The hallmark of PH is an increase in workload placed on the right side of the heart. The precise pathological changes leading to this are dependent on disease etiology. In IPAH (formerly known as primary PH), typical changes are seen in the small arterioles, with medial smooth muscle hypertrophy, thickening or fibrosis of the intima of the vessels with in situ thrombosis and, in some cases, plexiform lesions (histological features shared with other forms of PAH). In CTEPH, the increase in pulmonary vascular resistance is brought about partly by obstruction of the pulmonary vascular bed due to an “organized” thrombus that has become incorporated into the vessel wall, although, interestingly, these patients also have the histological changes of PAH. This is the rationale for treating some of these patients with pulmonary arterial vasodilators as a bridge to pulmonary endarterectomy or as a definitive treatment in inoperable disease. In patients with left-sided heart disease (such as impaired left ventricular function or mitral valve disease), the primary problem is that of pulmonary venous hypertension. In these patients, the clinical features and treatment is that of the underlying cardiac problem rather than directing therapy at the pulmonary arterial circulation. Pulmonary vaso-occlusive disease (previously known as pulmonary veno-occlusive disease, but renamed to reflect the involvement of both pulmonary arterial and venous circulations) is a rare cause of PH that has a particularly poor prognosis. This may be worsened by pulmonary arterial vasodilators that can precipitate pulmonary edema.
This chapter concerns the management of PH involving the precapillary circulation (i.e. PAH and CTEPH) and does not discuss the management of those patients with postcapillary PH (PH associated with left-sided heart disease), for which the natural history and management differ.
The cardinal symptoms of PH are fatigue and breathlessness due to inability of the right heart to generate a sufficient cardiac output. Initially, this may be mild and occur only on strenuous exertion, but it is progressive and may later be accompanied by chest pain (similar to angina, reflecting right ventricular ischemia) and presyncope or syncope on exercise. Syncope usually reflects a low cardiac output and indicates severe disease. As right heart failure develops, patients may develop ascites and ankle edema. It should be noted that ankle swelling occurs very late in the natural history of the disease, and in young patients may never occur. In patients with right to left shunts, either via a patent foramen ovale or intracardiac defect, “classical features” of right heart failure may not appear despite severe elevations of pulmonary artery pressure. Signs on examination may initially be very few, and PH may be difficult to identify in the pregnant patient in whom a hyperdynamic circulation, loud second heart sound, and flow murmurs are common (Table 15.1).
|Elevated jugular venous pressure|
|Right ventricle heave|
|Pansystolic murmur from tricuspid regurgitation|
|Pulmonary diastolic murmur from pulmonary regurgitation|
|Hepatomegaly ± pulsation|
P2 = pulmonary component of second heart sound; PH = pulmonary hypertension
The nonspecific nature of the symptoms and the subtle nature of the signs of pulmonary vascular disease often delay diagnosis. Chest X-ray and/or electrocardiography are said to be abnormal in >80% of patients with established disease. The most common noninvasive investigation suggesting PH is transthoracic echocardiography (TTE). With TTE, an estimate of the systolic pulmonary artery pressure can be derived from the jet of tricuspid regurgitation when present. This is calculated using the modified Bernoulli equation (p=4×v2, where, “p” = the pressure gradient between the right atrium and ventricle and “v” = the peak tricuspid jet velocity) with an added estimate of right atrial pressure. In addition, other features may suggest PH (dilated right-sided chambers, right ventricular hypertrophy) and other causes of PH can be diagnosed (valvar heart disease, left ventricular dysfunction, intracardiac shunts). Subsequent investigation aims to confirm or refute the presence of PH and, where present, identify the etiology and severity. This requires extensive investigation, including detailed imaging investigations such as computed tomography scanning and magnetic resonance imaging (which in addition to identifying a cause of PH may also be helpful in defining severity and prognosis [Table 15.2] [9–12]), and usually right heart catheterization. In pregnancy, one may instinctively wish to delay investigation and have concerns regarding radiation exposure and the finite risk of invasive investigation, particularly when the mother wishes to continue with the pregnancy regardless of the risk to her health. However, it should be remembered that PH in pregnancy carries a grave prognosis and having more information available will help in making difficult management decisions. The need for investigation can be tailored depending on the individual patient’s presentation. Where features are classical, this may negate the need for a standard battery of investigations. It is key that PH patients who are pregnant are managed in centers where there is experience in the management of PH and pregnancy.
