Expert consensus document on management of … · Expert consensus document on management of...

Guidelines

Expert consensus document on management of

cardiovascular diseases during pregnancy

The Task Force on the Management of Cardiovascular

Diseases During Pregnancy of the European Society of

CardiologyTask Force Members, Celia Oakley, Chairperson*, Anne Child,Bernard Iung, Patricia Presbitero, Pilar Tornos, CPGPC Members,Werner Klein, Chairperson, Maria Angeles Alonso Garcia,Carina Blomstrom-Lundqvist, Guy de Backer, Henry Dargie,Jaap Deckers, Marcus Flather, Jaromir Hradec, Gianfranco Mazzotta,Ali Oto, Alexander Parkhomenko, Sigmund Silber, Adam Torbicki,Hans-Joachim Trappe, ESC Staff, Veronica Dean,Dominique Poumeyrol-Jumeau

Preamble..............................................762Introduction ..........................................762Management of cardiac diseases during

pregnancy ..........................................762Haemodynamic modifications during

pregnancy .........................................763Congenital heart disease ...........................764

High-risk patients .................................764Pulmonary hypertension ......................764Severe left ventricular outflow tract

obstruction ...................................764Cyanotic heart disease ........................764

Treatment of high risk patients ...........764Low-risk patients..................................764Specific conditions................................765

Pulmonary valve stenosis .....................765Tetralogy of Fallot ..........................765Coarctation of the aorta ...................765

Intra-atrial repair for transposition ofgreat arteries (TGA) ......................765

Congenitally corrected transposition ofthe great arteries .........................765

Fontan procedure............................765Arrhythmias in pregnancy associated with

congenital heart disease (see alsoSection 11) .................................766

Foetal assessment ................................766Timing and mode of delivery....................766

Marfan syndrome and other inherited conditionsaffecting the aorta ...............................767Marfan syndrome..................................767

Maternal health.................................767Delivery ..........................................767Aortic dissection during pregnancy..........767Health of the newborn ........................768Genetic testing .................................768Ehlers–Danlos syndrome.......................768Familial thoracic aortic aneurysms and

dissections ....................................768Summary.........................................768

Acquired valvular heart disease ...................768Regurgitant valve disease........................769Stenotic heart valve disease ....................769

* Corresponding author. Task Force on the Management ofCardiovascular Diseases during Pregnancy of the EuropeanSociety of Cardiology, Hammersmith Hospitals, Du Cane Road,London W12 0HS, UK. Tel.: +44-81-383-3141/184-420-8246;fax: +44-181-740-8373/1844-202968E-mail address: [email protected] (C. Oakley).

European Heart Journal (2003) 24, 761–781

0195-668X/03/$ - see front matter © 2003 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved.doi:10.1016/S0195-668X(03)00098-8

Mitral stenosis ..................................769Aortic stenosis ..................................770Pregnancy in women with heart valve

prostheses ....................................770Mode of delivery................................771

Recommendations ................................771Coronary artery disease ............................772Cardiomyopathies....................................772

Peripartum cardiomyopathy.....................772Dilated cardiomyopathy..........................773

Recommendations..............................773Hypertrophic cardiomyopathy ..................773

Recommendations..............................774Infective endocarditis...............................774

Prophylactic antibiotics ..........................774Recommendations..............................774

Arrhythmias...........................................774Hypertensive disorders..............................775

Classification and definitions ...................775Chronic hypertension.............................776Management of low-risk hypertension .........776High-risk patients .................................776Pharmacological treatment .....................776Management of post partum hypertension....777Pre-eclampsia .....................................777Treatment of acute hypertension ..............777Summary ...........................................777

Summary ..............................................778References ............................................778

Preamble

Guidelines aim to present all the relevant evidenceon a particular issue in order to help physiciansto weigh the benefits and risks of a particulardiagnostic or therapeutic procedure. They shouldbe helpful in everyday clinical decision-making.

A great number of guidelines have been issued inrecent years by different organisations—EuropeanSociety of Cardiology (ESC), American HeartAssociation (AHA), American College of Cardiology(ACC), and other related societies. By means oflinks to web sites of National Societies severalhundred guidelines are available. This profusioncan put at stake the authority and validity of guide-lines, which can only be guaranteed if they havebeen developed by an unquestionable decision-making process. This is one of the reasons why theESC and others have issued recommendations forformulating and issuing guidelines, which arequoted as a preamble or appendix in the finalreports.

In spite of the fact that standards for issuinggood quality guidelines are well defined, recentsurveys of guidelines published in peer-reviewed

journals between 1985 and 1998 have shown thatmethodological standards were not compliedwithin the vast majority of cases. It is therefore ofgreat importance that guidelines and recommenda-tions are presented in formats that are easilyinterpreted. Subsequently, their implementationprogrammes must also be well conducted. At-tempts have been made to determine whetherguidelines improve the quality of clinical practiceand the utilisation of health resources. In addition,the legal implications of medical guidelines havebeen discussed and examined, resulting in positiondocuments, which have been published by aspecific Task Force.

The ESC Committee for Practice Guidelines andPolicy Conferences (CPGPC) supervises and coordi-nates the preparation of new Guidelines and ExpertConsensus Documents produced by Task Forces,expert groups or consensus panels. The Committeeis also responsible for the endorsement of theseguidelines or statements.

This document defines the procedure and rulesfor developing and issuing guidelines and expertconsensus documents, from the moment of concep-tion of the Task Force or expert group to the finalpublication of the document.

Introduction

This document is addressed to cardiologists whoseyoung and not so young female patients may desirepregnancy, seek advice once pregnant or in whomheart disease is first discovered during pregnancy.

The focus is on those conditions which threatenthe life or health of mother or baby with only shortmention of those that are well tolerated. Weemphasise the haemodynamic principles on whichdetermination of likely outcomes are based, stress-ing the importance of early consultation wherethere is doubt and of team work between allthose concerned: physicians, cardiologists, generalpractitioners, obstetricians, anaesthetists andgeneticists as appropriate.

Management of cardiac diseases duringpregnancy

Most women with heart disease have successfulpregnancies but most cardiologists and obstetri-cians see only small numbers. Pregnant womenseek local care but women with known or suspectedheart disease, unexplained shortness of breath orother symptoms in pregnancy or planning preg-nancy should be referred to a specialist centre.Experienced cardiologists working as a team with

762 Task Force on the Management of Cardiovascular Diseases

obstetricians, anaesthetists, clinical geneticistsand neonatologists will advise. Shared care canthen be organised with the local hospital and GPwith the extent of surveillance, site and mode ofdelivery arranged according to individual need.

The success of neonatal surgery has greatly in-creased survival and allowed infants with complexcongenital anomalies to reach adulthood. Womenwith congenital heart disease now far outnumberthose with rheumatic heart disease in pregnancyexcept in developing countries. Because rheumaticvalve disease is now rare in the West except inimmigrants, it can sometimes be missed and short-ness of breath wrongly ascribed to the pregnancyitself or to asthma rather than to mitral stenosis orpulmonary hypertension. Modern echo along withan ECG usually provides all the means needed forcompleting the clinical diagnosis. Chest X-raysshould be restricted during pregnancy and shieldprotection used but they can provide valuable in-formation not otherwise easily obtained. The likelyresponse to the haemodynamic changes in preg-nancy can then be assessed but if heart disease isnot even considered the patient will never get asfar as an echo study or a cardiologist.

Most, but importantly not all, patients withheart disease in NYHA classes I and II will have asuccessful outcome. Some conditions, like mitral oraortic stenosis, can give trouble even when symp-toms were absent or no problem was even realisedto exist before the pregnancy. The dangerous con-ditions are: pulmonary vascular disease (whateverits cause), fragile aortas as in Marfan syndrome, leftsided obstructions and already dilated poorly func-tioning left ventricles. The risk is obviously high inany woman in NYHA class III or IV.

Women with pre-existing disease are less able tocope with superimposed conditions acquired inpregnancy such as peripartum cardiomyopathy(PPCM) and are more at risk from complicationssuch as pulmonary embolism, arrhythmias andstroke. These and spontaneous dissection of a cor-onary artery (or indeed the aorta) can smite thepreviously healthy though they are rare.

Cardiologists rely more on evidence from ran-domised trials than any other speciality in medicinebut there is no such evidence base from which toguide management in pregnancy. Both cliniciansand patients would probably be reluctant to joinsuch trials and recruitment of adequate numberswould be difficult.

Drugs prescribed in pregnancy have crept intocommon use without trial and their use continuedas long as their track record remains good. Oralanticoagulants are the exception as they remain in

use for patients with mechanical valves but onlybecause there is no equally effective alternative.

Management is guided by observational studiesand these have been consistent in linking risk tofunctional class1,2 and emphasising the dangersfaced by women with pulmonary vascular dis-ease.3,4 A recent multi-centre study from Canadareported on 562 women referred between 1994 and1999 but did not describe large enough numbers ofindividual condition for statistical validity. Heartdisease was congenital in three quarters of theCanadian women (none of whom had severe pul-monary hypertension or the Eisenmenger syn-drome) and acquired in only a fifth.5 The studyreinforced the existing body of knowledge of risksin pregnancy and emphasised the differences incase mix between the West and the developingworld. Mitral stenosis is still a major cause of deathrelated to pregnancy in these countries in which thegreatest experience in both closed and balloonmitral valvotomy has been acquired.

Haemodynamic modifications duringpregnancy

Hormonal changes, which relax smooth muscle,followed by formation of the placenta and foetalcirculation, determine an increase in blood volumewhich starts to rise as early as the fifth week. Theincrease reaches 50% towards the end of pregnancyand is greater in multiple pregnancies than single-tons. Both systemic vascular resistance and bloodpressure decrease and the resting heart rate in-creases by 10–20 beats per minute. The result is anincrease of 30–50% in cardiac output, which ismainly achieved by an increase in stroke volume.6

Failure to achieve this is marked by a resting tachy-cardia which provides evidence of diminishedcardiovascular reserve and which is itself detrimen-tal in conditions in which left ventricular filling isslow.

Labour and delivery feature a further increase incardiac output and also in blood pressure particu-larly during uterine contractions and an increasein oxygen consumption. These haemodynamicmodifications are heavily influenced by the mode ofdelivery.7

Cardiac output is also increased during the earlypostpartum period because additional bloodreaches the circulation from the contracting uterusdetermining an increase in preload.8 That is whyat-risk patients often develop pulmonary oedemaat this stage. Haemodynamic conditions havelargely returned to normal within 1–3 days in mostcases but may take up to a week.

Management of cardiovascular diseases 763

Congenital heart disease

Haemodynamic changes during pregnancy can ex-acerbate the problems associated with congenitalheart disease.6 The outcome is related to func-tional class (NYHA classification), the nature of thedisease and previous cardiac surgery.

High-risk patients

Any patient who reaches functional class III or IVduring pregnancy is at high risk whatever the under-lying condition as this means that there is no re-maining cardiovascular reserve. The situationscarrying highest risk are as follows.

Pulmonary hypertension

Severe pulmonary vascular disease whether with(in the Eisenmenger syndrome) or without septaldefects has long been known to carry the highestrisk (maternal mortality 30–50%).4,9 This is mainlybecause of a life-threatening further rise in pul-monary vascular resistance due to pulmonarythrombosis or fibrinoid necrosis which developsparticularly fast in the peripartum and postpartumperiods and can determine a fatal outcome even inpatients who previously had little or no disability.In the Eisenmenger syndrome right to left shuntingincreases during pregnancy because of systemicvasodilatation and right ventricular overload withincreased cyanosis and decrease in pulmonaryblood flow.

