Over the past decades, advances in fiberoptics, light sources, high-resolution lenses, and endoscopic surgical instrumentation have made hysteroscopy an important diagnostic and therapeutic tool for patients with intrauterine diseases. Hysteroscopy permits a panoramic view of the uterine cavity and direct biopsy of lesions, thus increasing the accuracy of diagnoses. The common complications encountered during hysteroscopy are mainly related to the difficulty of cervical dilatation and include cervical tears, creation of false tracts, and uterine perforation.
Misoprostol is a stable, orally active, synthetic prostaglandin (PG) analogue used to treat peptic ulcers. Oral misoprostol has proved effective for cervical dilation before suction evacuation in the first trimester of pregnancy.1 Recently, oral misoprostol was shown to be effective for preoperative cervical dilation in nonpregnant women before hysteroscopy.2 Vaginal misoprostol has facilitated cervical dilation in pregnancy interruption before 10 weeks' gestation3 and in missed abortion.4 We propose that vaginal misoprostol also may facilitate cervical dilation in women undergoing hysteroscopy.
We designed this double-blind, randomized, placebo-controlled study to evaluate the effectiveness and side effects of vaginal misoprostol for cervical dilation in nonpregnant women before hysteroscopy. From a pilot study, 30 patients were randomized to receive misoprostol or placebo. The difference in cervical width after administration of vaginal misoprostol and placebo was the main indicator for calculating the sample size. The mean cervical widths in the misoprostol and placebo groups were 6.5 ± 1.9 and 4.9 ± 1.2 mm, respectively. With a type I error of 0.01 and a power of 0.99, 35 subjects were needed in each arm.
Materials and Methods
From October 1997 through September 1998, 91 women who underwent routine investigations for infertility were recruited for a randomized study at Ramathibodi Hospital. These patients were suspected of having intrauterine abnormalities (Table 1) that needed further definite diagnosis and treatment by hysteroscopy. Women who were suspected of being in early pregnancy or of having genital tract infections were excluded. The study was performed in a double-blind manner, and all subjects were outpatients. Randomization was done by a resident-in-training who picked random numbers to assign the subjects to receive either 200 μg of misoprostol or an identical-looking placebo (lactose filler). The drug or placebo was placed in the posterior vaginal fornix 9–10 hours before hysteroscopy. This study was approved by the departmental ethics committees, and written informed consent was obtained from each subject.
The subjects were told to fast from midnight and to come to the operating room in the morning. With the women in a lithotomy position, a number-1 Hegar dilator was inserted through the internal os, followed by successively larger Hegar dilators until resistance was met. The cervical width, assessed by the size of the Hegar dilator, was then noted. For diagnostic purposes, if the endocervical canal was tight, the cervix was dilated to Hegar number 6. If the operative sheath was required for a targeted biopsy or another minor operative procedure, the cervix was dilated to Hegar number 8. For hysteroscopic resection, the cervix was dilated to Hegar number 9.
Hysteroscopy usually was timed for the proliferative phase of the menstrual cycle. We used a standard, rigid 4-mm hysteroscope with a 30° forward-oblique lens and a 5.5-mm diagnostic sheath (Karl Storz GmbH, Tuttlingen, Germany). The women were placed under general anesthesia using propofol as total intravenous anesthesia. In most cases, the uterine cavity was distended with carbon dioxide by an electronic Hamou hysteroflator (Karl Storz GmbH), providing a flow rate of up to 50 mL/minute at a pressure of 100 mmHg. When visualization was inadequate because of excessive bleeding or when an operative procedure was needed, the distention medium was changed to 1.5% glycine solution at an insufflation pressure of 100 mmHg, with careful monitoring of the fluid balance. The technique of hysteroscopy is described elsewhere.5
The outcomes assessed included cervical width at hysteroscopy, number and percentage of women who needed cervical dilatation, duration of hysteroscopy from insertion of the hysteroscope until completion of hysteroscopic examination, and associated side effects. Any side effects such as nausea, vomiting, diarrhea, headache, increase in body temperature, lower abdominal pain, or vaginal bleeding also were recorded. Data were analyzed using the unpaired t test, χ2 test, and Fisher exact test where appropriate, with P < .05 considered statistically significant.
