Heart and blood medications

2.8.1

Arterial hypertension and pregnancy

The management of arterial hypertension differs greatly between pregnant and nonpregnant women. One reason is that certain antihypertensive agents are clearly harmful for the fetus – e.g. ACE inhibitors or angiotensin II receptor antagonists (sartans) during the fetal period (second and third) trimester – while others have not been sufficiently examined regarding their safety. Also, the aims of therapy differ.

Outside pregnancy the main objective of treatment is to lower the risk of cardiovascular sequelae such as heart attack or stroke. This risk reduction has been proven for thiazide diuretics (also for chlortalidone and indapamide), β-blockers, calcium antagonists, ACE inhibitors and sartans. In nonpregnant patients these medications belong to the drugs of first choice whereby each type of agent has its own advantages and disadvantages. Thus, people with diabetes or a high risk of diabetes should not usually receive β-blockers, diuretics, or combinations.

During pregnancy the main objective of antihypertensive management is to lower the risk of maternal complications during gestation and to ascertain the normal development of the fetus. The goal is to lower the risk for preeclampsia, placental abruption, prematurity and intrauterine growth retardation.

The following types of hypertension are distinguished in pregnancy:

• ▪

  Chronic arterial hypertension : diagnosed before, during, and after pregnancy.

• ▪

  Gestational hypertension : develops after the twentieth gestational week, no proteinuria, and disappears within 6 weeks after birth. About half of these patients develop preeclampsia.

• ▪

  Preeclampsia and eclampsiay : Proteinuria (>300 mg/24 hours) and first appearance of hypertension with possible edema.

• ▪

  Superimposed gestational hypertension : preeclampsia in pregnant patients with chronic hypertension, seen in 20–25% of pregnant women with chronic hypertension.

A blood pressure reading of 140/90 mmHg marks the border to hypertension during pregnancy. There is controversy surrounding the issue of when pregnant women should be treated, and if women who already have lower blood pressure elevations would benefit from a therapeutic intervention. This discussion is partially based upon the results of a meta-analysis that showed that a lowering of the mean blood pressure by 10 mmHg during pregnancy decreased the birth weight by 145 g. This reduction did not correlate with the type of antihypertensive medication or with the duration of treatment ( ). A pathophysiologic explanation for this finding might be that lowering of the pressure compromises utero-placental perfusion and thereby leads to a loss of weight (see also ).

A pilot study investigating the issue at what level of diastolic pressure arterial hypertension should be treated during pregnancy finds only minor differences between tight and less tight blood pressure control when looking at maternal complications and fetal outcomes ( ).

A population-based retrospective cohort study investigated 100,029 deliveries; of those 1,964 were babies born to mothers who experienced hypertension during pregnancy, with 620 neonates exposed to at least one antihypertensive medication. The authors found a higher rate of intrauterine growth restriction, small for gestational age, and preterm deliveries between both the treated and untreated groups, concluding that not only medication, but also hypertension itself is an independent risk factor for perinatal adverse outcomes ( ).

However, a large trial is still needed to identify optimal blood pressure goals and therapy of non-severe maternal hypertension in pregnancy ( ). Despite many studies and experiences there are still no uniform recommendations for pregnant women. Methyldopa remains the first line medication for long-term therapy of chronic hypertension during pregnancy. Metoprolol , nifedipine , and, with some reservation, dihyralazine/hydralazine are also considered to be well studied.

Blood pressure situations caused by preeclampsia that are more dangerous for mother and fetus are usually best managed with oral nifedipine or urapidil or, with reservations, dihyralazine/hydralazine IV. Also β-receptor blockers may be given, among them the well-studied labetalol.

2.8.2

α-Methyldopa

2.8.3

β-Receptor blockers

β-Receptor blockers inhibit the interaction of the neurotransmitters noradrenaline and adrenaline on the β-receptors of the relevant target organ. The heart contains primarily β1-receptors.

β1-selective agents are atenolol , acebutolol , betaxolol , bisoprolol , celiprolol , esmolol , nebivolol , metoprolol , and talinolol . Most of these are used for the treatment of hypertension, and some for other indications.