Severe left ventricular outflow tract obstruction

A fixed outflow tract resistance may not be able toaccommodate the increased cardiac output causedby increased plasma volume. This can lead to heartfailure with a detrimental rise in left ventricularand pulmonary capillary pressures, low output andpulmonary congestion.10

Cyanotic heart disease

The over all maternal mortality is around 2% withhigh risk of complications (30%) such as infectiveendocarditis, arrhythmias and congestive heartfailure (CHF).11 The foetal prognosis is also verypoor with a high risk of spontaneous abortion (50%),premature delivery (30–50%) and low-for-datesbirth weight because maternal hypoxaemia impairsfoetal growth.

Thromboembolism is one of the risks in high-riskpregnancies and the use of prophylactic heparinshould be considered especially after surgicaldelivery and in the puerperium.

Treatment of high-risk patientsPregnancy is not recommended. If pregnancy oc-curs, termination should be advised as the risks tothe mother are high (mortality 8–35%, morbidity50%). Even termination of pregnancy has its attend-ant risks because of vasodilatation and depressionof myocardial contractility due to anaesthesia.

Physical activity should be restricted and bedrest is recommended if symptoms occur. Oxygenshould be given if hypoxaemia is evident. Thepatient should be hospitalised by the end of thesecond trimester and low molecular weight heparinadministered subcutaneously, as prophylaxisagainst thromboembolism particularly in cyanoticpatients.

In severe aortic stenosis, it is especially import-ant to monitor systemic pressure and the ECG, aschanges can indicate the appearance or worseningof left ventricular overload. Balloon valvotomy canrelieve symptomatic and severe cases if the valve ispliable. This procedure is best performed in thesecond trimester when embryogenesis is completeand to avoid any negative effect of ionic contrastagents on the foetal thyroid late in gestation. Theradiation dose to the abdomen of the mother is low,between 0.05 and 0.2 rads.12 Ballooning is contra-indicated if the valve is calcified or there is alreadysignificant regurgitation. Surgery is the alternative.Cardiopulmonary by-pass has a foetal mortality of20%13 so every effort should be made to continuethe pregnancy until the foetus is viable and todeliver the baby by caesarean section before thecardiac surgery.

In severe cyanotic heart disease, monitoring ofoxygen saturation is very important. Haematocritand haemoglobin levels are not reliable indicatorsof hypoxaemia due to the haemodilution that oc-curs in pregnancy. If severe hypoxaemia is presentand termination of pregnancy is refused some kindof shunt should be implanted if feasible to improveoxygenation.

Low risk patients

Patients with small or moderate shunts withoutpulmonary hypertension or mild or moderate valveregurgitation benefit from the decrease of systemicvascular resistance that occurs during pregnancy.Patients with mild or moderate left ventricularoutflow tract obstruction also tolerate pregnancywell.10 In such cases the pressure gradient in-creases steadily as the stroke output rises. Evenmoderately severe right ventricular outflow tractobstruction (pulmonary stenosis) is well tolerated


and only rarely needs intervention duringpregnancy.

Most patients who have had cardiac surgery earlyin life without prosthetic valves can tolerate preg-nancy well. However, residual defects are presentin 2–50% of cases and need to be assessed clinicallyas well as with echocardiography. In these low riskcases it is reasonable to reassure the patientsand follow them with a cardiac assessmentevery trimester. Assessment of congenital heartdisease in the foetus should be done by foetalechocardiography.

Specific conditions

Pulmonary valve stenosis

Right ventricular outflow tract (RVOT) obstructiontends to be well tolerated during pregnancy despitethe gestational volume overload imposed on analready pressure-loaded right ventricle. No deathsand a low incidence of minor maternal complica-tions (about 15%) have been reported. When thestenosis is severe pregnancy may precipitate rightheart failure, atrial arrhythmias, or tricuspid regur-gitation, irrespective of the presence of symptomsprior to pregnancy. Patients with severe RVOT ob-struction should, therefore, be considered for itsrelief prior to conception. In cases of right ventricu-lar failure during pregnancy, balloon valvulotomy isthe option of choice for severe valve stenosis (fourcases have been reported with no complications).14

Tetralogy of FallotPregnancy in unoperated patients carries a risk ofmaternal and foetal complications, which is tied tothe degree of maternal cyanosis. The risk is highwhen oxygen saturation is <85%.11 The rise in bloodvolume and venous return to the right atrium with afall in systemic vascular resistance increases theright to left shunt and cyanosis. Close monitoring ofsystemic blood pressure and blood gases duringlabour is needed and any further systemic vasodila-tation (drug induced) avoided.

The risk of pregnancy in repaired patients de-pends on their haemodynamic status. The risk islow, approaching that of the general population, inpatients with good repairs. In patients with signifi-cant residual RVOT obstruction, severe pulmonaryregurgitation with or without tricuspid regurgita-tion and/or RV dysfunction, the increased volumeload of pregnancy may lead to right heart failureand arrhythmias. All patients with tetralogy shouldhave genetic counselling pre-conception with as-sessment in case of 22q11 deletion syndrome usingfluorescent in situ hybridisation (FISH). In its ab-sence the risk of defects in the foetus is low (about4%).15

Coarctation of the aortaCoarctation of the aorta should be repaired prior topregnancy. It is rare during pregnancy (9% of allcongenital defects). The management of hyperten-sion is difficult in the unoperated pregnant patient.Foetal growth is usually normal and in contrast toessential hypertension pre-eclamptic toxaemiadoes not occur but over enthusiastic treatment maycause too low a pressure in the distal segment. Thismay result in abortion or foetal death even thoughpressure in the proximal segment continues to riseon effort. Rupture of the aorta is the commonestreported cause of death16 and rupture of an aneu-rysm of the circle of Willis has also been reportedduring pregnancy. The increase in blood volumeand cardiac output increases the risk of aorticdissection or rupture during pregnancy and a beta-blocker should be prescribed.

Restriction of physical activity is the only way ofminimising potentially dangerous surges in bloodpressure. Surgical correction is only very rarelyindicated during pregnancy if systolic hypertensionis uncontrolled or heart failure is present. Balloonangioplasty is contra-indicated because of therisk of dissection or rupture. Whether this risk isavoidable with stenting is not known.Intra-atrial repair for transposition of greatarteries (TGA)Over 100 pregnancies have been reported in theliterature with no deaths. In women in functionalclass I–II pregnancy is usually well tolerated. Wors-ening of systemic ventricular function during orshortly after pregnancy occurred in 10% of thereported cases.17 ACE inhibitors should be stoppedbefore pregnancy or as soon as possible. Frequentreview is recommended.Congenitally corrected transposition of the greatarteriesWomen without significant other cardiac defectsusually do well but problems can develop throughfailure of the systemic right ventricle with increas-ing regurgitation through its tricuspid atrio-ventricular valve. Supraventricular arrhythmias,embolism and atrioventricular block are otherpotential complications.18

Fontan procedurePregnancy carries additional risk to the motherbecause of the increased haemodynamic burden onthe right atrium and the single ventricle. A ma-ternal death rate of 2% is reported.19 Increase invenous congestion and deterioration in ventricularfunction are the most common complications.Atrial arrhythmias tend to develop or worsen. Rightatrial thrombus formation may occur with a risk ofparadoxical embolism if the Fontan is fenestrated.


Spontaneous abortion is frequent and occurs in upto 40% of the cases, probably because of congestionof the intra-uterine veins. Only 45% of live births atterm have been reported. Careful patient selectionis important. The successful Fontan with a smallright atrium or total cavopulmonary connection(TCPC) in functional class II or I can probably com-plete pregnancy with a normal live birth. Fontanpatients with a large right atrium and some venouscongestion have to be monitored very carefully.They need anticoagulant treatment and conversionto TCPC before pregnancy is considered.Arrhythmias in pregnancy associated withcongenital heart disease (see also Section 11)The incidence of arrhythmias, both ventricularand supra ventricular, increases during pregnancydue to haemodynamic, hormonal and emotionalchanges.20 In most congenital heart diseases rightatrial and/or ventricular pressure or volumeincrease and arrhythmias, particularly supra-ventricular arrhythmias, occur in 10–60% of cases.During pregnancy arrhythmias become even morefrequent and develop in up to 80% of patients.Physiological changes in pregnancy may alter theabsorption and excretion and effective plasmaconcentration of all antiarrhythmic drugs.21

When chronic antiarrhythmic treatment isneeded to prevent episodes of arrhythmia digoxin isusually the first drug prescribed but it is ineffec-tive. Quinidine, verapamil and beta-blockers havebeen used for long-term treatment of supraven-tricular and ventricular arrhythmias in both motherand foetus without any evidence of teratogeniceffects.21 Amiodarone is a potent antiarrhythmicbut should be used only when other therapy hasfailed and then at the lowest effective dose.22 Allthese drugs have a depressive effect on myocardialcontractility so they have to be used with caution inthe presence of an impaired left or right ventricle.

Episodes of sustained tachycardia (particularlyatrial flutter which is the most common arrhythmiain adult congenital heart disease) that are not welltolerated can cause foetal hypo-perfusion andemergency DC conversion should be performed torestore sinus rhythm. If the tachycardia is haemo-dynamically well tolerated, drug therapy should beattempted.

Foetal assessment

In every pregnant woman with congenital heartdisease foetal cardiac assessment is necessary be-cause there is a 2–16% risk of congenital heartdisease in the foetus.15 The incidence of congenitalheart disease in the offspring is more common

in the foetus when the mother rather than thefather is affected particularly if the mother has acondition like bicuspid aortic valve which is morecommon in the male (see Table 1).

In a population at specific risk the detectionrate of congenital heart disease is high (75–85%).Affected foetuses benefit from delivery in a ter-tiary care centre, but the main importance of anearly (before 24 weeks of gestation) diagnosis is thepossibility of termination of the pregnancy (TOP).The two main determinants of foetal prognosis arematernal functional class and the degree of ma-ternal cyanosis. When the mother is in functionalclass III–IV, or in high-risk diseases such as severeaortic stenosis, Eisenmenger syndrome, etc earlydelivery is usually a good option. It will be obliga-tory in cyanosed women in whom monitoring offoetal growth is very important because it usuallyslows up and ceases before term. The survival ratefor pre-term neonates is high after 32 weeks (95%)and the risk of neurological sequelae is low so if thepregnancy is ≥32 weeks delivery should be expe-dited. Since the survival rate is low before 28 weeks(<75%) and the risk of brain damage in the survivingneonates is high (10–14%) surgery or percutaneousprocedures should if feasible be undertaken inorder to postpone delivery as long as possible.

The choice may be difficult between 28 and 32weeks, and decisions must be individualised. If thefoetus is going to be delivered at ≤34 weeks, lungmaturation must be induced by betamethasoneadministration to the mother.

Timing and mode of delivery

In the majority of patients, spontaneous delivery isindicated using epidural anaesthesia in order toavoid the stress of pain during delivery. In high-riskpatients, elective caesarean section should be per-formed. This allows the haemodynamics to be keptmore stable. Although the cardiac output increasesduring both general and epidural anaesthesia theincrease is less (30%) than during spontaneous de-livery (50%).7,23 Moreover, induction of labour at anearly gestational age often fails or takes a long

Table 1 Pregnancy and congenital heart disease incidenceof CHD in offspring

Total 4.10% Mother 5.00% Father 2.00%

Fallot 2.5% 1.5%LV obstruction 10–18% 3.00%VSD 6.00% 2.00%ASD 4.50% 1.50%


time. If heart surgery is needed caesarean sectioncan be performed immediately before it. Haemo-dynamic parameters and blood gases should bemonitored during delivery. In patients with con-genital heart disease in pregnancy, a multi-disciplinary approach in consultation withcardiologists, cardiac surgeons, anaesthesiologists,obstetricians, neonatologists and geneticists isneeded to minimise the risk to both mother andchild.

