- Open Access
Prediction and prognosis of adverse maternal and foetal/neonatal outcomes in pulmonary hypertension: an observational study and nomogram construction
Respiratory Research volume 23, Article number: 314 (2022)
Pregnant women with pulmonary hypertension (PH) have higher mortality rates and poor foetal/neonatal outcomes. Tools to assess these risk factors are not well established.
Predictive and prognostic nomograms were constructed using data from a “Development” cohort of 420 pregnant patients with PH, recorded between January 2009 and December 2018. Logistic regression analysis established models to predict the probability of adverse maternal and foetal/neonatal events and overall survival by Cox analysis. An independent “Validation” cohort comprised data of 273 consecutive patients assessed from January 2019 until May 2022. Nomogram performance was evaluated internally and implemented with online software to increase the ease of use.
Type I respiratory failure, New York Heart Association functional class, N-terminal pro-brain natriuretic peptide \(\ge\) 1400 ng/L, arrhythmia, and eclampsia with pre-existing hypertension were independent risk factors for maternal mortality or heart failure. Type I respiratory failure, arrhythmia, general anaesthesia for caesarean section, New York Heart Association functional class, and N-terminal pro-brain natriuretic peptide \(\ge\) 1400 ng/L were independent predictors of pulmonary hypertension survival during pregnancy. For foetal/neonatal adverse clinical events, type I respiratory failure, arrhythmia, general anaesthesia for caesarean section, parity, platelet count, fibrinogen, and left ventricular systolic diameter were important predictors. Nomogram application for the Development and Validation cohorts showed good discrimination and calibration; decision curve analysis demonstrated their clinical utility.
The nomogram and its online software can be used to analyse individual mortality, heart failure risk, overall survival prediction, and adverse foetal/neonatal clinical events, which may be useful to facilitate early intervention and better survival rates.
Pulmonary hypertension (PH) is a chronic, progressive cardiopulmonary disease with a significant risk of maternal and foetal/neonatal complications [1, 2]. According to the 2018 European Society of Cardiology (ESC) guidelines, the mortality rate of pregnant patients with PH is approximately 16–30% , with most deaths occurring in the first week postpartum [2, 4]. Foetal/neonatal outcomes are also reported to be poor, with mortality rates ranging from 7 to 13% . Therefore, the guidelines recommend against pregnancy, favour contraception, and encourage early pregnancy termination in these patients . Nevertheless, the pregnancy rate among women with PH is increasing , with some patients even declining termination. Moreover, 30% of women are diagnosed with PH during pregnancy, presenting a challenge for mothers and caregivers [7, 8].
Advances have been made in pharmacological and other treatments for PH, improving the overall quality of life and prognosis  and lowering mortality rate (3%) . However, because the maternal mortality rate remains high, pregnancy remains contraindicated in women with PH . A large-scale, multi-centre clinical study is needed to reconsider this general proscription. In the meantime, pregnant patients with PH must be informed regarding their options and pregnancy-associated risks.
To date, no objective, large-scale, multi-centre, machine learning-based clinical prediction models have been developed to evaluate the adverse maternal and foetal/neonatal outcomes in pregnant patients with PH. Given the risks, practical and reliable tools for early assessment of adverse foetal/neonatal events, maternal mortality or heart failure (HF), and overall survival (OS) are needed. Nomograms are based on core diagnostic indicators and are useful for comprehensive patient evaluation and early disease diagnosis. They have been effectively applied for various conditions, including soft-tissue sarcomas , small-cell lung cancer , coronavirus disease , and aortic dissection .
In this study, we aimed to develop and validate a machine-based model to predict maternal mortality or HF and adverse foetal/neonatal clinical events in pregnant patients with PH. The goal is to identify high-risk patients and make rapid, accurate clinical decisions. Additionally, we constructed and verified a prognostic model to guide treatment, advancing the capability for early intervention.
We reviewed obstetric records from six Chinese hospitals from 2009 to 2022. PH was defined according to the clinical diagnostic criteria, with pulmonary artery systolic pressure > 35 mmHg [14, 15] confirmed by echocardiography. Maternal mortality was defined as death during pregnancy or within 7 days postpartum. HF was defined according to the ESC and American Heart Association (AHA) guidelines [16, 17], as a complex clinical syndrome caused by any structural or functional damage resulting from ventricular filling or blood excretion. Patients with elevated right ventricular systolic pressure from outflow obstruction or pulmonary stenosis were excluded. The Development cohort included 420 patients who met these criteria between January 2009 and December 2018. An independent Validation cohort comprising 273 consecutive patients from January 2019 to May 2022 was created using the same criteria. Both cohorts were observed over time, with 117 (16.8%) patients lost to follow-up. The remaining 355 patients from the Development cohort were used as the Follow-up set for the OS nomogram construction, and the remaining 221 patients from the Validation cohort became the external validation set for the prognostic nomogram. Pregnancy was continued in 304 patients from the Development cohort and in 194 patients from the Validation cohort. The corresponding foetal/neonatal records were included in the Delivery and Validation groups, respectively. These groups were used to establish (and validate) another nomogram to predict adverse foetal/neonatal events. This study protocol was approved by the Medical Ethics Committee of the Guangdong Women and Children Hospital (reference number: 202101357). The requirement for informed consent was waived due to the retrospective nature of the study.
