Abstract
Pregnancies complicated by BCR::ABL1-negative myeloproliferative neoplasms (MPNs) are associated with increased maternal and thrombotic risk; however, large real-world comparative data remain limited. We performed a multicenter retrospective propensity score-matched cohort study using the TriNetX database to evaluate maternal, obstetric, and thrombotic outcomes among pregnant patients with MPN compared with matched non-MPN controls. Baseline demographics, comorbidities, and antithrombotic medication use were balanced after matching (standardized mean difference < 0.1). Among 4,467,127 pregnancies, 4,563 patients had MPN, most commonly essential thrombocytosis (85%). Compared with matched controls, MPN pregnancies had significantly higher all-cause mortality (1.1% vs. 0.2%; hazard ratio [HR] 5.4, 95% CI 2.7–10.8), gestational hypertension (11.1% vs. 7.8%; HR 1.3, 95% CI 1.1–1.5), preeclampsia/eclampsia (10.4% vs. 5.3%; HR 1.8, 95% CI 1.6–2.1), preterm labor (6.1% vs. 3.7%; HR 1.5, 95% CI 1.2–1.8), intrauterine growth restriction (7.3% vs. 5.1%; HR 1.3, 95% CI 1.1–1.7), postpartum hemorrhage (5.5% vs. 3.8%; HR 1.3, 95% CI 1.1–1.6), and venous thromboembolism (2.7% vs. 0.7%; HR 3.2, 95% CI 2.2–4.6). Arterial thrombotic events were not significantly different between groups. Pregnancies in patients with MPN are associated with significantly increased maternal, obstetric, and venous thrombotic complications, highlighting the need for multidisciplinary management and individualized thromboprophylaxis strategies.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s00277-026-07285-6.
Keywords: Myeloproliferative neoplasm, Pregnancy, Outcomes, Thrombosis
Introduction
The classical Philadelphia (Ph) or Breakpoint Cluster Region- Abelson murine Leukemia viral oncogene homolog 1 (BCR-ABL1) negative Myeloproliferative neoplasms (MPN) include polycythemia vera (PV), essential thrombocythemia (ET), and myelofibrosis (MF). These neoplasms have a distinct biology and are associated with an increased risk of thrombosis, bleeding, and transformation into leukemia [1, 2]. Over time, the reported incidence of MPNs has increased, largely attributable to advances in molecular diagnostics and the broader availability and sensitivity of these assays. MPNs are more frequently diagnosed in older individuals; however, these disorders can occur across all age groups. In parallel with the trend toward delayed childbearing and improved diagnostic capabilities, clinicians are encountering a growing number of pregnant patients with MPNs [3, 4]. Furthermore, emerging evidence suggests that driver mutations, particularly Janus Kinase 2 (JAK2), may be acquired early in life and persist for decades, in some cases originating in childhood. This raises the possibility that widespread implementation of genetic screening could lead to the identification of a substantially larger cohort of younger individuals harboring these mutations [5, 6].
Pregnancy in women with MPN presents unique maternal and fetal challenges that complicate clinical management. As both conditions contribute to a hypercoagulable state, this increases the risk of adverse clinical outcomes [7]. Additionally, pregnancy poses unique physiologic hemodynamic, hormonal, and hematologic stress, and these risks are often amplified, creating a particularly vulnerable clinical setting [8]. Adverse maternal outcomes, including vascular complications, hypertensive disorders, and postpartum hemorrhage (PPH), have been consistently reported, alongside fetal complications such as miscarriage, intrauterine growth restriction (IUGR), preterm delivery, and stillbirth [3, 4, 9]. It is, therefore, essential that this group of patients is regularly monitored throughout their pregnancy by a hematologist and obstetrician. To date, the available interventions for reducing the risk of these complications include phlebotomy, aspirin, heparin, and interferon therapy, or a combination of these, in addition to obstetric interventions such as early delivery when feasible [10]. There are no randomized clinical trials reported to date, and the data are limited to case series, systematic reviews, large collaborative studies, population registry-based studies, and 1 prospective study from the United Kingdom (UK) [11]. There is a wide variability in various outcomes and also the live birth rate (LBR) reported in the literature, depending on the disease and type of study [10]. Therefore, we conducted a population-level multicenter retrospective study using the TrinetX framework to analyze both obstetric and thrombotic outcomes in patients with Ph-negative MPNs.
Methods
Overview of study design
We used the TriNetX database to study the effect of MPN on pregnancy outcomes. Additionally, we incorporated propensity score matching (PSM) to balance baseline characteristics between both groups. PSM improves comparability between cohorts and reduces measured confounding commonly associated with observational studies. We selected a normal pregnant population as the control cohort. Supplemental Fig. 1 lists the key protocol components.