Marfan syndrome and other inheritedconditions affecting the aorta

Of the major inherited disorders affecting the heartand aorta during pregnancy, Marfan syndrome, witha population incidence of 1 in 5000, is the mostimportant worldwide. Eleven types of Ehlers–Danlos syndrome have now been characterised,with a combined incidence of 1 in 5000 births.Aortic involvement is seen primarily in Ehlers–Danlos syndrome type IV. Other familial formsof thoracic aortic aneurysm and dissection alsopresent problems of management in pregnancy.

Marfan syndrome

Marfan syndrome is the most serious dominantlyinherited fibrillin-1 deficiency disorder, affectingall systems, but mainly eyes, heart and skeleton.Signs of classical involvement in two out of threemain systems constitute the clinical diagnosticcriteria.24 In 25% of patients the syndrome resultsfrom a spontaneous mutation, but in 75% of casesthere is a family history of other affected relatives.The pregnancy history of affected females, and, ifavailable, their aortic root diameter at the time ofdissection or aortic root surgery is helpful in decid-ing a plan of management. The ages at which aorticaneurysms occur in other females is an approximateguide but there is great clinical variability evenwithin a family.

Maternal health

Eighty percent of Marfan patients have somecardiac involvement.25 The majority have mitralvalve prolapse with mitral regurgitation and poss-ibly associated arrhythmia. Mitral valve repair maybe necessary before pregnancy.

Aortic aneurysm, rupture and dissection are stillthe commonest causes of death in Marfan syn-drome. Pregnancy is a high-risk period for affectedfemales, with dissection occurring most often inthe last trimester or the early postpartum period.26

Full assessment should be performed before preg-

nancy and include ultrasound examination of theheart and entire aorta. Women with minimal car-diac involvement (aortic root diameter less than4.0 cm, and no significant aortic or mitral regurgi-tation) should be informed of a 1% risk of aorticdissection or other serious cardiac complicationsuch as endocarditis or congestive cardiac failureduring pregnancy.27 Patients with an aortic rootdiameter of more than 4 cm should be told that therisk of dissection during pregnancy is 10%.28,29 Thepros and cons of pregnancy should be fully dis-cussed as well as the alternatives (childlessness,adoption, surrogate pregnancy).26

The risk is lower for pregnancy following electiveaortic root replacement for aortic root diameters of4.7 cm and over. A number of patients have suc-cessfully undergone elective aortic root replace-ment, and subsequently had successful full termpregnancies without complication. One patientwent on to have a second successful pregnancy butfollowing this developed an aneurysm of the aorticarch which has since been successfully replaced.These patients need to be monitored by echocardi-ography of the remaining aorta at 6–8 week inter-vals throughout the pregnancy and for 6 monthspostpartum. Beta-blocker therapy should be con-tinued throughout the pregnancy. Each pregnancyshould be supervised by a cardiologist and obste-trician who are alert to the possible complications.

Delivery

If normal delivery is planned the second stageshould be expedited. The woman may be allowed tolabour on her left side or in a semi-erect position tominimise stress on the aorta. If the aortic rootdiameter is 4.5 cm or greater caesarean delivery isadvised.

Aortic dissection during pregnancy

Acute dissection of the ascending aorta is a surgicalemergency.30 Repair with a composite graft is theprocedure of choice.31 Preservation of the aorticvalve32 or its replacement with a homograft avoidsthe need for long-term anticoagulants. Normother-mic bypass, progesterone per vaginum and continu-ous foetal heart monitoring reduce the risk to thefoetus. Poor wound healing is a feature of Marfansyndrome, as is postpartum haemorrhage and anincreased tendency to prolapse of pelvic organs.Sutures should be left in longer than normal andantibiotic cover extended until the sutures areremoved.

Acute dissection with origin beyond the left sub-clavian artery and not involving the proximal aortashould be managed medically. This does not usually


need surgery and can be followed by serial MRI.Progressive dilatation to 5 cm or more, recurrentpain, or signs consistent with fresh dissection suchas the development of organ or limb ischaemia, areall indications for repair.33 The baby, if viable,should be delivered by caesarean section beforegoing on to bypass.

The anaesthetic management of caesarean sec-tion followed by repair of aortic dissection shouldminimise foetal exposure to depressant drugs whileensuring a well-controlled haemodynamic environ-ment for the mother.33 Epidural and spinalanaesthesia should be undertaken only after con-sideration of the possibility of dural ectasiaas arachnoid cysts could result in considerabledilution.34

Health of the newborn

Babies with Marfan syndrome tend to be long andthin with wise-looking faces, high palate and longfingers. They may be hypotonic and have difficultyfeeding. Ophthalmological examination for lensdislocation should be performed soon after birth.

Genetic testing

Almost 200 mutations have been reported in thefibrillin-1 gene, and almost every patient has aunique mutation. At present, if a mutation has beenidentified in an affected parent, diagnosis can bemade by chorionic villus biopsy at 13 weeks gesta-tion, by amniocentesis cell culture or postnatallyusing cord blood, or buccal rub sample from theinfant. If parents simply wish to know whether theinfant is affected, testing should be offered inthe newborn period. This avoids the 1% risk ofmiscarriage incurred during foetal sampling.

Ehlers–Danlos syndrome

This heterogeneous group of inherited connectivetissue disorders is characterised by articular hyper-mobility, skin hyperextensibility, and tissue fragil-ity. Together they occur in 1 in 5000 births. Aorticinvolvement occurs almost exclusively in EDS typeIV35 which is transmitted as an autosomal domi-nant. Affected women are usually of short, slimbuild with prematurely aged hands, triangularfaces, large eyes and small chins, thin pinchednoses and small lobeless ears. During pregnancy,women may show increased bruising, hernias, vari-cosities, or suffer rupture of large blood vessels.36

Aortic dissection may occur without dilatation. Thecourse of pregnancy and delivery should be closelymonitored. Postpartum haemorrhage may besevere. Incisions heal slowly and it is suggestedthat retention sutures be used, and not removed

for at least 14 days, to avoid wound dehiscence.Prematurity and precipitous deliveries are commonbecause of lax cervical connective tissue, and weakmembranes. Affected infants tend to be hyper-extensible and may have congenitally dislocatedhips. Floppiness and bleeding tendency may also befeatures.

Familial thoracic aortic aneurysms and

dissections

Some patients have a family history of aortic dissec-tion in the absence of overt Marfan syndrome.37

Close examination of affected surviving familymembers may indicate Marfanoid habitus to a minordegree. Frequently the surgical histopathology ofthe aorta indicates cystic medial necrosis, as inMarfan syndrome. Some patients demonstratemutations in the fibrillin-1 gene38 and recently,two other gene loci have been identified in suchfamilies.

Pregnancy in such patients should be managed inan identical fashion to that in Marfan syndromepatients.

Summary

Joint cardiac and obstetric management of high-risk pregnancies in women with inherited tendencyto aortic aneurysm and dissection should includeregular echocardiograms before, during and afterpregnancy. Hypertension and arrhythmia should beclosely controlled. Aortic surgery during pregnancybears a high risk of foetal mortality. It may beavoided through elective aortic root replacementwith preservation of valve or a homograft, prior topregnancy. Caesarean section should be reservedfor those with aortic roots over 4.5 cm, or delayedsecond stage. Beta-blocker therapy should be con-tinued throughout pregnancy. Postpartum haemor-rhage can be expected. The newborn should havecareful physical, echocardiographic and ophthal-mic examination. Alternatives to pregnancy shouldbe discussed with high-risk patients.

Acquired valvular heart disease

Rheumatic heart valve disease remains a majorproblem for public health in developing countries.In western countries, even though the prevalenceof rheumatic fever has declined considerably, rheu-matic heart diseases are still seen. This is particu-larly the case in immigrants who have not hadoptimal access to health care facilities. Besidesnative heart valve diseases, specific problems areencountered in women with heart valve prostheses


during pregnancy, mainly related to anticoagulanttherapy.

Regurgitant valve disease

Severe mitral or aortic regurgitation in youngwomen is frequently of rheumatic origin. Severedystrophic regurgitation is seldom encountered inyoung women in the absence of Marfan syndrome orprevious infective endocarditis.39 The prognosis ofpregnancy in women with mitral valve prolapse isexcellent unless the regurgitation is severe andpoorly tolerated.40

The increase in blood volume and cardiac outputwill increase the volume overload consequent uponthe valvular regurgitation but the decrease in sys-temic vascular resistance reduces the regurgitantfraction which compensates in part for this. Inaortic regurgitation the shortening of diastole con-sequent on tachycardia also contributes to a reduc-tion in the regurgitant volume. This explains whypregnancy is frequently well tolerated even inpatients with severe valve regurgitation. Haemo-dynamic tolerance is worse in the rare cases ofacute regurgitation because of the absence of leftventricular dilatation.41

Patients may develop progressive CHF, particu-larly during the third trimester. They need di-uretics plus vasodilators to reduce after-loadeven further unless blood pressure is low.42 Angio-tensin receptor antagonists and ACE inhibitorsare contra-indicated and since the withdrawal ofhydralazine from use during the first and secondtrimesters of pregnancy the only available vasodi-lators are nitrates and the dihydropyridinecalcium-channel blockers. Vaginal delivery may becarried out safely in most patients, even thosewho have experienced transient heart failure,using the same medication. Haemodynamicmonitoring is needed only in the severest cases.

Surgery should be avoided during pregnancy be-cause of the risk to the foetus and considered onlyin patients with refractory heart failure, which isvery infrequent in valvular regurgitation.43 Repairof the mitral valve is preferable whenever poss-ible but preservation of the aortic valve is rarelysuccessful (except in Marfan syndrome).

Stenotic heart valve disease

An increase in cardiac output across the stenosedvalve will determine a sharp increase in thetransvalvular gradient and pregnancy may bepoorly tolerated in patients with severe mitral or

aortic valve stenosis. The onset of functional wors-ening occurs most frequently during the secondtrimester.44

Mitral stenosis

Mitral stenosis is the most frequently encounteredvalve disease in pregnant women and it is nearlyalways of rheumatic origin.2,44 The transmitralgradient increases particularly during the secondand third trimester and tachycardia by shorteningdiastole contributes to a further rise in left atrialpressure.45 In patients with a mitral valve area<1.5 cm2 (or 1 cm2/m2 of body surface area) preg-nancy carries a risk of pulmonary oedema, CHF,arrhythmia and intra-uterine growth retardation.44

Close follow-up is necessary in every pregnantwoman with severe mitral stenosis, even if she wastotally asymptomatic before pregnancy or duringthe first trimester.44,46 The mean transmitral gra-dient and pulmonary artery pressure should bemeasured by Doppler echocardiography at threeand five months and monthly thereafter.

Medical treatment with beta blockers46,47 shouldbe started in patients who have symptoms orestimated systolic pulmonary artery pressure>50 mmHg. Choice of selective agents such as aten-olol or metoprolol limits the risk of interaction withuterine contractions.48 Adjustment of the dosageshould take into account mean gradient, pulmonaryartery pressure and functional tolerance. Highdoses are frequently required by the end of preg-nancy. Diuretics need to be added if signs of pul-monary congestion persist. If the patient still hassymptoms and/or pulmonary hypertension despitemedical therapy there is a high risk of pulmonaryoedema at delivery or during the postpartumperiod, threatening the life of both mother andfoetus49–51 and relief of the mitral stenosis isindicated.