Patient characteristics and outcome measures
The primary outcome was maternal mortality or HF. OS was calculated from diagnosis to all-cause death or final follow-up (May 2022). Foetal/neonatal death and adverse clinical events were a composite of foetal death (in utero), neonatal death (within 30 days of birth), and small for gestational age (SGA) (foetal/neonatal weight of small for gestational age < 10%) .
Stepwise regression based on the minimum value of the Akaike information criterion was used to select variables for nomogram inclusion . Discriminatory ability was assessed using the bootstrap concordance index (C-index) and area under the curve (AUC) of the receiver operating characteristic (ROC) curve . The Hosmer–Lemeshow test versus the calibration curve was used to evaluate the ability to calibrate . A C-index or AUC value > 0.7 indicated that the nomogram had good discriminatory ability . The calibration plot showed the predicted and actual probabilities for each patient in the nomogram model, with a line close to the ideal 45° indicating good correlation [23, 24]. Decision curve analysis (DCA) was performed to evaluate the clinical utility of the nomogram [25, 26]. Detailed and expanded methods are found in the Online Supplement.
Patient and disease characteristics
The maternal mortality rates in the Development and the Validation cohorts were 10.2% (43/420) and 4.4% (12/273), respectively, while the probabilities of HF were 16.2% (68/420) and 12.8% (35/273), respectively. The foetal/neonatal mortality rates in the Delivery and Validation groups were 6.6% (20/304) and 3.5% (8/228), respectively; the rates of SGA were 41.4% (126/304) and 32.9% (75/228), respectively. The demographic and clinical characteristics and univariate logistics analyses of the Development and Validation cohorts are summarised in Table 1, and those of the Follow-up and Validation sets are shown in Table 2 and Additional file 1: Table S1, respectively. Maternal and foetus/neonate characteristics in the Delivery and Validation groups are provided in Additional file 1: Table S2.
Nomogram development and validation (maternal mortality or HF)
We constructed a nomogram to predict the risks of maternal mortality or HF in the Development cohort (n = 420). Multivariate logistic analysis identified the following independent predictors: type I respiratory failure (RF), New York Heart Association (NYHA) functional class, N-terminal pro-brain natriuretic peptide (NT-proBNP) ≥ 1400 ng/L, arrhythmia, and eclampsia with pre-existing hypertension (HTN) (Additional file 1: Fig. S1A). These were presented in nomogram form (Fig. 1A). For application of the dynamic nomogram, a score was awarded to correspond with each variable, and the sum of scores was recorded as the total score. The total score corresponded with the predicted risk of maternal mortality or HF in patients with PH (Additional file 1: Fig. S1B).
To evaluate the nomogram’s efficacy and its potential for clinical use, we performed internal and external validation. The bootstrap-corrected C-indexes from the Development and Validation cohorts were 0.892 (95% CI: 0.818–0.966) and 0.877 (95% CI: 0.757–0.996), respectively; this indicated excellent discriminative ability. Furthermore, the calibration plot showed that the predicted probabilities were close to actual observations in both cohorts (Fig. 1B, a–b). The results of the Hosmer–Lemeshow test (χ2 = 43.225; P = 0.873) indicated a good nomogram fit. Moreover, the AUC of the nomogram’s ROC curve for both cohorts also had good discrimination ability. As shown in Additional file 1: Fig. S1C, the AUC of the training cohort was 0.892 (95% CI: 0.855–0.929); this was verified in the Validation cohort with an AUC of 0.899 (95% CI: 0.843–0.955). The DCA curve demonstrated that the optional nomogram threshold probability was high and relatively safe in both cohorts, with a high net benefit (Fig. 1C, a–b). The nomogram was further used to calculate the probability of maternal mortality or HF for all patients. Violin plot analysis showed that the predicted risks for maternal mortality or HF were markedly lower in the surviving patients without HF in both cohorts (the nomogram predicted probabilities of mortality/HF as 0.612 ± 0.030 vs. survivors/no HF as 0.140 ± 0.010, P < 0.001 in the Development cohort and 0.645 ± 0.051 vs 0.137 ± 0.012, respectively, P < 0.001 in the Validation cohort) (Fig. 1D, a–b).
