American Journal of Gastroenterology ISSN 0002-9270

C 2006 by Am. Coll. of Gastroenterology doi: 10.1111/j.1572-0241.2006.00359.x
Published by Blackwell Publishing

CME

Fiber for the Treatment of Hemorrhoids Complications:

A Systematic Review and Meta-Analysis
Pablo Alonso-Coello,1 Ed Mills,2 Diane Heels-Ansdell,2 Maite López-Yarto,3 Qi Zhou,2 and John F. Johanson,4
Gordon Guyatt2
1
Iberoamerican Cochrane Center, Hospital Sant Pau, Barcelona, Spain; 2 Department of Clinical Epidemiology
and Biostatistics, McMaster University, Hamilton, Ontario, Canada; 3 Servicio de Ginecologı́a y Obstetricia,
Hospital del Mar, Barcelona, Spain; and 4 Department of Medicine, University of Illinois College of Medicine,
Rockford, Illinois

OBJECTIVES: To evaluate the impact of laxatives on a wide range of symptoms in patients with symptomatic
hemorrhoids.
METHODS: We searched using the following sources: MEDLINE, EMBASE, CINAHL and CENTRAL, BIOSIS, AMED,
Papers First and Proceedings; study authors, industry, and experts in the field. We included all
published and unpublished parallel group randomized controlled trials comparing any type of
laxative to placebo or no therapy in patients with symptomatic hemorrhoids. Two reviewers
independently screened studies for inclusion, retrieved all potentially relevant studies, and extracted
data on study population, intervention, prespecified outcomes, and methodology.
RESULTS: Seven trials randomized 378 patients to fiber or a nonfiber control. Studies were of moderate quality
for most outcomes. Meta-analyses using random effects models suggested that fiber has an
apparent beneficial effect. The risk of not improving/persisting symptoms decreased by 47% in the
fiber group (RR = 0.53, 95% CI 0.38–0.73) and the risk of bleeding by 50% (RR = 0.50, 95% CI
0.28–0.89). Studies with multiple follow-ups, usually at 6 wk and at 3 months, showed consistent
results over time. Results are also compatible with large treatment effects in prolapse, pain, and
itching, but even in the pooled analyses confidence intervals were wide and compatible with no
effect (RR = 0.79, 95% CI 0.37–1.67; RR = 0.33, 95% CI 0.07-1.65; and RR = 0.71, 95% CI
0.24–2.10, respectively). One study suggested a decrease in recurrence. Results showed a
nonsignificant trend toward increases in mild adverse events in the fiber group (RR = 6.0, 95% CI
0.57–64.8).
CONCLUSIONS: Trials of fiber show a consistent beneficial effect for symptoms and bleeding in the treatment of
symptomatic hemorrhoids.
(Am J Gastroenterol 2006;101:181–188)

INTRODUCTION measures and medical treatment for symptomatic hemor-

rhoids. The initial approach aims to increase the amount of
Symptomatic hemorrhoids are a common medical condition water and fiber in the diet, or to introduce a laxative. Consti-
with a prevalence ranging from 4.4% in the general pop- pation may be due to low fluid intake (5), but the effective-
ulation, to 36.4% in general practice (1), and an increased ness of increasing fluid intake as a treatment for constipation
prevalence during pregnancy and in the postpartum (1). While remains unknown. Dietary fiber intake has been positively
experts have usually divided internal hemorrhoids into four associated with increases in bowel movement frequency and
categories depending on the degree of prolapse (I–IV), some fecal mass among individuals with occasional or mild con-
authors recommend that they now base their classification stipation (6, 7). Other types of laxatives (stimulant laxatives,
on the presence or absence of bleeding or prolapse (2). osmotic agents, and fecal softeners) have proved effective for
The pathophysiology is not completely understood; struc- the treatment of constipation in randomized trials (5, 8–10)
tural and/or vascular changes are involved (3) and chronic but the poor methodology of these studies weakens inferences
straining is inconsistently associated (4). about treatment effect.
Minimizing constipation, and the prolonged straining that Several small clinical trials have evaluated the effect of
may be associated, is one of the main purposes of lifestyle fiber compared with placebo in patients with hemorrhoids
To access a continuing medical education exam for this article, please visit (11, 12). Authors of narrative reviews (13, 14) and clin-
www.acg.gi.org/journalcme. ical practice guidelines (15–17) have found the evidence
181
182 Alonso-Coello