The risk of foetal death during open-heart sur-gery is estimated to lie between 20 and 30% and isunpredictable13,52–56 but short of death, signs ofdistress have been documented by foetal monitor-ing during cardiopulmonary bypass.57,58 For thisreason, closed mitral valvotomy has been consid-ered the procedure of choice during pregnancy.59 Itis safe for the mother but carries a foetal mortalityof between 2 and 12%, even in series published inthe eighties.59–61

Percutaneous balloon mitral valvotomy (PMV)has now replaced surgery. Its feasibility and safetyduring pregnancy are well established. Publishedseries now total more than 250 patients.62–69

Haemodynamic results are good because young


women usually have favourable anatomy. Func-tional status improves and pregnancy continuesuntil vaginal delivery of a healthy newborn. Radia-tion exposure is minimised by shielding the abdo-men and omitting haemodynamic measurementand angiography. The ease of use of the Inoueballoon is of particular importance in keeping theprocedure as short as possible.

Foetal safety has been demonstrated by peri-procedural foetal monitoring and measurement ofradiation exposure.64 There is a 5% risk of severetraumatic mitral regurgitation, which is generallypoorly tolerated and requires emergency surgeryunder cardiopulmonary bypass. This is particularlydangerous for the foetus.65,66 The risk of tampon-ade or embolic events during PMV is very low.

Because of these potential complications PMVshould only be performed in highly experiencedcentres12 and limited to pregnant patients whoremain symptomatic despite medical therapy. It isnot recommended prophylactically or in patientswho have severe mitral stenosis but no pulmonaryhypertension and good functional tolerance. Thesame is true for closed mitral valvotomy, whichfor economic reasons remains the most frequentintervention for mitral stenosis in developingcountries. In rare cases PMV needs to be performedin emergency as a lifesaving procedure in criticallyill pregnant patients.70

Aortic stenosis

Severe aortic stenosis is far less frequent thanmitral stenosis in pregnancy. Most cases arecongenital, less frequently, rheumatic when it isusually associated with mitral stenosis. Deliveryis safe in patients whose functional tolerance isgood.71

In rare cases where patients remain severelysymptomatic, in particular if they have signs ofheart failure, aortic stenosis should be relievedbefore delivery. Percutaneous balloon aortic

valvotomy should be attempted when possible toavoid aortic valve replacement12,72–75 but is riskyduring pregnancy and should only be considered inexperienced centres in very selected cases.

Pregnancy in women with heart valve

prostheses

The haemodynamic tolerance of pregnancy and de-livery is generally good in women who have under-gone heart valve replacement. The problem is theneed for anticoagulant therapy in patients withmechanical prostheses which can be summarised asfollows:

1. a hypercoagulable state exists throughout preg-nancy76 and

2. vitamin K antagonists cross the placenta andincrease the risk of early abortion, embryo-pathy and prematurity.

The incidence of embryopathy is still debated. Theoverall risk seems to be around 5% in patients whoreceive vitamin K antagonists between the sixthand the twelfth weeks,77 although lower rateshave been reported78 and the risk is dose related.Vitamin K antagonists should be withdrawn beforedelivery.79 Unfractionated heparin does not crossthe placenta but long-term heparin therapy duringpregnancy is difficult to manage and considerablyincreases the thromboembolic risk for themother.77

There are no randomised trials allowing for anaccurate comparison of different anticoagulationregimens during pregnancy. A recent review of theliterature reported a total of 1234 pregnancies in976 women with mechanical heart valve prosthe-ses, two-thirds of which were in the mitral position(Table 2).77 It indicated that the use of heparinthroughout pregnancy leads to a prohibitive inci-dence of thromboembolic events, even when usingadjusted doses. There is agreement to the use ofvitamin K antagonists during the second and third

Table 2 Frequency of foetal and maternal complications according to the anticoagulation regimen used during pregnancy inwomen with mechanical heart valve prosthesis. Adapted from Chan et al.77

Anticoagulation regimen Embryopathy(%)

Spontaneousabortion (%)

Thromboemboliccomplications (%)

Maternaldeath (%)

Vitamin K antagonists throughout pregnancya 6.4 25 3.9 1.8Heparin throughout pregnancy 0 24 33 15

Low dose 0 20 60 40Adjusted dose 0 25 25 6.7

Heparin during first trimester, then vitamin Kantagonistsa

3.4 25 9.2 4.2

aWith or without heparin prior to delivery.


trimester of pregnancy. The usual recommendationis that they should be replaced by percutaneous orintravenous heparin at the 36th week to avoid therisk of neonatal intracranial haemorrhage duringdelivery.79 The alternative is elective caesareansection at 36 weeks. This is frequently needed inany case because labour often begins prematurelywhile the foetus is still anticoagulated and it issensible because it minimises the period onheparin.

There is no consensus regarding treatment dur-ing the first trimester.80–85 The continuation ofvitamin K antagonists allows safe and stable anti-coagulation for the mother. Recent data suggestthat the risk of abortion or embryopathy is very lowin patients who take ≤5 mg of warfarin per day.86

The alternative is to use subcutaneous unfraction-ated heparin during the first trimester, particularlybetween the sixth and twelfth week. This regimendecreases the risk of embryopathy to zero only ifheparin is started before the sixth week.77 How-ever, in addition to the discomfort and the risks ofthrombocytopenia and osteoporosis subcutaneousheparin during the first trimester is associated witha high incidence of thromboembolic complicationsparticularly prosthetic thrombosis. Consistent datashow that continuation of vitamin K antagonistsduring the first trimester is the safest therapeuticoption for the mother.77,78,87–90

The choice should be made after clearly in-forming the patient and her partner of the risksinherent in the different anticoagulation modali-ties. Potential medico-legal concerns should alsobe considered as the label states that warfarinis contra-indicated during pregnancy. The targetINR is the same and the dose does not usuallychange.

Low-molecular weight heparin has advantagesover unfractionated heparin, notably, it provides amore stable anticoagulation level.91 Its efficacy hasbeen demonstrated during pregnancy in venousthromboembolism92 but it has been used in only asmall number of pregnant women with heart valveprostheses.91,93,94 The safety and efficacy of suchtreatment has not been documented in patientswith mechanical heart valves outside pregnancy.Although its use is mentioned in recent recommen-dations,95 our opinion is that low-molecular weightheparin should not be recommended at the presenttime in patients with heart valve prostheses duringpregnancy. Whatever the anticoagulation regimen,pregnancy in a patient with a mechanical prosthesisis associated with a maternal mortality between 1and 4%, mainly due to valve thrombosis while onheparin therapy.

Valve repair before conception should be per-formed whenever possible or biological substitutesconsidered. Although pregnancy per se may notaccelerate bioprosthesis degeneration96 durabilityis still poor in young adults and patients need toaccept the inevitability of re-operation in a fewyears time while their children are still young97 andto understand that it carries risk.

Mode of delivery

Despite greater haemodynamic stress vaginal deliv-ery under epidural analgesia is safe in patients withheart valve disease provided they are in stablecondition.44 Obstetrical procedures to shorten thetotal duration of labour particularly the secondstage may be helpful.98 Invasive haemodynamicmonitoring is indicated only in patients with severevalve stenosis or recent heart failure.49,99

Caesarean section has the advantage of avoidingthe physical stress of labour, but it is not free fromhaemodynamic consequences related to anaesthe-sia and assisted ventilation and the increased risk ofvenous thromboembolism needs to be countered.

In all cases the modality of delivery should bediscussed between cardiologists, obstetricians,anaesthetists and the patient. It is preferable to setthe date so that the whole medical team can beready.

In patients on anticoagulant therapy, heparinshould be withdrawn 4 h before caesarean sectionor at the onset of labour and resumed 6–12 h aftereither surgical or vaginal delivery.

In high-risk patients with previous endocarditisor heart valve prostheses, prophylactic antibiotictreatment should be given at the beginning oflabour and during delivery.

Breast-feeding can be encouraged in womentaking anticoagulants. Heparin is not secreted inbreast milk and the amount of warfarin is low.100

Recommendations

• Echocardiographic evaluation should be per-formed in any young woman who has valvularheart disease, even in the absence ofsymptoms.

• The management of the valve disease should,whenever possible, be discussed before theonset of pregnancy, particularly in cases of mi-tral stenosis <1.5 cm2 suitable for percutaneousmitral valvotomy and in cases of aortic stenosis<1.0 cm2.

• Close follow-up is mandatory after the begin-ning of the second trimester.


• In cases of poor functional tolerance, medicaltreatment should include beta-blockers in se-vere mitral stenosis, vasodilators in regurgitantvalve disease and diuretics.

• Percutaneous mitral valvotomy is indicatedduring pregnancy only if the patient remainssymptomatic despite medical therapy.

• Open heart surgery should be performed onlywhen the mother's life is threatened and ifviable the foetus should be delivered before-hand.

• In a pregnant patient with a mechanical pros-thesis, the choice of anticoagulant therapy dur-ing the first trimester should take into accountthe greater thromboembolic risk with heparinand the risk of embryopathy with vitamin Kantagonists. The use of vitamin K antagonistsduring the first trimester is the safest regimenfor the mother.

• Delivery should if possible be planned and itsmodality discussed in close collaboration withthe obstetricians and anaesthetists.

Coronary artery disease

Atheromatous coronary artery disease is un-common in pregnancy but not as rare as itwas. Apart from familial hypercholesterolaemia,smoking, obesity and diabetes as well as older ageat conception account for increasing numbers.Such women may develop angina during pregnancyand need treatment to provide them with suf-ficient coronary flow reserve to carry them safelythrough the pregnancy. Exercise testing is import-ant in assessing this. If beta-blockers and calciumantagonists are insufficient percutaneous inter-vention (PCI) can be performed with care to mini-mise the radiation dose to the foetus. The secondtrimester is the best time to do this. Patients withalready known coronary disease should of coursebe assessed and treated before conceiving. Pre-vious coronary bypass surgery is not a contra-indication if the woman is fit. The geneticconsequences of having a child who will be anobligate heterozygote need to be discussed inwomen with homozygous or combined hetero-zygotic hypercholesterolaemia. These patientsalso develop left ventricular outflow obstructiondue to a narrowed aortic root combined withimmobilisation of the aortic valve cusps byxanthomatous deposits in the aortic sinuses. Ifdetermined on pregnancy they should embark onit early.

Sudden severe chest pain in a previously fit preg-nant woman may be caused by dissection of the

aorta. If the pain is caused by myocardial infarctionit is most likely that spontaneous dissection of acoronary artery has occurred.101 Thrombolyticsshould therefore not be given (they are only rela-tively contra-indicated in pregnancy) but immedi-ate coronary angiography performed with a view toPCI with stenting. Dissection can occur in any ormore than one coronary artery and the indicationfor intervention depends on the site and apparentsize of the evolving infarct.

Congenital coronary anomalies are also encoun-tered occasionally. Coronary-cameral and coronarypulmonary artery fistulae do not usually causeany problem. Coronary arteritis due to previousKawasaki Disease with aneurysm formation andthrombosis (which may be new) may present withangina or infarction in pregnancy and need cor-onary grafting. This should preferably not be doneon bypass but it may be unavoidable. Coronaryarteritis may also be associated with on-going auto-immune vascular disease and present with infarc-tion in pregnancy or the puerperium.102,103

Coronary angiography is essential for recognitionof the mechanism and anatomy of the infarct toenable management of it to be appropriate. Mosttend to occur in the peripartum period and needdifferentiation from PPCM if heart failure hasoccurred.

