Human Reproduction vol.15 no.10 pp.2224–2227, 2000
Chemical ripening of the cervix with intracervical application of sodium nitroprusside: a randomized controlled trial
Fabio Facchinetti1, Federica Piccinini and Annibale Volpe Department of Gynaecologica, Obstetrics and Paediatric Sciences, University of Modena and Reggio Emilia, Modena, Italy 1To
whom correspondence should be addressed at: Clinica Ostetrica Ginecologica, Via del Pozzo 71, 41100 Modena, Italy. E-mail:
[email protected]
Nitric oxide (NO) has been found to be involved in the processes of cervical ripening. In a randomized, placebocontrolled study, cervical softening by an intracervical application of sodium nitroprusside, one of the most clinically potent and effective NO donor agents, was evaluated. A total of 36 primigravid women undergoing pregnancy termination between 9 and 12.5 weeks were enrolled. In one series, 18 patients were randomized to receive intracervically either placebo or 1% nitroprusside gel (5 mg), followed by uterine evacuation 6 h after treatment. In another series, 18 patients received either placebo or 2% nitroprusside gel (10 mg) into the cervical canal followed by uterine evacuation 3 h later. The cervical resistance, i.e. the force required to dilate the cervix from 3 to 10 mm, was the main outcome variable. It was recorded using a force sensing apparatus (dynamometer). Blood pressure was measured. Adverse events were recorded until 2 h after surgery. Women treated with both doses of nitroprusside gel showed values of cervical resistance significantly lower than those treated with placebo gel, at any tested diameter. No differences were found between subjects treated with the two different doses of nitroprusside. No significant consistent changes in blood pressure were induced by either dose of nitroprusside. No headaches were found in subjects treated with the NO donor. This study demonstrates that sodium nitroprusside applied into the cervical canal induces a rapid and significant softening of the cervix, thus reducing the force required to dilate it, compared with placebo-treated subjects. The chemical ripening of the cervix with sodium nitroprusside intracervical gel is an efficacious procedure in first-trimester pregnancy. Key words: cervical dilation/nitric oxide/pregnancy/sodium nitroprusside
Introduction The discovery that several tissues produce nitric oxide (NO) has stimulated investigators to look at the possible involvement of this agent in physiological as well as pathophysiological 2224
processes, including those pertaining to female reproduction and perinatology (Neri et al., 1995; Rosselli et al., 1998). Of paramount importance, in addition to the expected and well described effect that NO inhibits myometrial contractility (Izumi et al., 1993), Garfield et al. highlighted the possibility that NO is involved in the control of cervical ripening (Garfield et al., 1997). Inducible NO synthase activity was detected both in rats and guinea pig cervices where it is up-regulated during labour, contrary to what happens in the myometrium (Buhimschi et al., 1996). Moreover, the local application of NO donors induced cervical ripening associated with ultrastructural and functional changes in the guinea pig (Chwalisz et al., 1997). Furthermore, it is possible that NO also exerts its actions through the mediation of prostaglandins, as supported from in-vitro observations (Ledingham et al., 1999). A similar effect was demonstrated in women. It was reported that both glyceryltrinitrate (Thomson et al., 1997) and isosorbide mononitrate (Thomson et al., 1998) administered by vaginal route were able to reduce the cumulative force required to dilate the cervix in pregnant women during the first trimester. The current study was based on the hypothesis that NO donors applied intracervically could induce cervical softening without interfering with blood pressure which is known to be reduced in the case of uncontrolled NO release into general circulation. Therefore, the cervical ripening with intracervical application of a very potent NO donor, sodium nitroprusside, was evaluated.
