Nifedipine in pregnancy

Authors


Introduction

Nifedipine is a dihydropyridine calcium channel blocker that is widely used for the treatment of cardiovascular disorders in nonpregnant individuals. Over the last 15 years its favourable pharmacologic characteristics have resulted in its efficacy and safety being assessed in pregnancy. Its application both as a treatment for acute severe hypertension, as well as for long term use for hypertension in pregnancy, have been explored. The drug has been shown to have a tocolytic effect on uterine smooth muscle and hence its use in the prevention of preterm delivery has been investigated.

In this article, the mechanism of action of the drug, as well as the current understanding of its metabolism and pharmacokinetics in pregnancy, is reviewed, including assessment of the literature on the use of nifedipine in the management of hypertension in pregnancy and its use in suppressing preterm labour.

A literature search of MEDLINE and the Cochrane Library was conducted for the years 1975 to September 1997 concerning the use of nifedipine in pregnancy, both for the treatment of hypertension and for tocolysis. The keywords used were: nifedipine, calcium channel blockers, pregnancy, hypertension, pre-eclampsia, umbilical artery blood flow, uteroplacental blood flow, teratogenicity, tocolysis, preterm labour, preterm delivery. The reference lists of all identified articles were examined to find additional relevant studies.

Nifedipine

Nifedipine is a type 2 calcium channel blocker that inhibits the inward flow of calcium across the L-type slow channels of cellular membranes1,2. Its effect is primarily that of causing smooth muscle relaxation. Its applications have been in vascular, uterine and bladder smooth muscle relaxation. Unlike the type 1 agents, it has minimal effects on the cardiac conducting system3.

The ability of nifedipine to vasodilate the systemic and pulmonary vasculature with full reversibility on stopping the drug and its lack of tachyphylaxis1 has resulted in it becoming a widely used agent in the treatment of acute and chronic hypertension and angina pectoris. It causes a decrease in arterial vascular resistance but with minimal effect on the venous system. It results in a 20% lowering of the systolic, diastolic and mean arterial blood pressures1. An increase in cardiac output results from both reflex tachycardia secondary to stimulation of the sympathetic nervous system and an increase in stroke volume. Hepatic and renal blood flow are also increased1,4. This vascular relaxation, which is marked in hypertensive patients, does not occur significantly in normotensive patients, thus allowing the drug to be used in its other applications1.

Nifedipine has been shown to cause relaxation of the uterine smooth muscle and hence its use as a tocolytic agent has been evaluated. In vitro studies have shown the ability of nifedipine to inhibit myometrial contraction in pregnant and nonpregnant women5–7. It causes a reduction in basal tone and in the amplitude and frequency of uterine contractions7,8, whether these are spontaneous contractions or induced by potassium, oxytocin or prostaglandin E2 and F9,10.

The pharmacokinetics of the drug are well described1,11. Maximum serum concentrations of the drug are obtained most quickly if the capsule is first bitten and the drug then swallowed. With standard oral administration the peak concentration occurs slightly later. Sublingual administration with absorption through the buccal mucosa is poor and variable12,13. Blood pressure falls within five to ten minutes of a capsule being bitten and then swallowed, and ten to thirty minutes with oral administration. The rate of onset of action of nifedipine is similar in pregnant and in nonpregnant women. Metabolism occurs in the liver, 30% to 40% of the drug being metabolised in the first pass, and the inactive metabolites are excreted in the urine. In pregnancy the physiological changes in the cardiovascular system and in hepatic function result in an increased rate of metabolism and clearance of the drug. The peak serum concentration (Cmax), the elimination half-life and clearance rate in pregnancy and nonpregnancy of a standard 10 mg oral dose are shown in Table 114,15.

Table 1.  Peak serum concentration, elimination half-life and clearance rate of standard 10 mg dose in pregnancy and nonpregnant state. Values are given as mean (SD).
 PregnancyNonpregnancy
Peak serum concentration (ng/mL)38.6 (18)73.48 (17.48)
Elimination half-life (h)1.3 (0.5)3.43 (10.6)
Elimination clearance rate (L/h/kg)2 (0.8)0.49 (0.09)

Due to these differences in metabolism and clearance in pregnancy, the duration of action of nifedipine is reduced to four to six hours15. Consequently higher and more frequent doses may be required in pregnant women to obtain adequate antihypertensive effects. The recommended starting dose is 10 to 20 mg, six hourly.