Nomogram development and validation (overall survival)
In the Follow-up set, the independent prognostic factors type I RF, arrhythmia, general anaesthesia for caesarean sections (C-section), NYHA functional class, and NT-proBNP ≥ 1400 ng/L (Additional file 1: Fig. S2A) were used to create a prognostic nomogram for OS (Fig. 2A). In the nomogram and dynamic nomogram applications, each variable was assigned a value, the sum of which represents the total score. The predicted 1, 2, and 3-year OS rates corresponded with the overall score (Additional file 1: Fig. S2B). We then used a time-dependent AUC to verify the predictive capacity of the prognostic model. In the Follow-up set, the OS predictions at 1, 2, and 3 years were 0.936, 0.916, and 0.916, respectively (Fig. 2B.a), while those in the Validation set were 0.939, 0.861, and 0.877, respectively (Fig. 2B.b). These results indicated satisfactory predictive performance. On DCA, the nomogram offered a net benefit over the ‘treat-all’ or ‘treat-none’ strategy for both sets (Fig. 2C, a–b). The calibration curves of the nomogram demonstrated good correlation between the 1, 2, and 3-year predicted and actual survival probabilities for both sets (Fig. 3A, B).
Risk stratification based on the prognostic nomogram
Risk stratification was performed based on the total scores of the prognostic nomogram. Patients in the Follow-up and Validation sets were divided into low and high-risk groups, with the median risk score representing the cut-off value. Kaplan–Meier analysis was used to explore survival differences using the log-rank test; the OS curves demonstrated excellent discrimination between the two risk groups in the different sets. Survival time was significantly shorter in the high-risk than low-risk groups for all sets (Fig. 3C, D).
Nomogram development and validation (Foetal/neonatal adverse events)
For the Delivery group, multivariate logistic regression analysis identified seven variables (type I RF, NT-proBNP ≥ 1400 ng/L, arrhythmia, general anaesthesia for C-section, parity, platelets, fibrinogen, and left ventricular systolic diameter) as independent predictors for foetal/neonatal adverse clinical events (Additional file 1: Fig. S4A). These were used to build another nomogram (Fig. 4A). On ROC curves, fibrinogen reduction to ≤ 2.00 g/L, platelet count > 245 × 109/L, parity > 2, and maternal left ventricular systolic diameter \(\ge\) 35 mm were significant thresholds for predicting foetal/neonatal adverse events. The nomogram and dynamic nomogram applications are depicted in Additional file 1: Fig. S4B.
Internal validation was performed using bootstrapping resampling for 1000 repetitions, and the C-index value was 0.854 (95% CI: 0.770–0.938) in the Delivery group and 0.764 (95% CI: 0.639–0.889) in the Validation group. The AUC of the Delivery and Validation groups were 0.854 (95% CI: 0.812–0.896) and 0.764 (95% CI: 0.701–0.828), respectively (Additional file 1: Fig. S4C). Notably, the calibration plot showed that the predicted probabilities were close to the actual observed outcomes for the Delivery (Fig. 4B.a) and Validation groups (Fig. 4B.b). Results of the Hosmer–Lemeshow test of the model in the Delivery group (χ2 = − 0.544, P = 0.586) indicated a good nomogram fit. Moreover, DCA curves revealed that the nomogram may better predict the risk of foetal/neonatal adverse events than the ‘treat-all’ and ‘treat-none’ schemes, as it added more net benefits for both groups (Fig. 4C). The nomogram constructed by logistic regression was used to calculate the probability of foetal/neonatal adverse events for all pregnant patients with PH. Nomogram scores of the foetal/neonatal adverse events patients were markedly higher than those of the patients without foetal/neonatal adverse events (the nomogram predicted probabilities of 0.678 ± 0.021 vs 0.298 ± 0.018, P < 0.001 in the Delivery group and 0.477 ± 0.021 vs 0.242 ± 0.014, P < 0.001 in the Validation group, respectively) (Fig. 4D). This implies that this model may accurately predict a patient's risk probability of having foetal/neonatal adverse events.
Implementation of the web server
We designed three dynamic nomograms using practical online applications; each incorporated an independent predictor. The online application for predicting maternal mortality or HF is available at https://ph-666.shinyapps.io/maternal-D/ (Additional file 1: Fig. S1D), for maternal prognosis at https://ph-666.shinyapps.io/COX-pregnant/ (Additional file 1: Fig. S3A–C), and for neonatal adverse events at https://ph-666.shinyapps.io/AE-fetal/ (Additional file 1: Fig. S4D).