Materials and methods Protocol Thirty-six women, admitted to the hospital for a voluntary pregnancy termination during the first trimester, were enrolled. Inclusion criteria were: first pregnancy, pregnancy dating between 9 and 12.5 weeks (determined according to last menstrual period and clinical examination), filling requirements of Italian Law number 194/78 and undergoing uterine evacuation under local anaesthesia. Exclusion criteria were: inability to understand the consent form, multiple pregnancy, previous uterine surgery, known allergy to drugs and any chronic disease requiring medication. On the morning planned for uterine evacuation the procedure was explained to the women, who were enrolled after signing the informed consent. In the fasting condition, 18 women were then randomized to receive either placebo gel or 1% nitroprusside-based gel (equivalent to 5 mg) into the cervical canal, with the aid of a plastic catheter, after visualization of the cervix with a speculum. Just before, and 60, 120, 240 and 360 min later blood pressure was measured with an automatic device (Dinamap®; Critikon, Tampa, FL, USA), while the patients were in the supine position. A chart record of undesirable events including headache, vomiting, palpitation and abdominal pain was completed © European Society of Human Reproduction and Embryology
Nitroprusside and cervical ripening
2 and 4 h after treatment, and again 2 h after intervention. The uterine evacuation was performed ~6 h after gel administration. After the first series was completed, a further 18 women were enrolled using the same inclusion and exclusion criteria. The second series of women was randomized to receive either placebo gel or 2% nitroprusside-based gel (equivalent to 10 mg). Hysterosuction was done 3 h after gel administration, based on previously observed effects after 6 h and blood pressure was obtained just before, and 60, 120 and 180 min thereafter. Undesirable events were recorded 2 h after treatment and 2 h after intervention. In the surgical theatre, after the paracervical block was obtained with the injection of 6 ml of 2% Carbocaine® (Astra Farmaceutici, Milan, Italy), the force required to dilate the cervix to 10 mm was recorded with the use of a force sensing apparatus (a dynamometer) designed ad hoc by the Department of Clinical Physics and Bio-engineering (Glasgow University, Glasgow, UK) coupled to hegar dilators (Richardson et al., 1989). Cervical measurements and pregnancy termination were performed by one of us (F.P.), who was blinded to the treatment given. Primary outcome was the cervical resistance. The ‘cervical resistance to 8 mm’ was also calculated by adding the peak force required to dilate the cervix from 4 to 8 mm. In a previous study (Thomson et al., 1997) this parameter was reduced by 30% using NO donors with respect to placebo. Supposing intracervical application to be more efficacious than vaginal application as in the Thompson study, a sample of 36 subjects was found to be required for assessing a 30% significant difference of intracervical nitroprusside with respect to placebo. Cervical data were compared in a between-subject design, by using one way analysis of variance (ANOVA). Blood pressure changes after treatment were compared with baseline value in a within-subject design by using multiple ANOVA (MANOVA). In order to make cervical resistance to 8 mm measurement comparable to previous studies, the median value was also reported. Assignment There was only one unit of randomization which was produced from a computer-generated random list where pair and odd number allocated placebo or active treatment. Allocation was done after signing informed consent by one of us (F.F.), who did not perform cervical measurements. This protocol was approved by the Local Ethics Committee. Masking Placebo and active gel were similar in terms of syringes, consistency and volume. Active gel was brown, while placebo gel was white. However, F.P., who performed measurements, was unable to see the gel colour. Gel preparation Both placebo and nitroprusside gels were kindly prepared by Istituto Biochimico Italiano (Milano, Italy) under specific request and responsibility of one of us (F.F.). None of the authors have commercial interest in the present research. Sodium nitroprusside was obtained from Sigma (USA) and Carbopol Ultrez 10 B.F. was supplied from Goodrich (USA). Two gels at 1% and 2% (w/w) were prepared with the active compound. Sterility was obtained by using microfilters and working under sterile conditions. Prefilled syringes containing 0.5 g of nitroprusside or placebo (0.5 ml) were single packed in plastic envelopes and maintained in darkness, at 4°C. The gels were used within a period of 5 months. A gel titration 6 months after preparation indicated the persistence of NO releasing activity of more than 95%.