Among the 91 women examined, 46 were allocated to misoprostol and 45 to placebo. All recruited women were nulliparous. Most of them suffered from primary infertility. The mean age of the women in the misoprostol group was 29.6 ± 5.3 years (range 18–40). The mean age in the placebo group was 31.2 ± 5.0 years (range 19–40). The indications for diagnostic hysteroscopy were similar in the two groups (Table 1).
The effects of vaginal misoprostol and placebo administration and the intrauterine findings of both groups are shown in Table 2. Findings in the 91 women included intrauterine adhesions in 30 (33.0%), endometrial polyps in 29 (31.9%), submucous myoma in 12 (13.2%), müllerian defects in five (5.5%), endometrial hyperplasia in six (6.6%), and normal uterine cavities in nine (9.9%). The mean cervical widths for the misoprostol and the placebo groups were 7.0 ± 1.0 mm (range 6–8.5) and 3.8 ± 1.2 mm (range 2–5.5), respectively (P < .001). In the misoprostol group, only three women (6.5%) needed cervical dilation before hysteroscopy, compared with 14 women (31.1%) in the placebo group (P = .006). The mean durations of diagnostic hysteroscopy in the misoprostol and placebo groups were 90.0 ± 38.4 seconds (range 60–240) and 142.0 ± 38.7 seconds (range 60–270), respectively (P < .001).
Side effects in the 46 women given vaginal misoprostol were mild lower abdominal pain in 15 (32.6%, P < .001), vaginal bleeding in 12 (26.1%, P < .001), nausea in two (4.4%), watery diarrhea in three (6.5%), and perceived increase in body temperature in five (10.9%). Endometrial polyps were found in the vagina at hysteroscopy in five of 12 women with vaginal bleeding after misoprostol administration. There were no side effects in the control group. Cervical tears were detected in two women (4.4%) in the placebo group and required sutures at the end of hysteroscopy. All women were discharged a few hours after the procedure.
Prostaglandins are known to cause cervical softening and dilation in the first, second, or third trimester of pregnancy. At Ramathibodi Hospital, PGs have been used for many years for pregnancy termination and labor induction. Misoprostol, a commercially available synthetic derivative of PGE1, is used to treat peptic ulcers. It is very inexpensive (in Thailand, one tablet costs $0.25) and can be kept at room temperature. We recently showed that vaginal administration of misoprostol in women with missed abortions produced spontaneous expulsion of the pregnancies and reduced the need for surgical treatment.4 Creinin et al6 showed that vaginal misoprostol was more effective with fewer side effects than oral misoprostol for the uterine evacuation of early pregnancy failure at 8 weeks' gestation or less, as determined by ultrasound or physical examination. Ngai et al2 compared oral misoprostol with placebo for a cervical priming effect in nonpregnant women before hysteroscopy. They found that the cumulative force required for cervical dilation was significantly lower and the baseline cervical dilatation was significantly greater in the misoprostol group. To our knowledge, our study is the first randomized, double-blind study to compare the effectiveness of vaginal misoprostol with placebo for cervical priming before hysteroscopy.
Our study demonstrates that vaginal misoprostol is more effective than placebo for cervical dilation before hysteroscopy. Widths of the cervical canal were significantly different between the groups as evaluated by Hegar dilators. We also found that a softening of the uterine cervix takes place in women after misoprostol administration. The mechanism of this action is not known. We believe that the mechanism of cervical priming in nonpregnant women may be similar to that in pregnant women. Prostaglandin E2 has been shown to modulate biochemical events that underlie cervical ripening in pregnant women.7 Ding et al8 investigated the clinical effects and cervical tissue changes after treatment with PGE1 in the first trimester of pregnancy. They found that the mean cervical dilatation was 7.9 mm in treated women, compared with 3.9 mm in controls. The in vitro incorporation of radioactive proline in cervical tissue along with the hydrolytic activity against a synthetic collagenlike polypeptide was increased after treatment with PGE1. The cervical smooth-muscle sensitivity to PGs, as revealed by inhibition of muscle activity, was higher in treated women than in controls. Ding et al8 concluded that cervical dilation as induced by PGE1: involved an adaptation of both connective-tissue and smooth-muscle components. Recently, el-Refaey et al9 have shown clinical and histo-chemical evidence of change in the human pregnant cervix in response to the vaginal administration of misoprostol and gemeprost.