|Ventilation perfusion scanning|
|Lung hypofractionated conformal radiotherapy|
|Contrast helical CT pulmonary arteries|
|Cardiac magnetic resonance imaging and angiography|
|Pulmonary angiogram (in selected cases)|
|Arterial blood gases|
|Nocturnal oxygen saturation monitoring|
|Exercise test (6-min walk/shuttle)|
|Routine hematology and biochemistry|
CT = computed tomography; HIV = human immunodeficiency virus
PH is a challenging disease to diagnose, accurately classify, and treat. IPAH is a severe, progressive disease with an incidence of 1–2 cases per million of population per year and is three times more common in women. With “conventional” treatment, it has a median survival time of 2.8 years from diagnosis; however, with new therapies there has been a doubling of the survival time. Younger patients have seen the greatest improvement in outcome, with 5-year survival approximating 80% in patients with IPAH. In systemic sclerosis, PH has a profound effect on the prognosis, with a 2-year survival rate of 80% in the absence of PH, in contrast to 40% with the development of PH. In the setting of congenital heart disease (CHD), the presence of PH has an adverse effect on prognosis, although the presence of intracardiac defects (allowing blood to shunt from the right to left side in the face of high pulmonary artery pressures, thereby off-loading the right ventricle) and right ventricular hypertrophy confer a survival benefit compared to that of other forms of PAH.
Until the advent of transplantation in the 1980s, no specific treatment was available for patients with PH. The last two decades have seen the development of novel therapies that have been shown to improve the symptoms and survival of patients with PH. Patients with severe disease have a 5-year survival rate of only 27% with supportive treatment, which increases to 54% with certain targeted therapies. These therapies include orally active agents such as endothelin receptor antagonists (bosentan, macitentan, and ambrisentan ), phosphodiesterase inhibitors (such as sildenafil  and tadalafil ), and the cyclic GMP stimulator, riociguat. More complex therapies include prostaglandin therapy nebulized (iloprost ), subcutaneous (treprostinil ), or intravenous (epoprostenol [prostacyclin; 15], iloprost,  and treprostinil ). There is growing experience with the use of these therapies in pregnancy, although a number, such as endothelin receptor antagonists, are contraindicated because of concerns regarding teratogenicity.
Targeted PH treatments are often complex, and their use and monitoring requires significant expertise. As such, the investigation and treatment of certain forms of PH (particularly PAH and CTEPH) are often focused at nationally designated specialist centers. In the UK, centers currently exist in Sheffield, London (Royal Brompton, Hammersmith and Royal Free), Cambridge (Papworth), Newcastle, and Glasgow. There is also a specialist center in Dublin, Ireland.
The pre-existence or gestational occurrence of PH is considered to pose an extreme risk of maternal death because of the additional cardiovascular stresses that pregnancy places on an overburdened right ventricle. Although the physiological effects of pregnancy on the cardiovascular system are well documented, the low prevalence of PH coupled with the historically low survival of such patients means that clinical experience is very limited, particularly for IPAH. The sporadic nature of this illness and (until recently) the fragmented nature of the delivery of specialized pulmonary vascular care around the world means that reports have been largely anecdotal. Even in large pulmonary vascular centers, experience has been relatively limited. Owing to the improved prognosis of patients with Eisenmenger syndrome compared with those wth IPAH, much of the historical literature is based on this patient population, in which the physiology is very different from in IPAH; however, even in this group, the data are limited and point to a poor prognosis. Although systematic reviews exist on the management of pregnancy in patients with PH, there are inherent flaws in interpreting such a varied data set.
Over the last decade, with growing experience of targeted PH therapies in pregnancy, there is increasing consensus among the PH community regarding the management of PH in pregnancy. The Pulmonary Vascular Research Institute (an international organization aiming to improve the care of patients worldwide) is publishing a consensus statement on the management of PH in 2015.
