Sublingual nitroglycerin versus placebo as a tocolytic for external cephalic version: A randomized controlled trial in parous women Emmanuel Bujold, MD,a,b Marc Boucher, MD,a Denyse Rinfret, RN,a Susan Berman, MD,b Ema Ferreira, MSc, Pharm D,c and Ge´rald P. Marquette, MDa,d Montre´al, Que´bec, Canada, and Detroit, Mich OBJECTIVE: The purpose of this study was to evaluate the efficacy of sublingual nitroglycerin as a tocolytic agent for external cephalic version in parous women. STUDY DESIGN: A double-blinded randomized controlled trial was performed. Patients with parity of $1 at 36 to 40 weeks of gestation who were eligible for external cephalic version were included. Patients were randomized to receive either two sublingual sprays of 400 lg of nitroglycerin or two sprays of placebo 3 minutes before the trial of external cephalic version. Rates of successful external cephalic version and side effects were compared between groups. RESULTS: Of 99 patients in the study, 50 patients received sublingual nitroglycerin, and 49 patients received placebo. There were no differences in maternal age, gestational age, estimated fetal weight, amniotic fluid index, and placental location between the two groups. The success rate of external cephalic version was 48% in the nitroglycerin group compared with 63% in the placebo group (P = .13). There was a higher incidence of headaches in the nitroglycerin group (42% vs 4%, P < .001). CONCLUSION: Sublingual nitroglycerin was associated with a higher rate of headache and did not improve the rate of successful external cephalic version. (Am J Obstet Gynecol 2003;189:1070-3.)
Key words: Nitroglycerin, tocolysis, breech, external cephalic version
External cephalic version (ECV) is a safe and effective method to decrease the rate of cesarean deliveries that are performed for breech presentation at term.1,2 The average rate of successful ECV ranges from 25% to 80%, with the highest success rates in parous women.1,2 Tocolytics, such as b-mimetics (terbutaline, ritodrine), have been shown to improve the success rate of ECV.1,3 In the largest published randomized controlled trial, Marquette et al4 demonstrated that the use of intravenous ritodrine, compared with placebo, was associated with a higher rate of successful ECV (43% vs 25%, P = .03) in nulliparous women. This difference was not statistically significant in parous women (66% vs 58%, P = .39).4 Because of its ease of administration and rapid onset of action (2-4 minutes),5 the use of sublingual nitroglycerin, From the Department of Obstetrics and Gynecology, Sainte-Justine Hospital and the University of Montreal,a the Division of MaternalFetal Medicine, Department of Obstetrics and Gynecology, Wayne State University,b the Department of Pharmacy Sainte-Justine Hospital and University of Montreal,c and the Department of Obstetrics and Gynecology, Children’s and Women’s Health Centre of British Columbia.d Presented at the Twenty-Third Annual Meeting of the Society for Maternal-Fetal Medicine, San Francisco, Calif, February 3-8, 2003. Reprint requests: Emmanuel Bujold, MD, Department of Obstetrics/ Gynecology, Hutzel Hospital, 4707 St Antoine Blvd, Detroit, MI 48201. E-mail:
[email protected] Ó 2003, Mosby, Inc. All rights reserved. 0002-9378/2003 $30.00 + 0 doi:10.1067/S0002-9378(03)00850-0
1070
a donor of nitric oxide, has been proposed as a tocolytic agent in ECV. In a previous study when we compared the efficacy of sublingual nitroglycerin with intravenous ritodrine as tocolytics in nulliparous women, we found a higher success rate in the ritodrine group; however, this did not reach statistical significance (25% vs 45%, P = .08).6 Yanny et al7 compared the use of sublingual nitroglycerin versus placebo before a second attempt of ECV in patients who had a first failed attempt of ECV. They found no difference in the success rate of ECV between patients who received sublingual nitroglycerin and patients who received placebo. Of note, they did not stratify their data by parity. Finally, a recent retrospective study, which included >200 parous patients, demonstrated a potential benefit of sublingual nitroglycerin as a tocolytic for ECV.8 Given these conflicting results, the goal of this study was to compare prospectively the efficacy of sublingual nitroglycerin to placebo in parous patients who undergo a trial of ECV. Method A prospective, randomized, double-blinded study was performed at the antenatal assessment unit of SainteJustine Hospital between April 1999 and August 2002. The research ethics committee of our institution approved the study protocol. To be eligible to participate in this study,
