Jogc-may-06.vp

GYNAECOLOGY
GYNAECOLOGY
Use of Misoprostol Before Hysteroscopy:
A Systematic Review
Joan M. G. Crane, MD, FRCSC, Sarah Healey, MD, FRCSC
Department of Obstetrics and Gynecology, Memorial University of Newfoundland, St John’s NL
Abstract
Résumé
Objective: To evaluate the effectiveness of administering misoprostol
Objectif : Évaluer l’efficacité de l’administration de misoprostol avant
prior to hysteroscopy in achieving cervical dilatation and reducing une hystéroscopie en vue d’obtenir une dilatation cervicale et de complications including cervical laceration.
réduire les complications (y compris les lacérations cervicales).
Data Sources: Computerized searches of MEDLINE, PubMed and
Sources de données : Des recherches informatisées ont été menées
EMBASE were conducted using the key words “hysteroscopy” and dans MEDLINE, PubMed et EMBASE au moyen des mots clés “misoprostol.” References from identified publications were hysteroscopy et misoprostol. Les références des publications manually searched and cross-referenced to identify additional identifiées ont fait l’objet d’une recherche manuelle et ont été recoupées en vue d’identifier des articles pertinents additionnels.
Study Selection: We included randomized clinical trials that
compared women undergoing hysteroscopy who received Sélection des études : Nous avons inclus les essais cliniques
misoprostol before the procedure with those who received placebo.
randomisés qui comparaient les femmes subissant une Studies were excluded if there was no control group, if placebo hystéroscopie qui recevaient du misoprostol avant l’intervention à was not used, if women were not randomized, or if only the celles qui recevaient un placebo. Les études étaient exclues si abstract was available. Ten of 19 articles identified met the criteria elles ne comptaient pas de groupes témoins, si elles ne faisaient pas appel à un placebo, si leurs participantes n’étaient pasrandomisées ou si seul leur résumé était disponible. Dix des Data Extraction and Synthesis: The two co-authors separately
19 articles identifiés ont satisfait aux critères de l’analyse abstracted data. Any differences in data abstraction were resolved through discussion, and a consensus was reached. QUORUMguidelines for meta-analyses and systematic reviews of Extraction et synthèse des données : Les deux co-auteurs ont
randomized controlled trials were followed.
procédé séparément au résumé des données. Toutes les In premenopausal women, misoprostol before hysteroscopy différences quant au résumé des données ont été résolues au resulted in a reduced need for further cervical dilatation (relative moyen de discussions; ainsi, un consensus a pu être atteint. Les risk [RR] = 0.61; 95% confidence interval [CI] = 0.51, 0.73), a lower lignes directrices QUORUM visant les méta-analyses et les rate of cervical laceration (RR 0.22; 95% CI 0.09, 0.56) and analyses systématiques d’essais comparatifs randomisés ont été increased cervical dilatation (weighted mean difference 2.64; 95% CI 1.73, 3.54). In premenopausal women, misoprostol also resulted Chez les femmes préménopausées, l’administration de in a higher rate of side effects, including vaginal bleeding (RR misoprostol avant l’hystéroscopie a entraîné une diminution de la 11.09; 95% CI 3.08, 40.00), cramping (RR 7.98; 95% CI 3.38, nécessité de poursuivre la dilatation cervicale (risque relatif [RR] = 18.84), and elevated temperature (RR 5.24; 95% CI 1.37, 20.09).
0,61; intervalle de confiance [IC] à 95 % = 0,51, 0,73), un taux For every four premenopausal women who received misoprostol moindre de lacération cervicale (RR : 0,22; IC à 95 % : 0,09, 0,56) prior to hysteroscopy, one woman avoided the need for further et une dilatation cervicale accrue (différence moyenne pondérée : cervical dilatation. For every 12 premenopausal women receiving 2,64; IC à 95 % : 1,73, 3,54). Chez les femmes préménopausées, misoprostol, one cervical laceration was avoided.
l’administration de misoprostol a également entraîné un taux accru Conclusion: In premenopausal women, misoprostol appears to be
d’effets indésirables, y compris les saignements vaginaux (RR : promising as a cervical ripening agent prior to hysteroscopy, 11,09; IC à 95 % : 3,08, 40,00), les crampes (RR : 7,98; IC à although further research is needed to identify the ideal dose, 95 % : 3,38, 18,84) et une hausse de la température (RR : 5,24; IC route, and timing. Further research in postmenopausal women or à 95 % : 1,37, 20,09). Pour chaque groupe de quatre femmes those receiving GnRH agonists is also needed, to determine préménopausées ayant reçu du misoprostol avant une whether misoprostol is effective in cervical ripening in this hystéroscopie, une femme a pu éviter d’avoir besoin d’une dilatation cervicale accrue. Pour chaque groupe de 12 femmespréménopausées ayant reçu du misoprostol, un cas de lacérationcervicale a pu être évité.