Pregnancy results in major hormonal and hemodynamic changes to meet the metabolic demands of both the fetus and the mother. Hormonal changes, and the release of various vasoactive substances, result in a reduction in systemic and pulmonary vascular resistance in healthy pregnancy. The blood volume usually exceeds the nonpregnant level by 10% in the 8th week, reaching a peak of 40–50% above prepregnancy levels between the 32nd and 36th weeks, and then remaining unchanged until full term. Oxygen consumption rises by approximately 30% during pregnancy and cardiac output increases steadily as a result of an increase in both heart rate and stroke volume throughout pregnancy. Cardiac output reaches a peak at approximately 28 weeks, and in the final trimester may plateau, increase, or slightly decrease until the onset of labor. Each uterine contraction increases the cardiac output by a further 10–40%, resulting in a total increase of 60–80% above prepregnancy levels. This can be attenuated but not abolished by analgesia. These changes in cardiac output are affected by body posture, particularly in the third trimester. Delivery results in aortocaval decompression and the redistribution of blood volume, with a transient increase in cardiac output, although variable amounts of blood loss due to either vaginal delivery or cesarean section leads to unpredictable changes in circulating blood volume, cardiac output, and blood pressure. Whereas blood volume rapidly returns to normal, it takes up to 2 weeks for hemodynamics to return to prepregnancy levels after vaginal delivery and takes longer after cesarean section. It may take more than 6 months for subtle hemodynamic changes to return to normal.
In patients with severe PAH and a low cardiac output state (<3 l/min in some cases), the demands of increasing blood volume and cardiac output may not be met by an already compromised right ventricle. In addition, it is not known what direct effects pregnancy may have on the pulmonary vasculature, and whether hormonal changes or the release of vasoactive mediators may have adverse effects on pulmonary vascular tone.
It is clear that patients with severe PAH may fail to cope with these metabolic demands and be unable to respond appropriately from a hemodynamic perspective. In a series from France, including three patients who deteriorated between 12 and 23 weeks, two patients died and one patient survived following interruption of pregnancy. Deterioration early in pregnancy undoubtedly reflects an inability to cope with the hemodynamic requirements of pregnancy and/or intense pulmonary vasoconstriction occurring as a direct consequence of pregnancy. Supporting a role for the latter, at least in part, is the observation of hemodynamic improvement following prostaglandin therapy in some of these patients while continuing with the pregnancy.
The peri-/postpartum period is also a fraught time for these patients. A significant mortality risk exists immediately following delivery, which most probably reflects the large changes in blood volume (with concomitant shifts in fluid) and the inability of a compromised cardiovascular system to respond to this. Indeed, volume overload in the immediate postpartum period can result in a negative Starling response, which may be avoided by augmenting the physiological diuresis following delivery.
In an effort to minimize risks to the mother, a number of different approaches have been taken regarding timing of hospital admission, timing of delivery, and both mode of delivery (vaginal vs cesarean) and the choice of anesthesia (local vs general). The approach taken, however, is likely to have reflected the individual needs of the patient as well as the preference of the supervising clinician. The lack of systematic prospective evaluation, therefore, does not allow any definite conclusions to be made regarding the benefits of each strategy. It is clear, however, that even with expert care the mortality rate remains high and had not changed dramatically in the three decades preceding the use of targeted drug therapies.[26,31,33,34]
In a retrospective review spanning four decades to 1978 Gleicher et al. found that 30% of all pregnancies in patients with Eisenmenger syndrome resulted in maternal death. Lower mortality rates have been seen in single-center or two-center studies, with an 11% mortality in the London–Torino study,[35,36] similarly low mortality rates in an Indian study, and survival of all five patients with Eisenmenger syndrome in a study of eight patients in Canada. These series raised the possibility that mortality in this form of PH had improved, particularly when the experience was concentrated in large centers.