Bujold et al 1071
Volume 189, Number 4 Am J Obstet Gynecol
patients with parity of $1 had to be between 36 and 40 weeks of gestation with a singleton pregnancy and a fetus in breech presentation. Exclusion criteria included intrauterine growth restriction (defined as an estimated fetal weight [determined by ultrasound examination] < 10th percentile for gestational age), oligohydramnios (defined as an amniotic fluid index of #5 cm), the presence of a placenta previa or an abruptio placentae, a previous uterine scar other than a low transverse cesarean delivery, active labor, rupture of membranes, fetal anomalies incompatible with life, a nonmobile breech by abdominal palpation, any contraindication to vaginal delivery, or a medical/allergic contraindication to nitroglycerin. Before randomization, eligible patients underwent a nonstress test (NST) and ultrasonographic evaluation for estimated fetal weight, fetal morphologic features, amniotic fluid index, and placental location. After a reactive NST, and if the patients met study criteria, one of the maternal-fetal medicine faculty (G. P. M., M. B., R. J. G.) verified breech mobility by abdominal palpation. When all these criteria were met, the informed consent form was reviewed with the patient. On consent, patients were assigned by a computerized randomization table to receive either sublingual nitroglycerin (nitroglycerin group) or sublingual placebo (placebo group) by a block of 6. Sublingual nitroglycerin and placebo were supplied by the hospital pharmacy in identical 75-mL pumps that delivered a metered dose of 400 lg of nitroglycerin per spray (Nitrolingual pump spray; Rho ^ ne-Poulenc Rorer, Montre´al, Que´bec, Canada). The sublingual placebo was provided to the hospital pharmacy by the same company and had the same minty taste as the active drug. Moreover, the patients, nurses, and attending physicians were blinded to the contents of the sublingual spray. After consent was obtained, randomization was achieved, and vital signs were recorded, a research nurse administered two sublingual sprays. Three minutes after the administration of the sublingual nitroglycerin or placebo, one of the authors (G. P. M., M. B., R. J. G.) completed a maximum of four attempts at ECV under ultrasound guidance by either forward or back flip. An NST was performed after the maneuver, before the patient was discharged. Blood pressure was taken manually with a mercury sphygmomanometer on admission and 10 minutes after the administration of the study medication. The following data were collected: maternal age, parity, type of breech, placental location, position of fetal spine, estimated fetal weight, amniotic fluid index, duration of tocolysis, NST results, gestational age at the time of procedure and at delivery, maternal weight, maternal blood pressure, maternal and fetal obstetric complications, and mode of delivery. Symptomatic hypotension was defined as an episode of systolic blood pressure < 80 mm Hg that was associated with symptoms that required medical intervention. Prolonged fetal heart rate deceleration was
defined as a fetal heart rate < 90 beats/min for >90 seconds. A follow-up survey was conducted by telephone. All patients who were not delivered in our center were contacted approximately 6 weeks after the trial of ECV to obtain the date of delivery, mode of delivery, and birth weight. We compared the demographics, the rate of successful ECV (defined as a cephalic presentation of the fetus at the end of the procedure), the rate of vertex presentation at the time of delivery, and the rate of vertex vaginal delivery between the two groups. Statistical analysis was performed by v2 test, Mann Whitney U test, Student t test (independent and paired), and Fisher exact test when appropriate. A probability value of < .05 was considered statistically significant. A power analysis was performed on the basis of an a error of .05 (two-tailed), a b error of .2, and a baseline ECV success rate of 55%. Using these parameters, we estimated that to detect a 20% difference in ECV success rate, it would be necessary to randomly place 196 patients. Halfway through the trial, an interim analysis was performed by the data safety monitoring board. This board decided to stop the trial because of a statistically significant (P < .01) higher rate of side effects and a trend toward a lower rate of successful ECV in one group. This decision was based on