The non-selective β-blockers include carteolol , oxprenolol , penbutolol , pindolol , propranolol , timolol , and the antiarrhythmic agent sotalol (discussed in Section 2.8.17 ).

Labetalol has an additional α-receptor blocking component. Hardly any experience is found with the use of carvedilol, an α1- and non-selective β-receptor blocker (more on β-receptor blockers in Chapter 2.17.22 Ophthalmic Medications).

β-Receptor blockers cross the placenta and have no teratogenic effect as far as is known (e.g. ). These experiences refer primarily to the well-studied systemic β-blockers such as atenolol, bisoprolol, labetalol, metoprolol, and propranolol.

Using data from the Swedish Birth Registry a study compared 1,418 women without diabetes under various hypertensive drugs with 1,046, 842 women without hypertension and diabetes in early pregnancy. The risk of prematurity and small for gestational age (SGA) births was increased in the heterogeneous total group of users of hypertensive drugs, and major malformations were slightly increased (OR 1.63; 95% CI 1.26–2.12). No medication-specific risks or malformations were seen for subjects exposed to β-blockers (monotherapy n = 798) or to other hypertensive drugs ( ).

More undesirable effects have been reported with atenolol than with any other β-blocker. It is unresolved if this represents a specific effect of this agent, or if atenolol has been studied more closely, and those results might show up with all β-blockers ( ). Reports indicate a lower weight of the placenta, intrauterine growth retardation (IUGR), and a lower birth weight ( ). examined 491 pregnant women with hypertension in regard to the birth weight. As controls served 189 untreated women and patients treated with, among others, calcium antagonists ( n = 14). The result is interesting, yet its validity limited by the low case number: Newborns whose mothers had taken atenolol since conception or during the first trimester until birth ( n = 40) displayed a significantly lower birth weight. However, the use of atenolol during the second trimester did not show this effect. Independent of the type of hypertensive drug, a superimposed hypertension (second or third trimester) resulted in a lower birth weight. The use of IV labetalol is increasing, especially in a hypertensive crisis during pregnancy. Although there are reports of low birth weight associated with the use of this medication, most studies agree that it does not seem to offer greater risks in pregnancy ( ).

One report described a child with a retroperitoneal fibromatosis with compression of the medulla leading to a later scoliosis; this was linked to maternal treatment with atenolol. The authors consider that this association should be mentioned, as similar results had been described in adults after the use of atenolol ( ).

Pathophysiologically β-receptor blockers, especially perhaps atenolol, could decrease placental perfusion and lead to IUGR and lower placental and birth weight. A causative factor could be that β-blockers increase the tonus of the uterus, but also their hypoglycemic activity has been discussed. As these problems can be induced by severe hypertension alone, the disease itself has to be considered a co-factor as a minimum.

When considering the effects of the medication upon intrauterine growth, a distinction should be made between severe and lighter forms of maternal hypertension (Section 2.8.1 ). A meta-analysis of the side effects of antihypertensive drugs in lighter forms of hypertension found that the subgroup using β-receptor blockers only displayed a trend towards a lower birth weight (see ).

A neonatal β-receptor blockage due to maternal treatment can lead to hypoglycemia and a decrease in heart frequency. Respiratory depression has been observed in newborns when propranolol is given IV just prior to a caesarean delivery (review by ); this, however, is an unusual event.

Some authors discuss the discontinuation of medication 24–48 hours prior to birth. This approach, though, cannot be recommended, because the maternal blood pressure increases during labor, and the generally mild side effects of a β-blocker in the newborn subside without sequelae within 48 hours. Nevertheless, obstetricians and pediatricians should be aware of the maternal medication.

The postnatal growth during the first year of life and other developments of the children do not appear to be affected ( ). This is confirmed in a later investigation of 32 children who had been exposed to labetalol in utero . At the age of 3–7 years they were compared with two control groups ( ). While in this study the “Labetalol Children” had slightly better results than the “Methyldopa Children,” a Dutch study group found different results: Children with intrauterine exposure to labetalol ( n = 58) had a higher tendency to develop attention deficit hyperactivity disorder (ADHD) than methyldopa-exposed offspring ( ).

