Congress Proceedings: COGI

DISCUSSION 

Steer: Prof Cabrol, do you believe that atosiban will be widely used in France? 

Cabrol: Yes, I do. When preterm labour is diagnosed, any given treatment is likely to be unnecessary in one-half of women. Therefore, until we are able to identify accurately which woman will benefit from treatment it is mandatory to use the tocolytic agent with the least number of side effects. In this respect, atosiban could be such a compound because of its impressive safety profile. 

Steer: Prof Marsal, what is your opinion on the long-term outcome for infants following tocolytic administration? 

Marsal: It is most important to use a tocolytic agent that is safe for both the mother and fetus. Reports indicate that the prolongation of pregnancy can have a beneficial effect on long- term outcome since the extra time gained allows for the administration of corticosteroids. This is probably the most effective treatment for the fetus as corticosteroids decrease the incidence of respiratory distress syndrome and paraventricular haemorrhage. Long-term studies are needed on the effects of both corticosteroids and tocolytic agents. At a recent meeting of the Swedish Society of Clinical Medicine it was shown that children receiving corticosteroid treatment in utero had a lower rate of cerebral palsy, which, to my knowledge, has not been previously demonstrated. Therefore, the fetus can benefit from corticosteroids in these circumstances. 

Audience: Is there a difference in postpartum haemorrhage or maternal blood loss between those patients receiving atosiban or ß-agonists?

Moutquin: The multi-national trial comparing atosiban and ß-agonists reported no difference in the rate of postpartum haemorrhage or blood loss between treatments. 

Steer: Dr Goodwin, do you believe that this is due to the effects of atosiban having already worn off when the women deliver? 

Goodwin: Yes. There was a placental transfer study carried out in the USA in which eight women received intravenous atosiban prior to caesarean section and there was no increase in haemorrhage. The concentration of atosiban in the maternal serum was twelve times that in the cord blood but there was obviously atosiban at therapeutic levels at the time of delivery. Although only a small number of women were studied, this study also suggests that postpartum haemorrhage is not increased with atosiban. 

Audience: Is the maternal hyperglycaemia reported in the trials related to the corticosteroids given for fetal lung maturity or is it a confounding variable? 

Lamont: There is an increased incidence of hyperglycaemia in the ß-agonist group compared to atosiban but it was not statistically significantly different. 

Cabrol: Hyperglycaemia is a very common complication of preterm birth. Therefore, I do not believe that there is a relationship between the tocolytic agent and hyperglycaemia in the preterm baby. 

Moutquin: In Canada we experienced many cases of hyperglycaemia due to the administration of ritodrine. Ritodrine induces maternal hyperglycaemia and in some centres in Canada insulin was given to decrease this complication. Another complication with ritodrine was initial hyperglycaemia in the baby, which reverted to hypoglycaemia as a result of placental transfer of glucose. This rebound hypoglycaemia was not reported with atosiban. 

Audience: Dr Goodwin, it appears that the effect of intravenous infusion of atosiban was more marked in the earlier gestational age groups. Is there a gestational age at which you would not recommend subcutaneous infusion of atosiban? 

Goodwin: I consider the apparent benefit at lower gestational ages to be largely due to the fact that the later you start therapy, the less possible benefit you can show because you are approaching 36 weeks which represents the end of the preterm period. In the PTL-O96 study, all women were included in the analysis -some women started treatment at 34 weeks and so they could not remain pregnant for very long, while other women started treatment at 26 weeks. I would make the decision on whether to use a subcutaneous infusion independent of gestational age. 

Audience: Specifically referring to the ß-agonist atosiban trial, you permitted the local titration of ß-agonist within the protocol, which is influenced by cardiovascular side effects. Is this really a blind study and is a dummy worthwhile? 

Moutquin: During the study in Canada we performed a blinded trial in an attempt to identify which women were receiving atosiban or ritodrine. As reported in the trials, 10-12% of women on atosiban experienced tachycardia so we were unable to know categorically which treatment had been given to which woman. In fact, when the trial was unblinded, the results showed that we were unable to identify the tocolytic administered. 

Audience: Most studies of ritodrine have looked at dose response based on side effects and uterine contractions. In view of the absence of side effects with atosiban, can yon keep increasing the dose and get a better response? 

Goodwin: There are no data on any infusion rate in humans higher than 300 µg/min. This gives a concentration of 450 ng/ml which, in theory, should be enough to saturate every oxytocin receptor in the uterus. 

Lamont: In addition, when ß-agonists are administered, you start at a low concentration and increase the dose if the contractions remain whereas the opposite applies with atosiban, where you start with a bolus dose and then a 300 µg/min infusion followed by 1µg/min, so you actually give progressively less atosiban with time for the same effectiveness. 

Audience: We know that preterm labour is the major complication of multiple pregnancy. Can atosiban be given in multiple pregnancy and is the dose the same? 

Cabrol: We know from the studies that it works as well in multiple and singleton pregnancies and the dosage is exactly the same. 

Audience: Atosiban is very effective for the prolongation of pregnancy and has few maternal side effects. However, the data shows that neonatal outcome is the same in the placebo group and atosiban group. What could be the explanation for this?

Goodwin: The prolongation of pregnancy is very discreet. What has been emphasised today is that atosiban is equivalent to other tocolytic agents in terms of effectiveness. No tocolytic agent, not even the ß-agonists, has ever been shown to improve outcome. The FDA in the USA requested that a study was conducted that investigated neonatal outcome, with the aim of achieving larger babies with less morbidity. The number of women required for such a study would dwarf the capacity of any combination of institutions. Even so, the data presented today represent more than all of the data on tocolytic agents of any kind -ever. Most of our current tocolytic therapies are based on very scant evidence and there is almost no evidence of actual improvements in outcome. So, what we can show from these data is that pregnancy is prolonged, atosiban is safe, and presumably if we could make maximal use of that time period with adjuvant therapies we will eventually get improvements in infant outcomes. 

