Peppermint Oil for irritable Bowel Syndrome
A 2005 article in the journal Phytomedicine reviews 16 clinical trials that studied the effect of 180 200 mg of peppermint oil in irritable bowel syndrome or recurrent abdominal pain in children. Nine of the 16 studies were randomized double-blind crossover trials, five had a randomized double-blind parallel group design and two were open labeled studies. Eight of the 12 placebo-controlled studies show statistically significant effects of peppermint oil.
“Average response rates in terms of “overall success” are 58% (range
39-79%) for PO and 29% (range 10-52%) for placebo. The three studies
versus smooth muscle relaxants did not show differences between
treatments hinting for equivalence of treatments. Adverse events
reported were generally mild and transient, but very specific. PO
caused the typical GI effects like heartburn and anal/perianal burning
or discomfort sensations, whereas the anticholinergics caused dry mouth
and blurred vision. Anticholinergics and 5HT3/4-ant/agonists do not
offer superior improvement rates, placebo responses cover the range as
in PO trials. Taking into account the currently available drug
treatments for IBS PO (1-2 capsules t.i.d. over 2-4 weeks) may be the
drug of first choice in IBS patients with non-serious constipation or
diarrhea to alleviate general symptoms and to improve quality of life.”
[c] 2005 Elsevier GmbH. All rights reserved.
Keywords: Peppermint oil; Irritable bowel syndrome; Clinical trials; Efficacy; Safety; Treatment alternatives
Introduction
The lack of efficacious and safe medications for irritable bowel
syndrome (IBS) warrants a review of the quite extensive peppermint oil
(PO) clinical database in IBS, taking into account the significant
prevalence of the disease and its enormous costs as well as the
potential cost effectiveness of PO and its clinical safety.
The current definition of IBS (Rome II criteria) is listed below (Thompson et al., 1999).
At least 12 weeks, which need not to be consecutive, in the preceding
12 months of abdominal discomfort or pain that has two out of three
features:
* relieved with defecation; and/or
* onset associated with a change in frequency of stool; and/or
* onset associated with a change in form (appearance) of stool.
Symptoms that cumulatively support the diagnosis of IBS:
* abnormal stool frequency (for research purposes ‘abnormal’ may be
defined as greater than three bowel movements per day or less than
three bowel movements per week),
* abnormal stool form (lumpy/hard or loose/watery stool),
* abnormal stool passage (straining, urgency, incomplete evacuation),
* passage of mucus,
* bloating or feeling of abdominal distension.
PO exerts an antispasmodic action via interference of menthol, the main
component of PO, acting as a calcium antagonist (Hawthorn et al., 1988)
and anti-flatulent effects of currently unexplained nature (WHO, 2002).
Data base
A total of 15 published studies as
listed in Table 1 using PO in IBS were found in a literature search. A
further study (Kline and Barbero, 1997) refers to recurrent abdominal
pain in children. It was included due to the spasmogenic nature of the
underlying symptomatology. In Table 1 key information for all studies
is summarized.
Nine out of 16 studies were randomized double
blind cross over trials with (n = 5) or without (n = 4) run in and/or
wash out periods, five had a randomized double blind parallel group
design and two were open labeled studies. A total of 651 patients were
enrolled. A sex distribution for patients enrolled is available for
half of all studies. In general, there is a prevalence of women, which
is in line with current literature. Inclusion criteria appear to be
adequate, mentioning in 15 out of 16 studies IBS or relevant symptoms,
taking into account that some studies started in the 1980s before
definition of the Rome criteria. Qualitative efficacy rating by either
or both patient and physician included symptoms of IBS and overall
success judgments. Mostly per protocol analyses were done. Treatment
duration ranged from 2 to 11 weeks, in one open study it was 6 months.
As comparators served placebo (n = 12 studies; 1 study compares placebo
and an anticholinergic), smooth muscle relaxants (anticholinergics, n =
3 studies; 1 study compares placebo and an anticholinergic) and
psychotherapy (n = 1 study; stress management program). Placebo was
usually matching the active treatment or double dummy technique was
used were needed (anticholinergics). Out of these 13 trials refer to
investigations with enteric-coated PO and three to PO formulations
without galenic specification, the latter being presumably also enteric
coated. Treatments were administered 1-2 capsules t.i.d., each capsule
containing between 182 and 200 mg of PO, treatment duration was usually
2-4 weeks. Compliance was monitored in four studies. It was
approximately 60-75%.