Results Participant flow and follow-up The flow chart of subjects is reported in Figure 1. Three subjects (one in placebo and two in the nitroprusside groups) dropped out from the study because they refused to wait the time necessary for the protocol. Follow-up was done until discharge from the hospital which was the same day as hysterosuction. Analysis The mean (⫾SD) age of the four groups of women was similar (placebo: 24.9 ⫾ 5.2; nitroprusside 5 mg: 25.8 ⫾ 5.2; placebo: 25.9 ⫾ 5.7; nitroprusside 10 mg: 24.1 ⫾ 3.2). The mean gestational age of the four groups of subjects was also similar (placebo: 9.9 ⫾ 0.9; nitroprusside 5 mg: 10.4 ⫾ 0.8; placebo: 9.2 ⫾ 0.9; nitroprusside 10 mg: 10.0 ⫾ 1.4). The time intervals between gel administration and surgical evacuation were similar within the two matched groups (placebo: 6.2 ⫾ 0.5; nitroprusside 5 mg: 6.3 ⫾ 0.4; placebo: 3.3 ⫾ 0.7; nitroprusside 10 mg: 3.1 ⫾ 0.5). The force required to dilate the cervix using dilators of increasing diameter from 3 to 10 mm is reported in Figure 2. In placebo-treated subjects increasing amounts of force were required according to the increasing dilation of the cervix. Women treated with both doses of nitroprusside gel showed values significantly lower than those treated with placebo gel, almost at any tested diameter. Moreover, no differences were found between subjects treated with the two doses of nitroprusside or between the two groups of subjects receiving placebo. The cervical resistance at 8 mm was lower in women receiving either 5 mg (24.2 ⫾ 17.1, median ⫽ 24.5) or 10 mg (22.1 ⫾ 7.6, median ⫽ 20.5) nitroprusside gel with respect to those receiving placebo gels (60.6 ⫾ 32.6, median ⫽ 61 and 51.6 ⫾ 28.9, median ⫽ 50). The variations of blood pressure (BP) observed after treatments are reported in Table I. According to MANOVA no significant changes were induced by either nitroprusside doses (5 mg: systolic BP: F ⫽ 1.91, degrees of freedom (df) 28/4; diastolic BP: F ⫽ 1.48, df 28/4; 10 mg: systolic BP: F ⫽ 1.44, df 21/3; diastolic BP: F ⫽ 1.34, df 21/3). In both placebo groups, for 5 and 10 mg nitroprusside groups respectively, neither systolic BP (F ⫽ 0.28, df 32/4; F ⫽ 1.18, df 24/3) nor diastolic BP (F ⫽ 1.60, df 32/4; F ⫽ 1.19, df 24/3) values showed significant changes. Undesirable events were reported by a minority of subjects. A mild headache occurred in one subject of the placebo group and a short vomiting in another of the nitroprusside group. None reported palpitations. Two hours after intervention, abdominal pain similar to that of menses appeared in eight women in the two placebo groups and seven in the two nitroprusside groups.
Discussion This study demonstrates that, like other NO donors, sodium nitroprusside applied into the cervical canal induced a significant softening of the cervix in first-trimester pregnant women. 2225
F.Facchinetti, F.Piccinini and A.Volpe
Figure 1. Flow chart of subjects.
Figure 2. Force required to dilate the cervix from 3–10 mm after the intracervical application of placebo (filled bar) or sodium nitroprusside-based gel (open bar). Values are expressed as mean ⫾ SD. *P ⬍ 0.05; **P ⬍ 0.01 between groups.