Weiss et al. systematically reviewed all journals and book chapters in MEDLINE from 1978 to 1996. They identified reports of 73 patients with Eisenmenger syndrome, 30 patients with primary PH (now renamed IPAH), and 25 patients with “secondary vascular PH” (now reclassified). This represents a large but heterogeneous population from many countries and centers. Unfortunately, in contrast to a prospective study, no validation of the diagnosis was possible and there was no suggestion of uniformity in clinical practice. These investigators identified a mortality rate of 36% for Eisenmenger syndrome, of 30% for primary PH, and of 56% in patients with secondary vascular hypertension. In those with Eisenmenger syndrome, the majority delivered at term (47%), 33% delivered between 32 and 36 weeks, and 20% delivered at <31 weeks. Three patients died during pregnancy but mortality was highest postpartum. Twenty-three patients died within 30 days of delivery, with 14 patients dying at between 2 and 7 days following delivery. The cause of death was described as pulmonary hypertensive crisis and therapy-resistant heart failure in 13 cases, and as sudden death in 7. Autopsy identified thromboembolism in three cases, although it should be noted that chronic thrombosis may often be seen in proximal pulmonary arteries in clinically stable patients with long-standing Eisenmenger syndrome. This does not necessarily imply that these patients died of acute thrombosis. In patients with “primary PH,” 41% delivered at term, 37% at between 32 and 36 weeks, and 22% before 31 weeks. All eight deaths in this group within 35 days of delivery were due to “therapy resistant heart failure,” most at 2–7 days postdelivery. The mortality rate of patients with “secondary vascular PH” was high at 56%, although this group was very heterogeneous, including patients with connective tissue disease and also those with unusual disease processes such as Takayasu arteritis, peripheral pulmonary artery stenosis, hepatitis, and dwarfism with congenital hypothyroidism. This is a very different population of patients from those usually encountered with associated precapillary forms of PH in a pulmonary vascular clinic. Mortality was high in the immediate postpartum period, with 6 out of 14 deaths occurring within the first 24 h. Death was also due to progressive “heart failure” in seven patients and was sudden in the remaining seven.
Interestingly adverse events were noted in two patients in the Eisenmenger’s group who received oxytocic drugs, and “pulmonary hypertensive crises” were also noted in one patient from each of the other two groups. This class of drugs does have a potentially adverse effect on the pulmonary vasculature. Although oxytocin can be safely used in selected individuals, its use should be carefully controlled and small boluses given initially to assess the effect on systemic and pulmonary vasculatures.
Weiss et al. acknowledge the study’s limitations but identified a number of risk factors for death in this patient population. These include late diagnosis, elevated diastolic pulmonary artery pressure, and late hospital admission. It is difficult to validate the importance of these factors because the data set is incomplete, with a variety of confounding factors and physician preference at different centers influencing treatment. In particular, although operative delivery is identified as a risk factor for maternal death, this may simply reflect the fact that more symptomatic patients were delivered early because of the severity of their underlying condition. In addition, although the postpartum period is well recognized as a period in which patients are at a high risk of acute decompensation, it is unusual that death during pregnancy was rare. This may partly reflect publication bias.
A major pulmonary vascular center in France published a retrospective review of patients managed at their center between 1992 and 2002. Again, a heterogeneous population of patients was identified but, not surprisingly, the disease spectrum was more in keeping with that seen by physicians with a primary interest in pulmonary vascular disease. Of 14 women and 15 pregnancies managed during this period, 6 cases were associated with CHD, 4 with IPAH, and 2 with CTEPH, and 1 case each of mixed connective tissue disease associated PAH, HIV-associated PAH, and fenfluramine-associated PAH. In this population, two patients died during pregnancy at 12 and 23 weeks, and another who had deteriorated had a therapeutic abortion at 21 weeks and subsequently improved. All of these three patients had PAH with no evidence of an intracardiac shunt (which may be expected to confer a survival advantage in the setting of rising pulmonary artery pressures). There were three additional deaths at 1 and 3 weeks and at 3 months following delivery. The death at 1 week following delivery was in a patient with a systemic to pulmonary shunt. Importantly, the five patients who remained stable during pregnancy were in NYHA Class II with a relatively well-preserved exercise capacity, whereas the patients who deteriorated during pregnancy and died were either in NYHA Class III prior to pregnancy or presented for the first time during pregnancy in NYHA class IV (breathless at rest). Only one patient in this study was on targeted pulmonary vascular therapy with an oral prostaglandin analog, although three patients who deteriorated in the postpartum period were treated with intravenous epoprostenol (prostacyclin). In this series, a number of approaches were used to manage delivery: cesarean section and regional anesthesia in five cases; cesarean section and general anesthesia in four cases; and vaginal delivery with regional anesthesia in four cases. There was no significant difference in outcome using these different approaches. The authors concluded that despite expert treatment in a specialist center PH complicating pregnancy had a high mortality rate at 36% and that, although a scheduled cesarean delivery using a combined spinal/epidural remained an attractive option, there was no evidence of actual benefit.