Assessment of efficacy
In
11 out of 16 studies there was a daily patient rating of a whole set or
only selected symptoms (e.g. abdominal pain, distension, assessment of
winds, stool frequency, urgency, bloating, stool quality, frequency of
attacks, severity of attacks). The symptom catalogue essentially
followed the “Rome” criteria. In two studies rating by patients ensued
at regular intervals of 2 weeks, in two studies the interval is not
given and in one study there was the physician rating at the end of
each study week (open trial).
To make all these variations
comparable the variable “overall success” (overall benefit, global
improvement, overall assessment) was used (% of responders). This
parameter is either rated by investigators/patients or could be
calculated from the data available. Also, this variable is used in
recent IBS drug study meta-analyses (Poynard et al., 1994; Pittler and
Ernst, 1998) and review articles (Camilleri and Choi, 1997) to achieve
data comparability.
Results
In Table 2 all
overall success data from the 16 studies reviewed are summarized.
Placebo response is in the range from 10% to 52% (mean 29%) for all
studies. The open studies (Fernandez, 1990, Shaw et al., 1991) mark the
extremes of PO response, i.e. 18% or 93%. The double blind cross over
trials with (n = 5) or without (n = 4) run in and/or wash out periods
and the five randomized double blind parallel group design trials show
PO efficacy in the range from 39% to 79% (mean 58%). Eight out of 12
placebo controlled studies show statistically significant effects in
favor of PO. Where no response data versus placebo are recorded (n = 3
studies) statistical information reveals significance (n = 2) in favor
of or a borderline effect of PO. The three double blind cross over
studies versus smooth muscle relaxants did not show differences between
treatments hinting for equivalence of treatments, although a placebo
arm is missing with the exception of one trial (Carling et al., 1989).
There is reasonable evidence that enteric-coated PO. 180-200 mg t.i.d.,
given over 2-4 weeks, in IBS is efficacious as compared to placebo and
the smooth muscle relaxants investigated.
Adverse events
reported were generally mild and transient, but very specific. PO
caused the typical GI effects like heartburn (n = 14) and anal/perianal
burning or discomfort sensations (n = 26), whereas the anticholinergics
caused dry mouth (n = 21) and blurred vision (n = 14). There is no dose
or time dependent pattern for either active treatment. Tolerance in the
children study was good.
A total of 71 patients dropped out.
The vast majority (n = 58) due to events unrelated to study drugs (e.g.
protocol violations, failure to report back). Other reasons were: n = 6
worsening of symptoms (PO or placebo), n = 2 nausea and vomiting (PO),
n = 3 perianal burning (PO) and n = 2 peppermint taste and pyrosis.
Discussion
Data analysis of PO clinical data in IBS reveals that in a sufficient
number of studies with appropriate design the oil is safe and
efficacious as a symptomatic remedy. Average response rates are 58%
(range 39-79%) and 29% (range 10-52%) for placebo. This is in line with
the data compiled by Camilleri and Choi (1997) quoting 73% (range
39-89%) for various active treatments and 41% (range 13-69%) placebo
response. In a meta-analysis focusing on smooth muscle relaxants
Poynard et al. (1994) calculate 62% for active treatment and 35% for
placebo from 25 drug studies. Mertz (2003) quotes a 20 [right arrow]
50% response rate for placebo. The significant variability of response
to any treatment in IBS may underline the multi-causal etiology of the
disease.
It is interesting to note that also with recently
developed more targeted drugs like tegaserod (Mertz, 2003), alosetron
(SCRIP, 2000) or cilansetron (Pink Sheet, 2004) versus placebo response
rates cover the same range (52/42%, 41/29%, 60/45%). Only cilansetron
appears to be effective in both sexes. In 2000 (FDA Talk Paper, 2000)
alosetron was withdrawn from the US market because of ischemic colitis.
In 2002, the drug was re-launched for women with serious IBS under a
restricted distribution program (Pink Sheet, 2004). Also for tegaserod,
in an FDA Talk Paper (2004) new risk information was announced
referring to serious diarrhea, ischemic colitis and other forms of
intestinal ischemia. The drug is only approved for short-term treatment
of women. Also cilansetron, a new chemical in this group may have
comparable risks (Pink Sheet, 2004).
In a recent
meta-analysis assessing the global improvement of IBS by PO Pittler and
Ernst (1998) identified eight randomized placebo controlled trials.
Five studies were eligible for a meta-analysis and calculation shows a
beneficial effect of the oil (p < 0.001). This is in line with the
results of this review. The discussion touches a number of significant
issues (e.g. inclusion criteria, carry over effects) and it is stated
that definitive conclusions concerning efficacy cannot be inferred.
