Impact of pregnancy in patients with systemic lupus erythematosus-associated pulmonary arterial hypertension: case series and literature review

Subjects

This single-centre retrospective study was conducted at Peking Union Medical College Hospital. Consecutive pregnancy cases complicated by SLE-PAH were collected from May 2009 to September 2020. A search for inpatient and outpatient data in the hospital information system of Peking Union Medical College Hospital and the Chinese Lupus Treatment and Research Group registry was performed. All applicable medical records for patients were reviewed. Cases were included if (1) the patient fulfilled the Systemic Lupus International Collaborating Clinics (SLICC) 2012 classification criteria for SLE; (2) the patient had a diagnosis of PAH with mean pulmonary artery pressure (mPAP) ≥25 mm Hg confirmed by RHC; and (3) PAH was diagnosed before or during pregnancy. A completed pregnancy included pregnancy with induced labour or uninduced labour (including liveborn infants by vaginal or caesarean and stillbirth). Planned pregnancy was defined as planning of the pregnancy by the patient with awareness of potential risks in coordination with family members and physicians. Extracted patient demographics included age, disease duration, whether the pregnancy was planned and autoantibody levels. Obstetric characteristics were collected including pregnancy history, time of pregnancies completed, mode of delivery and anaesthesia, and maternal and neonatal outcomes. Clinical evaluation of SLE was based on the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI), organ involvement, the American College of Rheumatology/SLICC Damage Index and Physician Global Assessment. Assessment of PAH was according to the European Society of Cardiology/European Respiratory Society (ECS/ERS) risk stratification, including related symptoms and their progression, WHO functional class, echocardiography results, and N-terminal pro-brain natriuretic peptide (NT-proBNP) levels at baseline and within 6 months after delivery.21–23

Literature review of pregnancy cases complicated by SLE-PAH

We conducted a literature search in PubMed before 7 August 2021, with combined terms of PAH and pregnancy (online supplemental figure 1). Studies were eligible if (1) the patient was diagnosed with SLE using a validated standard diagnosis; (2) the patient had a diagnosis of PAH confirmed by either RHC or echocardiography (mPAP ≥25 mm Hg); (3) PAH was diagnosed before or during pregnancy; (4) the study was an original study published in English, including a case report, case series, case–control or retrospective cohort study; and (5) detailed medical information could be obtained. The exclusion criteria were as follows: (1) pulmonary hypertension was due to other causes, such as idiopathic PAH, heritable PAH, and pulmonary hypertension secondary to cardiac and interstitial lung diseases; and (2) there is a lack of detailed clinical records of the patient, including WHO functional class or mPAP level, and maternal outcomes. Eleven studies were finally included in the literature review, seven of which were case reports and four were retrospective studies; relevant medical information was then extracted.

Statistical methods

Descriptive and continuous data are shown in the tables. Descriptive data were presented as percentages. Continuous data were expressed as mean±SD or median (range) if they do not conform to a normal distribution. Data were analysed using SPSS V.23.0 and Prism V.9 (GraphPad Software, La Jolla, California, USA).

Baseline characteristics

Ten pregnant women who were diagnosed with SLE-PAH between April 2012 and August 2019 were identified at our hospital. Patient baseline characteristics and disease evaluation results are shown in table 1. The average age at the patients’ first visits during pregnancy was 30.0±5.7 years. The median disease duration of SLE and PAH at baseline was 34.5 (range 1–164) months and 2 (1–51) months, respectively. Among these 10 patients, 2 planned their pregnancy with a full understanding of the risks and possible harm, 7 patients developed PAH during their pregnancy, and 1 pregnancy was unplanned (case 5). Nine cases were multigravida, and five primiparas were enrolled in this study. None of these women experienced previous pregnancy-related complications, such as pre-eclampsia or intrauterine growth restriction. Detailed pregnancy histories are reported in table 1.

All cases were positive for ANA, seven were positive for anti-Sjogren's syndrome antigen A (anti-SSA), two were positive for anti-Sjogren's syndrome antigen B (anti-SSB) and two were positive for anti-double strand DNA (anti-dsDNA). Nine patients with reported laboratory results were all negative for antiphospholipid antibodies, including anticardiolipin, anti-beta(2)-glycoprotein I and lupus anticoagulant. Nine patients had low disease activity, with SLEDAI between 0 and 4. The SLEDAI score of case 1 was 14, indicating moderately active SLE.