Using the parameters of type I RF, NYHA functional class, NT-proBNP \(\ge\) 1400 ng/L, eclampsia with pre-existing HTN, and arrhythmia, this nomogram is a useful screening tool for patients at high risk of maternal mortality or HF. This personalised approach also represents a dynamic online tool with a user-friendly digital interface (https://ph-666.shinyapps.io/maternal-D/). Most maternal deaths occur during the first week postpartum due to postdelivery hemodynamic changes. Moreover, five variables (type I RF, arrhythmia, NT-proBNP \(\ge\) 1400 ng/L, pregnancy outcomes, and NYHA functional class) were incorporated into a nomogram to predict OS and monitor PH condition over time. To enable easy clinical implementation, we developed a dynamic online nomogram application (https://ph-666.shinyapps.io/COX-pregnant/). Pregnancy in women with PH has long been regarded as a high risk for both maternal and foetal/neonatal complications . We found that type I RF, NT-proBNP \(\ge\) 1400 ng/L, arrhythmia, delivery mode, parity, platelet, fibrinogen, and left ventricular systolic diameter were independently associated with increased adverse foetal/neonatal events. Further, these online dynamic nomograms are available at https://ph-666.shinyapps.io/AE-fetal/. Our online nomograms ascertained the probabilities of maternal mortality, survival, and the foetal/neonatal adverse events when the corresponding clinical factors were input.
Results in the context of what is known
PH affects pulmonary vasculature and the heart, and pregnancy places a significant burden on the cardiovascular system. Like other studies , we found that maternal mortality or HF were associated with NYHA functional class. Pregnant patients with NYHA functional class IV had poorer pregnancy outcomes. NT-proBNP is also recommended by international guidelines for risk assessment in patients with PH . Moreover, since pregnancy is a hypermetabolic state, oxygen consumption increases by approximately 20% . Increased minute ventilation leads to respiratory alkalosis, increased arterial oxygen tension, and dyspnoea. In our study, lower oxygen partial pressure was associated with maternal and foetal/neonatal complications . Pregnant patients with hypoxia can also experience placental hypoperfusion, which can affect foetal growth and lead to perinatal complications .
Consistent with previous studies, maternal arrhythmias were found to be associated with an increase in maternal cardiac events and foetal/neonatal adverse outcomes [30, 31]. In particular, hypertensive disorders of pregnancy are associated with higher rates of maternal mortality . Our results are consistent with recent studies where the administration of general anaesthesia during C-section is associated with an increased incidence of adverse maternal [1, 33], and foetal/neonatal outcomes [34, 35]. This may be related to increased intrathoracic and pulmonary artery pressure and thus, reduced venous return from endotracheal intubation and positive pressure ventilation . The latest guidelines from the ESC reflect existing evidence that planned caesarean or vaginal deliveries carry lower risk . Our study found lower risks of death or HF in women who delivered via C-section with spinal or epidural anaesthesia than via vaginal delivery, which is consistent with other literature favouring or even recommending planned C-sections [1, 2, 37]. Vaginal delivery can cause pain and increase thoracic pressure, which may reduce venous reflux . In contrast, C-section provides a more regulated delivery environment . It avoids long labour and allows for careful preparation of anaesthesia, haemodynamic optimisation, and contingency planning [1, 40]. However, more research is needed to determine the optimal delivery method. The current guidelines recommend strict contraception for patients with PH, and early pregnancy termination ; however, termination has also been associated with high maternal risks ; this was consistent with our findings. Our results suggested that a fibrinogen reduction to ≤ 2.00 g/L, platelet count > 245 × 109/L, parity > 2, and maternal left ventricular systolic diameter \(\ge\) 35 mm were the significant thresholds for predicting foetal/neonatal adverse events. Fibrinogen level before delivery may be a good predictor of placental abruption . Decreased fibrinogen level may be associated with the degree of sub-placental hematoma formation, resulting in acute/chronic foetal acidaemia and increased intrauterine growth restriction . Maternal platelet count has been used as an early predictor of neonatal respiratory distress and adverse foetal/neonatal outcomes . Consistent with previous studies, parity was associated with growth restriction and mortality in neonates . Interestingly, left ventricular mass has been shown to have a strong heritability component for cardiac features [45, 46].