Nifedipine crosses the placenta14. The ratio of nifedipine concentrations in umbilical cord blood compared with maternal serum is 0.9314,15. The fetal liver poorly metabolises the drug and it is excreted in the urine, resulting in an amniotic fluid to maternal serum concentration ratio of 0.5615. It does not cause any alteration in fetal blood pressure in animals16,17.

Hypertension in pregnancy

Hypertension in pregnancy may be chronic (essential or secondary to underlying maternal disease) or due to preeclampsia. The effect of antihypertensive agents is different in these conditions, which must be considered when interpreting randomised trials of the treatment of hypertension in pregnancy.

Normally in pregnancy there is a reduction in the systemic vascular resistance, such that chronic hypertension often improves in the second trimester and early third trimester, to be restored in late pregnancy18. Women with chronic hypertension have a 10% to 30% increased risk of developing pre-eclampsia19–21. If pre-eclampsia does not occur the course of mild and moderate essential hypertension is benign, and it is uncertain whether antihypertensive treatment improves the outcome of pregnancy19–24.

Pre-eclampsia is a progressive systemic disease due to inadequate trophoblastic invasion of the spiral arterioles25. It is associated with significant maternal and fetal morbidity and mortality. In pre-eclampsia hypertension is due to generalised arteriolar vasospasm and decreased plasma volume26–28. The spiral arteries of the placenta maintain their muscular tone and responsiveness to vasoactive substances29. There is associated lipid deposition and acute atherosis, leading to placental ischaemia and infarction29. Altered prostanoid metabolism and nitric oxide production in placental blood vessels also occurs30,31. The pathological processes in the placenta result in dissemination of abnormal substances throughout the circulation, leading to disturbed endothelial function, platelet consumption, and increased vascular resistance and hypertension32,33.

The only definitive treatment of pre-eclampsia is delivery of the woman, but immediate delivery is not desirable when the fetus is preterm and the pre-eclampsia is not severe. Bed rest and antihypertensive agents34 are most commonly advised, and plasma volume expansion and antioxidants have also been evaluated35,38. The efficacy of these treatments as regards maternal and fetal outcome in pre-eclampsia is uncertain.

The control of severe hypertension, however, is uncontroversial. It is well proven that mean arterial pressures > 140 mmHg can result in cerebral arteriolar damage. Histological changes are detectable as early as 10 minutes after the onset of severe hypertension, and clinically apparent arteriolar damage is seen within a few hours39,40. Cerebral haemorrhage is the commonest cause of maternal death in pre-eclampsia and eclampsia41–43.

A number of drugs have been used in pregnancy to treat hypertension in pregnancy. The most widely used are methyldopa and the beta-blockers, atenolol and labetalol. Methyldopa has been used extensively and appears to be effective, and safe, both for mother and fetus34,44. Maternal side effects include lethargy, drowsiness and depression, and drug-induced hepatitis can be confused with early HELLP syndrome45. Although betablockers are considered safe in pregnancy46, intrauterine growth retardation has been linked to their use and side effects of hypoglycaemia and blockade of the neonatal cardiac conducting system have been reported47.

Dihydralazine has been the most commonly used drug in the acute hypertensive emergency in pregnancy, but its side effects of headaches, palpitations and tachycardia, as well as its ability to cause too rapid a fall in blood pressure are well described48,49. It is possible that these hypotensive episodes may be a reflection of underlying intravascular volume status rather than any property of the drug itself. Sodium nitroprusside and diazoxide are known to produce precipitous falls in blood pressure, often with adverse effects in an already compromised fetus49,50.

The ideal drug for the treatment of pregnancy hypertension continues to be sought. It should be rapidly effective and long acting, with minimal side effects to the mother, and it should not compromise the fetus either directly or by a reduction in placental perfusion. With an ever increasing body of evidence describing the safety, efficacy and minimal side effects of the calcium channel blockers outside of pregnancy, a number of authors have investigated their use in pregnancy. Nifedipine has been the most commonly used of these agents.

The mode of action of nifedipine to reduce systemic vascular resistance and its ability to improve urine output by increasing renal blood flow and by inhibiting the release of anti-diuretic hormone make it a highly appropriate drug for use in hypertension in pregnancy.