Right heart function was assessed using clinical features and laboratory data. The percentages of patients with WHO functional class I, II, III and IV were 0%, 30%, 30% and 40%, respectively. RHC is the gold standard for the diagnosis of PAH and was used to evaluate PAH at all of the patients’ first visits. As listed in table 1, all patients had an elevated mPAP (≥25 mm Hg) and pulmonary vascular resistance (PVR ≥3 Woods unit, WU). The median mPAP, pulmonary artery wedge pressure, cardiac output and PVR were 35 (25–66) mm Hg, 7.5 (2–16) mm Hg, 5.3 (1–6.5) L/min and 5.67 (3.08–31.00) WU, respectively. Echocardiography was used to evaluate PAH and monitor patients’ cardiac function during pregnancy as it is a non-invasive measurement. The median pulmonary arterial systolic pressure (PASP) was 75 (39–102) mm Hg. The median tricuspid annular plane systolic excursion (TAPSE) was 16 (10–23) mm, indicating impaired right ventricular function in most patients. However, dilation of right ventricle was not observed in any of the patients and no reduction in ejection fraction (EF) was found in nine patients with recorded EF. Only one patient (case 2) was confirmed to have a moderate pericardial effusion via echocardiography. NT-proBNP levels were elevated in 70% of patients, with a median of 776 (56–18 023) pg/mL in all cases. Detailed baseline characteristics are presented in table 1.

Maternal and neonatal outcomes

Nine of 10 patients were administered corticosteroid therapy during pregnancy. Immunosuppressive agents, PAH-targeted medications and supportive treatments were adjusted according to patients’ manifestations and drug teratogenicity around pregnancy (online supplemental table 1).

Four out of 10 patients underwent induced labour. Among the other six pregnancies without induced labour (including both planned pregnancies, cases 3 and 9), five patients delivered their neonates via caesarean section and one delivered vaginally. The median time of pregnancies completed in all patients, patients with induced labour and patients without induced labour was 31 (15–38) weeks, 22 (15–25) weeks and 33 (31–38) weeks, respectively. Of the 10 patients, 40% were administered general anaesthesia and 60% were administered continuous epidural anaesthesia.

Table 2 shows that most patients in our study benefited from completed pregnancies regardless of its form. In general, SLE activity and PAH severity improved in 70% of patients and remained stable in 20% of patients within 6 months after delivery. One patient (case 6) died of shock caused by deterioration of cardiac function and infection 15 days after induced labour. In the other patients, SLEDAI remained 0–4, no disease flared and no new organ was involved within 6 months after delivery, indicating a lupus low disease activity state. According to ECS/ERS risk stratification, all alive patients were categorised at a lower risk state compared with their risk stratification during pregnancy, except for cases 3 and 5 who remained at intermediate risk. Regarding cardiac function evaluated using WHO functional class within 6 months after delivery, 80% of patients exhibited mild impairments (WHO functional class I or II). The median PASP of all surviving patients was 59 (23–83) mm Hg, indicating the improvement of PAH in this group. The median TAPSE was 15 (14–21) mm, which was similar to the level at baseline. Pericardial effusion could not be detected after delivery in case 2. In all surviving patients, the median of NT-proBNP level decreased from 293 (56–9098) pg/mL at baseline to 184 (32–4003) pg/mL within 6 months after delivery. Decreases in NT-proBNP levels were especially remarkable in patients with NT-proBNP levels higher than 10-fold upper limit of normal (ULN, 125 pg/mL) at baseline. It was also observed that the levels of NT-proBNP were lower than threefold ULN in eight out of nine surviving patients (figure 1A). However, several results of echocardiography indicated deterioration of cardiac function within 6 months after delivery. Right ventricular dilation was not detected in most patients, except for case 4, whose right ventricular transverse diameter was 50 mm. It was notable that reduction in EF was found in two out of eight patients with recorded EF.

Request permission

Despite prematurity and low birth weight in most of six neonates, the 5 min Apgar scores of all infants were in the normal range (7–10). Five of six infants were admitted to the neonatal intensive care unit for 4–30 days due to preterm birth and/or low birth weight. There was no neonatal mortality, and no severe diseases were reported in these children until the last follow-up in January 2022.

Literature review

A review of the literature included 14 pregnancy cases complicated by SLE-PAH in 11 studies from 2002 to 2021.13–18 24–28 Most patients (except for the case in Zhou et al24) delivered via caesarean section (76.92%); the others delivered vaginally (23.08%). The low proportion of induced labour was the most distinct characteristic of the cases in the literature compared with cases at our hospital. However, these patients shared some similar clinical features with the patients at our hospital. The average age of patients from our literature review was 30.09±5.37 years. The median time of their completed pregnancies was 36 (28–40) weeks, which was similar to the data for patients with uninduced labour at our hospital. Further, more pregnancies were planned in the 14 reviewed cases. Regarding obstetric outcomes, the survival rates of mothers and neonates were 85.71% and 92.86% respectively. Assessment of cardiac function at baseline and/or after delivery, including WHO functional class and levels of mPAP and NT-proBNP, was reported in some of the cases from the reviewed literature. WHO functional class at baseline and after delivery was available in seven cases in the literature and ten cases at our hospital. Figure 1B illustrates that in these 17 patients, the percentage of patients with WHO functional class I and II rose from 41.18% to 76.47%. Further clinical characteristics, outcomes and assessment of cardiac function are displayed in table 3.