

nd,

cont

roll

ed

46of

4782

.140

/62-

wk

run-

in(p

lace

bo)

foll

owed

by3-

wk

trea

tmen

tph

ase

SF,

SC

,di

sim

pact

ion,

adve

rse

effe

cts

SF

—81

%of

the

pati

ents

rece

ivin

gD

CS

impr

oved

1.75

–2.8

3B

M/w

k,(p

<0.

02).

DS

Sca

used

nosi

gnifi

cant

impr

ovem

ent

over

plac

ebo

inei

ther

the

qdor

b.i.d

.dos

ing

(1.5

–1.9

5B

M/w

kan

d1.

76–2

.29

BM

/wk,

resp

ectiv

ely)

No

sign

ifica

ntad

vers

eef

fect

sor

chan

ges

inla

bora

tory

mea

sure

men

tsw

ere

repo

rted

inan

yof

the

grou

ps

The

mea

nnu

mbe

rof

natu

ralb

owel

mov

emen

ts(w

itho

utot

her

laxa

tives

ordi

sim

pact

ion)

amon

gth

eD

CS

grou

pin

crea

sed

appr

oxim

atel

y62

%ov

erth

epl

aceb

ope

riod

,mor

eth

antw

ice

the

30%

incr

ease

seen

wit

hD

SS

adm

inis

tere

dei

ther

bid

orqd

455

Psy

lliu

m5.

1g

b.i.d

.and

docu

sate

100

mg

b.i.d

.

Mul

tice

nter

,ra

ndom

ized

,do

uble

-bli

nd,

para

llel

desi

gn

170

of18

737

.215

6/14

1-w

kw

asho

utfo

llow

edby

1-w

kba

seli

ne(p

lace

bo)

foll

owed

by2-

wk

trea

tmen

t

Sto

olw

ater

cont

ent,

stoo

lwei

ght,

tota

lsto

olou

tput

,S

F

Com

pare

dto

base

line

,ps

ylli

umin

crea

sed

SW

Cvs

docu

sate

(psy

lliu

m2.

33%

vsdo

cusa

te0.

01%

,p=

0.00

7and

)

Non

est

ated

Psy

lliu

mal

soin

crea

sed

SW

W(8

4.0

g/B

M)

vsdo

cusa

te(7

1.4

g/B

M);

p=

0.04


Tota

lsto

olou

tput

high

erw

ith

psyl

lium

(359

.9g/

wk)

vsdo

cusa

te(2

71.9

g/w

k);p

=0.

005)

Psy

lliu

mha

da

high

erO

’Bri

enra

nk-t

ype

scor

eco

mbi

ning

obje

ctiv

em

easu

res

ofco

nsti

pati

on(p

syll

ium

475.

1;do

cusa

te40

3.9;

p=

0.00

2)67

4D

ioct

ylso

dium

sul-

fosu

ccin

ate

(DS

S)

100

mg

t.i.d

.or

plac

ebo

Ran

dom

ized

,do

uble

-bli

nd,

plac

ebo-

cont

roll

ed,

cros

sove

r

34of

40el

derl

yG

eria

tric

—no

tot

herw

ise

spec

ified

Not

spec

-ifi

edTw

oco

nsec

utiv

e4-

wk

trea

tmen

tpe

riod

s

SF,

over

all

impr

ovem

enti

nco

nsti

pati

on

DS

Sin

crea

sed

SF

bya

mea

nof

1B

M/w

kco

mpa

red

wit

hpl

aceb

o(p

<0.

01)

Non

est

ated

The

diff

eren

cein

the

over

alli

mpr

ovem

enti

nco

nsti

pati

onw

asst

atis

tica

llysi

gnifi

cant

(p<

0.05

)66

4D

ioct

ylca

lciu

msu

l-fo

succ

inat

e(D

CS

)24

0m

gb.