the likelihood ( 95%) that the subsequently randomized patients would be exposed to an increased risk of adverse outcomes without potential benefit if the trial was completed. The investigators were informed, and the trial was stopped. Results Ninety-nine patients with breech presentation met all inclusion criteria. Fifty patients were randomly assigned to the nitroglycerin group, and 49 patients were randomly assigned to the placebo group. The demographic profile for both groups is depicted in Table I. There was no statistically significant difference between the two groups with regard to maternal age, gestational age, estimated fetal weight, amniotic fluid index, placental location, and type of breech. The rate of successful ECV was higher in the placebo group, but this difference did not reach statistical significance (P = .13, Table II). The median number of ECV attempts was comparable between the two groups. There was no difference in the rate of successful ECV between operators (57% vs 58% vs 50%, P = .79). The mean arterial blood pressure, maternal heart rate, and the frequency of side effects that reported by the patients are displayed in Table III. The mean blood pressure was statistically lower in the nitroglycerin group 10 minutes after administration of the sublingual spray when compared with the placebo group (Table III). There was no statistically significant difference in the maternal
1072 Bujold et al
October 2003 Am J Obstet Gynecol
Table I. Patient profile according to study groups Demographic Maternal age (y)* Gestational age at ECV (wk)* Maternal weight (kg)* Estimated fetal weight (g)* Amniotic fluid index (cm)* Frank breech Anterior placenta
Nitroglycerin (n = 50)
Placebo (n = 49)
P value
31.5 (21, 41) 37.6 (36.0, 40.0)
31.7 (21, 44) 37.4 (36.0, 40.7)
.65 .50
75 (58, 102)
73 (46, 102)
.28
3063 (2218, 4305) 2979 (2010, 4031) .44 11.2 (5.6, 29.3) 25 (50%) 21 (42%)
13.8 (5.0, 25.0) 16 (33%) 21 (43%)
.08 .08 .93
*Data are given as median (minimum, maximum).
Table II. Results of ECV attempts
ECV attempts* Successful ECV
Nitroglycerin (n = 50)
Placebo (n = 49)
P value
3 (1, 4) 24 (48%)
3 (1, 4) 31 (63%)
.65 .13
*Data are given as median (minimum, maximum).
heart rate before and after the administration of nitroglycerin. The frequencies of maternal tachycardia and sensation of palpitation that were reported by the patients were not significantly different between groups. However, the rate of symptomatic hypotension in the nitroglycerin group was 12% compared with 2% in the placebo group (P = .05). Of note, headaches after the procedure were reported more frequently by patients in the nitroglycerin group (P < .001). No patient required an emergency cesarean delivery after the procedure. Birth weight and mode of delivery are shown in Table IV. No statistically significant differences were found in terms of the rate of vertex presentation at delivery or the rate of vertex vaginal delivery. Comment When compared with placebo, sublingual nitroglycerin spray (as a tocolytic agent) did not improve the success rate of ECV. Although not statistically significant, the lower ECV success rate in the nitroglycerin group could be explained partially by a lower amniotic fluid index and a higher rate of frank breech in this group. Vaginal examinations were not performed before ECV. Three previous comparative studies evaluated the effect of sublingual nitroglycerin on the success rate of ECV. The first trial, which was prospective, compared nitroglycerin versus placebo as a tocolytic second attempt of ECV in 57 women who had shown no significant benefit of nitroglycerin (12% vs 29%, P = .19). The second study, which was a retrospective study, compared nitroglycerin versus no tocolytic before ECV in 287 women (including 126 parous patients) and showed no significant difference
Table III. Vital signs and side effects during and after ECV Nitroglycerin (n = 50) Mean arterial blood pressure (mm Hg) Before sublingual (T0)* 10 min after sublingual (T10)* Significance (T0 vs T10)y Maternal heart rate (beats/min) Before sublingual (T0)* 10 min after sublingual (T10)* Significance (T0 vs T10)y Hypotension episode Prolonged fetal heart rate deceleration Headaches Palpitations
Placebo (n = 49)
85.2 ± 8.9 85.2 82.6 ± 10.0 88.4 0.13 85.3 87.3
± 11.8 ± 13.4
P value
± 12.7 ± 10.9
.97 .01
0.04 85.6 83.3
± 11.8 ± 10.7
0.26 6 (12%) 2 (4%)
0.14 1 (2%) 5 (10%)
21 (42%) 0
2 (4%) 0
.93 .11 .05 .21 < .001
*Mean ± SD, studied by Student independent t test (2-tailed). yStudent paired t test (2-tailed).