Data source
This is a non-interventional, retrospective, real-world study conducted using TrinetX LLC. TriNetX is the global federated health research network that provides access to electronic medical records (diagnoses, procedures, medications, laboratory values, genomic information) from large healthcare organizations (HCOs) [12]. This report was run on the set of HCOs grouped into the Global Collaborative Network, which included 170 HCOs. This dataset includes comprehensive information on inpatient and outpatient clinical encounters, including diagnosis codes, procedures performed, medication prescriptions, and laboratory values. The TrinetX platform is widely used in clinical research and has been validated for its reliability in supporting observational research [13]. The database was accessed on April 13th, 2026.
Study population
We included all pregnant patients aged 18 to 40 years from January 1 st, 2000, to March 30th, 2025, in the study. The index event was defined as the first recorded pregnancy-related ICD-10 diagnosis code. Subsequently, pregnant patients with MPN were identified using the International Classification of Diseases, 10th revision, Clinical Modification (ICD-10-CM) codes. We excluded patients with a prior history of any other hematological malignancy, solid organ malignancy, inherited or acquired thrombophilia, multiple pregnancies, or in vitro fertilization, identified using specific ICD-10-CM codes (Supplementary Tables 1 and 3).
Outcomes
Outcomes are classified into obstetric and thrombotic outcomes. Obstetric outcomes included all-cause mortality, gestational hypertension, preeclampsia and eclampsia, abruptio placentae, oligohydramnios, polyhydramnios, preterm labor, PPH, and IUGR. Thrombotic outcomes are further classified into venous thromboembolic events (VTE) and arterial thromboembolic events (ATE). Occurrence of VTE, defined as a composite of pulmonary embolism (PE) or deep vein thrombosis (DVT), and occurrence of ATE is defined as a composite of stroke or myocardial infarction. Further, we also analyzed all the events separately (DVT, PE, other site thrombosis like portal vein thrombosis (PVT), cerebral venous sinus thrombosis (CVST), hepatic vein thrombosis (HVT), stroke, and myocardial infarction). These outcomes were identified using ICD-10-CM codes (Supplemental Table 1).
Follow-up
The index event was defined as the first recorded pregnancy-related ICD-10 diagnosis code. Each eligible patient was followed from the index pregnancy diagnosis until the occurrence of the outcome of interest, death, completion of 350 days of follow-up, or the last recorded encounter in the TriNetX network, whichever occurred first. Patients were censored at the time of disenrollment or death or the last recorded encounter in the TriNetX database, allowing for time-to-event analyses. This approach accounts for varying durations of follow-up among patients, with data considered up to the last available data Competing risks were not modeled given platform limitations. Similarly, pregnancy loss could not be incorporated as a censoring or competing event because gestational age and the timing and termination of individual pregnancy episodes could not be reliably ascertained from the available structured TriNetX data.
Statistical analysis
The Pregnant patients with MPN were matched with the normal pregnant population in a 1:1 ratio using predetermined variables that included age, sex, race, underlying comorbidities based on components of the Charlson Comorbidity Index, body mass index (BMI), socioeconomic status, and the use of aspirin and other antiplatelet medications (Supplemental Tables 3 and 4). We selected these variables because they alter the likelihood of occurrence of an outcome and may be potential confounders. Propensity score matching was performed using a nearest neighbor greedy matching algorithm with a caliper of 0.1 times the pooled standard deviation. Patients with missing data were still included in the propensity score matching. The missing values did not contribute to the matching process for the affected variables, as propensity scores were calculated using available data. We compared the differences in baseline characteristics between the cohorts using the standardized mean differences. We considered a standardized mean difference of less than 0.1 as an acceptably balanced distribution of covariates between the cohorts. Hazard ratios were estimated using Cox proportional hazards models implemented within the TriNetX survival analysis framework, and survival distributions between the two groups were compared using the log-rank test. We evaluated the association of MPN with each outcome using the Cox proportional hazards model. Continuous variables were expressed as mean ± standard deviation, while categorical variables were reported as proportions. A 2-sided P value < 0.05 was considered statistically significant. However, given platform limitations, proportional hazards assumptions could not be independently verified.
Table 3.