A critical view has to be made on: (1) The authors missed four
randomized trials versus placebo (Evans, 1982; Weiss and Kolbl, 1988;
Liu et al., 1997; Kline and Barbero, 1997 [children]), irrespective of
their eligibility for meta-analysis. (2) No wash out period in cross
over trials is addressed as a serious concern (carry over effect). It
is stated that only the study by Schneider and Otten (1990) had an
appropriate wash out period. But also the eligible double blind cross
over studies by Dew et al. (1984) and Rees (1979) must have had
variable wash out periods because the authors explicitly write that
“each treatment began when active symptoms developed”. (3) Whether the
study by Shaw et al. (1991) does not corroborate the positive findings
of other studies remains an open question because 6 months
psychotherapy are compared to drug treatment in an open design. There
is only 18% drug and 72% psychotherapy response. The drug rate is at
the very low end of placebo response, and it may be speculated that
drug patients aggravated their symptoms.
The data presented
in this review depict the current knowledge of PO in IBS, providing
evidence that the oil, administered orally in an enteric coated form,
is a safe, efficacious and cost-effective symptomatic short term
treatment in reducing global symptoms and pain due to its spasmolytic
and antiflatulent effect. The adverse event pattern of PO is distinct
from that of anticholinergics and 5HT3/4-ant/agonists with a low
frequency and mild symptoms, which in general do not require any
intervention, thus resulting in a positive benefit/risk ratio. Taking
into account the currently available drug treatments for IBS PO may be
the drug of first choice in IBS patients with non-serious constipation
or diarrhea to alleviate the general symptoms and to improve quality of
life, e.g. pain or bloating. Anticholinergics and 5HT3/4-ant/agonists
do not offer increased improvement rates, but the latter appear to be
useful under tight medical supervision in patients with serious
constipation or diarrhea.
References
Ambross,
N.S., 1990. Synthetic antispasmodic or peppermint for the irritable
bowel syndrome sufferer: a double blind comparative study. Abstract
World Congr Gastroenterol, Sydney.
Camilleri, M., Choi, M.G., 1997. Review article: irritable bowel syndrome. Aliment Pharmacol. Ther. 11, 3-15.
Carling, L., Svedberg, L.E., Hulten, S., 1989. Short term treatment of
the irritable bowel syndrome: a placebo-controlled trial of peppermint
oil against hyoscyamine. OPMEAR 34, 55-57.
Dew, M.J., Evans,
B.K., Rhodes, J., 1984. Peppermint oil for the irritable bowel
syndrome: a multicentre trial. Br. J. Clin. Pract. 38, 394-395.
Evans, B.K., 1982. Multicentre trial of peppermint oil capsules in
irritable bowel syndrome. Scand. J. Gastroenterol. 17 (Suppl. 78), 503.
FDA Talk Paper, 2000. Glaxo Welcome decides to withdraw Lotronex from the market. November 28.
FDA Talk Paper, 2004. FDA updates Zelnorm labeling with new risk information. April 20.
Fernandez, F., 1990. Peppermint in the treatment of irritable bowel
syndrome (Mentha piperita en al tratamiento de sindrome de colon
irritable). Investigacion Medica Intern. 17, 42-46.
Hawthorn, M., Ferrante, J., Luchowski, E., et al., 1988. The actions of
peppermint oil and menthol on calcium channel-dependent processes in
intestinal, neuronal and cardiac preparations. Aliment Pharmacol.
Therap. 2, 101-118.
Kline, R., Barbero, G., 1997. The use of
peppermint oil in the treatment of recurrent abdominal pain in
children. Gut 41(Suppl. 1), A90 P 046.
Lawson, M.J., Knight,
R.E., Tran, K., Walker, G., Roberts-Thomson, I.C., 1988. Failure of
enteric-coated peppermint oil in the irritable bowel syndrome: a
randomized, double-blind crossover study. J. Gastroenterol. Hepatol. 3,
235-238.
Lech, Y., Olesen, K.M., Hey, H., Rask-Pedersen, E.,
Vilian, M., Ostergaard, O., 1988. Treatment of irritable bowel syndrome
with peppermint oil (Behandling af colon irritabile med
pebermynteolie). Ugeskr Laeger 150 23, 88-89.
Liu, J.H.,
Chen, G.H., Yeh, H.Z., Huang, C.K., Poon, S.K., 1997. Enteric-coated
peppermint-oil capsules in the treatment of irritable bowel syndrome: a
prospective, randomized trial. J. Gastroenterol. 32, 765-768.
Mertz, H.R., 2003. Irritable bowel syndrome. N. Engl. J. Med. 349, 2136-2146.