inconclusive, but have still recommended use of fiber due Quantitative Data Synthesis
to its safety and low cost (13). To establish the strength of the Trials did not consistently use similar symptom measures;
available evidence, we conducted a systematic review of the all of them, however, recorded the proportion of patients ei-
impact of laxatives on a wide range of symptoms in patients ther free of symptoms, with symptom improvement, or still
with symptomatic hemorrhoids. symptomatic. We considered outcomes of patients free of
symptoms and patients with symptomatic improvement as
METHODS equivalent, and pooled each outcome of interest based on the
a priori expectation of a similar magnitude and direction of
We began by constructing a protocol that readers can obtain treatment effect.
by correspondence with the first author. We present results as the relative risk and risk difference
of being symptomatic or persisting symptoms. We calculated
Eligibility Criteria
pooled risk differences for being symptomatic/persisting
We selected all published and unpublished parallel group
symptoms for the different outcomes. Studies varied in their
randomized controlled trials of patients with symptomatic
duration of follow-up, the number of discrete measurements
hemorrhoids comparing any type of laxative to placebo or
they made, and the timing of their first follow-up measure-
no therapy, with any of the following outcomes recorded:
ment. Investigators’ first follow-up measurement occurred
individual or global symptom improvement, number of re-
from 6 wk to 3 months—we used this first measurement
currences in a time period, change in the degree of prolapse,
for all our pooled analyses. In studies with multiple follow-
need of surgical treatment, or other adverse effects. We also
ups we compared the different estimates across each study.
included crossover trials and quasi-randomized methods of
We calculated the pooled relative risks of re-treatment, pa-
treatment allocation. We contacted authors to provide addi-
tient satisfaction, need for additional treatment, and adverse
tional data and details about the key validity issues. There
effects.
were no language restrictions.
We undertook the analysis using the intention-to-treat prin-
Search Strategy ciple, including all patients in the study arm to which they
We searched OVID versions of MEDLINE (1966 to April were originally allocated. We used Review Manager 4.2 (The
Week 2, 2005), EMBASE (1980 to 2005 Week 17), CINAHL Cochrane Collaboration, Oxford, UK) to aggregate data for
(1982 to April Week 4 2005), limiting our searches to ran- each outcome using a random effects model (22). We present
domized controlled trials using a maximally sensitive strategy all pooled effect estimates with 95% confidence intervals; all
(18). We modified these searches for other databases as CEN- p values are two sided.
TRAL (the Cochrane Central Register of Controlled Trials, In crossover studies, we analyzed the data in the same way
The Cochrane Library, issue 2, 2005) BIOSIS, AMED (Al- as for parallel group studies, comparing treatment periods
lied and Alternative Medicine Database), Papers First and to control periods. We tested for between-study heterogene-
Proceedings. Two reviewers screened reference lists from ity for each pooled comparison using the Cochran Q statis-
all retrieved articles and from reviews and clinical practice tic. We also report the I 2 statistic, which is the proportion
guidelines to identify additional studies (13–16). We sought of the total variation among studies that is likely to be ex-
additional trials from pharmaceutical companies and experts plained by between-study heterogeneity rather than chance
in the field. We also searched for on-going trials in the Meta (23). Irrespective of the results of the formal statistical test
Register of Controlled Trials (mRCT), U.S. NIH register, and for heterogeneity, we tested whether our a priori hypotheses
the Register of the Center for Clinical Trials and Evidence- could explain variability in the magnitude of treatment effects
Based Healthcare. across studies. For each hypothesis, we tested the difference
in estimates of treatment effect between the two subgroups
Data Abstraction
using a Z test and considered p < 0.05 to be statistically
Two reviewers (E.M., P.A.) independently screened studies
significant (24).
for inclusion, retrieved all potentially relevant studies, and
Our a priori hypotheses to explain heterogeneity were:
extracted data on study population, intervention, prespeci-
(1) severity: smaller treatment effect in hemorrhoids grade
fied outcomes, and methodology from included trials. In both
III–IV compared to grade I–II; (2) condition: smaller treat-
phases, we resolved disagreements by consensus between re-
ment effect in thrombosed hemorrhoids versus nonthrom-
viewers, if unsolved after contacting study authors. We used
bosed; (3) intervention: smaller treatment effect in studies
Cohen’s κ to assess agreement between the two reviewers on
that used another treatment for hemorrhoids in both treat-
the selection of articles for inclusion (19).
ment arms (e.g., venotonic in both arms comparing fiber
Validity Assessment versus no fiber or placebo) (4) methodology: smaller treat-
We extracted methodological information for the assessment ment effect in studies with adequate allocation concealment
of internal validity (20): existence and method of generation and in studies with appropriate blinding of caregivers and
of the randomization schedule, and method of allocation con- smaller treatment effect in cross-over compared to parallel
cealment (21); blinding of caregivers and outcomes assessors; trials.
number and reasons of patients lost to follow-up; and use of An expanded version of this review will appear in the
validated outcome measures. Cochrane Library.
 Fiber for the Treatment of Hemorrhoids Complications 183

