Education And Debate

Grand Rounds - Hammersmith Hospital (beta) Adrenergic agonists and pulmonary oedema in preterm labour Must be used with care

BMJ 1994; 308 doi: (Published 22 January 1994) Cite this as: BMJ 1994;308:260
  1. G J Clesham
  1. Department of Medicine, Hammersmith Hospital, London W12 0NN.
  • Accepted 28 September 1993

Ritodrine and other β adrenergic agonists relax uterine smooth muscle and have been widely used to manage preterm labour. The effect of these drugs on birth weight and perinatal mortality remains controversial. Several side effects, including maternal pulmonary oedema, have been described.

Case history

A 34 year old woman with primary infertility for 10 years had in vitro fertilisation, resulting in a twin pregnancy. She was normally fit and well with no history of cardiac disease. At 24 weeks of pregnancy she developed intermittent vaginal bleeding and was admitted to hospital. Initial assessment showed no evidence of uterine activity, but later the same day she developed contractions and the cervix was found to be 2 cm dilated, although the membranes were intact. Preterm labour was diagnosed and a ritodrine infusion was started at a dose of 200 μg/min, given in 500 ml of normal saline every four hours. She was also given dexamethasone (two 12 mg doses) and thyrotrophin releasing hormone (400 μg eight hourly over 48 hours) to improve fetal lung maturation and reduce the risk of hyaline membrane disease.

Over the next 24 hours the contractions became less frequent and although she had a resting tachycardia, her chest was clinically clear. During the next day the contractions became more frequent and she was given ritodrine in 500 ml of normal saline every four hours. The contractions continued and on the fourth day she became increasingly breathless with a tachycardia of 135 beats/min, bilateral basal crackles, and some peripheral oedema. Pulmonary oedema was diagnosed clinically and she responded to oxygen and a bolus of intravenous frusemide. The ritodrine infusion was stopped and twins were delivered a few hours later.

Thirty minutes after delivery she became acutely breathless and started coughing up pink frothy sputum. She had severe tachycardia with chest crepitations and pulmonary oedema was confirmed by chest radiography (fig 1). An electrocardiogram showed minor ST segment changes. She again improved with oxygen and intravenous frusemide but had a further less severe episode four hours later.


Chest radiograph taken during episode of acute breathlessness showing upper lobe blood diversion, alveolar shadowing, and septal lines. These signs confirm the diagnosis of acute pulmonary oedema

The next day her respiratory symptoms had resolved, and she had a pulse rate of 70 beats/min, normal heart sounds, and a clear chest. An echocardiogram showed no valvular abnormalities, good biventricular function, normal chamber dimensions, and unremarkable Doppler filling velocities (fig 2). An electrocardiogram appeared normal.


Apricol four chamber view from an echocardiogram taken day after delivery. Pulsed wave Doppler signal of left ventricular filling shows the characteristic early (E) and atrial (A) components of normal diastolic function. No other abnormalities were seen. RV=right ventricle, LV=left ventricle, LA=left artrium

The twins were transferred to the neonatal unit. The first twin weighed 780 g and died one week later. The second twin weighed 690 g and was treated with surfactant for the respiratory distress syndrome. He was extubated six weeks later and was growing well when he left hospital aged four months. His mother remains well.


Ritodrine was developed solely for obstetric use and its efficacy has been the subject of controversy for many years. In the United States it became widely used after it was approved by the Food and Drug Administration in 1980. By 1985 it was used in over 100 000 cases of preterm labour a year in America alone.

Preterm delivery is an important cause of low birth weight and must be addressed in measures designed to reduce perinatal mortality and morbidity. Ritodrine acts on uterine β2 receptors to induce relaxation of smooth muscle cells. Despite numerous studies of its use, convincing data about the effects of ritodrine on birth weight and perinatal mortality and morbidity are lacking. Many of the early trials of β mimetic drugs were poorly randomised or too small to draw meaningful conclusions. However, an analysis of the percentage live births in various birthweight categories in the United States between 1979 and 1985, a period during which there was a widespread increase in the use of ritodrine, showed no change in the number of low birthweight babies.1

Individual trials showed no effect of ritodrine on perinatal outcome, although pregnancy was prolonged by 1-2 days.2,3 A meta-analysis of the randomised trials concluded that the main effect of β mimetic drugs was to delay delivery by 24-48 hours but found no effect on neonatal respiratory disease or perinatal mortality. The largest randomised trial of ritodrine in preterm labour, which included over 700 women with preterm labour, also showed no overall effect on birth weight, perinatal mortality, or the number of pregnancies progressing to term.5 However, ritodrine was clearly shown to prolong pregnancy by 48 hours. Unfortunately, the use of measures to improve fetal lung maturation, such as steroids and thyrotrophin releasing hormone, was not systematically studied. Many of the women in this study may have been too close to term to benefit from tocolytic treatment, but there was a suggestion of reduced fetal loss between 24 and 27 weeks' gestation.