In summary, a review of published series and systematic reviews going back over 50 years of managing pregnant patients with PH using conventional therapy found that maternal mortality of patients with PH remains high. Authors suggest that the mortality was unchanged over recent decades; however, large series in centers expert in the management of CHD report a lower mortality rate than that seen in a systematic review of the experience of various centers. This seems to support the strongly held belief that expert management and assessment can improve the prognosis of these patients.
A retrospective review of the literature (1997–2007) by Bedard et al.  compared the outcome of patients with PH managed during this era with those described by Weiss et al. (1978–1996). This study identified 73 parturients, emphasizing that this is a relatively uncommon condition, but overall noted a significant improvement in outcome with a mortality of 25% compared with 38% (p=0.04) for all forms of PH, with a maternal mortality rate of 17% in IPAH, 28% in PAH-CHD, and 33% in other forms of PH. This study confirmed the findings of historical studies that the highest mortality occurs in the first week following delivery. Interestingly, they found that general anesthesia was associated with a higher mortality rate compared with regional anesthesia (OR 4.37, p=0.02), primigravidae had a higher mortality rate (OR 3.70, p=0.03); there was a higher proportion of premature delivery and a lower proportion of vaginal delivery, reflecting changes in practice. Furthermore, they found no improvement in outcome in patients receiving targeted therapy, although for many of these patients this included using therapies such as nitric oxide in the peripartum period. The authors concluded that patients with PAH who become pregnant warrant a multidisciplinary approach and consideration of advanced/targeted PAH therapies.
A literature review of the outcome of pregnancy in patients with PH found that despite different approaches mortality remains high, particularly in patients with severe PAH. Interestingly, patients may deteriorate early (usually in the second trimester, when the mortality risk without the intervention of early delivery is extremely high, presumably reflecting an inability to cope with the increasing cardiovascular demands) or in the immediate postpartum period (when fluid shifts and withdrawal of the vasodilatory effects of pregnancy may result in more intense vasoconstriction). Although the mode of death is variable in patients who deteriorate during pregnancy, the vast majority of those dying in the immediate postpartum period are identified as dying from a “pulmonary hypertensive crisis” or “heart failure.” This raises the possibility of directing treatment directly at the pulmonary vasculature to reduce the degree of pulmonary vasoconstriction.
Over the last 5 years since the retrospective review of the literature by Bedard et al., a number of publications have summarized the experiences of various centers around the world that use multidisciplinary approaches and targeted drug therapies for PAH. These include retrospective reviews summarizing single-centre and multicenter experiences and a prospective registry of pregnant patients with PH (Table 15.3).[39–49]
|Variable||Kiely et al. (2010) ||Rosengarten et al. (2012) ||Jais et al. (2012) ||Duarte et al. (2013) |
|Time period||2002–2009||Not stated||2007–2010||1999–2009|
|Design||Retrospective, single center||Retrospective, single center||Prospective, multicenter: 18 centers||Retrospective, multicenter: 5 centers|
|Continued with pregnancy||12||9||20||12|
|Miscarriages||2||0||2 (2 deaths)||0|
|Maternal deaths or requiring transplant||1||2||4||2|
CS = cesarean delivery; GA = general anesthesia; LA = local anesthesia; NVD = normal vaginal delivery; PAH = pulmonary arterial hypertension