This was the first study to develop and validate a practical nomogram to predict maternal mortality or HF, OS, and adverse foetal/neonatal events in patients with PH. The nomogram was based on large-scale data analysis, and its intuitive icon model was convenient for clinical application. It can also be developed into an online application for intelligent patient management as well as auxiliary diagnosis and treatment. Nomograms have achieved good clinical application results in the fields of cancer, cardiovascular, and other diseases. The nomograms constructed in this study had good application results, including primary medical institutions' preliminary judgement of the prognostic risk of patients with pregnant women with PH and early identification of high-risk patients for early referral, evaluation, and treatment. It can also provide clinicians with accurate prediction tools for individual mortality, heart failure, overall survival, and adverse foetal/neonatal clinical events. The online tools developed based on the nomograms can be used for the self-management by pregnant patients with PH through an intelligent communication terminal operation programme and can be integrated with a hospital case management system for artificial intelligence-assisted diagnosis. The machine learning and feedback optimisation in the later application process will further improve the efficiency and accuracy of clinical prediction.
Strengths and limitations of the study
This study has some limitations, including limited cohort size and potential selection bias. Furthermore, data interpretation revealed some limitations to echocardiography as a diagnostic method: although RHC is the gold standard for PH diagnosis, radiation exposure may lead to foetal teratogenicity, carcinogenicity, or mutation. Spontaneous abortion, growth restriction, and intellectual disabilities may occur with high exposure levels. Regardless of the dose, the cancer risk also increases; therefore, the Swan–Ganz catheter is not recommended for routine perinatal monitoring . Another limitation is that the patient populations of the Development and external Validation cohorts differed, and the sample size of the external Validation cohort was relatively small. Although the external validation proves a good calibration, its validation efficiency is limited. Lastly, a larger prospective multi-centre study of patients with PH is needed to determine the exact risks associated with pregnancy, the role of supportive care and late PH treatment, and prognostic factors for each subgroup.
The nomograms created in this study may be used to accurately predict maternal mortality or HF, adverse foetal/neonatal outcomes, and survival in pregnant patients with PH. Additionally, these tools can guide more effective clinical decision-making. Our nomograms and accompanying software can assess individualised risks and anticipate OS more easily in pregnant patients with PH. The online nomogram software was shown to be a useful tool for the management of pregnant patients with PH for primary hospitals or community centres.
Availability of data and materials
All data included in this study are available upon request through contact with the corresponding author.
American Heart Association
Activated partial thromboplastin time
Area under the curve
Decision curve analysis
European Society of Cardiology
Left ventricular (systolic) diameter
N-terminal pro-brain natriuretic peptide
New York Heart Association
Pulmonary artery systolic pressure
Red blood cells
Right heart catheterisation
Receiver operating characteristic
Right ventricular diameter
Small for gestational age
Bédard E, Dimopoulos K, Gatzoulis MA. Has there been any progress made on pregnancy outcomes among women with pulmonary arterial hypertension? Eur Heart J. 2009;30:256–65. https://doi.org/10.1093/eurheartj/ehn597.
Sliwa K, van Hagen IM, Budts W, Swan L, Sinagra G, Caruana M, Blanco MV, Wagenaar LJ, Johnson MR, Webb G, et al. Pulmonary hypertension and pregnancy outcomes: data from the Registry Of Pregnancy and Cardiac Disease (ROPAC) of the European Society of Cardiology. Eur J Heart Fail. 2016;18:1119–28. https://doi.org/10.1002/ejhf.594.
Regitz-Zagrosek V, Roos-Hesselink JW, Bauersachs J, Blomström-Lundqvist C, Cífková R, De Bonis M, Iung B, Johnson MR, Kintscher U, Kranke P, et al. 2018 ESC Guidelines for the management of cardiovascular diseases during pregnancy. Eur Heart J. 2018;39:3165–241. https://doi.org/10.1093/eurheartj/ehy340.
Meng ML, Landau R, Viktorsdottir O, Banayan J, Grant T, Bateman B, Smiley R, Reitman E. Pulmonary hypertension in pregnancy: a report of 49 cases at four tertiary north american sites. Obstet Gynecol. 2017;129:511–20. https://doi.org/10.1097/aog.0000000000001896.
Olsson KM, Channick R. Pregnancy in pulmonary arterial hypertension. Eur Respir Rev. 2016;25:431–7. https://doi.org/10.1183/16000617.0079-2016.
Ladouceur M, Benoit L, Radojevic J, Basquin A, Dauphin C, Hascoet S, Moceri P, Bredy C, Iserin L, Gouton M, Nizard J. Pregnancy outcomes in patients with pulmonary arterial hypertension associated with congenital heart disease. Heart. 2017;103:287–92. https://doi.org/10.1136/heartjnl-2016-310003.
Thomas E, Yang J, Xu J, Lima FV, Stergiopoulos K. Pulmonary hypertension and pregnancy outcomes: insights from the national inpatient sample. J Am Heart Assoc. 2017. https://doi.org/10.1161/jaha.117.006144.