Nifedipine in the management of hypertension in pregnancy

Acknowledging the uncertainty surrounding the benefits of antihypertensive treatment in mild and moderate hypertension, a number of small studies have been performed to evaluate the use of nifedipine in pregnancy remote from term. It has been compared with bed rest51, placebo52, methyldopa53 and dihydralazine54 and appears to be an effective antihypertensive agent but shows no benefit in significantly prolonging pregnancy or improving maternal or fetal outcome. No studies have been done to investigate the long term effects on infants of mothers who have received nifedipine in pregnancy.

Sibai et al.51 compared bed rest with oral nifedipine in 200 women with pre-eclampsia between 26 and 36 weeks of gestation. They showed a significant reduction in the systolic and diastolic blood pressures with nifedipine, as well as a reduction in the number of deliveries for severe hypertension. However, the mean duration of hospitalisation in the two groups was similar and the use of nifedipine made no difference to prolongation of pregnancy. It also did not affect renal function, urinary excretion of protein or platelet count. No difference in fetal outcome was noted between the two groups. Abnormal fetal heart rate traces, Apgar scores, incidence of suspected small for gestational age infants and admission to the special care nursery were all similar.

Comparing nifedipine 20 mg (8 hourly) with placebo, given in a similar regimen, Ismail et al.52 reaffirmed its antihypertensive efficacy in pregnancy without any adverse fetal outcome. In the 20 women given nifedipine he was able to show an improvement in serum urea and creatinine levels, as well as in 24 hour urinary protein excretion with prolonged use of nifedipine. Similar results were found when nifedipine was compared with placebo in the management of women with pre-eclampsia in the immediate postpartum period. There was both a significant reduction in the mean arterial pressure as well as a mean increase in urine output of over 800 mL/24 hours. However, no difference was found in urinary protein excretion, creatinine clearance or serum levels of urea or uric acid.

Nifedipine has been compared with methyldopa for the treatment of pregnancy-induced hypertension and been shown to be comparable in its antihypertensive action but achieved no improvement in prolongation of pregnancy or fetal outcome53.

Fenakel et al.54 suggested that nifedipine was superior to hydralazine in the treatment of pre-eclampsia remote from term. In a group of 54 women with pre-eclampsia they were able to control the blood pressure significantly more often with nifedipine than with hydralazine and found that it was more predictable in its action and did not cause precipitous falls in blood pressure. They also noted an increase in urine output and associated decrease in peripheral oedema in 12 of the 24 women on this drug. No difference was found in urinary protein excretion. Perinatal outcome was similar with the two drugs but pregnancy was prolonged on average a week longer than with hydralazine, with a resultant increase in birthweight and reduced admission to the neonatal intensive care unit.

In a descriptive study Constantine et al.55 investigated the use of nifedipine as a second line drug in women with hypertension in pregnancy not controlled by either atenolol or methyldopa. They were able to control the hypertension effectively in 20 of the 23 women by the administration of 40 to 120 mg of slow release nifedipine daily. However, a high rate of perinatal morbidity and mortality was disturbing. Due to the uncontrolled nature of the study and the different causes of the hypertension, the contribution of the drug regimen was uncertain.

It appears that nifedipine is at least as good as dihydralazine in its antihypertensive effect, but its onset of action is less likely to be precipitous. In the short term it may increase urine output by reduction of the resistance of the renal arteries and by decreasing the release of anti diuretic hormone. No improvement in outcome or adverse effect on the fetus has been shown but it may have a benefit by prolonging pregnancy.

Control of acute severe hypertension

Nifedipine has been used successfully in nonpregnant women to reduce the blood pressure in acute severe hypertension. Research using animals has confirmed its efficacy in pregnancy without any significant effect on placental blood flow or compromise to the fetus56. Some studies have described its use in human pregnancy, comparing it with dihydralazine as regards its efficacy and its side effects57–60.

In a descriptive study of 21 pregnant women with severe hypertension in the antepartum or postpartum period, Walters and Redman57 showed that nifedipine resulted in a significant reduction in systolic and diastolic blood pressure. Its action was prompt and an oral 10 mg dose resulted in control for four hours. Side effects were minimal and there appeared to be no significant potentiation of other antihypertensive agents.