Cardiomyopathies

Peripartum cardiomyopathy

This is a form of dilated cardiomyopathy (DCM) thatoccurs in the peripartum period in previouslyhealthy women. It is defined as unexplained leftventricular systolic dysfunction confirmed echo-cardiographically which develops in the last monthprenatally or within five months of delivery.104 Thisdefinition is intended to exclude pre-existingDCM105 which may have been present but un-suspected before the pregnancy as DCM is likely tobe exacerbated by the pregnancy and to presentbefore the last month. There are few accounts ofDCM in pregnancy probably because overt cases arediscouraged from becoming pregnant. Case reportsusually describe marked deterioration.

Women who have developed a PPCM usuallypresent with heart failure with marked fluid reten-tion, less often with embolic stroke or arrhythmia.The worst cases tend to develop during the first fewdays postpartum. Failure may be fulminating andrequire inotropes, a ventricular assist device oreven transplantation. As ventricular function


usually (but not always) eventually improves, im-plantation of a device is much preferable to trans-plantation if they can be got through the worstperiod. As in acute myocarditis outside pregnancythe most fulminant cases show the most capacity toimprove (as well as to die) and in them the useof a device as a bridge to recovery is particularlyappropriate.

Less severe cases require standard therapy forheart failure and their left ventricular functionfollowed carefully. Anticoagulants are important asthe risk of systemic embolism is high. Improvementmay be delayed but continue for a year or more butin some cases function deteriorates and transplan-tation is then appropriate. A survey of 44 womenwith a history of PPCM who had a total of 60subsequent pregnancies showed a high relapserate in subsequent pregnancies.106 This was notconfined to women with residual left ventriculardilatation. It was seen also among women whosefunction had apparently returned to normal butthere were no fatalities in this group. Other experi-ence has been more encouraging107 but reportednumbers are small.

Cardiac biopsy usually shows acute myocarditis ifit is performed early after onset. The cause isunknown but possibly an immune reaction to the‘foreign’ foetus. Immunosuppressive therapy maytherefore be appropriate but there are only obser-vational data to support this. Immune globulin hasalso been tried with apparent benefit in a smallnumber of women.108

The most frequent time of presentation is duringthe first few days postpartum. Haemodynamicstress should be abating except that this is a periodof hypervolaemia in those women who have suf-fered little blood loss during delivery. Hyper-hydration may be a factor after operative deliverywith which PPCM is particularly associated. How-ever, when PPCM in milder form is first seen later inthe puerperium it can only be blamed on the preg-nancy itself or on the unlikely and coincidentaldevelopment of DCM at this time. PPCM also some-times affects women with pre-existing heart dis-ease and diminished cardiovascular reserve butwhose left ventricular function had previously beendocumented to be normal.109,110

Dilated cardiomyopathy

It is only very rarely that DCM is well documentedbefore the pregnancy. In most cases pregnancy isavoided on medical advice and patients with di-lated left ventricles are only occasionally first diag-nosed in early or mid gestation. If symptoms first

develop in the last month of pregnancy the title“peripartum” is given with the question concerningprevious left ventricular function unanswered andunanswerable.

If there is a family history of DCM this may bea clue to pre-existing but occult dysfunction in apatient who develops first symptoms within theartificial time envelope necessarily assigned thedesignation “peripartum”. The often explosiveearly postpartum onset or later, quieter, presenta-tion at a time of no haemodynamic load isso distinctive that PPCM deserves its separatecategory.

Patients with DCM should be advised againstpregnancy because of the high chance of deteriora-tion both during gestation and peripartum. If preg-nancy occurs, termination should be advised if theejection fraction is <50% and/or the LV dimensionsare definitely above normal.

If termination is refused the patient must beseen frequently and LV function checked by echo.Early admission to hospital is wise especially asboth ACE inhibitors and angiotensin II antagonistsare contra-indicated and treatment options aremuch more limited than outside pregnancy.

Recommendations

• Echocardiography should be performed, beforeconception if possible, in all patients in whomDCM is known or suspected or who has a familyhistory of DCM or of PPCM

• Pregnancy should be discouraged if left ven-tricular contractile function is reduced becauseof a high risk of deterioration.

• In patients with a family history of DCM agreater risk of PPCM should be given consider-ation.

• Pregnant patients with DCM are at high risk andshould be admitted to hospital if there is anyevidence of deterioration.

Hypertrophic cardiomyopathy

Women with hypertrophic cardiomyopathy usuallytolerate pregnancy well as the left ventricle seemsto adapt in a physiological way.111 This is especiallyadvantageous in this condition in which the cavitydimensions tend to be small. Fatalities have beenreported during pregnancy but are rare. A casereport of systolic dysfunction developing post-partum may well have been PPCM.112

Women with a murmur and outflow tract gradi-ent are especially likely to be first diagnosed inpregnancy. Considerable distress may be caused by


the diagnosis and by the genetic implications. Thisis not helped by the considerable publicity given inthe lay press to the risk of sudden death. In theabsence of a family history of sudden death asymp-tomatic patients can be told that their risks are lowand that pregnancy is usually completed success-fully. After the diagnostic echo and an ECG, exer-cise testing and ambulatory ECG monitoring andgenetic counselling are carried out as in the non-pregnant patient.

Women with severe diastolic dysfunction cancause concern with pulmonary congestion or evensudden pulmonary oedema. This may develop onexertion or emotion but is most likely to occurperipartum. Beta-blockers should be continued anda small dose of diuretic may help but rest in con-junction with the beta-blocker to prevent tachycar-dia is essential in these rare high-risk patients. It iswise to give low dose heparin.

If atrial fibrillation (AF) develops anticoagulationis imperative. Low molecular weight heparin issuitable. If new onset AF fails to revert to sinusrhythm, DC reversion may be necessary after ex-cluding left atrial thrombus by transoesophagealecho. A beta-blocker will help control the ventricu-lar rate in AF and to prevent recurrence. Digoxin isnot contra-indicated as these patients rarely haveoutflow gradients.

Patients with persistent arrhythmias, particu-larly symptomatic ventricular arrhythmias, whichhave developed during pregnancy, may need amio-darone despite the risk of inducing foetal hypothy-roidism. It is particularly effective in conjunctionwith a beta-blocker.

Normal delivery on a selected date should beconducted with continuation of a beta-blocker andavoidance of systemic vasodilatation. Any bloodloss should be replaced but care should be takennot to cause fluid overload in high-risk patientswith very labile left atrial pressures.

The genetic risk needs to be discussed includingthe phenomenon of anticipation, which determinesan earlier onset and more severe form in succeed-ing generations in some families.

Recommendations

• Most aymptomatic patients with HCM do well.• Medication should be confined to treatment of

symptoms.• Patients with severe diastolic dysfunction will

need rest and medication in hospital.• Pulmonary congestion or oedema are most

likely to occur in the third stage, delivery shouldalways be in hospital and the date planned.

Infective endocarditis

Infective endocarditis is rare in pregnancy but canpresent some difficult problems in management.The increase in blood volume and output can pre-cipitate failure caused by fever and worseningstructural damage. Antibiotics must be chosen tosave the life of the mother but also to try to avoiddamage to the foetus. The need for surgical treat-ment must be weighed against the risk of losing theinfant but should not be delayed if the indication isacute valvar regurgitation or conduit or shunt ob-struction nor if the organism is a virulent staphylo-coccus in a non-responding toxic patient. In suchcases the temptation to delay surgery until afterdelivery should be resisted. If the baby is viable itshould be delivered before the cardiac surgery.

Prophylactic antibiotics

The indications for antibiotic prophylaxis are thesame as in the non-pregnant state to cover dentalor other procedures or conditions likely to causeGram positive bacteraemia.

The incidence of bacteraemia following normaldelivery is between 0 and 5%.113,114 Moreover thebacteraemia when it occurs is low grade and may bedue to many different organisms. The risk of infec-tive endocarditis complicating normal delivery isextremely low.115 Nevertheless prophylaxis isindicated in patients with prosthetic valves orprevious endocarditis and may be chosen inother patients with anticipated normal delivery be-cause complications are unpredictable. Antibioticsshould of course be given before surgical delivery orcardiac surgery.

Recommendations

• Diagnosis and treatment are the same as out-side pregnancy.

• If gentamicin has to be used its levels need to bechecked with particular care because of the riskof causing foetal deafness.

• Decisions for surgery should be made early asthe foetal risk is dependent on the maternalcondition.

• Antibiotic prophylaxis is discretionary for nor-mal delivery but should be given to patientswith prosthetic valves or a history of previousendocarditis.

Arrhythmias

Both ectopic beats and sustained arrhythmias be-come more frequent during pregnancy when they


may even develop for the first time.20 In generalthey are treated in the same way as outside preg-nancy but as conservatively as possible reservingdefinitive treatment for later if it is safe to do so.116

All commonly used antiarrhythmic drugs crossthe placenta. The pharmacokinetics of drugs arealtered in pregnancy and blood levels need to bechecked to ensure maximum efficacy and avoidtoxicity.

Patients worried about ectopic beats can usuallybe reassured unless the frequency increases onexercise. Supraventricular tachycardias are cor-rected by vagal stimulation or, failing that, intra-venous adenosine. Electrical cardioversion is notcontra-indicated and should be used for anysustained tachycardia causing haemodynamic in-stability and therefore threatening foetal security.Beta blocking drugs with beta-1 selectivity are thefirst choice for prophylaxis. Verapamil is constipat-ing, many patients do not feel well on sotolol andverapamil and though they may be effective theytend to cause foetal bradycardia. Radio frequencyablation for AV nodal re-entry or certain AV re-entry tachycardias can if necessary be performedduring pregnancy with suitable lead shielding andmaximal use of echo rather than X-ray fluoroscopy.

If a class III agent is needed amiodarone is pref-erable to sotolol. Lesser amounts of amiodaronecross the placenta (the foetal concentration is only20% of maternal concentration), it has a less de-pressant effect on ventricular function than otheragents and has little pro-arrhythmic or lethal riskcompared with other antiarrhythmic drugs. Long-term use can cause neonatal hypothyroidism (9%of newborns), hyperthyroidism and goitre, soit should only be used when other therapy hasfailed and the arrhythmia causes haemodynamicinstability with risk of foetal hypoperfusion.22

Potentially life threatening ventricular tachy-arrhythmias are much less common and should beterminated by electrical cardioversion. Beta-1 sel-ective beta-blockers alone, amiodarone alone orthe combination, may be effective in preventionbut if ineffective an ICD will be needed. The pres-ence of an ICD does not itself contra-indicate futurepregnancy.117

A pacemaker for the alleviation of symptomaticbradycardia can be implanted at any stage ofpregnancy using echo guidance.

Hypertensive disorders

Hypertension is the most commonly occurring com-plication of pregnancy. Hypertensive disordershave remained one of the leading causes of both

maternal and perinatal morbidity and mortality.Management has not changed significantly formany years because of little progress in our under-standing and lack of an evidence base for theintroduction of new therapies.

Classification and definitions118

• Chronic hypertension, pre-existing hyper-tension +/− proteinuria in a patient with pre-existing disease diagnosed prior to, during orafter pregnancy.

• Pre-eclampsia–eclampsia. Proteinuria (>300 mgover 24 h or ++ in two urine samples) in additionto new hypertension. Oedema is no longerincluded in the diagnosis of pre-eclampsiabecause of its poor specificity.

• Pre-eclampsia superimposed on chronic hyper-tension. Increased blood pressure above thepatient’s baseline, a change in proteinuria orevidence of end-organ dysfunction.

• Gestational hypertension. New hypertensionwith a blood pressure of 140/90 on two separateoccasions, arising de novo after the 20th weekof pregnancy.

Korotkoff V is now recommended for the measure-ment of diastolic blood pressure in pregnancy asit corresponds most closely to the intra-arterialpressure.118,119

In women with pre-existing hypertension raisedblood pressure is the main feature. By contrast, inthe more ominous condition of pre-eclampsia,raised blood pressure is one sign of a syndromeresulting from an underlying systemic endothelialdisorder with vasospasm, reduced organ perfusionand activation of the coagulation cascade.

