Trends in Pregnancy Outcomes in Patients with Pulmonary Hypertension: Still a Long Way to Go

This article refers to ‘Pulmonary hypertension and pregnancy outcomes: data from the Registry Of Pregnancy and Cardiac Disease (ROPAC) of the European Society of Cardiology’, by K. Sliwa et al., published in this issue on pages 1119–1128. Pulmonary hypertension (PH) is an abnormal haemodynamic condition defined by a mean pulmonary arterial pressure ≥25 mmHg at rest as measured by right heart catheterization.1 Rather than being a disease itself, PH is the result or complication of a broad spectrum of cardiac, (pulmonary) vascular, and respiratory diseases, and is most of all an indicator of an adverse outcome during the clinical course of these entities. Due to this heterogeneity, the diseases leading to PH have been classified into five groups based on pathological findings, clinical presentation, haemodynamic constellation, and therapeutic strategy.1, 2 Although the majority of patients with PH are classified into group 2 (PH due to left heart disease) or 3 (PH due to lung diseases and/or hypoxia), the PH group that has received most attention since the beginning of this century is pulmonary arterial hypertension (PAH; group 1), a pre-capillary form of PH for which specific effective treatment has become available. The haemodynamic fluctuations accompanying pregnancy, and particularly those during labour, delivery, and the early post-partum period, may lead to intractable right heart failure and have lethal consequences for patients with any form of severe PH. In early case reports of patients with PAH, mortality during pregnancy or in the puerperium was absolutely prohibitive, reaching levels as high as 56%.3 Even in the first years following the introduction of specific PAH treatment, maternal mortality rates of up to 36% persisted, as documented in a series of 15 consecutive patients.4 More recently, however, trends towards declining mortality rates (between 17% and 33%) were published,5, 6 and a multinational 3-year prospective survey in 13 centres with 26 reported pregnancies found that ‘only’ 12% of the patients died and another 4% required urgent heart–lung transplantation.7 In the latter study, very high (>1500 dyn s−1 cm−5) levels of pulmonary vascular resistance and emergency caesarian section under general anaesthesia were associated with the worst outcomes, but the authors concluded that pregnancies can overall be successfully managed in patients with well-controlled PAH.7 While acknowledging the progress made in the management of P(A)H and the positive trends in outcomes as suggested by the above-mentioned data, current guidelines1, 8 and consensus documents9 continue to emphasize the high level of pregnancy-associated risk for mother and child, and strongly recommend effective contraception to avoid pregnancy, particularly in the setting of group 1 PH. Now, in the present issue of the journal, the authors of the multinational Registry Of Pregnacy and Cardiac Disease (ROPAC) describe the outcome of 151 pregnant patients with various forms (groups 1 and 2) of PH; this is the largest population yet reported.10 The results of the registry appear impressive: no deaths during pregnancy and very low (3.3%) mortality up to 1 week after delivery, with another 2.6% of the followed patients dying within 6 months. Do these registry results finally indicate a breakthrough in ensuring good or at least acceptable pregnancy outcomes among patients with PH? Hardly. Do they imply that the warning to avoid pregnancy in patients with PH and particularly PAH is outdated and therefore no longer to be taken ‘so seriously’? By no means. Being a global registry in countries with different standards of care, the present study unavoidably has numerous important limitations that must be emphasized, since they dilute the results and preclude definitive conclusions. The most important limitation lies in the fact that, in many of the cases, PH was ‘roughly’ estimated by echocardiography alone using a rather arbitrary criterion (right ventricular systolic pressure above 30 mmHg), without confirmation by right heart catheterization and also without central adjudication of the echocardiograms. Consequently, the patients' haemodynamic profile at baseline remains obscure to a large extent, and major flaws in the classification, severity assessment, and even the diagnosis of PH itself are indeed likely. Without catheterization, it is not possible to reliably distinguish pre-capillary from post-capillary or combined forms of PH.2 Moreover, the estimated elevation of systolic pulmonary artery pressure, which apparently was mild in most cases, is not a substitute for crucial prognostic parameters such as cardiac index and pulmonary vascular resistance. There are further caveats beyond the above methodological issues. First, the results of ROPAC cannot be directly compared with those of previous series, since only a minority of patients were classified as having PAH in the present registry; in fact, mortality rates were disproportionately high (43%) in the very few patients with confirmed (or presumed) idiopathic PAH. Secondly, pregnancies had a seriously complicated (even if ultimately non-fatal) course in many of the patients followed. For example, heart failure occurred in 28% and thrombo-embolic events in 3.2% of the women during pregnancy, caesarean section was performed in 63% (in 24% it was an emergency procedure), and ‘therapeutic’ abortion was necessary in 4.0%. Importantly, it should also not go unnoticed that foetal or neonatal death was recorded in as many as 10% of pregnancies, with another 12 neonates (8%) having cardiac or other anomalies.10 So, what can we learn from ROPAC, the latest observational study of pregnancy outcomes in PH, keeping all the above limitations and caveats in mind? First, and above all, that the recommended algorithms and strategies which have been elaborated in current guidelines1 and aim at the correct and definitive diagnosis, classification, and risk stratification of patients with PH, must be strictly followed. If this is true for every patient with suspected PH, it is absolutely imperative for women at childbearing age, since effective contraception should be advised and the interruption of pregnancy should be discussed with the patient, particularly if she is diagnosed with PAH and is at high risk based on her clinical and haemodynamic status. Secondly, if pregnancy is continued, the patient should immediately be referred to an expert PH centre for interdisciplinary follow-up and care, which should include the measures outlined in Table 1. Although these measures are still far from guaranteeing a favourable outcome for mother and child, they can help to reduce life-threatening right heart decompensation and, possibly, neonatal death. Finally, prospective observational studies such as ROPAC should continue to enrol pregnant patients, since the analysis of increasingly large, properly diagnosed and treated populations improves our experience and efficacy in managing this particularly vulnerable, young patient group. Anticoagulation needed in third trimester; also earlier if patient is bedridden or hospitalized No direct oral anticoagulants in pregnancy! Weekly visits in the puerperium, particularly in high-risk patientsa. Follow-up should be long enough: reversal of pregnancy-related volume expansion may take up to 6 months! The work of S.V.K. was supported by the German Federal Ministry of Education and Research (BMBF 01EO1003). The author is responsible for the contents of this publication. Conflict of interest: The author reports having received lecture fees and advisory board honoraria from Bayer Health Care, Boehringer Ingelheim, Pfizer–Bristol-Myers Squibb, Daiichi-Sankyo, and Actelion; and research grants to his institution from Bayer Health Care, Boehringer Ingelheim, and Actelion.