Double-blind randomised controlled trial of glyceryl trinitrate spray for external cephalic version


  • Heba Yanny,

    Consultant Anaesthetist
    1. Walsall Manor Hospital, Walsall
    Search for more papers by this author
  • Richard Johanson,

    Senior Lecturer (Perinatology), Corresponding author
    1. Academic Department of Obstetrics and Gynaecology, North Staffordshire Hospital, Stoke-on-Trent
      Correspondence: Dr R. Johanson, Academic Department of Obstetrics and Gynaecology, North Staffordshire Hospital, London Road, Stoke-on-Trent ST4 6QG, UK.
    Search for more papers by this author
  • K. Joanne Baldwin,

    Research Registrar
    1. Academic Department of Obstetrics and Gynaecology, North Staffordshire Hospital, Stoke-on-Trent
    Search for more papers by this author
  • Linda Lucking,

    Clinical Trials Co-ordinator
    1. Academic Department of Obstetrics and Gynaecology, North Staffordshire Hospital, Stoke-on-Trent
    Search for more papers by this author
  • Ray Fitzpatrick,

    Senior Lecturer (Medicines Management)
    1. Academic Department of Obstetrics and Gynaecology, North Staffordshire Hospital, Stoke-on-Trent
    Search for more papers by this author
  • Peter Jones

    Professor (Statistics)
    1. Academic Department of Obstetrics and Gynaecology, North Staffordshire Hospital, Stoke-on-Trent
    Search for more papers by this author

Correspondence: Dr R. Johanson, Academic Department of Obstetrics and Gynaecology, North Staffordshire Hospital, London Road, Stoke-on-Trent ST4 6QG, UK.


A placebo controlled trial of glyceryl trinitrate was carried out in women whose first attempt at external cephalic version without tocolysis had been unsuccessful. No significant differences between groups were found and no side effects were noted. It is recommended that a larger, possibly multicentre study, should be undertaken.


A large proportion of women with breech presentations at term currently undergo elective caesarean delivery. External cephalic version has been advocated as an alternative to breech delivery and has been subjected to rigorous scientific appraisal in six randomised controlled trials1–6. There is a significant reduction in the risk of caesarean section in the group of women where there is an intention to undertake external cephalic version (OR 0.4, 95% CI 0.3 to 0.6) without any increased risk to the baby7. This procedure can be facilitated by the use of tocolytic agents to provide uterine relaxation and increase the success rate. Tocolysis has been shown to be effective both when used routinely and when used selectively8.

The standard tocolytics are the intravenous selective beta-2 adrenoreceptor agonists, (e.g. ritodrine hydrochloride, salbutamol and terbutaline), but these are known to have adverse side effects, including tachycardia, palpitations, chest pain, chest tightening, anxiety, nausea, vomiting and hyperglycaemia.

Glyceryl trinitrate has recently come to the attention of obstetricians as a potent tocolytic agent. Glyceryl trinitrate, a donor for nitric oxide, is a potent endogenous smooth muscle relaxant in vasculature, gut, genitourinary tract and the uterus. Nitric oxide acts by increasing the intracellular production of cyclic guano-sine monophosphate, which acts as a mediator for dephosphorylation of the myosin light chain. Recent reports have suggested that administration of glyceryl trinitrate may be useful in relaxing uterine smooth muscle, possibly aiding fetal delivery. For example, it has been used successfully for urgent tocolysis in cases of breech extraction of the second twin9. It has also been used intravenously to provide tocolysis for intrapartum external cephalic version10. A review of the use of this agent in a variety of circumstances found no adverse reports with respect to safety for the fetus or newborn infant11. Given that the pharmacodynamics of sublingual glyceryl trinitrate show a rapid onset of action12, and that it has been used in the acute cardiac setting in place of the IV formulations, we proposed a placebo controlled trial of this agent with successful external cephalic version as the main end-point.


All women who choose external cephalic version attend a routine session run twice a week on the labour ward. On admission a baseline cardiotocography (CTG) is carried out and there is an ultrasound examination to demonstrate fetal position, adequate amniotic fluid, a flexed fetal head, a free loop of cord, and the position of fetal legs. Routine management thereafter includes a single gentle attempt at external cephalic version under ultrasound control without tocolysis. Within the population from whom the study was drawn, the initial external cephalic version success rate without tocolysis was 18% (20/114).