206 screened for relevance

Results were consistent across studies (heterogeneity p =
0.48, I 2 = 0%). Pooled risk difference for being symp-
188 articles excluded tomatic/persisting symptoms for the overall assessment was
Kappa: 0,67
25% (95% CI 0.36–0.13). The range of absolute percentages
between trials of those not improved was 0.16 to 0.40 for fiber
versus 0.23 to 0.61 for placebo.
18 for full text

8 excluded as unrelated Bleeding

Kappa: 1 Four studies (251 patients) that compared fiber to placebo
reported bleeding as an individual outcome (Fig. 3) (12, 25,
10 studies selected 31, 33). All results showed either a trend or a significant
difference in favor of the fiber group. The pooled analysis
3 excluded as double
publication, wrong topic
showed a 50% relative risk reduction in the active treatment
or retrospective study arm (RR = 0.50, 95% CI 0.28–0.89). No statistically signif-
icant heterogeneity was present but I 2 was moderate (p =
0.14, I 2 = 45.6%). Pooled risk difference for being symp-
7 studies included
tomatic/persisting symptoms for bleeding was 0.26 (95% CI
Figure 1. Flowchart of search results. 0.44 to 0.07). The range of absolute percentages between
trials of those being symptomatic/persisting symptoms was
0.07 to 0.31 for fiber versus 0.38 to 0.76 for placebo.
RESULTS One of the included studies provided the number of bleed-
ing episodes during the first 15 days, from day 15 to 30 and
The two reviewers achieved good agreement in the initial
from 30 to 45 days. These data could not be pooled with the
selection of trials from the titles for inclusion (κ = 0.67, 95%
rest of the studies as the authors no longer had access to the
CI 0.48–0.85) and excellent agreement on the final stage of
raw data (11). This study demonstrated a significant benefit
inclusion from full text articles (κ = 1.0) (Fig. 1). Six of the
in the treatment group compared to placebo but only in the
seven authors provided additional information regarding key
last two periods (5.5 ± 3.2 vs 3.1 ± 2.7 days and 5.5± 2.9
validity issues.
(p < 0.05) vs 1.1 ± 1.4 days (p < 0.001), respectively). There
Seven studies, comparing fiber versus placebo, met the in-
was no significant difference in the number of patients with
clusion criteria (Table 1). Six were parallel and one of them
hemorrhoids bleeding on contact with the anoscope or finger
used a crossed-over design (25). We excluded three retrieved
after 40 days of treatment (RR = 0.13, 95% CI 0.01–2.29)
studies for the following reasons: partial duplicate publica-
(11).
tion (26–28), wrong topic or retrospective study (29). Three
of the included studies were abstracts and were both pub-
lished later in full text (26–28). All trials included adults with
Prolapse
symptomatic hemorrhoids (grades I to III) and most patients
The pooled analysis of the three studies (223 patients) showed
presented with rectal bleeding as their main complaint. All
a nonsignificant difference between treatment and placebo
articles comparing laxatives evaluated the use of fiber versus
for persistent prolapse (RR = 0.79, 95% CI 0.37–1.67)
placebo. We did not identify any studies using other types of
(Fig. 4) (12, 25, 33). Pooled risk difference for being symp-
laxatives.
tomatic/persisting symptoms for prolapse was 0.08 (95% CI
The apparent quality from the published reports was gen-
0.22–0.06). The range of absolute percentages between trials
erally low with little detail provided concerning key validity
of those not improved was 0.03 to 0.35 for fiber versus 0.22
issues such as allocation concealment. When contacted di-
to 0.35 for placebo. No statistically significant heterogeneity
rectly the majority of authors provided additional information
was present but I 2 was moderate (p = 0.21, I 2 = 35.7%)
that generally indicated they had met methodological criteria
Perez-Miranda et al. similarly reported no differences in the
(Table 2). This finding is in agreement with recent data sug-
degree of prolapse by hemorrhoidal grade within arms com-
gesting authors typically use concealment of randomization
pared with baseline.
and blinding despite the failure to report these methods (30).
None of the included studies used validated questionnaires
to assess study outcomes.
Pain
We pooled two studies evaluating pain or discomfort (12, 32).
Global Ratings The pooled estimate showed a nonsignificant trend in favor
The pooled analysis for overall improvement showed a 47% of fiber (RR = 0.33, 95% CI 0.07–1.65). No statistically
reduction in the risk of not improving/persistent symptoms significant heterogeneity was present but I 2 was moderate
(RR = 0.53, 95% CI 0.38–0.73) (Fig. 2) (12, 25, 31, 33). (p = 0.14, I 2 = 53%).
 184