Table IV. Pregnancy outcomes after attempted EVC Nitroglycerin (n = 50)
Placebo (n = 49)
Gestational age at 39.3 (37.7, 41.1) 39.4 (36.3, 41.7) delivery (wk)* Birth weight (g)* 3395 (2215, 4224) 3402 (2300, 4700) Vertex 24 (48%) 32 (65%) presentation at delivery Vertex vaginal 19 (38%) 24 (49%) delivery
P value .81 .78 .08 .27
*Data are given as median (minimum, maximum).
between the two groups (nulliparous, 42% vs 39%; P = .21; parous, 67% vs 81%; P = .09). Finally, we recently reported a randomized controlled trial that compared the efficacy of intravenous ritodrine versus sublingual nitroglycerin as a tocolytic before ECV in 74 nulliparous women.6 Although the difference was not statistically significant, we found a higher rate of successful ECV in the ritodrine group (45% vs 25%, P = .075). Even though the two first reports suggest a potential positive effect of nitroglycerin before ECV, our previous trial and this current study, which were both double-blinded randomized trials, do not support the use of nitroglycerin as a tocolytic before ECV. The use of sublingual nitroglycerin was associated with a higher frequency of headaches (P < .001) and a higher rate of symptomatic hypotension (P = .05). This latter observation is consistent with previous studies on the use of nitroglycerin during pregnancy.9,10 Finally, we found no improvement in the rate of vertex presentation at the time of delivery or the rate of vertex vaginal delivery.
Bujold et al 1073
Volume 189, Number 4 Am J Obstet Gynecol
Our results are consistent with previous randomized controlled trials that focused on the clinical effect of nitroglycerin on the human myometrium when acute tocolysis is needed.11 David et al12 recorded the subjective level of uterine relaxation for fetal breech extractions during cesarean delivery. In that study, there was no difference in the degree of uterine tone that was evaluated by the surgeon between patients who received intravenous nitroglycerin and patients who received intravenous placebo. Those results are in agreement with the results reported by Ouellet et al13 who found no significant difference in uterine relaxation that was measured by an intrauterine catheter after the administration of 0.4 mg of sublingual nitroglycerin to patients who were in labor. We found no study that compared the tocolytic effect of sublingual versus intravenous nitroglycerin on the uterine muscle. However, in a recent review of the literature by O’Grady et al,14 the suggestion was made that sublingual or intravenous nitroglycerin has been associated with positive outcomes in situations in which acute tocolysis was needed. After careful evaluation of these observational reports, we believe that the benefit of sublingual nitroglycerin was probably the result of its effect on the uterine cervix. Internal podalic version of a second twin, retained placenta, and uterine inversion represent most of the clinical situations in which sublingual or intravenous nitroglycerin is thought to be useful.14,15 We hypothesize that the potential benefit of nitroglycerin is to increase the compliance of the cervix and the lower uterine segment and thus to facilitate intrauterine maneuvers. Further evidence to support this hypothesis is proposed by Ekerhovd et al,16 who demonstrated a significant relaxation of the cervical tissue after nitroglycerin, with human cervical biopsy specimens obtained after delivery. These studies suggest that the benefit of nitroglycerin as a tocolytic agent is questionable. Nitroglycerin may be beneficial only when uterine cervical relaxation is required. Because of the higher rate of side effects and a lower rate of successful ECV with the use of sublingual nitroglycerin, we do not recommend the use of sublingual nitroglycerin as a tocolytic agent in ECV in parous patients. We believe that prospective trials should be per-
formed before nitroglycerin is used in clinical situations that require acute uterine relaxation. We thank Dr Tinnakorn Chaiworapongsa for his assistance in the statistical analysis and data interpretation.