Where the first attempt at external cephalic version was unsuccessful, women were eligible for inclusion in the study. The study had local ethical committee approval and all women gave written informed consent. Women who agreed to participate were randomised (using opaque sealed envelopes, prepared by L.L. from a list of computer generated random numbers) to receive either Spray A or Spray B. One group therefore received no tocolysis (placebo spray) and the other glyceryl trinitrate sublingual spray 800 μg (2 sprays). A record was made of any side effects noted by the mother and of discomfort experienced. Changes in blood pressure or pulse were recorded. A second attempt at external cephalic version was then attempted. The fetal heart rate was monitored during the procedure on ultrasound and afterwards with a repeat CTG. Where this attempt was unsuccessful and uterine relaxation was not considered adequate, a slow intravenous infusion of salbutamol was administered and then a final attempt at external cephalic version was undertaken. The principal end points of the study were success rate (overall currently 43% in North Staffordshire Hospital) and the presence of side effects (e.g. headache, dizziness or flushing)

With 30 patients in each arm of the study, it was calculated that the power should be 80%, with a one-tailed significance level of P= 0.05, in terms of detecting an increase in success rate of 30% from a baseline of approximately 20%, or a difference in incidence of side effects of 30%. Normally distributed continuous data are expressed as mean (SD). Non-normally distributed data are presented as medians and ranges and evaluated with the Mann-Whitney U test. Frequency data are analysed using the StatXact Turbo package (CYTEL, Cambridge, Massachusetts, USA) and are presented as odds ratios (95% confidence interval).


All the women who had a failed first attempt at external cephalic version and who were invited to participate in the study agreed to be included. Fifty-seven women were included in the study: 31 received the active agent, glyceryl trinitrate spray and 26 the placebo. We found that women were very ready to be recruited into this study and were therefore able to achieve a larger number of participants than we had expected. The results are shown in Table 1 and the baseline characteristics of the two groups were similar. The success rate was greater in the glyceryl trinitrate group [9/31 (29%)] than in the placebo group [3/26 (12%)], but the difference was not statistically significant [2 (1 d.f.) = 2.6 (P= 0.19)]. The odds ratio (95% CI) for the difference is 3.08 (0.65-19.98). No side effects were noted in either group. No women complained of dizziness, flushing or headache, and there were no episodes of faintness, hypotension or tachycardia. In addition, there were no differences recorded between the two groups in terms of patient discomfort or in terms of obstetricians' comments about uterine relaxation. The proportions of women in each group requiring additional salbutamol were not significantly different (14/26 (54%) in the active group and 13/31 (42%) in the placebo group). The median dose of salbutamol used within each group was the same: 100 μg (range 50–100 μg) in the active group, and 100 μg (range 50–250 μg) in the placebo group.

Table 1.  Table of results. Values are given as n (%), mean [SD] or median [range]. ECV = external cephalic version.
Baseline characteristics  
Maternal age (years)(n= 26)(n= 31)
 28.1 [5]28 [4.1]
Parity(n= 26)(n= 31)
  013 (50)15 (48)
  111 (42)9(29)
Extended legs(n= 24)(n= 31)
  Yes12 (50)14 (45)
  No10 (42)16 (55)
  1 flexdl1 extended2(8)1(3)
Liquor volume (cm)(n= 24)(n= 29)
 4.69 [0.79]4.62 [0.95]
ECV successful(n= 26)(n= 31)
Patient discomfort(n= 21)(n= 30)
  No discomfort2(9)3(10)
Uncomfortable12(57)18 (60)
Obstetrician's comments on uterine relaxation(n= 25)(n= 30)
  Reasonable8(32)11 (37)
No. requiring salbutamol(n= 26)(n= 31)
 14 (54)13 (42)
Dose of salbutamol (μg)100 {50–250}100 {50–100}


This study is the first placebo-controlled evaluation of glyceryl trinitrate in its use as a tocolytic for external cephalic version. Although we enrolled more patients into the study than we had originally expected, the differences seen were not statistically significant, but we consider the difference to be clinically worthwhile. Were the proportions of success to be a reflection of the true effect of glyceryl trinitrate, the study would have needed 61 patients in each half to show significance. It was reassuring that no complications of treatment were found and that women did not report side effects.

It is important that further studies are now undertaken to further evaluate this potentially effective and safe drug. A larger study of its use in external cephalic version needs to be undertaken and further work should be undertaken in other areas of obstetrics where tocolysis is of value (e.g. retained placenta, inverted uterus and acute uterine hypertonia). Reconsideration of the mode of administration may be useful. Alternative glyceryl trinitrate formulations may be faster-acting (e.g. buccal glyceryl trinitrate) and may mimic intravenous infusion.

Current evidence would continue to support the use of tocolytics for external cephalic version. It may be that alternative dosages and methods of administration for standard tocolytics may render them agents of first choice. A recent study13 using the subcutaneous route of administration found a low incidence of side effects. If glyceryl trinitrate were to be shown to be an effective tocolytic agent it would need to be compared with the current standard agents.


E Merck provided the trial drug and placebo. We would like to thank the patients who participated, and the midwifery and medical colleagues who assisted with external cephalic versions. Mrs G. Masson, Mr R. Burr and Mr P. Young contributed patients to the study. L.L. is supported by a grant from the NHS (E) West Midlands Research and Development Programme.