Subgroup analysis based on prior history of thrombosis or atrial fibrillation*#
| Number of patients with outcome (n = 4672)% | ||||
|---|---|---|---|---|
| Outcomes | Pregnancy with the MPN with prior history of VTE or Atrial fibrillation (n = 261) | Pregnancy with MPN without prior history of VTE or atrial fibrillation (n = 4411) | HR (95% CI) | Log-rank p-value |
| Obstetric outcomes | ||||
| Maternal Mortality | 27 (10.3) | 27 (0.6) | 6.6 (3.9–10.9) | < 0.001 |
| Gestational HTN | 26 (9.9) | 502 (11.3) | 0.8 (0.5–1.2) | 0.431 |
| Preeclampsia and eclampsia | 37 (14.1) | 448 (10.1) | 1.4 (1.1–1.9) | 0.037 |
| Preterm labor | 20 (7.6) | 272 (6.1) | 1.2 (0.7–1.9) | 0.369 |
| IUGR | 18 (6.8) | 326 (7.3) | 0.9 (0.5–1.4) | 0.738 |
| PPH | 20 (7.6) | 238 (5.3) | 1.4 (0.9–2.2) | 0.127 |
| Thrombotic outcomes | ||||
| Venous thrombotic outcomes | ||||
| VTE | 66 (25.2) | 44 (0.9) | 28.5 (19.5–41.4) | < 0.001 |
#Subgroup analysis was done to identify a subgroup with higher baseline thrombotic risk and likely exposure to therapeutic anticoagulation based on prior history of VTE or atrial fibrillation. *Arterial thrombotic events, Abruptio placenta, Oligohydramnios, Polyhydramnios; individual VTE events could not be reported as their number is less than 10. MPN myeloproliferative neoplasm, HR hazard ratio, HTN hypertension, IUGR intrauterine growth retardation, PPH postpartum hemorrhage, VTE venous thromboembolism, DVT deep venous thrombosis, PE pulmonary embolism. %The total number here in this subgroup analysis is different (n = 4672 vs. n = 4563) compared to the main group analysis due to the difference in the number of active health care organizations at the time of performing the analysis within the database
Subgroup analysis
We performed various additional subgroup analyses within the MPN group.
Based on prior history of thrombosis or atrial fibrillation. We included these two indications as surrogate markers of higher baseline thrombotic risk and likely exposure to therapeutic anticoagulation.
Secondly, we performed a subgroup analysis by whether these patients were on Interferon therapy (IFN) to assess whether IFN has a disease-modifying effect on obstetric and thrombotic outcomes.
All analyses were conducted on the TriNetX platform, which utilizes Survival 3.2 to 3 in R 4.0.2 for survival analysis and incorporates data science and statistical libraries in Python 3.7 and Java 11.0.16 for various analytical functions, including propensity score matching and comparative cohort analysis.
Results
Among 4,467,127 pregnant patients, MPN was identified in 4563 patients (0.1%). Essential thrombocytosis was the most common MPN (85%, n = 3878), followed by myelofibrosis (11%, n = 502) and polycythemia vera (4%, n = 183). 62 patients (1.3%) had JAK2V617F status reported, out of which 75.8% (n = 47) were positive. Before propensity matching, the mean age of the MPN cohort was 27.2 years (± 5.0) as compared to 26.3 years (± 5.3) in the non-MPN group. Before matching, the groups were significantly different in their baseline comorbidities, such as type 2 diabetes, hypertension, BMI, ischemic heart disease, liver disease, and chronic kidney disease, and medications like antiplatelets and anticoagulants, as reflected in their SMD of more than 0.1. After propensity matching, both groups were adequately balanced in all their baseline demographics, comorbidities, and use of medications like antiplatelets and anticoagulants. (All post-match SMD values were less than 0.1) (Table 1).
Table 1.
Baseline characteristics of the study cohort
| Before propensity score matching | After propensity score matching | |||||
|---|---|---|---|---|---|---|
| Characteristic feature | Pregnancy with MPN (n = 4563) | Pregnancy without MPN (n = 4,266,392) | SMD | Pregnancy with MPN (n = 4529) | Pregnancy without MPN (n = 4529) | SMD |
| Basic demographics | ||||||