2.8.4

Calcium channel blockers

Most calcium channel blockers (CCBs) are approved for the management of hypertension, the majority for the treatment of coronary heart disease, and some as antiarrhythmic agents. Nifepidine is also used off-label as a tocolytic ( Chapter 2.14.6 ).

In regard to use in pregnancy, nifedipine and verapamil are the best evaluated CCBs, followed by amlodipine and diltiazem .

Experience with the following agents is inadequate and at best limited to a few single case reports: felodipine , gallopamil , isradipine , lercanidipine (a vaso-selective blocker), manidipine , nicardipine , nilvadipine , nisoldipine , and nitrendipine . The CCB nimodipine is used to manage severe hypertension in pregnancy. In a Cochrane meta-analysis the drug was found to be useless ( ).

Contrary to the animal experience, there is no evidence that CCBs reduce utero-placental perfusion in humans. observed no increased risks of malformations in 78 pregnancies (nifedipine n = 34, verapamil n = 32, diltiazem n = 10), but saw a higher rate of miscarriages and an earlier delivery. The birth weight of the newborns tended to be lower. These effects were, according to the authors, not caused by the medication. Also, found no evidence of teratogenicity in 25 children exposed during the first trimester. An analysis of data of the Hungarian Malformation Registry found no evidence of limb defects or a higher malformation risk in general ( ) after exposure to nifedipine, verapamil, or felodipine. Another publication presented 56 retrospective reports concerning undesirable side effects after exposure to nifedipine – most during the second or third trimester ( ). Malformations were noted in 15 cases, four of these on the extremities, and among them one case with a defect of the end phalanges and a syndactyly. However, as the timing of the exposure had not been indicated, nifedipine cannot be causally linked to the problem. Also, a retrospective analysis does not allow an assessment of the frequency of malformations.

A prospective multi-center study with 299 women treated in the first trimester did not demonstrate an increase in malformations or an accumulation of more limb defects. Agents used included nifedipine ( n = 76) and verapamil ( n = 62), less common were diltiazem ( n = 41) and amlodipine ( n = 38). Significant differences were noted in the increased rate of premature births when compared to the controls. Further, there was a tendency toward a lower birth weight for premature as well as term births. These effects could be explained by the type and severity of the usually underlying placental disorder and not the medication ( ).

Nifedipine showed a faster response in a clinical trial that compared the effectiveness of nifedipine administered orally and intravenously administered labetalol for acute blood pressure control in hypertensive emergency of pregnancy in 60 pregnant women; this study identified no serious adverse maternal or perinatal side effects ( ).

followed 94 children exposed to nifedipine during pregnancy at the age of 18 months. They did not differ from the 96 controls (children born from mothers not treated) in gross or fine motor measures, hearing, vision or language, based on maternal response to questionnaires. The use of sublingual nifedipine can lead to a rapid blood pressure fall ( ).

Verapamil used to manage fetal supraventricular tachycardia (Section 2.8.15 ) can induce hyperprolactinemia and galactorrhea.

In summary, current publications about CCBs do not suggest that there is an appreciable teratogenic risk in humans.

2.8.5

ACE inhibitors

ACE inhibitors (ACEI) reduce activity of the angiotensin-converting enzyme and lower blood pressure; they can also be used in cases of cardiac failure, coronary disease, and diabetic nephropathy. Their effectiveness is based essentially on a suppression of the plasma renin-angiotensin-aldosterone system. ACE inhibitors have not shown teratogenic properties, but are fetotoxic. Experience covering more than 1,000 pregnant women with exposure during the first trimester has been published as case reports, case series, and analytic studies. In a somewhat older case series with more than 200 pregnant women treated during the first trimester no clear evidence was seen concerning human teratogenicity (e.g. ). This result was confirmed by a new two-center study with 224 ACEI-treated pregnancies that were compared with two control groups, one managed with other antihypertensive drugs and the other a cohort of healthy pregnant women ( ). The women treated with ACEI or angiotensin antagonists ( n = 28) had a less favorable profile: they tended to be older and suffered more from diabetes. The investigation found, in both antihypertensive cohorts, more premature births and a significantly lower, yet still normal birth weight. analyzed 71 ACEI-exposed pregnancies as well as data of the United Kingdom Adverse Drug Reaction Reporting System. They did not find an association with major malformations in either case.