REFERENCES 1. American College of Obstetricians and Gynecologists. External cephalic version. Washington (DC): The College; 2000. Practice bulletin No.: 13. 2. Hofmeyr GJ, Kulier R. External cephalic version for breech presentation at term (Cochrane Review). In: The Cochrane Library, Issue 3, 2002. Oxford: Update Software. 3. Hofmeyr GJ. External cephalic version facilitation for breech presentation at term (Cochrane Review). In: The Cochrane Library, Issue 3, 2002. Oxford: Update Software. 4. Marquette GP, Boucher M, The´riault D, Rinfret D. Does the use of a tocolytic affect the success rate of external cephalic version? Am J Obstet Gynecol 1996;175:859-61. 5. Abrams J. Pharmacology of nitroglycerin and long-acting nitrates. Am J Cardiol 1985;56:12-8A. 6. Bujold E, Marquette GP, Fereirra E, Gauthier RJ, Boucher M. Sublingual nitroglycerin versus intravenous ritodrine as tocolytic for external cephalic version: a double-blinded randomized trial. Am J Obstet Gynecol 2003;188:1454-7. 7. Yanny H, Johanson R, Baldwin KJ, Lucking L, Fitzpatrick R, Jones P. Double-blind randomised controlled trial of glyceryl trinitrate spray for external cephalic version. BJOG 2000;107:562-4. 8. Bujold E, Sergerie M, Masse A, Verschelden G, Bedard MJ, Dube J. Sublingual nitroglycerine as a tocolytic in external cephalic version: a comparative study. J Obstet Gynaecol Can 2003;25:203-7. 9. Reddick LF, Livingston E, Bell E. Sublingual aerosol nitroglycerin for uterine relaxation in attempted external version. [letter]. Am J Obstet Gynecol 1997;176:496-7. 10. Lees C, Campbell S, Jauniaux E, Brown R, Ramsay B, Gibb D, et al. Arrest of preterm labour and prolongation of gestation with glyceryl trinitrate, a nitric oxide donor. Lancet 1994;343:1325-6. 11. Morgan PJ, Kung R, Tarshis J. Nitroglycerin as a uterine relaxant: a systematic review. J Obstet Gynaecol Can 2002;24:403-9. 12. David M, Halle H, Lichtenegger W, Sinha P, Zimmermann T. Nitroglycerin to facilitate fetal extraction during cesarean delivery. Obstet Gynecol 1998;9:119-24. 13. Ouellet A, Blouin D, Lafranchise L, Moutquin JM, Te´treault JP. Effets de 0, 4 mg de nitroglyce´rine sublinguale sur le rela^chement ute´rin de parturientes a` terme. J Soc Obstet Gynaecol Can 2001;23:21-6. 14. O’Grady JP, Parker RK, Patel SS. Nitroglycerin for rapid tocolysis: development of a protocol and a literature review. J Perinatol 2000;1:27-33. 15. Smith GN, Brien JF. Use of nitroglycerin for uterine relaxation. Obstet Gynecol Surv 1998;53:559-65. 16. Ekerhovd E, Bra¨nnstro¨m M, Weijdega˚rd B, Norstro¨m A. Nitric oxide synthases in the human cervix at term pregnancy and effects of nitric oxide on cervical smooth muscle contractility. Am J Obstet Gynecol 2000;183:610-6.