Three randomised controlled trials have compared the efficacy and safety of nifedipine and dihydralazine in acute severe hypertension in pregnancy58–60. Visser and Wallenberg58 compared the haemodynamic effects of 10 mg oral nifedipine and an intravenous hydralazine infusion of 1 to 3 mg per hour in 20 women with severe pre-eclampsia. They showed that both drugs had similar ability to reduce blood pressure to acceptable levels due to a significant reduction in systemic vascular resistance, accompanied by similar increases in cardiac output and heart rate. However, pulmonary capillary wedge pressure decreased significantly less with nifedipine than with dihydralazine. Five women treated with hydralazine developed fetal distress, compared with none treated by nifedipine. Similar efficacy of the two drugs has been shown by Jegasothy and Parathaman59 in a group of 200 severely hypertensive women where 5 mg sublingual nifedipine was compared with 5 mg of hydralazine given intravenously. In this trial no difference in adverse fetal outcome was found between the two groups. In a smaller study of 33 primigravidae with severe pre-eclampsia, Seabe et al.60 confirmed the similarity of effectiveness of the two drugs but showed an earlier onset of action with oral nifedipine. They reported equivalent increases of maternal heart rate with both drugs. One woman required delivery for fetal distress following a precipitous fall in blood pressure with nifedipine; otherwise fetal outcome was similar.

The comparable efficacy, possible earlier onset of action, minimal adverse effects and ease of administration of the oral nifedipine tablet, or of a punctured, then swallowed capsule of nifedipine, make it a suitable alternative to hydralazine in acute hypertensive emergencies. The Canadian Hypertension Society as well as the Australasian Society for Study of Hypertension have now both recommended its use as one of the first line drugs in severe hypertension in pregnancy61,62.

Attenuation of the hypertensive response at intubation

In pregnant women who are hypertensive there is an exaggeration of the pressor response to laryngoscopy and intubation63. This transient, severe hypertension has been associated with increased intracranial pressure, cerebral haemorrhage and cardiac failure with pulmonary oedema. Drugs, such as magnesium sulphate and alfentanil, have been used to attenuate this response. Nifedipine has been used to suppress the hypertensive response to intubation in both normotensive patients, as well as in patients with coronary vascular disease and found to be effective64,65. Hence its potential role in this regard in pregnancy has been investigated.

Kurmer et al.66 evaluated the use of 10 mg of nifedipine in 30 women with pre-eclampsia requiring caesarean section. Nifedipine resulted in a smooth, gradual and significant reduction in systolic, diastolic and mean arterial pressure over the 20 minutes before the general anaesthetic was administered. When compared with placebo, it significantly attenuated the hypertensive response to laryngoscopy and intubation. It did, however, result in a significant increase in maternal heart rate prior to and with intubation. No adverse fetal effects were noted. No controlled comparisons with other drugs used for this purpose exist.

The use of nifedipine in tocolysis

The most commonly used agents for the treatment of preterm labour have been the beta-sympathomimetic drugs, magnesium sulphate and indomethacin. These drugs appear to be effective in prolonging pregnancy in the short term but have no effect on delaying delivery in the long term67–69. They have no effect on fetal morbidity or mortality, and they all have significant side effects68. There have been instances of beta-sympathomimetic drugs being associated with pulmonary oedema, myocardial ischaemia and cardiac arrhythmias70. Their effects on the cardiovascular system, renal function and glucose homeostasis have made their use difficult in women with cardiac disease and with diabetes mellitus70,71. Magnesium sulphate may cause depression of the nervous system in both the mother and the newborn infant72; and long term indomethacin has been shown to cause spasm of the ductus arteriosus and renal arteries in the fetus73. Both the beta-sympathomimetic drugs and magnesium sulphate are used intravenously in preterm labour and are therefore associated with the risks of intravenous infusions.

The action of the calcium channel blockers to inhibit smooth muscle contractility with apparent minimal side effects to the mother and fetus has resulted in their being evaluated as tocolytic agents. Animal studies have shown the ability of nifedipine to prolong pregnancy and prevent preterm delivery in both rats74–76 and ewes77. Several studies in humans have evaluated the tocolytic action of nifedipine in preterm labour. There are no published randomised trials comparing nifedipine to placebo but it has been compared with ‘no treatment’78, ritodrine78–83 and terbutaline84 as well as to magnesium sulphate85.