It is believed that pre-eclampsia is caused byplacental hypoperfusion due to failure of re-modelling (dilatation) of the maternal spiral ar-teries and release of a (still unknown) circulatingfactor which causes changes in systemic endo-thelial function. The HELLP syndrome is defined byhaemolysis, elevated liver enzymes and low plate-lets. Headache, visual disturbance and pulmonaryoedema may also occur.120

Superimposed pre-eclampsia develops in 20–25%of women with chronic hypertension and carriesrisk to both mother and baby.

Gestational hypertension is distinguished frompre-eclampsia by lack of proteinuria and is termedtransient hypertension of pregnancy if the bloodpressure has returned to normal by 12 weeks post-partum and as chronic hypertension if it is stillraised. Gestational hypertension mandates close


attention as about half will develop pre-eclampsia121 and if symptoms or abnormal haema-tological or biochemical markers are found pre-eclampsia is likely even if proteinuria is absent.119

Chronic hypertension is present before the 20thweek of gestation whereas the pregnancy specificcondition of pre-eclampsia is rarely seen before20 weeks except in the presence of trophoblastdiseases such as hydatidiform mole.

Chronic hypertension

Maternal complications of hypertension includeabruptio placentae and cerebral haemorrhage aswell as superimposed pre-eclampsia. Foetal com-plications include prematurity and dysmaturity,still-birth and neonatal death.

Management of low-risk hypertension

Control of hypertension should begin before con-ception. Low risk patients have essential hyperten-sion with BP between 140–160/90–110, normalphysical examination, normal ECG and echo andno proteinuria. Several studies have shown thatantihypertensive drugs are effective in preventingexacerbation of high blood pressure in preg-nancy122,123 but antihypertensive treatment has notbeen shown to be effective in preventing super-imposed pre-eclampsia nor is perinatal mortalityaffected regardless of the drugs used. Only a fewrandomised trials have been performed. None ofthe drugs tested had adverse effects on outcome.Atenolol has been associated with an increasedincidence of small for dates babies and a lowerplacental weight but there was no difference at1 year.124

As in non-hypertensives the blood pressure tendsto fall during pregnancy so it may be possible todiscontinue drug treatment. Frequent supervisionis essential because the patient may become highrisk through development of severe hypertension orpre-eclampsia. Drug treatment will be needed formaternal protection if the BP rises but providedfoetal growth is normal the pregnancy can continueuntil term. Hospital admission or delivery will beindicated if pre-eclampsia develops or foetalgrowth slows.

High-risk patients

Conditions associated with micro-vascular diseasecan affect placentation and carry an increased riskof pre-eclampsia. Maternal and foetal genotypesalso contribute.125 High-risk patients have severe

hypertension with evidence of end-organ involve-ment, a poor obstetric history or co-morbidity fromrenal impairment, diabetes or collagen vasculardisease.125 These women need individual assess-ment, counselling and frequent assessment ofblood and urine chemistry and of foetal growth.

For ethical reasons there are no placebo con-trolled trials of pharmacological regimes for thetreatment of severe hypertension in pregnancy.Both maternal and foetal mortality used to be highin severe hypertensives largely because of super-imposed pre-eclamptic toxaemia whose mortalityhas since been reduced by anticipation and earlyrecognition rather than by effective treatment.Antihypertensive therapy is indicated for themother and may also benefit the foetus by securingprolongation of the pregnancy.

Pharmacological treatment

• Methyl dopa remains the first line agent be-cause it has the best safety record with noevidence of adverse effects in mothers orbabies including long term paediatric follow-up.The dose is 750 mg to 4 g per day in three or fourdivided doses.126

• Beta blocking drugs have had extensive use. Thealpha beta-blocker, labetalol has the advantageof vasodilatation. The dose is 100 mg twicedaily up to 2400 mg per day. None of thebeta-blockers have been associated withteratogenicity.When given only late in pregnancy atenolol,metoprolol, pindolol and oxprenolol have notbeen associated with any adverse effects.Like atenolol labetalol has been linked to lowweight for gestational age but no such associ-ation was found in one large trial in whichlabetalol was started at between 6 and 13weeks gestation.124

• Calcium channel blockers, mainly nifedipine,have not been found either beneficial or detri-mental127 but if given sublingually or intra-venously rapid and excessive BP reductionhas caused myocardial infarction or foetal dis-tress. Myocardial depression may follow combi-nation of a calcium blocker with intravenousmagnesium.128

• Clonidine has been used mainly in the thirdtrimester without report of adverse outcome.The usual dose is 0.1–0.3 mg per day in divideddoses up to 1.2 mg per day.123

• The use of diuretics is controversial becausethey reduce plasma volume expansion so caus-ing concern that their use might promote theoccurrence of pre-eclampsia. Although there is


no evidence of this diuretics should only be usedin combination with other drugs particularlywhen vasodilators exacerbate fluid retention asthey markedly potentiate the response to otherantihypertensive agents. Diuretics are contra-indicated as utero-placental circulation per-fusion is already reduced in pre-eclampsia withfoetal growth retardation (remembering thatthe changes in vascular reactivity and plasmavolume antedate even by weeks the clinicalfeatures of pre-eclampsia129). If needed athiazide should be chosen. Furosemide has beenused safely in pregnancy complicated by renalor cardiac failure.Pregnant women with renal disease are usuallyhypertensive. Foetal survival is markedlyreduced and birth weight decreases with in-creases in creatinine. Volume overload mayincrease and reduce drug responsiveness requir-ing salt restriction, loop diuretics or dialysis.Increasing proteinuria masks pre-eclampsia.Low birth weight and premature delivery arethe rule.

• ACE inhibitors are contra-indicated during thesecond and third trimesters because they causerenal dysgenesis.130

• Hydralazine has been widely used to controlsevere pre-eclampsia and no adverse effectsfollowed its use for chronic hypertension in thesecond and third trimesters but it was found tobe inferior to other agents.120

Management of post partum hypertension

Resolution of hypertension may be delayed andrenal failure, encephalopathy and pulmonaryoedema may develop during the postpartum periodparticularly in patients with chronic renal or car-diac disease and superimposed pre-eclampsia. Thisreflects the time needed for endothelial repair.Gestational hypertension usually resolves quickly.Breast-feeding is to be encouraged. Although mostantihypertensive agents are excreted in milk thereare few data on any effects on the neonate. Aten-olol, metoprolol and nadolol are concentrated inbreast milk and diuretics reduce milk volumeso these agents should be avoided in nursingmothers.131

Pre-eclampsia

Lack of knowledge of its cause precludes preven-tion of pre-eclampsia. Identification and earlydiagnosis of women at high risk permits closemonitoring and carefully timed delivery.

Pre-eclampsia is completely reversible andusually begins to abate with delivery which is al-ways appropriate treatment for the mother but notfor the foetus in whom maturation is the main aim.The key question is always whether the foetus ismore likely to survive in utero or in the nursery.

The aim is to reduce maternal vascular compli-cations without critically reducing utero-placentalblood flow and thereby exacerbating the condition.Restriction of activity is usual. Randomised trialshave shown no improvement in foetal outcomefrom antihypertensive therapy.122,127

Treatment of acute hypertension

The most commonly used parenteral therapies arenifedipine, labetalol and hydralazine.

The use of magnesium sulphate for severe pre-eclampsia and eclampsia is now well establishedthough little is understood about its mode of ac-tion. Treatment with antihypertensive drugs andmagnesium sulphate in hospital may be followed bysome improvement and such management may pro-long pregnancy and decrease perinatal mortalityand morbidity. Close maternal and foetal surveil-lance are essential as prompt delivery is indicatedby worsening of the maternal condition, laboratoryevidence of end-organ dysfunction or foetal dis-tress. Delivery is the only definitive treatment forpre-eclampsia.

Steroids should be given for 48 h to acceleratelung maturation if gestation is <34 weeks.

• The use of low molecular weight heparin inpatients with a known coagulopathy or previoushistory of pre-eclampsia remains controversial

• AspirinA recent Cochrane review showed a 15% reduc-tion in the incidence of pre-eclampsia and a 7%fall in deliveries before 37 weeks but littleoverall improvement in foetal outcome and thedata are conflicting.132,133

• AntioxidantsIf free radicals promote endothelial dysfunctionantioxidants might help. Endothelial functioncan be improved in vitro by ascorbic acid but arandomised trial of vitamins C and E showed nodifference in perinatal outcome.134

Summary

Pregnant women with hypertension are at risk.Careful management has reduced maternal andfoetal complications. Drug treatment does not im-prove perinatal outcome in women at low risk but


antihypertensive treatment should be used to pro-tect women with high-risk hypertension. Althoughunderstanding of pre-eclampsia has advanced thereis still no specific treatment for it. Therapeuticstrategies are aimed at ameliorating the maternalresponse but the only intervention available toimprove perinatal outcome is timely delivery.

Summary

• Women at low risk in pregnancy are those whohave few or no symptoms and good ventricularfunction without haemodynamically compro-mising or potentially life threatening arrhyth-mias. They lack severe left ventricular inflow oroutflow obstruction, do not have significantpulmonary or systematic hypertension and donot need to take anticoagulants.

• After full cardiac assessment, low risk patientscan be managed locally while maintaining po-tential links with the obstetric cardiac centreshould any question or problem arise.

• Patients at higher risk need to be managedwithin or from the cardiac centre and thehighest risk patients will need admission fromabout 20 weeks.

• The mode and time of delivery should be dis-cussed and decisions made well in advance.Vaginal delivery is usually advised. Exceptionsare patients with dilated Marfan aortic roots oraortic dissections, uncorrected coarctation,pulmonary vascular disease (including Eisen-menger syndrome) and/or cyanosis and patientswith mechanical valve prostheses in order tominimise the period of heparin withdrawal. Epi-dural anaesthesia is favoured but vasodilatationshould be avoided in patients with cyanosis orwhen stroke output is compromised. Adequatefluid volume loading is important but should notbe overdone in patients with left ventricularobstruction or severe hypertrophic cardiomy-opathy. Invasive monitoring is rarely justifiedby its inherent risks.

• Antibiotic prophylaxis is discretionary for an-ticipated normal delivery. The risk of endo-carditis has been shown to be very low and thebenefits have not been proved but cover islogical and wise for surgical deliveries, forpatients with intracardiac prostheses of anysort and for patients who have had previousendocarditis.

• In patients with pulmonary hypertension post-partum monitoring should continue for up to aweek and be conducted in the CCU for high-riskpatients with continuous pulse oximetry as thisis their period of highest risk when an increase

in pulmonary vascular resistance needs to becombated most aggressively.

References

1. Whittemore R, Hobbins J, Engle M. Pregnancy and its out-come in women with and without surgical treatment ofcongenital heart disease. Am J Cardiol 1982;50:641–51.

2. Mc Faul P, Dornan J, Lamki H et al. Pregnancy complicatedby maternal heart disease. A review of 519 women. Br JObstet Gynaecol 1988;95:861–7.

3. Morgan Jones A, Howitt G. Eisenmenger syndrome inpregnancy. Br Med J 1965;1:1627.

4. Weiss BM, Hess OM. Pulmonary vascular disease andpregnancy: current controversies, management andperspectives. Eur Heart J 2000;21:104–5.

5. Siu SC, Sermer M, Harrison DA et al. Risk and predictors forpregnancy related complications in women with heartdisease. Circulation 1997;96:2789–94.