Table 1. Characteristics of the Included Studies

Author Type of Publication Interventions Participants Follow-up Outcomes Funding
Alonso-Coello

Broader Parallel randomized Stercullia vs placebo 40 outpatients with anal bleeding, Three months -Bleeding, prolapse, discomfort, Industry: provided
JH, controlled trial. Full (starch, <20 g/day) prolapse or discomfort. and overall impression medication, envelopes
1974 text available. Three months of Hemorrhoids grade I–III. -Adverse events and travel expenses to
treatment. -Bowel habit present results in
meetings∗ .
Webster Cross-over randomized Isphaghula husk (7 g/day) 67 outpatients with symptomatic Assessment at six and -Pruritus, prolapse, bleeding, and Industry (provided
DJT, controlled trial. Full vs placebo Two periods hemorrhoids referred to an 12 weeks overall symptoms medication)
1978 text available. of 6 wk of treatment. outpatient surgery clinic -Days of laxatives used
Hemorrhoids grade I–III, Age -Consistency of faces and
23–71, 37% women. frequency of defecation
-Proctoscopic evaluation
Foster GE, Parallel randomized Isphagula husk vs 41 patients with hemorrhoids. One month -Overall symptomatic Not stated
1979 controlled trial. placebo. One month of improvement.
Abstract treatment. -Anal and rectal pressure
Hunt PS, Parallel randomized Isphaghula husk vs 28 patients with bleeding Evaluation at three -Symptomatic improvement Industry (minimal)∗
1981 controlled trial. placebo. Six weeks of hemorrhoids. Hemorrhoids and six weeks -Proctoscopic improvement
Abstract treatment grade I–II. Data provided only at -Bowel habit (ease of defecation)
six weeks -Overall symptomatic
improvement
-Adverse events∗
Moesgaard Parallel randomized Psyllium seed dietary 52 outpatients with symptomatic Evaluation at three -Bleeding, pain at defecation, Industry (provided
F, 1982 controlled trial. Full fiber (20 g/day) vs hemorrhoids. Hemorrhoids and six weeks pruritus and/or anal secretion, medication)
text available. placebo. Six weeks of grade I–II. Mean age 54, 26% prolapse, and overall
treatment. of women. assessment
-Adverse events∗
Pérez- Parallel randomized Plantago ovata (11.6 50 outpatients with internal Assessed at 40 days -Average number of bleeding Undertaken without
Miranda controlled trial. Full g/day vs placebo bleeding hemorrhoids referred plus patient diary episodes per time period funding∗
M, 1996 text available. (vitamin B preparation) to colorectal outpatient clinic. -Number of congested
40 days of treatment Hemorrhoids grade I–IV. Mean hemorrhoidal cushions.
age 48, 42% women. Hemorrhoids bleeding on
contact (during anoscopy)
-Degree of prolapse
-Adverse events
Jensen SL, Parallel randomized Unprocessed bran (20 92 patients with hemorrhoids 18 months (at 6 -Number or recurrences after No funding was available∗
1988 controlled trial. Full g/day) vs no treatment. grade III after rubber band months interval) RBL (symptoms and
text available. ligation. Median age 47, 47% protrusions)
women.
18 months of treatment. -Severity of symptoms
-Laxatives intake
-Adverse events
RBL = rubber band ligation; IT = intention to treat.
∗ Data provided by authors.
 Fiber for the Treatment of Hemorrhoids Complications 185