Nifedipine versus other tocolytic agents

The efficacy of nifedipine in suppressing preterm labour appears to be as good as, and possibly better than, ritodrine and terbutaline78–84,86. It also appears equivalent in its ability to prolong pregnancy once the premature contractions have abated79,83,86. The temporary effects on contractions and the immediate delay in delivery are not linked to an improvement in perinatal mortality.

Only one randomised trial has compared the efficacy of nifedipine to ‘no treatment’78. This trial showed that nifedipine was superior to no treatment but, similar to other studies, was not large enough to show an effect on important outcomes. In this trial Reid and Wellby78 allocated 60 women in preterm labour with singleton pregnancies between 20 and 35 weeks of gestation to receive either oral nifedipine, an intravenous infusion of ritodrine or no treatment. Preterm labour was defined as one contraction or more every ten minutes, with the cervix dilated < 5 cm. Thirty milligrammes of nifedipine were given orally, followed by 20 mg every eight hours. The primary outcomes were suppression of contractions for 48 hours, and no further dilatation of the cervix. There was a significantly higher success rate with nifedipine, compared with ritodrine or no treatment. They also found that nifedipine significantly prolonged the time from admission to hospital to delivery, compared with ritodrine or no treatment: the average times were 36.3 days with nifedipine; 25.1 days with ritodrine and 19.3 days with no treatment.

Papatsonis et al.86 randomised 185 women in preterm labour between 20 and 33 weeks of gestation to receive either intravenous ritodrine or oral nifedipine. The authors defined preterm labour as either one contraction or more every ten minutes or prelabour rupture of the membranes. The dosage regimen was 10 mg nifedipine every 15 minutes for the first hour until contractions stopped (to a maximum of 40 mg), then a daily dose of slow-release nifedipine of 60–160 mg, depending on the amount needed in the first hour. Significantly fewer women taking nifedipine were delivered within 24 hours, within 48 hours and within seven days compared with ritodrine. Separate analysis of the group of women with prelabour rupture of membranes showed that neither drug was superior in prolonging pregnancy. Twelve women discontinued ritodrine because of side effects, compared with none taking nifedipine. Fetal outcome was similar, with no difference in Apgar scores or in arterial or venous pH measurements at birth. There were significantly fewer admissions to the neonatal intensive care unit in the nifedipine group.

Kupferminc et al.79 undertook a randomised trial including 71 women to compare nifedipine and ritodrine, and showed a similar prolongation of pregnancy for 48 hours, seven days and until 36 weeks79. Preterm labour was defined as regular uterine contractions at least once every six minutes with progressive cervical dilatation. Nifedipine was administered in an initial dose of 30 mg followed by a further 20 mg 90 minutes later if uterine contractions persisted. A maintenance dose of 20 mg nifedipine was then given orally every eight hours. Eleven women with twin pregnancies were included in this trial; of the six that received nifedipine, three were delivered after 36 weeks of gestation, compared with three of the five women who received ritodrine. These results are in keeping with three smaller prospective trials in the literature that have compared these two drugs in similar doses81–83. All have shown comparable efficacy of nifedipine and ritodrine in suppressing preterm delivery.

Smith and Woodland84 compared oral nifedipine with subcutaneous terbutaline in 52 women in preterm labour between 20 and 35 weeks of gestation. They found a similar success rate of 68% for nifedipine and 71% for terbutaline in completely stopping contractions within two hours of starting treatment. Glock and Morales85 found nifedipine to be as effective as magnesium sulphate in delaying delivery for 48 hours, with success rates of 92% and 93%, respectively. McLaughin et al.87 showed these two drugs had similar efficacy in prolonging pregnancy and preventing birth before 37 weeks of gestation.

A major advantage noted in these studies is the reduced amount of maternal side effects reported with the use of nifedipine compared with the other drugs78–80,82–84,86,88. A lowering of blood pressure and an increase in heart rate were noted but were less common with nifedipine than with ritodrine or terbutaline78,79,84. No deleterious fetal effects have been noted with the use of nifedipine as a tocolytic agent. A systematic review in the Cochrane Library, comparing nifedipine and the beta-sympathomimetic drugs, showed that nifedipine reduces the number of deliveries of newborn infants weighing less than 2500 g, but the number of infants admitted to the neonatal intensive care unit was increased67. There was no difference in the stillbirth or neonatal death rate. It must be noted that collectively only a small number of women have received nifedipine as tocolysis, and the absence of adverse fetal effects needs further evaluation.