Hopkins WE, Ochoa LL, Richardson GW, Trulock EP. Comparison of the hemodynamics and survival of adults with severe primary pulmonary hypertension or Eisenmenger syndrome. J Heart Lung Transplant. 1996;15:100–5.
Anand V, Roy SS, Archer SL, Weir EK, Garg SK, Duval S, Thenappan T. Trends and outcomes of pulmonary arterial hypertension-related hospitalizations in the United States: analysis of the nationwide inpatient sample database from 2001 through 2012. JAMA Cardiol. 2016;1:1021–9. https://doi.org/10.1001/jamacardio.2016.3591.
Callegaro D, Miceli R, Bonvalot S, Ferguson P, Strauss DC, Levy A, Griffin A, Hayes AJ, Stacchiotti S, Pechoux CL, et al. Development and external validation of two nomograms to predict overall survival and occurrence of distant metastases in adults after surgical resection of localised soft-tissue sarcomas of the extremities: a retrospective analysis. Lancet Oncol. 2016;17:671–80. https://doi.org/10.1016/s1470-2045(16)00010-3.
Wang S, Yang L, Ci B, Maclean M, Gerber DE, Xiao G, Xie Y. Development and validation of a nomogram prognostic model for SCLC patients. J Thorac Oncol. 2018;13:1338–48. https://doi.org/10.1016/j.jtho.2018.05.037.
Dong YM, Sun J, Li YX, Chen Q, Liu QQ, Sun Z, Pang R, Chen F, Xu BY, Manyande A, et al. Development and validation of a nomogram for assessing survival in patients With COVID-19 pneumonia. Clin Infect Dis. 2021;72:652–60. https://doi.org/10.1093/cid/ciaa963.
Wu J, Qiu J, Jiang W, Qiu J, Zhang L, Zhao R, Yu C. Development and validation of a nomogram predicting the probability of type a aortic dissection at a diameter below 55 mm: a retrospective cohort study. Int J Surg. 2018;60:266–72. https://doi.org/10.1016/j.ijsu.2018.11.024.
Humbert M, Kovacs G, Hoeper MM, Badagliacca R, Berger RMF, Brida M, Carlsen J, Coats AJS, Escribano-Subias P, Ferrari P, et al. 2022 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension. Eur Heart J. 2022;43:3618–731. https://doi.org/10.1093/eurheartj/ehac237.
Seeger W, Adir Y, Barberà JA, Champion H, Coghlan JG, Cottin V, De Marco T, Galiè N, Ghio S, Gibbs S, et al. Pulmonary hypertension in chronic lung diseases. J Am Coll Cardiol. 2013;62:D109-116. https://doi.org/10.1016/j.jacc.2013.10.036.
McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, Burri H, Butler J, Čelutkienė J, Chioncel O, et al. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: Developed by the Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). With the special contribution of the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail. 2022;24:4–131. https://doi.org/10.1002/ejhf.2333.
Heidenreich PA, Bozkurt B, Aguilar D, Allen LA, Byun JJ, Colvin MM, Deswal A, Drazner MH, Dunlay SM, Evers LR, et al. 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure: a Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2022;145:e895–1032. https://doi.org/10.1161/cir.0000000000001063.
ACOG Practice Bulletin No. 204: Fetal Growth Restriction. Obstet Gynecol 2019, 133:e97–e109. https://doi.org/10.1097/aog.0000000000003070.
Wu J, Zhang H, Li L, Hu M, Chen L, Xu B, Song Q. A nomogram for predicting overall survival in patients with low-grade endometrial stromal sarcoma: a population-based analysis. Cancer Commun (Lond). 2020;40:301–12. https://doi.org/10.1002/cac2.12067.
Heagerty PJ, Lumley T, Pepe MS. Time-dependent ROC curves for censored survival data and a diagnostic marker. Biometrics. 2000;56:337–44. https://doi.org/10.1111/j.0006-341x.2000.00337.x.
Núñez E, Steyerberg EW, Núñez J. Regression modeling strategies. Rev Esp Cardiol. 2011;64:501–7. https://doi.org/10.1016/j.recesp.2011.01.019.
Gail MH, Pfeiffer RM, Cai T. Six articles related to risk assessment and prediction based on work presented at the October 12–14, 2011 Conference on Risk Assessment and Evaluation of Predictions in Silver Spring. Maryland Lifetime Data Anal. 2013;19:139–41. https://doi.org/10.1007/s10985-013-9253-9.
Steyerberg EW, Vickers AJ, Cook NR, Gerds T, Gonen M, Obuchowski N, Pencina MJ, Kattan MW. Assessing the performance of prediction models: a framework for traditional and novel measures. Epidemiology. 2010;21:128–38. https://doi.org/10.1097/EDE.0b013e3181c30fb2.