Conclusion : Chez les femmes préménopausées, le misoprostol
Key Words: Misoprostol, hysteroscopy, meta-analysis
semble prometteur en tant qu’agent de maturation du col avantune hystéroscopie. Cependant, de plus amples recherches s’avèrent requises pour en identifier la dose, la voie d’administration et la posologie idéales. De plus amplesrecherches chez les femmes postménopausées ou chez les femmes recevant des agonistes de la gonadolibérine s’avèrent MAY JOGC MAI 2006 l
GYNAECOLOGY
également requises, et ce, afin de déterminer l’efficacité du outcomes. Any differences in data abstraction were misoprostol pour la maturation cervicale chez cette population.
resolved through discussion, and a consensus was reached.
J Obstet Gynaecol Can 2006;28(5):373–379 QUORUM guidelines for meta-analyses and systematic INTRODUCTION
reviews of randomized controlled trials were followed.8 Sta-tistical analyses were conducted using the computer pro- Hysteroscopy is a common gynaecologic procedure gram Review Manager 4.27 (Cochrane Collaboration, used to investigate and treat menstrual disorders, Oxford, UK). Heterogeneity among trials for each outcome postmenopausal bleeding, infertility, and recurrent was evaluated by chi-square analysis. A random effects pregnancy loss.1,2 The procedure requires that the cervical model was used if there was significant heterogeneity (P < canal be dilated enough to allow passage of the 0.10). For categorical data the relative risk of each outcome hysteroscope. Complications of hysteroscopy include cervi- for each study was calculated, and a summary relative risk cal laceration, uterine perforation, and creation of a false and 95% CI were calculated. For continuous data a passage during the attempt to dilate the cervix, which may weighted mean difference and 95% CI were calculated. A P value of less than 0.05 was considered significant. Analyses Misoprostol, a prostaglandin E1 analogue used in the pre- were planned in advance for (1) premenopausal women, vention and treatment of gastric ulcers induced by (2) postmenopausal women and premenopausal women non-steroidal anti-inflammatory drugs (NSAIDs), has been receiving GnRH agonists, and (3) both groups of women increasingly used in obstetrics and gynaecology as a cervical combined. The number of women needed to treat to avoid ripening agent.5–7 The purpose of this meta-analysis was to the need for cervical dilatation prior to hysteroscopy was evaluate the effectiveness of misoprostol prior to hysteroscopy in achieving preprocedural cervical dilatation premenopausal nulliparous women and of premenopausal and reducing the need for mechanical dilatation with its women undergoing operative hysteroscopy.
Sample size of the meta-analysis for the primary outcome ofneed for cervical dilatation prior to hysteroscopy was based on a reduction in the primary outcome from 50% to 33%,with a two-tailed á = 0.05 and â = 0.20, requiring 143 We performed searches of MEDLINE, PubMed, andEMBASE using the keywords “hysteroscopy” and “misoprostol” to identify studies involving human subjects Of the 19 studies2–5,10–24 identified from the literature published in English between January 1980 and April 2005.
search, ten10–19 were included in the meta-analysis (Figure).
References from these publications were searched manually Nine studies were excluded: two were reviews,5,23 two were and cross-referenced to identify additional relevant articles.
letters to the editor commenting on previous randomized Because of lack of details regarding study methods and clinical trials,21,22 one study was not randomized,20 one results, abstracts and unpublished works were not included.
study compared misoprostol with laminaria,24 and three Study Selection
studies did not use misoprostol.2–4 One author was con- Studies considered eligible were randomized clinical trials tacted to clarify information presented in the published that compared the effects of misoprostol and placebo given before the procedure in women undergoing hysteroscopy.
Table 1 summarizes the ten randomized clinical trials Studies were ineligible if there was no control group, if pla- included in the meta-analysis. Five of the studies included cebo was not used, if treatment was not randomized, or if only the abstract was published. The primary outcome was postmenopausal women or those receiving GnRH the need for further mechanical cervical dilatation. Second- agonists,15–17,19 and one study included both groups of ary outcomes were complications (particularly cervical laceration and uterine perforation), side effects (nausea, diarrhea, vaginal bleeding, cramping, and elevated tempera- scopy,11,13–15,18,19 and one study evaluated both.12 ture), and the continuous measure of cervical dilatation.