A methodologically somewhat flawed prescription study was published in 2006, the results of which had not been seen by other investigators, namely an increased risk for heart septum defects ( n = 7) and for CNS anomalies ( ). Quite unusually, a coloboma was counted as a CNS anomaly. A Finnish registry study of 137 pregnant women indicated a slightly increased rate of malformation after ACEI exposure (corrected OR 2,20; 95% CI 1.19–4.08) and suggested that this was related to maternal diabetes ( ). Data of the Swedish Birth Registry showed 1,418 women without diabetes but intake of various antihypertensive drugs during early pregnancy. Their data were compared to those of 1,046,843 pregnant women without diabetes and hypertension. The corrected odds ratio for cardiovascular defects, especially septum defects, was higher overall in the hypertensive group with 2.59 (95% CI 1.92–3.51). This result applied to both the ACEI ( n = 157) and the β-blocker subgroups ( n = 1,013). The authors judged the effect not to be related to the medications ( ).

It has been known for some time that during the second part of pregnancy ACEIs can lead to reduced placental circulation ( ), fetal hypotension, oligohydramnios, and neonatal anuria requiring dialysis ( ). Such developmental problems have also been seen in animal experiments when high doses were used. The following pathophysiologic mechanism is considered to be at work: fetal kidney and urine production starts at the end of the first trimester. ACEIs reduce the vascular tonus of renal vessels resulting in fetal renal compromise with longer exposure. The result of a decreased perfusion is lower urine production with oligohydramnios and eventually renal failure and anuria. In addition, hypoplasia of skull bones, contractures, and pulmonary hypoplasia have been noted. The case reports were supported by data from an analysis of conspicuous pregnancy courses after use of enalapril that had been reported to the FDA ( ).

There are case reports that oligohydramnios subsides once ACIS had been discontinued (e.g. ). Concerning later sequelae, there are reports of four children who had been anuric postnatally after ACEI exposure during the third trimester, but recovered fully during the first three months of life. Later, during childhood or adolescence three of the four children developed renal insufficiency with proteinuria, and some also developed arterial hypertension ( ).

Since 1992 a so-called black box inscription warns against the administration of ACEIs during the second part of pregnancy in the USA. Nevertheless, the exposure rate tripled in a comparison of the late 1980s to 2003 ( ).

2.8.6

Angiotensin II receptor blockers (ARBs; Sartans)

Candesartan , eprosartan , irbesartan , losartan , olmesartan , telmisartan , and valsartan block the AT1 receptor selectively and competitively so that the formation of angiotensin II is inhibited. Azilsartan was approved as a new drug in the USA in 2011. Sartans are utilized to treat hypertension and cardiomyopathy. In patients with diabetic nephropathy the agents reduce proteinuria and increase the glomerular filtration rate.

Angiotensin II receptor blockers (ARBs) have not shown teratogenicity, but are fetotoxic.

Experiences covering more than 200 women with exposure in the first trimester are primarily found in case series or subgroups within antihypertensive drug studies. A follow-up study of 37 pregnancies with first trimester exposure reports about 30 live births and two pregnancies with major malformations, one of them was terminated because of exencephaly ( ). present 10 pregnancies with the exposure to sartans, seven of them until the thirteenth gestational week. The six live births did not display any malformations. In a case series with five children who had been exposed to ARBs in utero , one child with a negative family history showed a sixth right finger and a sixth left toe ( ). A two-center study failed to find malformations in 28 pregnant women who had used ARBs in the first trimester ( ). A further study investigated the effect of antihypertensive medications on the risk of cardiac malformation using the data of the Swedish Birth Registry. It detected 45 ARB-exposed pregnancies, 45 of these exclusively treated with sartans, and none of the offspring showed a cardiac anomaly ( ). When the results of the Diabetic Retinopathy Candesartan Trials (DIRECT) Program were analyzed, at least 40 women were identified who had taken candesartan during the first trimester. In none of their offspring was a malformation found ( ).