Harrison WT, Yeates RM, Phillips ML, Nenoff TM. New framework connectivity patterns in templated networks: the creatinine zinc phosphites C4N3OH7ZnHPO3, C4N3OH7.Zn(H2O)HPO3, and (C4N3OH7)2.ZnHPO3.H2O. Inorg Chem. 2003;42:1493–8. https://doi.org/10.1021/ic020593+.
Kerr KF, Brown MD, Zhu K, Janes H. Assessing the clinical impact of risk prediction models with decision curves: guidance for correct interpretation and appropriate use. J Clin Oncol. 2016;34:2534–40. https://doi.org/10.1200/JCO.2015.65.5654.
Vickers AJ, Cronin AM. Traditional statistical methods for evaluating prediction models are uninformative as to clinical value: towards a decision analytic framework. Semin Oncol. 2010;37:31–8. https://doi.org/10.1053/j.seminoncol.2009.12.004.
Galiè N, Humbert M, Vachiery JL, Gibbs S, Lang I, Torbicki A, Simonneau G, Peacock A, Vonk Noordegraaf A, Beghetti M, et al. 2015 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension: The Joint Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS): Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC), International Society for Heart and Lung Transplantation (ISHLT). Eur Respir J. 2015;46:903–75. https://doi.org/10.1183/13993003.01032-2015.
Prowse CM, Gaensler EA. Respiratory and acid-base changes during pregnancy. Anesthesiology. 1965;26:381–92. https://doi.org/10.1097/00000542-196507000-00003.
Liu Y, Li Y, Zhang J, Zhang D, Li J, Zhao Y, Liu K, Ma X, Bai C, Gu H, et al. Maternal and fetal outcomes of pregnant women with pulmonary arterial hypertension associated with congenital heart disease in Beijing, China: a retrospective study. Pulm Circ. 2022;12:e12079. https://doi.org/10.1002/pul2.12079.
Blomström-Lundqvist C, Scheinman MM, Aliot EM, Alpert JS, Calkins H, Camm AJ, Campbell WB, Haines DE, Kuck KH, Lerman BB, et al. ACC/AHA/ESC guidelines for the management of patients with supraventricular arrhythmias—executive summary. a report of the American college of cardiology/American heart association task force on practice guidelines and the European society of cardiology committee for practice guidelines (writing committee to develop guidelines for the management of patients with supraventricular arrhythmias) developed in collaboration with NASPE-Heart Rhythm Society. J Am Coll Cardiol. 2003;42:1493–531. https://doi.org/10.1016/j.jacc.2003.08.013.
Pillutla P, Nguyen T, Markovic D, Canobbio M, Koos BJ, Aboulhosn JA. Cardiovascular and neonatal outcomes in pregnant women with high-risk congenital heart disease. Am J Cardiol. 2016;117:1672–7. https://doi.org/10.1016/j.amjcard.2016.02.045.
Hutcheon JA, Lisonkova S, Joseph KS. Epidemiology of pre-eclampsia and the other hypertensive disorders of pregnancy. Best Pract Res Clin Obstet Gynaecol. 2011;25:391–403. https://doi.org/10.1016/j.bpobgyn.2011.01.006.
Sun X, Feng J, Shi J. Pregnancy and pulmonary hypertension: an exploratory analysis of risk factors and outcomes. Medicine (Baltimore). 2018;97:e13035. https://doi.org/10.1097/md.0000000000013035.
Algert CS, Bowen JR, Giles WB, Knoblanche GE, Lain SJ, Roberts CL. Regional block versus general anaesthesia for caesarean section and neonatal outcomes: a population-based study. BMC Med. 2009;7:20. https://doi.org/10.1186/1741-7015-7-20.
Kearns RJ, Shaw M, Gromski PS, Iliodromiti S, Pell JP, Lawlor DA, Nelson SM. Neonatal and early childhood outcomes following maternal anesthesia for cesarean section: a population-based cohort study. Reg Anesth Pain Med. 2021;46:482–9. https://doi.org/10.1136/rapm-2020-102441.
Canobbio MM, Warnes CA, Aboulhosn J, Connolly HM, Khanna A, Koos BJ, Mital S, Rose C, Silversides C, Stout K. Management of pregnancy in patients with complex congenital heart disease: a Scientific Statement for Healthcare Professionals From the American Heart Association. Circulation. 2017;135:e50–87. https://doi.org/10.1161/cir.0000000000000458.