Table 2 summarizes the outcomes of misoprostol adminis- If an abstract described a study that did not meet eligibility tration before hysteroscopy for all women, revealing a sig- criteria, it was not reviewed further. Papers for all other nificant reduction in the need for further dilatation prior to abstracts were reviewed in detail. The two co-authors sepa- hysteroscopy and a lower rate of cervical laceration. There rately abstracted the data for the primary and secondary were higher rates of side effects with misoprostol, including lMAY JOGC MAI 2006
Use of Misoprostol Before Hysteroscopy: A Systematic Review Studies included in the meta-analysis
nausea, diarrhea, cramping, and fever. The use of Further post hoc subgroup analyses were performed evalu- misoprostol appears to be beneficial in premenopausal women (Table 3), who had significant reductions in rates of menopausal women undergoing operative hysteroscopy. As cervical laceration and need for dilatation, and there was only one study evaluated diagnostic hysteroscopy in greater preprocedural cervical dilatation (weighted mean premenopausal women,10 we did not have adequate sample difference = 2.6; 95% CI 1.7, 3.5, P < 0.0001, random size to comment further on this particular subgroup.
effects model) in this group. These reductions were not evi- Among premenopausal nulliparous women, the use ofmisoprostol reduced the need for further cervical dila- dent when postmenopausal women or those receiving tion12,13 (RR = 0.70; 95% CI = 0.59, 0.82, P = 0.007) and the GnRH agonists were studied (Table 4). In this group of occurrence of cervical laceration10,12,13 (RR = 0.14; women there was no significant difference in cervical 95% CI = 0.03,0.74, P = 0.02). The mean cervical dilatation dilatation with the use of misoprostol (weighted mean dif- was also greater in the misoprostol group (weighted mean ference = 0.1; 95% CI 0.4, 0.6, P = 0.62, random effects difference 3.36; 95% CI = 3.12, 3.59, fixed effects model).
model). For every four premenopausal women receiving Among premenopausal women undergoing operative lapa- misoprostol prior to hysteroscopy, one woman avoided the roscopy, the use of misoprostol reduced the need for cervi- need for further cervical dilatation.9 For every 12 cal dilatation11,13 (RR = 0.68; 95% CI = 0.58,80, P = 0.001) premenopausal women receiving misoprostol, one cervical and the occurrence of cervical laceration11,13,14 (RR = 0.22; 95% CI = 0.08, 0.60, P = 0.003), and resulted in an MAY JOGC MAI 2006 l
GYNAECOLOGY
Table 1. Studies of misoprostol prior to hysteroscopy
increased mean cervical dilatation13 (weighted mean differ- determined a priori the sample size required to detect a 33% ence = 3.50; 95% CI = 3.21, 3.79, fixed effects model). As reduction in the need for further cervical dilation (from only one study presented data specific for premenopausal 50% to 33%). We found that misoprostol in premenopausal nulliparous women undergoing operative hysteroscopy,13 women was an effective cervical ripening agent because it we are unable to comment further on this subgroup.
reduced the need for further cervical dilatation and loweredthe rate of cervical laceration. There were higher rates of When further subgroup analyses assessing the types of complications in women who used misoprostol, including hysteroscopy (operative or diagnostic) were performed for vaginal bleeding, cramping, and elevated temperature, but postmenopausal women or those receiving GnRH agonists, these side effects were transient. We found subgroups of no benefits with misoprostol were seen.
premenopausal women in whom misoprostol was of bene-fit: women undergoing operative hysteroscopy and women DISCUSSION
who were nulliparous. This is not surprising, as operative Studies of misoprostol suggest that it is very effective in ini- hysteroscopy requires greater preprocedural dilation, and tiating uterine contractions and inducing labour in the sec- nulliparous women are more likely to have a narrow cervical ond and third trimesters of pregnancy.6,7,25,26 It has also been shown to be useful in cervical ripening before surgical It is important to address the shortcomings of this abortion, making the procedure easier to perform by meta-analysis. A variety of protocols with different doses, increasing cervical dilatation and ease of dilatation.5,25–30 routes, and frequencies of administration were used in the Previous studies of misoprostol as a cervical ripening agent studies included in the meta-analysis. The population of before hysteroscopy found conflicting results depending on women evaluated also varied. Some studies evaluated the population studied. The small sample sizes of the indi- premenopausal women,10–14 others evaluated post- vidual studies meant that they lacked the statistical power to demonstrate significant differences in some of the out- agonists,15,19 and others studied a combination of these comes evaluated, such as cervical laceration and the need groups.18 Hysteroscopes of different sizes also were used, and some studies evaluated misoprostol before operative lMAY JOGC MAI 2006
Use of Misoprostol Before Hysteroscopy: A Systematic Review Table 2. Outcomes of misoprostol prior to hysteroscopy (all women)
CI: Confidence interval. * Random effects model; ** Fixed effects model Table 3. Outcomes of misoprostol administration prior to hysteroscopy in premenopausal women
Table 4. Outcomes of misoprostol prior to hysteroscopy in postmenopausal women or those receiving GnRH
agonists
CI: Confidence interval. * Fixed effects model MAY JOGC MAI 2006 l
GYNAECOLOGY
hysteroscopy and others before diagnostic hysteroscopy.