i.d.o

rpl

aceb

o

Ran

dom

ized

,do

uble

-bli

nd,

plac

ebo-

cont

roll

ed,

cros

sove

r

15of

2278

yr4/

112-

wk

run-

in,t

wo

3-w

ktr

eatm

ent

sepa

rate

dby

a2-

wk

was

hout

SF,

SC

,UO

LN

osi

gnifi

cant

diff

eren

ces

inth

est

oolf

requ

ency

,vo

lum

e,an

dne

edfo

rad

diti

onal

laxa

tives

betw

een

the

docu

sate

and

plac

ebo

grou

ps

Non

est

ated

Pati

entp

erce

ptio

nof

cons

tipa

tion

and

disc

omfo

rtw

ith

defe

cati

onw

asal

sono

tsi

gnifi

cant

lydi

ffer

ent

betw

een

the

grou

ps


Table 16. Methodologic Scores of the Studies Evaluating Stool Soft-eners


65 1 1 1 345 2 2 1 567∗ 1 2 1 466 1 2 1 4

∗Because this study is double-blind and placebo controlled, the total score here is highbased of the criteria set out; the study description reveals very little information, and assuch the overall quality is not as good as the score suggests.A score of 1 or 2 was givenfor randomization (2 for appropriate randomization technique and concealed allocationexplicitly stated or described, 1 for study simply described as “randomized”). Scores of0–2 were given for blinding (2 when both subjects and investigators were blinded to thetreatment by use of identical placebo or other technique, 1 when the study is describedas “double-blind,” and 0 when the study was not double-blind). Score o f 0 or 1 wasgiven for frequency of withdrawals (1 when the number of withdrawals and reason forwithdrawals were stated, and 0 when no statement was made pertaining to withdrawals).

trials of DSS versus placebo (66, 67) with quality scoresof 4, 4; one trial of DSS versus psyllium (45) with a qual-ity score of 5. Evidence poor, Grade C recommendation.Psyllium superior.

Stimulant Laxatives(i) Senna (Sennokot, Perdiem overnight therapy)—one trial

versus sodium picosulfate (not available in the UnitedStates) (29) with a quality score of 2; one trial versus bran(40) with a quality score of 2. Evidence poor, Grade Crecommendation.

(ii) Bisacodyl (Gentlax)—one trial versus bisoxatin (notAvailable in the United States) (32) with a quality scoreof 2. Evidence poor, Grade C recommendation.

(iii) “Irritant”—one trial versus lactulose (31) with a qualityscore of 2. Lactulose better.

OthersTegaserod (Zelnorm) one very large trial with a quality scoreof 5. Evidence good. Category A recommendation.

WHAT IS KNOWN AND WHAT THIS PAPER ADDS

(i) Constipation is a common problem for which a widerange of medicines are used.

(ii) Many of the current recommendations for clinical prac-tice are based on evidence from trials.

(iii) Although metaanalysis of drug treatment of constipationhas been reported, there has been no systematic reviewof the quality or number of clinical trials or an in-depthevaluation of the safety and efficacy of these agents.

(iv) This review could serve to enlighten practitioners aboutthe evidence pertaining to various agents that are tradi-tionally used to treat constipation, and thus serve as aguide for selecting appropriate agents.

(v) In addition, this review highlights the fact that there isminimal or weak evidence to support the use of manycommonly used drugs for the treatment of constipation.

ACKNOWLEDGMENT

The authors would like to thank Dr. Philip Schoenfeld for hisadvice and the invaluable critique and thoughtful appraisalof Dr. W. Peterson in the preparation of this manuscript. Theauthors also wish to thank Novartis Pharmaceuticals for anunrestricted educational grant.

Reprint requests and correspondence: Satish S.C. Rao, M.D.,Ph.D., Neurogastroenterology, University of Iowa Carver Collegeof Medicine, Iowa City, IA 52242.

Received June 8, 2004; accepted November 15, 2004.

REFERENCES

1. Thompson WG, Heaton KW. Functional bowel disorders inapparently healthy people. Gastroenterology 1980;79:283–8.

2. Talley NJ, Weaver AL, Zinsmeister AR, et al. Functionalconstipation and outlet delay: A population-based study.Gastroenterology 1993;105:781–90.

3. Drossman DA, Li Z, Andruzzi E, et al. U.S. house-holder survey of functional gastrointestinal disorders. Preva-lence, sociodemography, and health impact. Dig Dis Sci1993;38:1569–80.

4. Johanson JF, Sonnenberg A, Koch TR. Clinical epidemiol-ogy of chronic constipation [comment]. J Clin Gastroenterol1989;11:525–36.

5. Harari D, Gurwitz JH, Minaker KL. Constipation in the el-derly [comment]. J Am Geriatr Soc 1993;41:1130–40.

6. Jones MP, Talley NJ, Nuyts G, et al. Lack of objective ev-idence of efficacy of laxatives in chronic constipation. DigDis Sci 2002;47:2222–30.