Konstantinides SV. Trends in pregnancy outcomes in patients with pulmonary hypertension: still a long way to go. Eur J Heart Fail. 2016;18:1129–31. https://doi.org/10.1002/ejhf.616.
Banerjee D, Ventetuolo CE. Pulmonary hypertension in pregnancy. Semin Respir Crit Care Med. 2017;38:148–59. https://doi.org/10.1055/s-0037-1602242.
Duarte AG, Thomas S, Safdar Z, Torres F, Pacheco LD, Feldman J, deBoisblanc B. Management of pulmonary arterial hypertension during pregnancy: a retrospective, multicenter experience. Chest. 2013;143:1330–6. https://doi.org/10.1378/chest.12-0528.
Chakravarty EF, Khanna D, Chung L. Pregnancy outcomes in systemic sclerosis, primary pulmonary hypertension, and sickle cell disease. Obstet Gynecol. 2008;111:927–34. https://doi.org/10.1097/01.AOG.0000308710.86880.a6.
Wang L, Matsunaga S, Mikami Y, Takai Y, Terui K, Seki H. Pre-delivery fibrinogen predicts adverse maternal or neonatal outcomes in patients with placental abruption. J Obstet Gynaecol Res. 2016;42:796–802. https://doi.org/10.1111/jog.12988.
Parker SE, Werler MM. Epidemiology of ischemic placental disease: a focus on preterm gestations. Semin Perinatol. 2014;38:133–8. https://doi.org/10.1053/j.semperi.2014.03.004.
Mishra S, Jaiswar S, Saad S, Tripathi S, Singh N, Deo S, Agarwal M, Mishra N. Platelet indices as a predictive marker in neonatal sepsis and respiratory distress in preterm prelabor rupture of membranes. Int J Hematol. 2021;113:199–206. https://doi.org/10.1007/s12185-020-03025-2.
Gibbs CM, Wendt A, Peters S, Hogue CJ. The impact of early age at first childbirth on maternal and infant health. Paediatr Perinat Epidemiol. 2012;26(Suppl 1):259–84. https://doi.org/10.1111/j.1365-3016.2012.01290.x.
Lam CS, Liu X, Yang Q, Larson MG, Pencina MJ, Aragam J, Redfield MM, Benjamin EJ, Vasan RS. Familial aggregation of left ventricular geometry and association with parental heart failure: the Framingham Heart Study. Circ Cardiovasc Genet. 2010;3:492–8. https://doi.org/10.1161/circgenetics.110.941088.
Verhaaren HA, Schieken RM, Mosteller M, Hewitt JK, Eaves LJ, Nance WE. Bivariate genetic analysis of left ventricular mass and weight in pubertal twins (the Medical College of Virginia twin study). Am J Cardiol. 1991;68:661–8. https://doi.org/10.1016/0002-9149(91)90361-n.
Hemnes AR, Kiely DG, Cockrill BA, Safdar Z, Wilson VJ, Al Hazmi M, Preston IR, MacLean MR, Lahm T. Statement on pregnancy in pulmonary hypertension from the Pulmonary Vascular Research Institute. Pulm Circ. 2015;5:435–65. https://doi.org/10.1086/682230.
This work was supported in part by the grants from the National Natural Science Foundation of China (82170069, 82120108001, 81870046, 82000045), Department of Science and Technology of China Grants (2016YFC0903700, 2016YFC1304102, 2018YFC1311900), Changjiang Scholars, and Innovative Research Team in University (IRT0961), Local Innovative and Research Teams Project of Guangdong Pearl River Talents Programme (2017BT01S155), Guangdong Department of Science and Technology (2022A0202030017, 2022A1515012052), Basic Science and Application of Guangzhou Science and Technology Plan (202201010069), Guangzhou Basic Research Programme Municipal School (Hospital) Joint Funded Foundation and Application Basic Research Project (ZNSA-2020003), the Independent Project of State Key Laboratory of Respiratory Disease (SKLRD-Z-202101), the State Key Laboratory of Respiratory Disease, and Guangdong-Hong Kong-Macao Joint Laboratory of Respiratory Infectious Disease (GHMJLRID-Z-202110).
Ethics approval and consent to participate
This study protocol was approved by the Medical Ethics Committee of the Guangdong Women and Children Hospital (reference number: 202101357). The requirement for informed consent was waived due to the retrospective nature of the study.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Chen, Y., Zhou, D., Xiong, M. et al. Prediction and prognosis of adverse maternal and foetal/neonatal outcomes in pulmonary hypertension: an observational study and nomogram construction. Respir Res 23, 314 (2022). https://doi.org/10.1186/s12931-022-02235-y
- Pulmonary hypertension
- Prediction model
- Prognostic model
- Overall survival
- Machine-based model