6. Robson SC, Hunter S, Boys RJ et al. Serial study of factorsinfluencing changes in cardiac output during humanpregnancy. Am J Physiol 1989;256:H1060–5.

7. Robson SC, Dunlop W, Boys RJ et al. Cardiac output duringlabour. Br Med J 1987;295:1169–72.

8. Elkayam U, Gleicher N. Hemodynamics and cardiac functionduring normal pregnancy and the puerperium. In: Elkayam,Gleicher N, editors. Cardiac problems in pregnancy. NewYork: Wiley; 1998, p. 3–19.

9. Oakley C. Pulmonary hypertension in heart disease inpregnancy. London: BMJ Publishing Group, 1997, p. 97–111.

10. Oakley C. Acyanotic congenital heart disease. Heart diseasein pregnancy. London: BMJ Publishing Group; 1997,p. 63–82.

11. Presbitero P, Somerville J, Stone S et al. Pregnancy incyanotic congenital heart disease. Outcome of mother andfetus. Circulation 1994;89:2673–6.

12. Presbitero P, Prever SB, Brusca A. Interventional cardiologyin pregnancy. Eur Heart J 1996;17:182–8.

13. Pomini F, Mercogliano D, Cavalletti C et al. Cardiopulmo-nary bypass in pregnancy. Ann Thorac Surg 2000;61:259–68.

14. Vajifdar BU, Gupta AK, Kerkar PG et al. Pre-natal echo-cardiographic diagnosis and neonatal balloon dilatation ofsevere valvar pulmonic stenosis. Indian J Pediatr 1999;66:799–803.

15. Burn J, Brennan P, Little J et al. Recurrence risks in off-spring of adults with major heart defects. Results from firstcohort of British collaborative study. Lancet 1998;351:311–6.

16. Deal K. Woolley coarctation of the aorta and pregnancy. AnnIntern Med 1973;78:706–10.

17. Genoni M, Jenni R, Hoerstrup SP et al. Pregnancy after atrialrepair for transposition of the great arteries. Heart 1999;81:276–7.

18. Connnolly HM, Grogan M, Warnes CA. Pregnancy amongwomen with congenitally corrected transposition of thegreat arteries. J Am Coll Cardiol 1999;84:820–4.

19. Canobbio M, Mair D, van der Velde M et al. Pregnancyoutcomes after the Fontan repair. J Am Coll Cardiol 1996;28:763–7.

20. Tan HL, Lie KI. Treatment of tachyarrhythmias during preg-nancy and lactation. Eur Heart J 2001;22:458–64.

21. Bloomfield TH, Hawkins DF. The effect of drugs on thehuman fetus. In: Setchell, Ginsberg, editors. Scientific foun-dations of obstetrics and gynaecology. Oxford: ButterworthHeinemann; 1991, p. 320–36.


22. Magee LA, Downar E, Sevner M et al. Pregnancy outcomeafter gestational exposure to amiodarone in Canada. Am JObstet Gynecol 1995;172:1307–11.

23. James C, Banner T, Caton D. Cardiac output in womenundergoing caesarean section with epidural or generalanesthesia. Am J Obstet Gynecol 1989;160:1178–82.

24. De Paepe A, Devereux R, Dietz H et al. Revised diagnosticcriteria for the Marfan syndrome. Am J Med Genet 1996;62:417–26.

25. Child AH. Marfan syndrome — current medical and gen-etic knowledge: how to treat and when. J Card Surg1997;12(Suppl):131–6.

26. Child AH. Pregnancy management in Marfan syndrome andother connective tissue disorders. In: Oakley C, editor.Management of pregnancy in women with cardiac disease.London: BMA publications; 1997, p. 153–62.

27. Pyeritz R. Maternal and fetal complications of pregnancy inthe Marfan syndrome. Am J Med 1981;71:784–90.

28. Rossiter J, Repke JT, Morales AJ et al. A prospectivelongitudinal evaluation of pregnancy in women withMarfan syndrome. Am J Obstet Gynecol 1995;183:1599–606.

29. Lipscomb KJ, Smith JC, Clarke B et al. Outcome of preg-nancy in women with Marfan syndrome. Br J ObstetGynaecol 1997;104:201–6.

30. Erbel R, Alfonso F, Boileau C et al. Task force report:diagnosis and management of aortic dissection. Eur Heart J2001;22:1642–81.

31. Gott VL, Pyeritz RE, Magovern GJ Jr. et al. Surgical treat-ment of aneurysms of the ascending aorta in the Marfansyndrome: results of composite-graft repair in 50 patients.N Engl J Med 1986;314:1070–4.

32. Yacoub MH, Gehle P, Chandrasekaran V et al. Late results ofa valve-preserving operation in patients with aneurysms ofthe ascending aorta and root. J Thorac Cardiovasc Surg1998;115(5):1080–90.

33. Pinosky ML, Hopkins RA, Pinckert TL et al. Anaesthesia forsimultaneous cesarean section and acute aortic dissectionrepair in a patient with Marfan syndrome. J CardiothoracVasc Anesth 1994;8:451–4.

34. Pyeritz R, Fishman EK, Bernhardt BA et al. Dural ectasia is acommon feature of Marfan syndrome. Am J Hum Genet1988;43(5):726–32.

35. Steinmann B, Royce PM, Superti-Furga A. The Ehlers–Danlossyndrome. In: Royce PM, Steinmann B, editors. Connectivetissue and its heritable disorders. New York: Wiley; 1993,p. 351–457.

36. Pope FM, Narcisi P, Nicholls AC et al. COL3A1 mutationscause variable clinical phenotypes including acrogeria andvascular rupture. Br J Dermatol 1996;135:163–81.

37. Coady M, Davies R, Roberts M et al. Familial patterns ofthoracic aortic aneurysms. Arch Surg 1999;134:361–7.

38. Furthmayr H, Francke U. Ascending aortic aneurysm with orwithout features of Marfan syndrome and other fibrillinopa-thies: new insights. Semin Thorac Cardiovasc Surg 1997;9:191–205.

39. Iserin L, Jondeau G, Sidi D et al. Manifestations cardiovas-culaires et indications therapeutiques. Arch Mal Coeur Vaiss1997;90(Suppl):1701–5.

40. Chia YT, Yeoh SC, Lim MC et al. Pregnancy outcome andmitral valve prolapse. Asia Oceania J Obstet Gynaecol 1994;20:383–8.

41. Hagay ZJ, Weissman A, Geva D et al. Labor and deliverycomplicated by acute mitral regurgitation due to rupturedchordae tendineae. Am J Perinatol 1995;12:111–2.

42. Sheikh F, Rangwala S, DeSimone C et al. Management of theparturient with severe aortic incompetence. J CardiothoracVasc Anesth 1995;9:575–7.

43. Khandelwal M, Rasanen J, Ludormirski A et al. Evaluation offetal and uterine hemodynamics during maternal cardiopul-monary bypass. Obstet Gynecol 1996;88:667–71.

44. Hameed A, Karaalp IS, Tummala PP et al. The effect ofvalvular heart disease on maternal and fetal outcome duringpregnancy. J Am Coll Cardiol 2001;37:893–9.

45. Bryg RJ, Gordon PR, Kudesia VS et al. Effect of pregnancy onpressure gradient in mitral stenosis. Am J Cardiol 1989;63:384–6.

46. Avila WS, Grinberg M, Decourt LV et al. Clinical course ofwomen with mitral valve stenosis during pregnancy andpuerperium. Arq Bras Cardiol 1992;58:359–64.

47. Al Kasab SM, Sabag T, Al Zaibag M et al. Beta-adrenergicreceptor blockade in the management of pregnant womenwith mitral stenosis. Am J Obstet Gynecol 1990;163:37–40.

48. Butters L, Kennedy S, Rubin PC. Atenolol in essential hyper-tension during pregnancy. Br Med J 1990;301:587–9.

49. Clark SL, Phelan JP, Greenspoon J et al. Labor and deliveryin the presence of mitral stenosis: central hemodynamicobservations. Am J Obstet Gynecol 1985;152:984–8.

50. Jakobi P, Adler Z, Zimmer EZ et al. Effect of uterinecontractions on left atrial pressure in a pregnant womanwith mitral stenosis. Br Med J 1989;298:27.

51. Ducey JP, Ellsworth SM. The hemodynamic effects of severemitral stenosis and pulmonary hypertension during labor anddelivery. Intensive Care Med 1989;15:192–5.

52. Zitnik RS, Brandenburg RO, Sheldon R et al. Pregnancy andopen-heart surgery. Circulation 1969;39(Suppl 1):257–62.

53. Bernal JM, Miralles PJ. Cardiac surgery with cardiopulmo-nary bypass during pregnancy. Obstet Gynecol Surv 1986;41:1–6.

54. Sullivan HJ. Valvular heart surgery during pregnancy. SurgClin North Am 1995;75:59–75.

55. Rossouw GJ, Knott-Craig CJ, Barnard PM et al. Intracardiacoperation in seven pregnant women. Ann Thorac Surg 1993;55:1172–4.

56. Weiss BM, von Segesser LK, Alon E et al. Outcome of cardio-vascular surgery and pregnancy: a systematic review of theperiod 1984–1996. Am J Obstet Gynecol 1998;179:1643–53.

57. Levy DL, Warriner RA, Burgess GE. Fetal response to cardio-pulmonary bypass. Obstet Gynecol 1980;56:112–5.

58. Lamb MP, Ross K, Johnstone AM et al. Fetal heart monitoringduring open heart surgery. Two case reports. Br J ObstetGynecol 1981;88:669–74.

59. El Maraghy M, Abou Senna I, El-Tehewy F et al. Mitralvalvotomy in pregnancy. Am J Obstet Gynecol 1983;145:708–10.

60. Vosloo S, Reichart B. The feasability of closed mitral val-votomy in pregnancy. J Thorac Cardiovasc Surg 1987;93:675–9.

61. Pavankumar P, Venugopal P, Kaul U et al. Closed mitralvalvotomy during pregnancy. A 20-year experience. Scand JThor Cardiovasc Surg 1988;22:11–5.

62. Esteves CA, Ramos AIO, Braga SLN et al. Effectiveness ofpercutaneous balloon mitral valvotomy during pregnancy.Am J Cardiol 1991;68:930–4.

63. Patel JJ, Mitha AS, Hassen F et al. Percutaneous balloonmitral valvotomy in pregnant patients with tight pliablemitral stenosis. Am Heart J 1993;125:1106–9.

64. Iung B, Cormier B, Elias J et al. Usefulness of percutaneousballoon commissurotomy for mitral stenosis duringpregnancy. Am J Cardiol 1994;73:398–400.


65. Kalra GS, Arora R, Khan JA et al. Percutaneous mitralcommissurotomy for severe mitral stenosis duringpregnancy. Cathet Cardiovasc Diagn 1994;33:28–30.

66. Gupta A, Lokhandwala Y, Satoskar P et al. Balloon mitralvalvotomy in pregnancy: maternal and fetal outcomes. J AmColl Surg 1998;187:409–15.

67. Ben Farhat M, Gamra H, Betbout F et al. Percutaneousballoon mitral commissurotomy during pregnancy. Heart1997;77:564–7.

68. Mangione JA, Lourenco RM, dos Santos ES et al. Long-termfollow-up of pregnant women after percutaneous mitralvalvuloplasty. Catheter Cardiovasc Interv 2000;50:413–7.

69. De Souza JAM, Martinez EE, Ambrose JA et al. Percutaneousballoon mitral valvuloplasty in comparison with open mitralvalve commissurotomy during pregnancy. J Am Coll Cardiol2001;37:900–3.

70. Patel JJ, Munclinger MJ, Mitha AS et al. Percutaneous bal-loon dilatation of the mitral valve in critically ill youngpatients with intractable heart failure. Br Heart J 1995;73:555–8.