Causes of pulmonary oedema

The association between the use of β mimetic drugs for preterm labour and maternal pulmonary oedema has been recognised for some years. This complication has been reported in up to 9% of cases6 and has been responsible for at least 15 maternal deaths. Pulmonary oedema can occur even some hours after delivery. Several mechanisms have been proposed to explain the risk of pulmonary oedema.

Fluid overload is probably the most important factor in the development of pulmonary oedema. Ritodrine stimulates the release of antidiuretic hormone and renin, producing water and sodium retention. Fluid overload may be worsened if the ritodrine is given with large volumes of intravenous solutions, particularly normal saline. A study of 23 pregnant women in 1981 showed raised plasma volume in women given ritodrine in dextrose but features of pulmonary oedema developed only when the drug was given in saline.7 The patients were noted to be less able to diurese their fluid load when the vehicle was saline. Our patient clearly had fluid overload but this may have been exacerbated by the greater rise in plasma volume that is seen in twin pregnancies. This the fluid retaining state produced by ritodrine may be compounded by the use of normal saline and result in pulmonary oedema.

A prolonged increase in heart rate caused by β mimetic drugs may result in cardiac failure because of decreased diastolic filling and systolic ejection times. Heart rate should be carefully monitored in patients receiving ritodrine, and this is particularly important when there is underlying cardiac disease such as mitral stenosis. Electrocardiographic changes have been reported in a large proportion of women given tocolytic treatment with β mimetics,8 although they are not usually accompanied by symptoms. The importance of these changes is not known, but β mimetic treatment in preterm labour should be avoided in women with risk factors for coronary artery disease. Heart failure after very prolonged treatment with these drugs may have a similar basis to the catecholamine induced cardiomyopathy seen in some patients with a phaeochromocytoma.

Altered pulmonary capillary membrane permeability has been suggested as another mechanism of pulmonary oedema. This hypothesis was originally based on anecdotal reports of women who developed pulmonary oedema with ritodrine infusions in whom Swan-Ganz catheterisation before treatment with diuretics had not shown raised pulmonary capillary wedge pressures. In a study of pregnant baboons ritodrine was shown to cause changes in pulmonary capillary wedge pressure but not in lung water, suggesting that membrane permeability was not a factor.9 There is no hard evidence of a primary permeability injury produced by β mimetics in humans.

In conclusion, β adrenergic agonists seem to be able to prolong pregnancy in preterm labour by about 48 hours. The beneficial effects of these drugs on perinatal outcome is probably related to measures taken to promote fetal lung maturation during this delay, particularly in very premature labour. Pulmonary oedema can usually be prevented by giving the β mimetic in small volumes of dextrose while the patient is carefully monitored. These drugs should not be used when there is underlying heart disease or in the later stages of pregnancy when treatment is unlikely to confer benefit.


AMC: The Canadian study of preterm labour has gone some way to resolving the role of ritodrine. Many of the trials of β mimetics have studied women who were up to 36 weeks pregnant, when no benefit can be expected. Even in the Canadian study only a small proportion of the women would have benefited from a small delay in delivery, and the benefit seen in very premature labour can be lost in a large study. There seemed to be an increase in survival for fetuses of 24 to 27 weeks' gestation. Most obstetricians would also use ritodrine in the 28 to 32 week period and give steroids to help fetal lung maturation.

MJW: Given that almost 10% of patients given β mimetics develop pulmonary oedema, are the benefits greater than the risks?

JS: There are some parts of the United States where these drugs are not used because of concern about safety. How clearly is the risk of pulmonary oedema related to the fluid in which the ritodrine is given?

AMC: In my experience pulmonary oedema is seen in less than 5% of cases. Ritodrine infusions should be made up in small volumes of dextrose and given with a syringe pump. Fluid overload seems to be a feature in many but not all cases of pulmonary oedema.

JMBH: Pulmonary oedema can be caused by increased pulmonary capillary pressure or by increased capillary permeability. A combination of the two is a particularly potent cause of pulmonary oedema as small pressure changes cause great increases in lung water. There really isn't any evidence in humans that β2 agonists increase capillary permeability. In fact they have a favourable effect helping clear fetal lung liquids at birth by increasing sodium transport.

JS: How useful is thyrotrophin releasing hormone in preterm labour?

MS: There is increasing evidence that thyrotrophin releasing hormone is effective in enhancing lung maturation and reducing the risk of hyaline membrane disease. It is generally used in conjunction with corticosteroids. Interestingly, the hormone itself can cause a tachycardia but I don't know how haemodynamically important it is in this situation.

CMO: β2 Agonists have been used to treat failing left ventricles as they produce peripheral vasodilation but have a small inotropic effect on the myocardium. The action of ritodrine on the myocardium was probably not unfavourable in this patient but fluid overload was responsible for the pulmonary oedema. Ritodrine should have been given in dextrose rather than saline. The subject has been confusing for years as women in labour are difficult to study. The 48 hour prolongation of pregnancy with ritodrine is valuable as steroids given during this time will help maturation of the fetal lungs.


View Abstract

Sign in

Log in through your institution