Itching

Inadequate / Per protocol

The two studies that evaluated itching did not find a signif-

Inadequate / Unclear
Inadequate / Unclear
icant difference between the groups (12, 25) (RR = 0.71,

Lost to Follow-

Yes / Per protocol

Yes / Per protocol

Yes / Per protocol

95% CI 0.24–2.10) One of the studies evaluated a composite

Up / IT
outcome with itching and/or anal secretion but authors could
not provide the data for its components (12). No statistically

Yes / Yes
significant heterogeneity was present but I 2 was moderate
(p = 0.21, I 2 = 36.4%) The range of absolute percentages be-
tween trials of those being symptomatic/persisting symptoms

Surgeons and patients blinded (Described as double

Everyone blinded except Head pharmacist who was

Health providers and patients not blinded∗ . Patients

was 0.03 to 0.40 for fiber versus 0.16 to 0.43 for placebo.

Patients and observer blinded (Described as double

Evaluation of follow-up by independent observer

Patients and investigators blinded (Described as
aware of the list∗ (Described as double blind)
Recurrences or Need for Further Treatment
Only one study comparing fiber with placebo looked at the

Treatment and placebo looked alike∗

Unclear (Described as double blind)
Treatment and placebo looked alike

not aware of therapeutic effect∗

number of recurrences in the long term (34). Jensen et al.

Endoscopist and analyst blinded∗

Blinding
reported less overall recurrences in the fiber group (15% vs
45%) at 18 months in patients with third-degree hemorrhoids
after rubber band ligation (RR = 0.34, 95% CI 0.15, 0.77).
During the follow-up period there were fewer recurrent pro-
trusions in the treatment group (10% vs 38%) In the same

double blind)∗
study, the number of rubber band ligations required until dis-

Open study
appearance of symptoms was lower in the fiber group (median

blind)∗
blind)
2, range 1–4 vs 3, range 1–5).

Adverse Effects (opened when patient agreed

Adequate. Sealed envelopes∗

Adequate. Sealed envelopes∗

Adequate. Sealed envelopes
Allocation Concealment

The most common adverse effects with fiber consisted of

gastrointestinal symptoms, typically starting at the beginning
of the study, and were generally not severe enough for patients
to enter study)∗

to stop taking the treatment. Adverse effects were inconsistent

with some studies reporting a 50% incidence of bloating, the
Inadequate∗

Inadequate∗
Adequate∗

most common complaint, in the treatment group versus none

Unclear

in the placebo group (34). Two of the studies did not observe
any adverse effects (information provided by authors) (12,
30). The pooled estimate showed a nonsignificant increase in
Adequate. Randomization schedule (table)∗
Random list prepared by throwing a coin∗