Munst, G.J., Schlaurl, H., Nuesch, H.J., Vetter, D., Holik, K., Hammer,
B., Hohler, S., 1987. Pfefferminzol (Colpermin[R]) im Vergleich zu
Mebeverin (Duspatalin[R]) also Therapeutikum des Colon irritable
[Peppermint oil (Colpermin[R]) by comparison with mebeverine
(Duspatalin[R]) as treatment for irritable colon–a double blind
study]. Therapiewoche Schweiz 9, 853-858.
Nash, P., Gould,
S.R., Barnardo, D.E., 1986. Peppermint oil does not relieve the pain of
irritable bowel syndrome. Br. J. Clin. Pract. 40, 292-293.
Pink Sheet, 2004. Solvay cilansetron irritable bowel NDA will include appropriate use plan. May 24.
Pittler, M.H., Ernst, E., 1998. Peppermint oil for irritable bowel
syndrome: a critical review and metaanalysis. Am. J. Gastroenterol. 93,
1131-1135.
Poynard, T., Naveau, S., Mory, B., Chaput, J.C.,
1994. Meta-analysis of smooth muscle relaxants in the treatment of
irritable bowel syndrome. Aliment Pharmacol. Ther. 8, 499-510.
Rees, W.D., 1979. Treating irritable bowel syndrome with peppermint oil. Br. Med. J. 2, 835-836.
Schneider, M.M.E., Otten, M.H., 1990. Efficacy of Colpermin in the
treatment of patients with the irritable bowel syndrome.
Gastroenterology 98, A3 89.
SCRIP, 2000. No. 2526, March 29, p. 19.
Shaw, G., Srivastava, E.D., Sadlier, M., Swann, P., James, J.Y.,
Rhodes, J., 1991. Stress management for irritable bowel syndrome: a
controlled trial. Digestion 50, 36-42.
Thompson, W.G.,
Longstreth, G.F., Drossman, D.A., Heaton, K.W., Irvine, E.J.,
Muller-Lissner, S.A., 1999. Functional bowel disorders and functional
abdominal pain. Gut 45(Suppl. 2), II 43-7.
Weiss, W., Kolbl,
Ch., 1988. Therapie des Colon irritable mit Pfefferminzol-kapseln
[Treatment of irritable bowel syndrome with peppermint oil capsules].
Therapiewoche Osterreich 3, 3-8.
WHO Collaborating Center
for International Drug Monitoring: Draft herbals ATC index, 2002.
Uppsala Monitoring Center, p. 4-5.
Wildgrube, H.J., 1988.
Untersuchung zur Wirksamkeit von Pfefferminzol auf Beschwerdebild und
funktionelle Parameter bei Patienten mit Reizdarm-Syndrom (Studie)
[Investigation of the efficacy of peppermint oil on the symptoms and
functional parameters of patients with irritable bowel syndrome
(study)]. Naturheilpraxis 41, 2-5.
H.-G. Grigoleit*, P. Grigoleit
Johann-Sebastian-Bach-Str.27, 65193 Wiesbaden, Germany
Received 13 September 2004; accepted 26 October 2004
*Corresponding author. Tel.: 
+49611 520509
; fax: +49611 5990443.
E-mail addresses: Dr.Grigoleit@t-online.de (H.-G. Grigoleit), Dr.Grigoleit@t-online.de (P. Grigoleit).
<br />Table 1. Summary of study information <br /> <br /> Study drug(s) <br />Study no./Ref. Design Peppermint oil <br /> <br /> 1 Rees (1979) db, co, wash outOne to two <br /> period capsules <br /> t.i.d. (b) <br /> 2 Evans et al. (1982) db, co, randomized, One to two <br /> wash out? capsules <br /> t.i.d. (d) <br /> 3 Dew et al. (1984) db, co, wash outOne to two <br /> period capsules <br /> t.i.d. (b) <br /> 4 Nash et al. (1986) db, co, no wash out,Two capsules <br /> randomized t.i.d. (a) <br /> 5 Munst et al. (1987) db, co, wash out, One capsule <br /> double dummy, t.i.d. (a) <br /> randomized <br /> 6 Weiss and Kolbl (1988) db, pg, randomized One capsule <br /> t.i.d. (a) <br /> 7 Lawson et al. (1988)db, co, no wash out One capsule <br /> t.i.d. (b) <br /> 8 Lech et al. (1988) db, pg, randomized One capsule <br /> t.i.d. (d) <br /> 9 Wildgrube (1988)Matched pairs, db pg, Capsules (c) <br /> randomized <br />10 Carling et al. (1989) db, 3 way co, wash out One to two <br /> capsules <br /> t.i.d. (a) and <br /> matching placebo <br />11 Schneider and Otten (1990) db, co, wash out, One capsule <br /> randomized t.i.d. (a) <br />12 Fernandez (1990)OpenOne capsule <br /> t.i.d. (b) <br />13 Ambross (1990) db, co, randomized Not specified (d) <br />14 Shaw et al. (1991) Open, pg, randomizedOne capsule <br /> t.i.d. (a) <br />15 Liu et al. (1997) db, pg, randomized One capsule <br /> t.i.d. or <br /> q.i.d. (a) <br />16 Kline and Barbero (1997)db, pg, randomized One to two <br /> capsules <br /> t.i.d. (a) <br /> <br /> Study drug(s) Treatment Patients <br />Study no./Ref. Comparator(s) weeks enrolled <br /> <br /> 1 Rees (1979) Placebo 1-23/ 18 <br /> capsules t.i.d.treatment <br /> 2 Evans et al. (1982) Placebo2/ 20 <br /> treatment <br /> 3 Dew et al. (1984) Placebo 1-22/ 29 <br /> capsules t.i.d.treatment <br /> 4 Nash et al. (1986) Pacebo 2 2/ 41 <br /> capsules t.i.d.treatment <br /> 5 Munst et al. (1987) Matching 3/ 16 <br /> mebeverine treatment <br /> 135 mg 1 tablet <br /> t.i.d. <br /> 6 Weiss and Kolbl (1988) Placebo, 1 3 60 <br /> capsule t.i.d. <br /> 7 Lawson et al. (1988)Placebo, 1 4 25 <br /> capsule t.i.d. <br /> 8 Lech et al. (1988) Placebo, 1 4 47 <br /> capsule t.i.d. <br /> 9 Wildgrube (1988)Matching placebo 2 40 <br /> capsules <br />10 Carling et al. (1989) Hyoscyamine 0.22/ 40 <br /> mg and matchingtreatment <br /> placebo, 1-2 <br /> tablets t.i.d. <br />11 Schneider and Otten (1990) Placebo 1 capsule 6/ 60 <br /> t.i.d. treatment <br />12 Fernandez (1990) 4 50 <br />13 Ambross (1990) Alverine citrate 11/ 18 <br /> treatment <br />14 Shaw et al. (1991) Stress management 24 35 <br /> program, median 6 <br /> psychotherapy <br /> sessions of each <br /> 40 min/patient <br />15 Liu et al. (1997) Placebo 1 capsule 4 110 <br /> t.i.d. or q.i.d. <br />16 Kline and Barbero (1997)Placebo 1-22 42 <br /> capsules t.i.d. <br /> <br />db = double blind, co = cross over, pg = parallel groups. <br />(a) Colpermin[R]. <br />(b) Enteric-coated PO capsule. <br />(c) Mentacur[R]. <br />(d) Unspecified PO formulation. <br /> <br />Table 2. Summary of "overall success" data for peppermint (PO) oil in <br />IBS <br /> <br />Study Overall success (%) Overall success <br />no.peppermint oil Comparator comparator (%) <br /> <br /> 1 50 Placebo 13 <br /> 2 No numerical dataPlacebo No numerical data <br /> 3 41 Placebo 10 <br /> 4 39 Placebo 52 <br /> 5 No numerical dataMebeverine No numerical data <br /> 6 74 Placebo 17 <br /> 7 Placebo <br /> 8 68 Placebo 26 <br /> 9 No numerical dataPlacebo No numerical data <br />10 57 Placebo 37 <br />Hyoscyamine 38 <br />11 57 Placebo 39 <br />12 93 <br />13 No numerical dataAlverine No numerical data <br />14 18 Stress 72 <br />management <br />program <br />15 79 Placebo 32 <br />16 70 Placebo 43 <br /> <br />Study <br />no.Comments <br /> <br /> 1 p < 0.01 <br /> 2 Overall success in favor of PO (p < 0.025) <br /> 3 p < 0.001 <br /> 4 n.s. <br /> 5 Except for "fullness" no difference between treatments <br /> 6 p < 0.001 <br /> 7 Increase in stool frequency (p < 0.05), formulation problem <br /> 8 p < 0.02 <br /> 9 All symptoms improved in favour of peppermint oil (p < 0.05) <br />10 Symptom score before/after PO p < 0.01; placebo and hyoscyamine <br /> p>0.05 <br />11 Difference n.s. p = 0.08 <br />12 Open study <br />13 No difference between treatments <br />14 Strongly in favour of psychotherapy after 6 months <br />15 Overall success calculated from mean improvement values of <br /> symptoms, single symptoms all p < 0.05 <br />16 Children/recurrent abdominal pain, p < 0.002 <br />