Drug interactions

Magnesium sulphate is commonly used in eclampsia to prevent further seizures and in preterm labour as a tocolytic. Concern has been raised about the concurrent use of magnesium sulphate and nifedipine and the possible potentiation of the antihypertensive action and neuromuscular blockade of magnesium sulphate. In rats nifedipine exacerbates both the antihypertensive action89 and the cardiac toxicity90 of magnesium sulphate.

Marked hypotension with magnesium sulphate and nifedipine has been reported. Weisman et al.91 reported two women who experienced significant hypotension after receiving 10 mg oral nifedipine while on an intravenous infusion of magnesium sulphate (20 g/day)91. These women were also receiving methyldopa 2 g daily. On stopping the magnesium sulphate the blood pressure returned to previous levels within 30 minutes. Ben-Ami et al.92 reported a woman with pre-eclampsia who was receiving an intravenous infusion of magnesium sulphate (2 g/h) and who experienced muscle weakness and hypotension after taking 20 mg nifedipine orally92. Her symptoms improved within 15 minutes of receiving 1 g calcium gluconate intravenously. Another case, reported by Snyder et al.93, described significant muscle weakness in a woman who received an intravenous infusion of 0.5 g/h magnesium sulphate while taking 20 mg nifedipine 8hrly as a tocolytic agent93. Her symptoms disappeared rapidly when magnesium sulphate was stopped.

Other authors have found no adverse side effects when nifedipine and magnesium sulphate are used simultaneously54,57. Although two drugs with negative inotropic actions may not compromise a woman with normal ventricular function, this may not be true in a woman with heart disease.

The fear of an exaggerated antihypertensive effect when nifedipine is used with other, commonly used, antihypertensive agents such as methyldopa and the beta-blockers appears to be unfounded55,57. Beta-adrenergic blocking drugs may reduce the tachycardia induced by nifedipine.

Nifedipine causes an increase in blood flow to the liver which may alter the metabolism of other drugs, and has been shown to affect blood levels of digoxin, phenytoin, and theophylline. These drugs require closer monitoring of their blood levels if nifedipine is also taken1.

Effects on the fetus

Teratogenic effects on the fetus are a concern with nifedipine, but there may also be adverse effects of nifedipine on uteroplacental blood flow and also on fetal blood flow. Nifedipine is rated as a Category C drug with respect to its use in pregnancy94. This means that its teratogenic potential is uncertain, and hence it is recommended that it is used only where maternal benefit is seen to outweigh potential fetal effects. No specific congenital defects in humans have been recorded that are attributable to its use. However, digital abnormalities have been noted resulting from very high doses adminstered to animals. The development of hyperphalangeal bones in the fingers and toes has occurred in rats given doses of > 150 mg/kg95. In rabbits, distal digital defects secondary to cartilage necrosis have been recorded96.

Magee et al.97 undertook a prospective multicentre cohort study to assess the risk of abnormalities in fetuses after exposure to calcium channel blocking drugs in the first trimester. Forty-four women had been exposed to nifedipine in pregnancy, and 43 infants were normal at birth. One woman was delivered of an infant with multiple congenital abnormalities and developmental delay; the patient had epilepsy and systemic lupus erythematosus and was also taking carbamazepine, cyclophosphamide, prednisone, atenolol and ibuprofen throughout her pregnancy.

Haemodynamic changes

Fetal growth restriction secondary to placental insufficiency commonly occurs in pregnancies complicated by hypertension98. Any reduction in maternal blood pressure may be accompanied by a decrease in uterine perfusion pressure which may diminish uterine blood flow and fetal oxygen supply. This effect may be exacerbated where there is decreased blood volume, as occurs in pre-eclampsia. Dilatation of the uterine arteries may also reduce the critical perfusion pressure necessary for an adequate supply of oxygen to the fetus. Nifedipine crosses the placenta and hence its direct effects on fetal haemodynamics must be assessed14.

The results of studies in animals concerning the effects of nifedipine on uterine perfusion have been contradictory. Experiments in hypertensive rats and in pregnant goats and ewes have not shown any significant alteration in uterine blood flow with nifedipine, despite a reduction in placental vascular resistance56,99,100.