the number of adverse events in the fiber group (RR = 6.0,

Adequate. Randomization centrally and

95% CI 0.57–64.84).
Pérez-Miranda M, 1996 Adequate. Computer generated list∗
Adequate. Randomization schedule

placed in sequentially numbered

Inadequate (by date of referral)∗

Randomization
Table 2. Methodological Quality of Included Studies

Variability in Study Results

envelopes by sponsor∗

In studies that measured symptoms on more than one visit—

usually at 6 wk and at 3 months—the results for later time
points were similar to earlier time points. Tests for het-
erogeneity all failed to reach statistical significance, but I 2
ranged from 1.1%, in the overall assessment, to 45.6% (sub-
Unclear

stantial heterogeneity exists when I 2 exceeds 50%). None

of our a priori hypotheses explained the variability in results
between the studies. Crossover estimates for the different out-
comes were consistently, though not significantly, closer to
Data provided by authors.
Moesgaard F, 1982
Webster DJT, 1978

1 than the parallel group estimates, suggesting a potential

IT = intention to treat.
Broader JH, 1974

Jensen SL, 1988

Foster GE, 1979

carry-over effect that decreased the size of the estimate. We

Hunt PS, 1981

found insufficient information in the studies for an adequate

evaluation of co-interventions (local treatments, bathing, and
Author

compliance with an increase of fiber in the diet). None of these

were part of our a priori hypotheses to explain heterogeneity.
∗
186 Alonso-Coello

Figure 2. Relative risk of being symptomatic/persisting symptoms for overall improvement.

DISCUSSION ponent of prolapse. Most trials evaluated grade I–II hem-

orrhoids, and those that included mixed populations failed
In this systematic review, we found that fiber shows an to provide data according to grade of severity. Although
apparent beneficial effect in the treatment of symptomatic fiber might also be effective in patients with more advanced
hemorrhoids. The risk of not improving/persisting symptoms stages of hemorrhoidal disease, this issue remains largely
decreased by 47% in the fiber group (RR = 0.53, 95% CI unaddressed.
0.38–0.73) and the risk of bleeding showed a significant dif- For all the major outcomes of this review, we would rate
ference in favor of the fiber too (RR = 0.50, 95% CI 0.28– the quality of the evidence as moderate (35). Publication bias
0.89) We also found that in studies with multiple follow- and funding remain issues of some concern. We contacted
ups, usually at 6 wk and at 3 months, the results for later authors and had access to the methodology for the major-
time points were very similar to earlier time points. Results ity of trials, and the information provided improved apparent
are also compatible with large treatment effects in prolapse, quality in comparison to the published articles. We found
pain, itching, but even in the pooled analyses confidence in- too few trials for the funnel plot to be of use. Our efforts to
tervals were wide and compatible with no effect (RR = 0.79, locate unpublished studies—we contacted authors, experts,
95% CI 0.37–1.67; RR = 0.33, 95% CI 0.07– 1.65; and RR and the pharmaceutical industry—and our success—we lo-
= 0.71, 95% CI 0.24–2.10, respectively). Results showed a cated two previously unpublished abstracts—make serious
nonsignificant trend toward increases in mild adverse events publication bias less likely. To the extent that we failed to
in the fiber group (RR = 6.0, 95% CI 0.57–64.8). identify additional unpublished studies with small or absent
Fiber is generally used in patients suffering from first- and treatment effects, our results may represent an overestimate
second-degree hemorrhoids, i.e., those with a lesser com- of the true underlying effect of treatment. There is evidence

Figure 3. Relative risk of being symptomatic/persisting symptoms for bleeding.