7. Schoenfeld P, Cook D, Hamilton F, et al. An evidence-based approach to gastroenterology therapy. Evidence-Based Gastroenterology Steering Group. Gastroenterology1998;114:1318–25.

8. Jadad AR, Moore RA, Carroll D, et al. Assessing the qualityof reports of randomized clinical trials: Is blinding neces-sary? Control Clin Trials 1996;17:1–12.

9. Harris RP, Helfand M, Woolf SH, et al. Current methodsof the US Preventive Services Task Force: A review of theprocess. Am J Prev Med 2001;20(3 Suppl):21–35.

10. Andorsky RI, Goldner F. Colonic lavage solution (polyethy-lene glycol electrolyte lavage solution) as a treatment forchronic constipation: A double-blind, placebo-controlledstudy. Am J Gastroenterol 1990;85:261–65.

11. Attar A, Lemann M, Ferguson A, et al. Comparison of alow dose polyethylene glycol electrolyte solution with lactu-lose for treatment of chronic constipation. Gut 1999;44:226–30.

12. Cleveland MV, Flavin DP, Ruben RA, et al. New polyethy-lene glycol laxative for treatment of constipation in adults: Arandomized, double-blind, placebo-controlled study. SouthMed Assoc J 2001;94:478–81.

13. Corazziari E, Badiali D, Habib FI, et al. Small volumeisosmotic polyethylene glycol electrolyte balanced solution(PMF-100) in treatment of chronic nonorganic constipation.Dig Dis Sci 1996;41:1636–42.

14. Corazziari E, Badiali D, Bazzocchi G, et al. Long term effi-cacy, safety, and tolerabilitity of low daily doses of isosmoticpolyethylene glycol electrolyte balanced solution (PMF-100) in the treatment of functional chronic constipation. Gut2000;46:522–26.


15. DiPalma JA, DeRidder PH, Orlando RC, et al. A random-ized, placebo-controlled, multicenter study of the safety andefficacy of a new polyethylene glycol laxative [comment].Am J Gastroenterol 2000;95:446–50.

16. Freedman MD, Schwartz HJ, Roby R, et al. Tolerance and ef-ficacy of polyethylene glycol 3350/electrolyte solution ver-sus lactulose in relieving opiate induced constipation: Adouble-blinded placebo-controlled trial. J Clin Pharmacol1997;37:904–7.

17. Chaussade S, Minic M. Comparison of efficacy and safetyof two doses of two different polyethylene glycol-based lax-atives in the treatment of constipation. Aliment PharmacolTher 2003;17(1):165–72.

18. Christie AH, Culbert P, Guest JF. Economic impact of lowdose polyethylene glycol 3350 plus electrolytes comparedwith lactulose in the management of idiopathic constipationin the UK. Pharmacoeconomics 2002;20:49–60.

19. Bass P, Dennis S. The laxative effects of lactulose in normaland constipated subjects. J Clin Gastroenterol 1981;3:23–28.

20. Sanders JF. Lactulose syrup assessed in a double-blindstudy of elderly constipated patients. J Am Geriatr Soc1978;26:236–39.

21. Wesselius-De Casparis A, Braadbaart S, Bergh-BohlkenGE, et al. Treatment of chronic constipation with lactulosesyrup: Results of a double-blind study. Gut 1968;9:84–86.

22. Kinnunen O, Winblad I, Koistinen P, et al. Safety and ef-ficacy of a bulk laxative containing senna versus lactulosein the treatment of chronic constipation in geriatric patients.Pharmacology 1993;47:253–55.

23. Lederle FA, Busch DL, Mattox KM, et al. Cost-effectivetreatment of constipation in the elderly: A randomizeddouble-blind comparison of sorbitol and lactulose. Am JMed 1990;89:597–601.

24. Passmore AP, Wilson-Davies K, Stoker C, et al. Chronicconstipation in long stay elderly patients: A comparison oflactulose and a senna-fibre combination [comment]. Br MedJ 1993;307:769–71.

25. Passmore AP, Davies KW, Flanagan PG, et al. A comparisonof Agiolax and lactulose in elderly patients with chronicconstipation. Pharmacology 1993;47:249–52.