Table I. Changes (mean ⫾ SD) of blood pressure after either placebo or nitroprusside cervical application
Systolic blood pressure Placebo Nitroprusside 5 mg Placebo Nitroprusside 10 mg Diastolic blood pressure Placebo Nitroprusside 5 mg Placebo Nitroprusside 10 mg
2226
Baseline
60 min
120 min
180 min
240 min
660 min
108.8 ⫾ 7.4 110.5 ⫾ 18.5 115.0 ⫾ 9.0 110.6 ⫾ 9.0
107.2 ⫾ 7.1 103.1 ⫾ 15.1 116.1 ⫾ 6.9 105.6 ⫾ 9.4
107.2 ⫾ 7.5 106.2 ⫾ 17.8 113.3 ⫾ 5.6 105.0 ⫾ 7.0
– – 113.8 ⫾ 6.0 106.8 ⫾ 7.5
108.3 ⫾ 5.6 107.5 ⫾ 8.4 – –
108.3 ⫾ 5.6 109.1 ⫾ 8.4 – –
69.4 ⫾ 8.8 68.7 ⫾ 8.7
70.5 ⫾ 7.6 71.8 ⫾ 7.5
66.1 69.3 73.3 63.7
⫾ ⫾ ⫾ ⫾
8.2 10.1 7.9 5.1
68.8 68.1 71.6 58.1
⫾ ⫾ ⫾ ⫾
6.9 12.8 6.6 9.6
67.2 63.7 70.5 60.0
⫾ ⫾ ⫾ ⫾
4.4 10.2 7.6 8.8
– –
73.3 ⫾ 5.5 63.7 ⫾ 6.9
– –
– –
Nitroprusside and cervical ripening
As a consequence, the force required to dilate the cervix was significantly less compared with placebo-treated subjects. This finding confirms the observations previously done in both animals and humans that NO is a potent agent for cervical ripening. According to Chwalisz and Garfield (1998), NO is the final mediator of the different processes allowing extracellular matrix degradation with the result of softening and effacement of the cervix. Moreover, it has also been demonstrated in humans that cervical ripening is associated with a strong increase of cervical expression of inducible nitric oxide synthase (Tschuggel et al., 1999). Electron microscopy observations of the guinea pig model indicate that sodium nitroprusside induces a disorganization of collagen fibres, with extended intercellular spaces of the stroma, stromal oedema and softening of the connective tissue of the cervix (Chwalisz et al., 1997). Due to organizational limitations it was not possible to obtain cervical specimens in this series of subjects. However, the very low cervical resistance found after both doses of sodium nitroprusside suggests that similar changes should also have occurred in these subjects. In terms of cervical resistance, Thomson et al. (1997, 1998) compared isosorbide mononitrate to Gemeprost after intravaginal applications of tablets and found the NO donor less effective than the prostaglandin analogue. Since the same apparatus was used in the current study to evaluate cervical resistance it is possible to compare the data of these studies; there were also similar clinical features of subjects under investigation. The cumulative median force required to dilate the cervix to 8 mm using 5 mg and 10 mg of sodium nitroprusside (24.5 and 20.5 Newtons respectively) is far less than previously reported using 40 mg and 80 mg of isosorbide mononitrate (40 and 36 Newtons respectively) and similar to that measured after Gemeprost (21 Newtons). Although a formal, randomized study is required before reaching any conclusion, it seems that among the NO donors glyceryl trinitrate, isosorbide mononitrate and sodium nitroprusside, sodium nitroprusside may be the most efficacious, its effect being similar to the prostaglandin analogue. This is not surprising since sodium nitroprusside is the most powerful among NO donors available, lacking any intermediate metabolism in order to release NO (Moncada et al., 1991). Despite the well known vascular activity of sodium nitroprusside only minor and not significant changes were recorded in blood pressure of women receiving the active compound. Another major concern with using NO donors is the onset of severe migraine headaches, as reported in several clinical studies (Thomsen et al., 1994; Pittrof et al., 1996; Lees et al., 1999). Contrary to the previous reports administering NO donors through the vaginal route, in the current study headache was not reported by women treated with sodium nitroprusside. The apparently reduced incidence of such side-effects may be related to the low absorption of the drug into the general circulation. Moreover, intracervical nitroprusside did not interfere with the appearance of menses-like pelvic cramps just after uterine evacuation. However, on the limited amount of evidence reported here, conclusions cannot be drawn on the safety and tolerability of intracervical nitroprusside because of the few subjects studied.