 Fiber for the Treatment of Hemorrhoids Complications 187

Figure 4. Relative risk of being symptomatic/persisting symptoms for prolapse.

that funding by the pharmaceutical industry can bias results in Reprint requests and correspondence:Pablo Alonso-coello,
favor of the intervention of interest (36), however, this kind Iberoamerican Cochrane Centre, Hospital Sant Pau, Sant Antonio
of funding other than providing the study medication was Maria Claret 171, Barcelona 08041, Spain.
present in only two of the studies (one of them declared min- Received June 14, 2005; accepted August 3, 2005.
imal funding). We believe that the limitations outlined above
leave inferences concerning the effects of fiber in ameliorat- REFERENCES
ing hemorrhoid symptoms moderately strong.
1. Johanson JF, Sonnenberg A. The prevalence of hemorrhoids
and chronic constipation. An epidemiologic study. Gas-
CONCLUSIONS troenterology 1990;98:380–6.
2. Beck DE. Hemorrhoidal disease. In: Beck DE, Wexner SD,
Fiber is an effective treatment for symptomatic hemorrhoids eds. Fundamentals of anorectal surgery. 2nd Ed. London:
WB Saunders edition, 1998:237–53.
(overall symptom improvement and bleeding). Results are
3. Abcarian H, Alexander-Williams J, Christiansen J,
also compatible with large treatment effects in prolapse, pain, et al. Benign anorectal disease: Definition, characterization,
itching, but even in the pooled analyses confidence intervals and analysis of treatment. Am J Gastroenterol 1994;89(8
were wide and compatible with no effect. Moderate study Suppl):S182–93.
quality leads to moderately strong inferences concerning the 4. Johanson JF, Rimm A. Optimal nonsurgical treatment of
hemorrhhoids: A comparative analysis of infrared coagula-
benefits of fiber. Thus, while future trials will likely confirm
tion, rubber band ligation, and injection sclerotherapy. Am
the observed effect, the relatively small number of patients J Gastroenterol 1992;87:1600–6.
enrolled in trials to date could argue for the need for addi- 5. Petticrew M, Rodgers M, Booth A. Effectiveness of laxatives
tional larger trials. Certainly trials that explore head to head in adults. Qual Health Care 2001;10:268–73.
comparisons with common first line treatments like venoton- 6. Bennett WG, Cerda JJ. Dietary fiber: Fact and fiction. Dig
Dis 1996;14:43–58.
ics (e.g., flavonoids) or topical treatments (anesthetic and/or
7. Spiller RC. Pharmacology of dietary fiber. Pharmacol Ther
steroids) would be informative, and most helpful, if they en- 1994;62:407–27.
rolled relatively large numbers of patients. The use of similar 8. Kenny KA, Dkelly JM. Dietary fiber for constipation in older
validated scales in future trials would facilitate comparisons adults: A systematic review. Clin Effect Nurs 2001;5:120–
and increase the validity of the results. 8.
9. Tramonte SM, Brand MB, Mulrow CD, et al. The treatment
of chronic constipation in adults: A systematic review. J Gen
ACKNOWLEDGMENTS Intern Med 1997;12:15–24.
10. Jones MP, Talley NJ, Nuyts G, et al. Lack of objective ev-
We would like to thank all authors of studies included in this idence of efficacy of laxatives in chronic constipation. Dig
review who provided additional information about their tri- Dis Sci 2002;47:2222–30.
11. Perez-Miranda M, Gomez-Cedenilla A, Leon-Colombo T,
als. The work of Dr. Alonso-Coello is partly funded by grant et al. Effect of fiber supplements on internal bleeding haem-
01/F070 of the Instituto de Salud Carlos III, Subdirección orrhoids. Hepatogastroenterology 1996;43:1504–7.
General de Investigación Sanitaria and by the Spanish Soci- 12. Moesgaard F, Nielsen ML, Hansen JB, et al. High-fiber
ety of Family Practice (semFYC) and the Red Temática de diet reduces bleeding and pain in patients with haemor-
Medicina Basada en la Evidencia G03/090. He is a Ph.D. rhoids: A double-blind trial of Vi-Siblin. Dis Colon Rectum
1982;25:454–6.
candidate at the Pediatrics, Obstetrics and Gynecology, and 13. Johanson JF. Evidence-based approach to the treatment
Preventive Medicine Department (Universidad Autónoma de of hemorrhoidal disease. Evidence-Based Gastroenterol
Barcelona, España). 2002;3:26–31.
188 Alonso-Coello