During use in the second and third trimester similar risks are present as with the ACE inhibitors. About 40 case reports describe oligo- and anhydramnios, renal dysfunction including anuria, pulmonary hypoplasia, and contractions of the extremities, hypoplasia of the skull, stillbirths and neonatal deaths (e.g. ).

There have been reports of five children who had been exposed until, or during, the third trimester and developed a thrombosis in utero or an obliteration of the inferior vena cava, respectively. In one case, thrombosis of both renal veins was described ( ). The pathogenesis has not been clarified; decreased renal blood flow may be of importance.

Oligohydramnios and compromised renal function may partially or completely reverse after medication has been discontinued (e.g. ). In these publications, medication change occurred twice in the twenty-second and once in the twenty-fourth gestational week. However, a series of 20 cases of pregnant women exposed to ARB identified a poor neonatal outcome associated with oligohydramnios due to these agents, even when the medication was discontinued ( ).

No studies have been undertaken so far as to whether children who have been exposed to sartans in utero may develop late sequelae such as kidney disease and hypertension, as seen in some children after ACEI exposure.

An analysis of 28 prospectively studied pregnancies with ARB exposure past the thirteen gestational week found that related fetal disease developed only when treatment continued past the twentieth week. The risk of oligohydramnios as the first visible sign of a sartan fetopathy was 31% in this scenario ( ).

A systematic review comparing neonatal outcomes between ARBs exposed babies with ACEIs showed poorer outcomes in the ARBs exposed group, and included renal failure, oligohydramnios, death, arterial hypotension, intrauterine growth retardation, respiratory distress syndrome, pulmonary hypoplasia, hypocalvaria, limb defects, persistent patent ductus arteriosus, or cerebral complications. The long-term outcome is described as positive in only 50% of the exposed children. The authors propose the term “fetal renin-angiotensin system blockade syndrome” to describe the related clinical findings ( ).

2.8.7

Dihydralazine

Dihydralazine , an antihypertensive agent with central and peripheral action, belongs to the medications that have been used in pregnancy the longest. Eighty percent is resorbed and after oral administration the liver inactivates about 2/3. The half-life is 2–8 hours.

There are hardly any documented experiences about its use in the first trimester. A higher risk of malformation has not been observed ( ).

Most investigations describe its application in the third trimester. In a few cases it was noted that patients with preeclampsia developed some liver toxicity ( ).

A pseudo-lupus syndrome has been known for some time to be a possible side effect of dihydralazine. Thus, a case description should be mentioned with a lupus-like syndrome in mother and offspring; the newborn died ( ).

In a meta-analysis examined maternal, fetal, and perinatal sequelae of the use of hydralazine in cases of severe hypertension – typically in the second or third trimester. As a comparison, the examined studies usually utilized nifedipine or labetalol. Three newer randomized studies compared hydralazine or dihydralazine with labetalol ( ), urapidil ( ) and diazoxid ( ). While results vary, it can be concluded that dihydralazine should not be the medication of first choice for the management of severe gestational hypertension.

2.8.8

α-1 Blockers (peripherally acting adrenergic antagonists)

Urapidil, prazosin, bunazosin, doxazosin, and terazosin belong to the peripherally effective α-1 adrenergic receptor blockers.

Urapidil can be given orally or by IV injection. It is used foremost in the management of emergency hypertensive situations during pregnancy. In a comparative clinical study, concluded that urapidil represents an equally effective alternative to dihydrazaline in preeclamptic women. This is confirmed by another study with 42 patients. Side-effects and control, however, were better with urapidil ( ). Urapidil is recommended as an alternative to dihydralazine in the management of preeclampsia. Compared to dihydralazine it may have the advantage that it does not increase the intracranial pressure. One case report notes a temporary respiratory depression in a premature infant of the thirtyfifth week in conjunction with high levels of urapidil in his urine ( ).

Prazosin crosses the placenta. Few publications from the 1980s describe that it is well tolerated in late pregnancy. There is no evidence of a teratogenic potential, but reports are not well documented. Even less documented experiences are available for bunazosin , doxazosin , and terazosin . In the management of hypertension of nonpregnant patients, these agents are typically used as medications of second choice and combined with other antihypertensive drugs; they are primarily useful in men with an accompanying prostatic hyperplasia.

2.8.9