Lamont: Every study that has ever been conducted on tocolytic agents has been underpowered to show a difference in perinatal mortality and morbidity. If you are going to show improvement in outcome you need to be using the agent before 28 weeks gestation. We know that the survival rate of babies improves by between 1% and 3% for every day gained between 23 and 26 weeks gestation. In the worldwide atosiban study, 140 (20% ) babies were born below 28 weeks. Therefore, you would need thousands of babies to demonstrate any benefit in terms of outcome in perinatal mortality and morbidity. 

Steer: We know that corticosteroids and tocolytic agents work in the short term. Therefore, we have to combine these facts and assume that there is a benefit from using a tocolytic agent and if you are going to use one it needs to be the most acceptable to the woman. 

Marsal: As a clinician, if I have a tocolytic agent which I am going to use in a defined group of women it should be as safe as possible and effective. Atosiban is at least as effective as ß-agonists and with regard to maternal safety, there is a clear difference -the safety profile is very advantageous for atosiban. I believe that ß-agonists also have a negative effect on the fetus and evidence from experimental studies investigating the cardiovascular effects of ß-agonists on the fetus support this suggestion. For example, experimental work on lamb fetuses has shown that the combination of hypoxia in the fetus and exposure to terbutaline led to heart failure. In addition, Doppler studies in the human fetus exposed to terbutaline have shown significant increases in the heart rate and also in blood pressure and flow in the aorta of the fetus. This means that in some cases, when the fetus is in a hypoxic state, it could have disastrous consequences. 

Lamont: When you compare all of the tocolytic agents that have been used in the past they can all be shown in vitro to stop uterine contractions and in vivo to reduce the number of contractions, but they have not demonstrated a reduction in perinatal mortality or morbidity, With respect to ß-agonists, the Canadian Preterm Labor Investigators Group (1992) and the meta-analysis by King et al (1988), an accumulated total of 1200 women (90% received ritodrine and 10% received terbutaline), showed that ß-agonists prolonged labour for up to 48 hours but were not associated with a reduction in perinatal mortality and morbidity. This is perhaps because we are not using the time gained in the best way possible. One option is to administer corticosteroids and the other is to arrange transfer to centres with neonatal intensive care facilities. Another point that has not been mentioned is the question of infection. I believe it may well be an advantage to us in the future to be able to create enough time to allow the identification of those women in preterm labour caused by infection, which represents approximately 40% of women in spontaneous preterm labour. If we are allowed time to identify those women who are infected then we can take the appropriate decision not to interfere with the pregnancy. This may be a very important factor which might effect how we manage preterm labour in the future. 

Audience: Atosiban has very few side effects and is consequently considered a safe tocolytic agent. Are you aware of any trials investigating the use of atosiban as a prophylactic? 

Cabrol: There is no evidence of any tocolytic agent used as a prophylactic treatment that was effective. Furthermore, there has been no evidence that maintenance treatment with tocolytic agents is effective either. The important point is that the US studies using atosiban indicate that maintenance treatment could be effective. I am unable to answer your question about prophylaxis, but if atosiban is effective in maintenance treatment for long-term therapy, perhaps it could be useful in particular groups of women, for example, in multiple pregnancies. 

Audience: How often can you repeat the bolus dose of atosiban and, if necessary, how long can atosiban treatment continue? 

Goodwin: To my knowledge, there are no data on the length of infusion of atosiban for longer than described here, which is an intravenous infusion of 48 hours. We are not aware of any side effects accumulating during that time period so it is quite possible that it could be continued for much longer. This kind of variation is something that would be learned in practice. 

Audience: The dosing regimen starts with a bolus dose of 6.75 mg atosiban. What happens if the woman improves but labour does not cease completely? Can the dose be repeated the following day or after a week? 

Goodwin: The atosiban bolus achieves the same concentration in the blood as the infusion but at a more rapid rate -it takes over an hour if given as an infusion only. So, the bolus is really designed to help get control of uterine contractions very rapidly with high concentrations of atosiban. 

Audience: The success rate for both tocolytic agents is almost the same, the only benefit is that atosiban is a safer drug. What about the need to improve the success rate? 

Goodwin: Every woman included in the studies was treated the same despite the fact that we now know that the aetiology of preterm labour is multifactorial. It is quite remarkable that a highly specific agent like atosiban was able to show the clinical effect that it did. We are now entering an era where we will be able to diagnose various subsets of women in preterm labour and select out those women more likely to respond to atosiban. 

Lamont: It has been emphasised many times today that the aetiology of preterm labour is multifactorial and therefore the approach to managing it has to be multifactorial. There are two ways to do this: either try everything possible in the hope that it treats the multifactorial process or, alternatively, try to separate women with preterm labour into the different aetiologies and then specifically treat those with infection with antibiotics and those requiring tocolytics with tocolytics. 

Steer: There is a large study going on throughout Europe called Oracle on the use of antibiotics in threatened and early preterm labour. If the results indicate that antibiotics have a definite effect in reducing preterm labour then we may be able to combine them with something like atosiban to stop the immediate uterine activity and then add an antibiotic to eliminate subclinical infections. The danger with any totally effective tocolytic is that if there is an infectious process occurring which you do not treat then this could progress into a major intrauterine infection which would be dangerous to both the mother and baby. If we are going to use combination therapy, I suspect it will be a combination of a tocolytic agent such as atosiban with antibiotics rather than just another tocolytic agent. I believe we are all united in the view that preterm labour is not a single syndrome, but a whole group of different aetiologies in which tocolytic agents are suitable in certain cases but not all, and different combinations may be the answer in the future.