14. Johanson JF. Nonsurgical treatment of hemorrhoids. J Gas- 26. Maté J, Gómez A, Correa JA, et al. Therapeutic fiber and
trointest Surg 2002;6:290–4. bleeding haemorrhoids. Gut 1996;39:A140.
15. Madoff RD, Fleshman JW. Clinical Practice Commit- 27. Hunt P, Stewardson A, Korman M. Double-blind trial of
tee, American Gastroenterological Association. American fybogel (isphaghula husk) in the treatment of haemorrhoids.
Gastroenterological Association technical review on the Aust & N Z J Med 1981;11:221–2.
diagnosis and treatment of hemorrhoids. Gastroenterology 28. Craven JL. A controlled, double-blind study of dietary fibre
2004;126:1463–73. in patients with hemorrhoids. Symposium on dietary fibre
16. Alonso P, Marzo M, Mascort JJ, et al. Clinical practice 1976, St. James Hospital, Leeds.
guidelines for the management of patients with rectal bleed- 29. Gorgul A, Mentes BB, Tascilar O, et al. The results and
ing. Gastroenterol Hepatol 2002;25:605–32. comparison of rubber band ligation and injection sclerother-
17. Abramowitz L, Godeberge P, Staumont G, et al. Clinical apy supplemented by high-fibre diet in the treatment of
practice guidelines for the treatment of hemorrhoid disease. second-degree internal hemorrihoids. Turk J Gastroenterol
Gastroenterol Clin Biol 2001;25:674–2. 1999;10:66–71.
18. Dickersin K, Manheimer E, Wieland S, et al. Develop- 30. Devereaux PJ, Choi PT, El-Dika S, et al. An observational
ment of the Cochrane Collaboration’s CENTRAL Register study found that authors of randomized controlled trials fre-
of controlled clinical trials. Eval Health Prof 2002;25:38– quently use concealment of randomization and blinding, de-
64. spite the failure to report these methods. J Clin Epidemiol
19. Fleiss JL, Cohen J. The equivalence of weighted kappa and 2004;57:1232–6.
the intraclass correlation coefficient as measures of reliabil- 31. Hunt PS, Korman MG. Fybogel in haemorrhoid treatment.
ity. Educ and Psychol Measurement 1973;33:613–9. Med J Aust 1981;2:256–8.
20. Clarke M, Oxman AD, eds. Cochrane Reviewers’ Hand- 32. Foster GE, Bolwel JS, Wright J, et al. Controlled trial of
book 4.1.5 [updated April 2002]. In: The Cochrane Library. bulk forming evacuants in the treatment of patients with
Oxford: Update Software. Updated quarterly 2002. haemorrhoids. GUT 1979;20(Suppl 2):A452.
21. Juni P, Altman DG, Egger M. Systematic reviews in health 33. Broader JH, Gunn IF, Alexander-Williams J. Evaluation of a
care: Assessing the quality of controlled clinical trials. BMJ bulk-forming evacuant in the management of haemorrhoids.
2001;323:42–46. Br J Surg 1974;61:142–4.
22. DerSimonian R, Laird N. Meta-analysis in clinical trials. 34. Jensen SL, Harling H, Tange G, et al. Maintenance bran ther-
Control Clin Trials 1986;7:177–88. apy for prevention of symptoms after rubber band ligation of
23. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring in- third-degree haemorrhoids. Acta Chir Scand 1988;154:395–
consistency in meta-analyses. BMJ 2003;327:557–60. 8.
24. Fleiss JL. The statistical basis of meta-analysis. Stat Meth- 35. The GRADE Working Group. Grading quality of evidence
ods Med Res 1993;2:121–45. and strength of recommendations. BMJ 2004;328:1490–4.
25. Webster DJ, Gough DC, Craven JL.The use of bulk evac- 36. Lexchin J, Bero LA, Djulbegovic B, et al. Pharmaceuti-
uant in patients with haemorrhoids. Br J Surg 1978;65:291– cal industry sponsorship and research outcome and quality:
2. Systematic review. BMJ 2003;326(7400):1167–70.