There was a significant difference between the groups in the need for cervical dilation before hysteroscopy. In the placebo group, cervical dilation was required during diagnostic hysteroscopy in 14 of 45 women (31.1%), but in the misoprostol group, only three of 46 women (6.5%) needed cervical dilation. The results of this study show that 200 μg of vaginal misoprostol can reduce the need for cervical dilation. In fact, the three women who needed cervical dilation after misoprostol administration had severe intrauterine adhesions with cervical stenosis. We postulate that intrauterine adhesions with scar-tissue formation at the cervix may inhibit the action of misoprostol.
The mean duration of diagnostic hysteroscopy was significantly shorter in the treated group than in the control group. Misoprostol increased the baseline cervical dilatation and reduced the need for mechanical dilation of the cervix. Hysteroscopic procedures performed with a dilated cervical canal are much easier and safer than those done with a stiff or narrow canal. We spent less time doing hysteroscopy and had no complications in the treated group, but in the control group, two patients had cervical tears that needed suturing.
We found a relatively high frequency of side effects after vaginal misoprostol administration, particularly mild lower abdominal pain and slight vaginal bleeding. The leading side effects in 15 women reported by Hald et al,10 using vaginal meteneprost potassium (PGE2) for cervical priming before hysteroscopy, were mostly gastrointestinal symptoms such as stomach pain (60%), nausea (40%), and diarrhea (33%). Five women (10.9%) in our treatment group felt feverish, but we did not measure body temperatures. Surprisingly, five of 12 women with vaginal bleeding after misoprostol administration had endometrial polyps found in the vagina at hysterscopy, but none were found in the 45 controls. It is possible that these polyps were expelled from the uterine cavity by uterine contractions. Most of the treated group (27 of 46 patients) had some portion of misoprostol left in the vagina on the day of hysteroscopy. Further study of misoprostol dosing is needed to maintain good efficacy in cervical priming and to reduce side effects.
In conclusion, our study shows that vaginal misoprostol as a cervical priming agent in nonpregnant women appears to be safe, effective, and inexpensive, with mild side effects. Pretreatment of nonpregnant women needing hysteroscopy should be considered.
1. Ngai SW, Tang OS, Lao T, Ho PC, Ma HK. Oral misoprostol versus placebo for cervical dilatation before vacuum aspiration in first trimester pregnancy. Hum Reprod 1995;5:1220–2.
2. Ngai WN, Chan YM, Liu KL, Ho PC. Oral misoprostol for cervical priming in non-pregnant women. Hum Reprod 1997;12:2373–5.
3. Ficicioglu C, Tasdemir M, Tasdemir S. Effect of vaginal misoprostol application for cervical softening in pregnancy interruption before ten weeks of gestation. Acta Obstet Gynecol Scand 1996;75:54–6.
4. Herabutya Y, O-Prasertsawat P. Misoprostol in the management of missed abortion. Int J Gynaecol Obstet 1997;56:263–6.
5. Preutthipan S, Linasmita V, Theppisai U. Diagnostic hysteroscopy: A result of 125 patients at Ramathibodi Hospital. J Med Assoc Thai 1997;80:575–9.
6. Creinin MD, Moyer R, Guido R. Misoprostol for medical evacuation of early pregnancy failure. Obstet Gynecol 1997;89:768–72.
7. Norstrom A. Influence of prostaglandin E2 on the biosynthesis of connective tissue constituents in the pregnant human cervix. Prostaglandins 1982;23:361–7.
8. Ding JQ, Granberg S, Norstrom A. Clinical effects and cervical tissue changes after treatment with 16, 16 dimethyl-trans delta 2 PGE1
, methylester in the first trimester. Prostaglandins 1990;39:281–5.
9. el-Refaey H, Calder L, Wheatley DN, Templeton A. Cervical priming with prostaglandin E1
analogues, misoprostol and gemeprost. Lancet 1994;343:1207–9.
10. Hald F, Kristoffersen SE, Gregersen E. Prostaglandin vaginal suppositories in non pregnant women requiring cervical dilatation prior to hysteroscopy. Acta Obstet Gynecol Scand 1988;67:219–22.