There are several possible applications of such specific chemical ripening exerted by nitroprusside. First- and secondtrimester terminations of pregnancy could be done more effectively since the cervix represents the main obstacle, often being insensitive to prostaglandins. Moreover, in spontaneous miscarriage, either incomplete or missed, chemical ripening could allow uterine evacuation with fewer, if any, use of anaesthetic agents. Finally, in future prospective studies, the induction of parturition at any gestational age, including term, would be easier if one removes the cervical resistance factor, without inducing inappropriate contractions as is the case with prostaglandins. In conclusion, the local application of sodium nitroprusside induces a striking reduction of cervical resistance. Such pharmacological treatment instead of mechanical dilatation of the cervix opens several perspectives but further studies are required to ensure its safety and support the possible exploitation in clinical practice. References Buhimschi, I., Ali, M., Jain, V. et al. (1996) Differential regulation of nitric oxide in the uterus and cervix during pregnancy and labour. Hum. Reprod., 11, 1755–1766. Chwalisz, K. and Garfield, R.E. (1998) New molecular challenges in the induction of cervical ripening. Hum. Reprod., 13, 245–252. Chwalisz, K., Shi, S-Q., Garfield, R.E. et al. (1997) Cervical ripening in guinea pigs after a local application of nitric oxide. Hum. Reprod., 12, 2093–2101. Garfield, R.E., Buhimschi, I., Buhimschi, C. et al. (1997) Regulation of uterine and cervical function by nitric oxide. In Lancaster, J.R. Jr and Parkinson, J.F. (eds), Nitric Oxide, Cytochromes P450, and Sexual Steroid Hormones. Springer, Berlin, pp. 141–180. Izumi, H., Yallampalli, C. and Garfield, R.E. (1993) Gestational changes in L-arginine-induced relaxation of pregnant rat and human myometrial smooth muscle. Am. J. Obstet. Gynecol., 169, 1327–1337. Ledingham, M.A., Denison, F.C., Kelly, R.W. et al. (1999) Nitric oxide donors stimulate prostaglandin F2α and inhibit thromboxane B2 production in the human cervix during first trimester of pregnancy. Mol. Hum. Reprod., 5, 973–982. Lees, C.C., Lojacono, A., Thompson, C. et al. (1999) Glyceryl trinitrate and ritodrine in tocolysis: an international multicenter randomized study. Obstet. Gynaecol., 94, 403–408. Moncada, S., Palmer, R.M.J. and Higgs, E.A. (1991) Nitric oxide: physiology, pathophysiology and pharmacology. Pharmacol. Rev., 43, 109–120. Neri, I., Di Renzo, G.C., Caserta, G. et al. (1995) Impact of the L-Arginine/ nitric oxide system in pregnancy. Obstet. Gynecol. Surv., 50, 851–858. Pittrof, C., Lees, C., Thompson, C. et al. (1996) Crossover study of glyceryl trinitrate patches for controlling pain in women with severe dysmenorrhoea. Brit. Med. J., 312, 884. Richardson, W., Smith, D.C., Evans, A.L. et al. (1989) A novel cervical dilatation force measurement instrument. J. Med. Eng. Technol., 13, 220– 221. Rosselli, M., Keller, P.J. and Dubey, R. (1998) Role of nitric oxide in biology, physiology and pathophysiology of reproduction. Hum. Reprod. Update, 4, 3–24. Thomsen, L.L., Kruuse, C., Iversen, H.K. et al. (1994) A nitric oxide donor (nitroglycerin) triggers genuine migraine attacks. Eur. J. Neurol., 1, 73–80. Thomson, A.J., Lunan, C.B., Lendingham, M. et al. (1998) Randomised trial of nitric oxide donor versus prostaglandin for cervical ripening before firsttrimester termination of pregnancy. Lancet, 352, 1093–1096. Thomson, A.J., Lunan, C.L., Cameron, A.D. et al. (1999) Nitric oxide donors induce ripening of the human cervix: a randomised controlled trial. Brit. J. Obstet. Gynaecol., 104, 1054–1057. Tschuggel, W., Schneeberger, C., Lass, H. et al. (1999) Human cervical ripening is associated with an increase in cervical inducible nitric oxide synthase expression. Biol. Reprod., 60, 1367–1372. Received on March 28, 2000; accepted on June 30, 2000
2227