26. Rouse M, Chapman N, Mahapatra M, et al. An open, ran-domised, parallel group study of lactulose versus ispaghulain the treatment of chronic constipation in adults. Br J ClinPract 1991;45:28–30.

27. Brocklehurst JC, Kirkland JL, Martin J, et al. Constipationin long-stay elderly patients: Its treatment and prevention bylactulose, poloxalkol-dihydroxyanthroquinolone and phos-phate enemas. Gerontology 1983;29:181–84.

28. Kinnunen O, Salokannel J. Constipation in elderly long-staypatients: Its treatment by magnesium hydroxide and bulk-laxative. Ann Clin Res 1987;19:321–23.

29. MacLennan WJ, Pooler A. A comparison of sodium picosul-phate (“Laxoberal”) with standardised senna (“Senokot”) ingeriatric patients. Curr Med Res Opin 1974;2:641–47.

30. Marlett JA, Li BU, Patrow CJ, et al. Comparative laxationof psyllium with and without senna in an ambulatory con-stipated population. Am J Gastroenterol 1987;82:333–37.

31. Connolly P, Hughes IW, Ryan G. Comparison of “Dupha-lac” and “irritant” laxatives during and after treatment ofchronic constipation: A preliminary study. Curr Med ResOpin 1974;2:620–5.

32. Rider JA. Treatment of acute and chronic constipation withbisoxatin acetate and bisacodyl. Double-blind crossoverstudy. Curr Ther Res Clin Exp 1971;13:386–92.

33. Mundow L. Danthron/poloxalkol and placebo in puerperalconstipation. Br J Clin Pract 1975;29:95–96.

34. Williamson J, Coll M, Connolly M. A comparative trial ofa new laxative. Nurs Times 1975;71:1705–7.

35. Griffenberg L, Morris M, Atkinson N, et al. The effect ofdietary fiber on bowel function following radical hysterec-tomy: A randomized trial. Gynecol Oncol 1997;66:417–24.

36. Hamilton JW, Wagner J, Burdick BB, et al. Clinical eval-uation of methylcellulose as a bulk laxative. Dig Dis Sci1988;33:993–918.

37. Anderson AS, Whichelow MJ. Constipation during preg-nancy: Dietary fibre intake and the effect of fibre supple-mentation. Hum Nutr Appl Nutr 1985;39:202–7.

38. Badiali D, Corazziari E, Habib FI, et al. Effect of wheat branin treatment of chronic nonorganic constipation. A double-blind controlled trial. Dig Dis Sci 1995;40:349–56.

39. Graham DY, Moser SE, Estes MK. The effect of branon bowel function in constipation. Am J Gastroenterol1982;77:599–603.

40. McCallum G, Ballinger BR, Presly AS. A trial of bran andbran biscuits for constipation in mentally handicapped andpsychogeriatric patients. J Hum Nutr 1978;32:369–72.

41. Mantle J. Research and serendipitous secondary findings.Can Nurs 1992;88:15–18.

42. Finlay M. The use of dietary fibre in a long-stay geriatricward. J Nutr Elder 1988;8:19–30.

43. Cheskin LJ, Kamal N, Crowell MD, et al. Mechanisms ofconstipation in older persons and effects of fiber comparedwith placebo. J Am Geriatr Soc 1995;43:666–69.

44. Ashraf W, Park F, Lof J, et al. Effects of psyllium therapyon stool characteristics, colon transit and anorectal functionin chronic idiopathic constipation. Aliment Pharmacol Ther1995;9:639–47.

45. McRorie JW, Daggy BP, Morel JG, et al. Psyllium is supe-rior to docusate sodium for treatment of chronic constipa-tion. Aliment Pharmacol Ther 1998;12:491–917.

46. Dettmar PW, Sykes J. A multi-centre, general practice com-parison of ispaghula husk with lactulose and other laxativesin the treatment of simple constipation. Curr Med Res Opin1998;14:227–33.

47. Odes HS, Madar Z. A double-blind trial of a celandin, aloev-era and psyllium laxative preparation in adult patients withconstipation. Digestion 1991;49:65–71.

48. Fenn GC, Wilkinson PD, Lee CE, et al. A general practicestudy of the efficacy of Regulan in functional constipation.Br J Clin Pract 1986;40:192–97.