In research using sheep, Harake et al.16 showed that when there was a significant drop in maternal blood pressure there was a decrease in uterine blood flow. Similar findings by Blea et al.17 found this decrease to be temporary. In both of these studies it was also noted that there was a tendency to fetal hypoxaemia and acidosis during the infusion of nifedipine.

Where nifedipine has been used in acute severe hypertension in pregnancy, Doppler measurements of velocity waveforms in the uterine artery have remained unchanged101,102. Similar findings have been shown by Moretti et al.103 with chronic use of the drug in pre-eclampsia. Using radio isotopes Lindow et al.104 were also unable to show any significant change in uteroplacental blood flow after a single dose of nifedipine.

In vitro studies have shown nifedipine has a vasodilatory effect on umbilical and chorionic plate vessels, suggesting that nifedipine reduces vascular resistance in the fetal placental circulation6,105. In humans, there have been no changes shown in umbilical artery flow52,82,101,103,106,107 or in the fetal descending aorta or internal carotid artery flow103,106 with nifedipine, whether used to treat severe hypertension or preterm labour.

Acute hypotension may be associated with fetal distress, and this may occur with nifedipine. Impey108 has reported a case of acute fetal distress in a woman with pregnancy induced hypertension who became hypotensive after receiving 10 mg of nifedipine sublingually. Hata et al.109 have also recorded a case of fetal distress at 32 weeks of gestation in a pregnancy with severe hypertension following the administration of sublingual nifedipine: the woman's blood pressure fell from 208/122 mmHg to 136/96 mmHg, and was accompanied by severe variable and late decelerations. The fetus had intrauterine growth retardation and abnormal flow velocity waveforms. When compared with dihydralazine however, nifedipine is associated with fewer episodes of acute fetal distress, as suggested by nonreassuring fetal heart rate changes in pregnancies with54,110 and without intrauterine growth restriction54,58,101,103,110.

A number of studies have investigated the cardiovascular effects of nifedipine in women whose blood pressure is normal. Mari et al.111 studied women receiving nifedipine for preterm labour and found no change in the uterine artery pulsatility index111. Pirhonen et al.106 however, found a significant reduction in the systolic to diastolic ratio in the uterine artery of normotensive women, but this decrease in vascular resistance did not extend to the arcuate artery. In neither of these studies were any adverse effects on fetal heart rate patterns noted, and fetal umbilical, middle cerebral and renal artery velocity waveforms remained normal. No alteration was found in fetal cardiac output111 or fetal heart rate. A review of all women who had received nifedipine for tocolysis at St Josephs Hospital, Denver, Colorado, was undertaken by Murray et al. in 199288. One hundred and two women had taken nifedipine, and there were 265 nonstress test or biophysical profiles performed during nifedipine therapy. There was only one abnormal test and one neonatal death of a child with a congenital abnormality and intrauterine growth restriction.

In summary, nifedipine is an effective tocolytic drug, has fewer maternal side effects than the currently used drugs, and appears to have no adverse fetal effects, but the studies are small and further research is required in this area.

Maternal side effects

Maternal side effects with nifedipine seldom occur, and there is rarely a need to stop the drug as a result of these. Side effects are related to the dose and include flushing, headaches, and sweaty palms15,54,57,58,78,79,88; rarely, constipation, diarrhoea, heartburn, dyspnoea, and chest pain have been described. Tachycardia occurs commonly but is seldom associated with palpitations and appears to have no adverse effects1,71,78,79,85. A reduction in blood pressure in normotensive women in whom the drug is being used for tocolysis generally occurs but is asymptomatic and appears to be clinically insignificant71,78,79,85.

Conclusions

Nifedipine is an effective drug to treat severe hypertension in pregnancy and preterm labour. Because it is given in a tablet or capsule by mouth, it is easier to use than intravenous drugs. The described side effects of nifedipine to the pregnant woman and her infant appear minimal.

However, the studies of nifedipine are small and larger randomised trials are required before any recommendations can be made about the use of nifedipine in routine clinical practice.

Acknowledgements

The authors would like to acknowledge the financial support of Bayer Pharmaceuticals for departmental research into preterm labour. The clinical trials of P. S. and R. J. were supported by a grant from the NHS(E) West Midlands R & D Programme.

Ancillary