nulliparous women or those undergoing operative Despite these heterogeneous groups, a significant differ- hysteroscopy, because it reduces the need for further ence was found in the primary outcome in the mechanical cervical dilatation and reduces the rate of cervi- premenopausal group. This meta-analysis suggests that in cal laceration. Further research is needed to identify the premenopausal women (and, in particular, nulliparous ideal dose, route, and timing of misoprostol prior to the women or those undergoing operative hysteroscopy), procedure and to determine if misoprostol is effective in misoprostol administration before hysteroscopy improves cervical ripening in postmenopausal women or those preprocedural cervical dilatation and reduces cervical lacer- ation. Because of variations in protocols, however, we arenot able to determine the ideal dosing regimen. Other REFERENCES
shortcomings of the studies included in this meta-analysis 1. Baggish MS. Operative hysteroscopy. In: Rock JA, Thompson JD, editors.
are also evident. The studies did not provide further details TeLinde’s operative gynecology. Philadelphia: Lippincott – RavenPublishers;1997, p. 415–42.
on specific dilators used, or the cause of cervical lacerations(i.e., the tenaculum or the dilators). The hysteroscopies 2. Bradley LD. Complications in hysteroscopy: prevention, treatment and legal risk. Curr Opin Obstet Gynecol 2002;14:409–15.
were performed in a variety of settings, ranging from outpa- 3. Loffer FD. Complications of hysteroscopy - their cause, prevention and tient offices without analgesia to operating rooms with use correction. J Am Assoc Gynecol Larparosc 1995;3:11–26.
of general anaesthesia. Studies in premenopausal women 4. Vilos GA, Vilos EC, King JH. Experience with 800 hysteroscopic did not specify the timing of hysteroscopy in relation to the endometrial ablations. J Am Assoc Gynecol Laparosc 1996;4:33–8.
menstrual cycle. The rate of serious complications such as 5. Goldberg AB, Carusi DA, Meckstroth KR. Misoprostol in gynecology.
uterine perforation or creation of a false passage was low Curr Womens Health Rep 2003;3:475–83.
(1.9% in the control group) and so one would need a very 6. Alfirevic Z. Oral misoprostol for induction of labour (Cochrane review). In large sample size (n = 2515 per group) to detect a 50% the Cochrane Library Issue 4, 2004. Chichester, UK: John Wiley & Sons,Ltd.
reduction (2% to 1%) in this outcome. However, the need 7. Hofmeyr GJ, Gulmezoglu AM. Vaginal misoprostol for cervical ripening for further cervical dilatation and cervical laceration may be and induction of labour (Cochrane review). In: The Cochrane Library, Issue 4, 2004. Chichester, UK: John Wiley & Sons, Ltd.
8. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of meta-analyses of randomized controlled trials: the QUORUMstatement. Quality of reporting of meta-analyses. Lancet The benefit of using misoprostol in postmenopausal women or in those receiving GnRH agonists is less clear. In 9. Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based this group of women, only one study suggested benefit,18 medicine – how to practice and teach EBM. New York: Churchill and when these studies were combined in this meta-analysis there was no evident benefit. Unfortunately, data from the 10. Ngai SW, Chan YM, Liu KL, Ho PC. Oral misoprostol for cervical priming only study suggesting benefit were not presented in a way in non-pregnant women. Hum Reprod 1997;12:2373–5.
that allowed incorporation as a dichotomous outcome in 11. Atay V, Duru NK, Pabuccu R, Ergun A, Tokac G, Aydin BA. Vaginal the meta-analysis.18 It is possible that the dosing protocol of misoprostol for cervical dilatation before operative office hysteroscopy.