49. Pers M, Pers B. A crossover comparative study with twobulk laxatives. J Int Med Res 1983;11:51–53.

50. Mamtani R, Cimino JA, Kugel R, et al. A calcium salt ofan insoluble synthetic bulking laxative in elderly bedriddennursing home residents. J Am Coll Nutr 1989;8:554–6.

51. Muller-Lissner SA. Treatment of chronic constipation withcisapride and placebo. Gut 1987;28:1033–38.

52. Muller-Lissner SA. Cisapride in chronic idiopathic consti-pation: Can the colon be re-educated? Bavarian ConstipationStudy Group. Eur J Gastroenterol Hepatol 1995;7:69–73.

53. Verheyen KVM, Demyttenaere P, Va Mierlo FJ. Double-blind comparison of two cisapride dosage regimens withplacebo in the treatment of functional constipation. CurrTher Res 1987;41:978–85.

54. Graham JR. Cisapride for severe non-ulcer dyspepsia, pseu-doobstruction and constipation. Med J Aust 1989;150:667.

55. Jost WH, Schimrigk K. Cisapride treatment of constipationin Parkinson’s disease. Mov Disord 1993;8:339–43.

56. Jost WH, Schimrigk K. Long-term results with cisapride inParkinson’s disease. Mov Disord 1997;12:423–25.

57. Longo WE, Woolsey RM, Vernava AM, et al. Cisapride forconstipation in spinal cord injured patients: A preliminaryreport. J Spinal Cord Med 1995;18:240–44.


58. Wang SJ, Lan JL, Chen DY, et al. Effects of cisapride oncolonic transit in patients with progressive systemic sclero-sis. Clin Rheumatol 2002;21:271–74.

59. Frame PS, Dolan P, Kohli R, et al. Use of colchicine to treatsevere constipation in developmentally disabled patients. JAm Board Fam Pract 1998;11:341–46.

60. Verne GN, Eaker EY, Davis RH, et al. Colchicine isan effective treatment for patients with chronic consti-pation: An open-label trial. Dig Dis Sci 1997;42:1959–63.

61. Verne GN, Davis RH, Robinson ME, et al. Treatment ofchronic constipation with colchicine: Randomized, double-blind, placebo-controlled, crossover trial. Am J Gastroen-terol 2003;98:1112–6.

62. Sandyk R, Gillman MA. Colchicine ameliorates constipa-tion in Parkinson’s disease. J R Soc Med 1984;77:1066.

63. Rajapakse R, Warman J, Korelitz BI. Colchicine for per-sistent constipation after total abdominal colectomy withileorectostomy for colonic inertia. J Clin Gastroenterol2001;33:81–84.

64. Soffer EE, Metcalf A, Launspach J. Misoprostol is effectivetreatment for patients with severe chronic constipation. DigDis Sci 1994;39:929–33.

65. Fain AM, Susat R, Herring M, et al. Treatment of constipa-tion in geriatric and chronically ill patients: A comparison.South Med Assoc J 1978;71:677–80.

66. Castle SC, Cantrell M, Israel DS, et al. Constipation preven-tion: Empiric use of stool softeners questioned. Geriatrics1991;46:84–86.

67. Hyland CM, Foran JD. Dioctyl sodium sulphosuccinate asa laxative in the elderly. Practitioner 1968;200:698–9.

68. Hurdon V, Viola R, Schroder C. How useful is docusatein patients at risk for constipation? A systematic review ofthe evidence in the chronically ill. J Pain Symptom Manage2000;19:130–36.

69. Johanson JF, Wald A, Tougas G, et al. Effect of tegaserodin chronic constipation: A randomized, double-blind, con-trolled trial. Clin Gastroenterol Hepatol 2004;2:796–805.

70. Sharma SS, Bhargava N, Mathur SC. Effect of oral ery-thromycin on colonic transit in patients with idiopathic con-stipation. A pilot study. Dig Dis Sci 1995;40:2446–49.

71. Meier R, Beglinger C, Thimshirn M, et al. Therapeuticeffects of loxiglumide, a cholecystokinin antagonist, onchronic constipation in elderly patients: A prospective, ran-domized, double-blind controlled trial. J Gastrointest Mot1993;5:129–35.

Documents

Efﬁcacy and Safety of Traditional Medical Therapies for ...Consti-pation was combined with the following terms: osmotic laxatives, irritant laxatives, stimulant laxatives, bulk lax-