71. Lao TT, Sermer M, MaGee L et al. Congenital aortic stenosisand pregnancy—a reappraisal. Am J Obstet Gynecol 1993;169:540–5.

72. McIvor RA. Percutaneous balloon aortic valvuloplasty duringpregnancy. Int J Cardiol 1991;32:1–3.

73. Lao TT, Adelman AG, Sermer M et al. Balloon valvuloplastyfor congenital aortic stenosis in pregnancy. Br J ObstetGynaecol 1993;100:1141–2.

74. Banning AP, Pearson JF, Hall RJ. Role of balloon dilatation ofthe aortic valve in pregnant patients with severe aorticstenosis. Br Heart J 1993;70:544–5.

75. Ben-Ami M, Battino S, Rosenfeld T et al. Aortic valve re-placement during pregnancy. A case report and reviewof the literature. Acta Obstet Gynecol Scand 1990;69:651–3.

76. De Boer K, Ten Cate JW, Sturk A et al. Enhanced thrombingeneration in normal and hypertensive pregnancy. Am JObstet Gynecol 1989;160:95–100.

77. Chan WS, Anand S, Ginsberg JS. Anticoagulation of pregnantwomen with mechanical heart valves: a systematic reviewof the literature. Arch Intern Med 2000;160:191–6.

78. Sbarouni E, Oakley CM. Outcome of pregnancy in womenwith valve prostheses. Br Heart J 1994;71:196–201.

79. Sareli P, England MJ, Berk MR et al. Maternal and fetalsequelae of anticoagulation during pregnancy in patientswith mechanical heart valve prostheses. Am J Cardiol 1989;63:1462–5.

80. Gohlke-Barwolf C, Acar J, Oakley C et al. Guidelines forprevention of thromboembolic events in valvular heart dis-ease: Study Group of the Working Group on valvular heartdisease of the European Society of Cardiology. Eur Heart J1995;16:1320–30.

81. Prendergast B, Banning AP, Hall RJC. Valvular heart disease:recommendations for investigation and management. Sum-mary of guidelines produced by a working group of theBritish Cardiac Society and the Research Unit of the RoyalCollege of Physicians. J R Coll Physicians Lond 1996;30:309–15.

82. Bonow RO, Carabello B, De Leon AC et al. ACC/AHA guide-lines for the management of patients with valvular heartdisease. A report of the American College of Cardiology/American Heart Association Task Force on practiceguidelines. J Am Coll Cardiol 1998;32:1486–588.

83. Oakley CM. Pregnancy and prosthetic heart valves. Lancet1994;344:1643–4.

84. Elkayam UR. Anticoagulation in pregnant women with pros-thetic heart valves: a double jeopardy. J Am Coll Cardiol1996;27:1704–6.

85. Hanania G. Management of anticoagulants duringpregnancy. Heart 2001;86:125–6.

86. Vitale N, De Feo M, De Santo LS et al. Dose-dependent fetalcomplications of warfarin in pregnant women with mech-anical heart valves. J Am Coll Cardiol 1999;33:1637–41.

87. Salazar E, Izaguirre R, Verdejo J et al. Failure of adjusteddoses of subcutaneous heparin to prevent thromboembolicphenomena in pregnant patients with mechanical cardiacvalve prostheses. J Am Coll Cardiol 1996;27:1698–703.

88. Hanania G, Thomas D, Michel PL et al. Pregnancy andprosthetic heart valves: a French cooperative retrospec-tive study of 155 cases. Eur Heart J 1994;15:1651–8.

89. Menschengieser SS, Fondevila CG, Santarelli MT et al.Anticoagulation in pregnant women with mechanical heartvalve prostheses. Heart 1999;82:23–6.

90. Sadler L, McCowan L, White H et al. Pregnancy outcomesand cardiac complications in women with mechanical,bioprosthetic and homograft valves. Br J Obstet Gynaecol2000;107:245–53.

91. Montalescot G, Polle V, Collet JP et al. Low molecularweight heparin after mechanical heart valve replacement.Circulation 2000;101:1083–6.

92. Sanson BJ, Lensing AWA, Prins MH et al. Safety of low-molecular-weight heparin in pregnancy: a systematicreview. Thromb Haemost 1999;81:668–72.

93. Lee LH. Low molecular weight heparin for thromboprophy-laxis during pregnancy in 2 patients with mechanical mitralvalve replacement. Thromb Haemost 1996;76:628–9.

94. Tenconi PM, Gatti L, Acaia B. Low molecular weightheparin in a pregnant woman with mechanical heart valveprosthesis: a case report. Thromb Haemost 1997;79:733.

95. Ginsberg JS, Greer I, Hirsh J. Use of antithrombotic agentsduring pregnancy. Chest 2001;119:122S–31S.

96. North RA, Sadler L, Stewart AW et al. Long-term survivaland valve-related complications in young women with car-diac valve replacements. Circulation 1999;99:2669–76.

97. Piehler JM, Blackstone EH, Bailey KR et al. Reoperation onprosthetic heart values. Patient-specific estimates of in-hospital events. J Thorac Cardiovasc Surg 1995;109:30–48.

98. Kee WD, Shen J, Chiu AT et al. Combined spinal-epiduralanalgesia in the management of labouring parturients withmitral stenosis. Anaesth Intensive Care 1999;27:523–6.

99. Brian JE Jr., Seifen AB, Clark RB. Aortic stenosis, cesareandelivery, and epidural anesthesia. J Clin Anesth 1993;5:154–7.

100. McKenna R, Cole ER, Vasan V. Is warfarin sodium contra-indicated in the lactating mother? J Pediatr 1983;103:325–7.

101. Bac DJ, Lotgering FK, Verkaaaik APK et al. Spontaneouscoronary artery dissection during pregnancy andpostpartum. Eur Heart J 1995;16:136–8.

102. Rallings P, Exner T, Abraham R. Coronary artery vasculitisand myocardial infarction associated with antiphospholipidantibodies in a pregnant woman. Aust NZ J Med 1989;19:347–50.

103. Parry G, Goudevenos J-C, Williams DO. Coronary thrombo-sis postpartum in a young woman with Still's disease. ClinCardiol 1992;15:305–7.

104. Pearson GD, Veille J-C. Peripartum cardiomyopathy:National Heart, Lung and Blood Institute and Office of RareDiseases (National Institutes of Health) workshop recom-mendations and review. JAMA 2000;283:1183–8.


105. Van Hoeven KH, Kitsis RN, Katz SD et al. Peripartum versusidiopathic dilated cardiomyopathy in young women—acomparison of clinical, pathological and prognosticfeatures. Int J Cardiol 1993;40:57–65.

106. Elkayam U, Tummala PP, Rao K et al. Maternal and fetaloutcome of subsequent pregnancies in women with peri-partum cardiomyopathy. New Engl J Med 2001;344:1567–71.

107. De Souza JL Jr., de Carvalho Frimm C, Nastari L et al. Leftventricular function after a new pregnancy in patients withperipartum cardiomyopathy. J Card Fail 2001;7:36–7.

108. Bozkurt B, Villaneuva FS, Holubkov R et al. IntravenousImmune globulin in the therapy of peripartum cardio-myopathy. J Am Coll Cardiol 1999;34:177–80.

109. Purcell IF, Williams DO. Peripartum cardiomyopathy com-plicating severe aortic stenosis. Int J Cardiol 1995;52:163–6.

110. Oakley CM, Nihoyannopoulos P. Peripartum cardio-myopathy with recovery in a patient with coincidentalEisenmenger ventricular septal defect. Br Heart J 1992;67:190–2.

111. Oakley CM. Hypertrophic cardiomyopathy in heart diseasein pregnancy. New York: BMJ Publishing, 1997, p. 201–209.

112. Kazimuddin M, Vashist A, Basher AW et al. Pregnancyinduced severe left ventricular systolic dysfunction in apatient with hypertrophic cardiomyopathy. Clin Cardiol1998;21:848–50.

113. Redhead, Fadell EJ. Bacteraemia during parturition. JAMA1959;160:1284–5.

114. Baker TH, Hubbell R. Reappraisal of asymptomatic puer-peral bacteraemia. Am J Obstet Gynecol 1967;97:575–6.

115. Sugrue D, Blake S, Troy P. Antibiotic prophylaxis againstinfective endocarditis after normal delivery—is itnecessary? Br Heart J 1980;44:499–522.

116. Joglar JA, Page RI. Treatment of cardiac arrhythmias dur-ing pregnancy; safety considerations. Drug Saf 1999;20:85–94.

117. Natale A, Davidson T, Geiger MJ. Implantable cardioverterdefibrillators and pregnancy. A safe combination?Circulation 1997;96:2808–12.

118. Report of the National High Blood Pressure Education Pro-gram Working Group on high blood pressure in pregnancy.Am J Obstet Gynecol 2000;183:S1–S22.

119. Higgins JR, de Swiet M. Blood pressure measurement andclassification in pregnancy. Lancet 2001;357:131–5.

120. Walker JJ. Pre-eclampsia. Lancet 2000;356:1260–5.

121. Barton JR, O’Brien JM, Bergauer NK et al. Mild gestationalhypertension remote from term: progression and outcome.Am J Obstet Gynecol 2001;184:979–83.

122. Sibai BM, Mabie WC, Shamsa F et al. A comparison of nomedication versus methyl dopa or labetalol in chronichypertension during pregnancy. Am J Obstet Gynecol 1990;162:960–5.

123. Sibai BM. Treatment of hypertension in pregnant women.N Engl J Med 1996;335:257–65.

124. Magee LA, Ornstein MP, von Dadelszen P. Fortnightly re-view: management of hypertension in pregnancy. BMJ1999;318:1332–6.

125. Roberts JM, Cooper DW. Pathogenesis and genetics ofpre-eclampsia. Lancet 2001;357:53–6.

126. Cockburn J, Moar VA, Ounsted M et al. Final report of studyon hypertension during pregnancy: the effects of specifictreatment on the growth and development of the children.Lancet 1982;1:647–9.

127. Nifedipine versus expectant management in mild to mod-erate hypertension in pregnancy. Gruppo di Studio iperten-sione in Gravidanza. Br J Obstet Gynaecol 1998;105:718–22.

128. Magee LA, Schick B, Donnenfeld AE et al. The safety ofcalcium channel blockers in human pregnancy: a prospec-tive multicenter cohort study. Am J Obstet Gynecol 1996;174:823–8.

129. Horvath JS, Phippard A, Korda A et al. Clonidinehydrochloride—a safe and effective antihypertensive agentin pregnancy. Obstet Gynecol 1985;66:634–8.

130. Lain Y, Roberts JM. Contemporary concepts of the patho-genesis and management of pre-eclampsia. JAMA 2002;287:3183–6.

131. Hanssens M, Keirse MJ, Vankelecom F et al. Fetal andneonatal effects of treatment with angiotensin-convertingenzyme inhibitors in pregnancy. Obstet Gynecol 1991;78:128–35.

132. Heyborne KD. Pre-eclampsia prevention: lessons from thelow dose aspirin therapy trials. Am J Obstet Gynecol 2000;183:523–8.

133. Duley L, Henderson-Smart DE, Knight M et al. Anti-plateletdrugs for prevention of pre-eclampsia and its conse-quences: a systematic review. BMJ 2001;322:329–33.

134. Chambers JC, Fusi L, Malik IS et al. Association of maternalendothelial dysfunction with pre-eclampsia. JAMA 2001;285:1607–12.


Expert consensus document on management of … · Expert consensus document on management of...

Documents

Transcript of Expert consensus document on management of … · Expert consensus document on management of...