Gynecol Endoscopy 1997;6:47–9.
the study that suggested improvement in cervical dilatation(400 ìg by mouth, given at 24 hours and 12 hours prior to 12. Preutthipan S, Herabutya Y. A randomized controlled trial of vaginal misoprostol for cervical priming before hysteroscopy. Obstet Gynecol hysteroscopy) may provide the appropriate timing and dose to allow cervical ripening in these women. The overall lack 13. Preutthipan S, Herabutya Y. Vaginal misoprostol for cervical priming before of benefit seen in hypoestrogenic women suggests that operative hysteroscopy: a randomized controlled trial. Obstet Gynecol either estrogen or progesterone plays an important role in prostaglandin-induced cervical ripening. When further sub- 14. Fernandez H, Alby JD, Tournoux C, Cheuveaud-Lambling A, DeTayrae R, group analyses were performed (postmenopausal women Frydman R, et al. Vaginal misoprostol for cervical ripening before operative hysteroscopy in premenopausal women: a double-blind, placebo-controlledtrial with three dose regimes. Hum Reprod 2004; 9:1618–21.
hysteroscopy only), no subgroup of postmenopausalwomen was found to benefit, although with small sample 15. Cooper KG, Pinion SB, Bhattacharya S, Parkin DE. The effects of the gonadotropin releasing hormone analogue (goseralin) and prostaglandin E sizes we did not have adequate statistical power to evaluate (misoprostol) on cervical resistance prior to transcervical resection of the the outcomes of interest in these subgroups.
endometrium. Br J Obstet Gynaecol 1996;103:375–8.
16. Ngai SW, Chan YM, Ho PC. The use of misoprostol prior to hysteroscopy CONCLUSION
in postmenopausal women. Hum Reprod 2001;16:1486–8.
17. Fung TM, Lam MH, Wong SF, Ho LC. A randomized placebo-controlled Misoprostol has promise for use as a cervical ripening agent trial of vaginal misoprostol for cervical priming before hysteroscopy in before hysteroscopy in premenopausal women, particularly postmenopausal women. BJOG 2002;109:561–5.
lMAY JOGC MAI 2006
Use of Misoprostol Before Hysteroscopy: A Systematic Review 18. Thomas JA, Leyland N, Durand N, Windrim RC. The use of oral 25. Goldberg AB, Greenberg MB, Darney PD. Misoprostol and pregnancy.
misoprostol as a cervical ripening agent in operative hysteroscopy: a double-blind, placebo-controlled trial. Am J Obstet Gynecol 26. Blanchard K, Clark S, Winikoff B, Gaines G, Kaoni G, Shannon C.
Misoprostol for women’s health: a review. Obstet Gynaecol 19. Bisharah M, Al-Fozan H, Tulandi T. A randomized trial of sublingual misoprostol for cervical priming before hysteroscopy. J Am Assoc Gynecol 27. Fong YF, Singh K, Prasad RN. A comparative study using two dose regimens (200 ug or 400 ug) of vaginal misoprostol for pre-operative 20. Darwish A. Modified hysteroscopic myomectomy of large submucous cervical dilatation in first trimester nulliparae. Br J Obstet Gynaecol fibroids. Gynecol Obstet Invest 2003;56:192–6.
21. Sharma S, El-Rafaey H. The use of misoprostol as a cervical ripening agent 28. Singh K, Fong YF, Prasad RN, Dong F. Randomized trial to determine in operative hysteroscopy. Am J Obstet Gynecol 2003;18:297–8.
optimal dose of vaginal misoprostol for preabortion cervical priming.
22. Scott P, Magos A. Vaginal misoprostol for cervical priming before operative hysteroscopy: a randomized controlled trial. Obset Gynecol 2001;97:640–1.
29. MacIsaac L, Grossman D, Balistreri E, Darney P. A randomized controlled 23. Lichtenberg ES. Complications of osmotic dilators. Obstet Gynecol Surv trial of laminaria, oral misoprostol, and vaginal misoprostol before abortion.
24. Darvish AM, Ahmad RM, Mohammad AM. Cervical priming prior to 30. Singh K, Fong YF, Prasad RN, Dong F. Evacuation interval after vaginal operative hysteroscopy: a randomized comparison of laminaria versus misoprostol for preabortion cervical priming: A randomized clinical trial.
misoprostol. Hum Reprod 2004; 19:2391–4.
MAY JOGC MAI 2006 l

Source: http://jogc.com/abstracts/full/200605_Gynaecology_1.pdf