| Age at index, mean | 27.7 ± 5.0 | 26.3 ± 5.3 | 0.26 | 27.7 ± 5.0 | 27.7 ± 5.1 | 0.009 |
| White | 2586 (57.0) | 1,976,320 (46.6) | 0.20 | 2584 (57.1) | 2589 (57.2) | 0.002 |
| Black or African American | 1222 (26.9) | 716,269 (16.9) | 0.24 | 1218 (26.9) | 1206 (26.6) | 0.006 |
| Examination values | ||||||
| Body mass index, mean | 31.2 ± 14.6 | 28.5 ± 7.4 | 0.33 | 26.9 ± 5.3 | 27.1 ± 5.7 | 0.026 |
| Underlying comorbidities | ||||||
| Type 2 Diabetes mellitus | 247 (5.4) | 21,002 (0.5) | 0.29 | 245 (5.4) | 247 (5.5) | 0.002 |
| Hypertension | 400 (8.8) | 28,911 (0.7) | 0.39 | 395 (8.7) | 414 (9.1) | 0.015 |
| Hyperlipidemia | 246 (5.4) | 22,236 (0.5) | 0.29 | 243 (5.4) | 281 (6.2) | 0.036 |
| Ischemic heart disease | 47 (1.0) | 1416 (0.05) | 0.13 | 45 (1.0) | 29 (0.6) | 0.039 |
| Cerebrovascular diseases | 48 (1.1) | 2136 (0.1) | 0.13 | 48 (1.1) | 45 (1.0) | 0.007 |
| Atrial fibrillation and flutter | 19 (0.4) | 830 (0.03) | 0.08 | 19 (0.4) | 10 (0.2) | 0.035 |
| Diseases of the liver | 188 (4.1) | 9037 (0.2) | 0.27 | 184 (4.1) | 183 (4.0) | 0.001 |
| Chronic lower respiratory diseases | 618 (13.6) | 73,936 (1.7) | 0.45 | 614 (13.6) | 603 (13.3) | 0.007 |
| Chronic kidney disease | 56 (1.2) | 2381 (0.1) | 0.14 | 54 (1.2) | 39 (0.9) | 0.033 |
| Personal history of thrombosis and embolism | 83 (1.8) | 2,074 (0.1) | 0.18 | 80 (1.8) | 50 (1.1) | 0.056 |
| Overweight and obesity | 837 (18.5) | 69,487 (1.6) | 0.58 | 831 (18.3) | 878 (19.4) | 0.027 |
| Pulmonary HTN | 112 (2.0) | 1439 (0.05) | 0.16 | 110 (2.4) | 108 (2.3) | 0.008 |
| Antiplatelets | ||||||
| Any platelet aggregation inhibitor | 343 (7.6) | 25,220 (0.6) | 0.35 | 337 (7.4) | 323 (7.1) | 0.012 |
| Aspirin | 342 (7.5) | 25,041 (0.6) | 0.35 | 336 (7.4) | 323 (7.1) | 0.011 |
| Any therapeutic anticoagulants | 616 (13.6) | 39,617 (0.9) | 0.50 | 610 (13.5) | 559 (12.3) | 0.034 |
| Socioeconomic status | ||||||
| Patients with potential health hazards related to socioeconomic and psychological circumstances | 181 (4.0) | 17,549 (0.4) | 0.24 | 178 (3.9) | 171 (3.8) | 0.008 |
MPN myeloproliferative neoplasm, HTN hypertension, BMI body mass index, SMD standardized mean difference
Obstetric outcomes
Overall, the all-cause mortality of the cohort during follow-up is 0.7% (n = 64). The all-cause mortality rates are significantly higher in the MPN cohort as compared to the non-MPN cohort (1.1% vs. 0.2%, HR: 5.4 (95%CI: 2.7–10.8)). Other Obstetric outcomes, like the incidence of gestational hypertension (11.1% vs. 7.8%, HR: 1.3 (1.1–1.5)); preeclampsia and eclampsia (10.4% vs. 5.3%, HR: 1.8 (1.6–2.1)); preterm labor (6.1% vs. 3.7%, HR: 1.5 (1.2–1.8)); IUGR (7.3% vs. 5.1%, HR:1.3 (1.1–1.7)); PPH (5.5% vs. 3.8%, HR:1.3 (1.1–1.6)) were significantly higher in the MPN cohort as compared to the non-MPN cohort. The incidence of abruptio placenta, oligohydramnios, and polyhydramnios is not significantly different between the two groups (p > 0.05) (Table 2).
Table 2.
Obstetric and Thrombotic outcomes in pregnant MPN population
| Number of patients with outcome | ||||
|---|---|---|---|---|
| Outcomes | Pregnancies with MPN (n = 4529) (%) | Pregnancies without MPN (n = 4529) (%) | HR (95% CI) | Log-rank p-value |
| Obstetric outcomes | ||||
| All-cause mortality | 54 (1.1) | 10 (0.2) | 5.4 (2.7–10.8) | < 0.001 |
| Gestational HTN | 506 (11.1) | 355 (7.8) | 1.3 (1.1–1.5) | < 0.001 |
| Preeclampsia and eclampsia | 473 (10.4) | 243 (5.3) | 1.8 (1.6–2.1) | < 0.001 |
| Abruptio Placenta | 47 (1.03) | 30 (0.6) | 1.4 (0.9–2.3) | 0.102 |
| Oligohydramnios | 79 (1.7) | 59 (1.3) | 1.2 (0.8–1.7) | 0.199 |
| Polyhydramnios | 109 (2.4) | 77 (1.7) | 1.3 (0.9–1.7) | 0.065 |
| Preterm labor | 278 (6.1) | 171 (3.7) | 1.5 (1.2–1.8) | < 0.001 |
| IUGR | 335 (7.3) | 231 (5.1) | 1.3 (1.1–1.7) | < 0.001 |
| PPH | 251 (5.5) | 174 (3.8) | 1.3 (1.1–1.6) | 0.002 |
| Thrombotic outcomes | ||||
| Venous thrombotic outcomes | ||||
| VTE | 123 (2.7) | 36 (0.7) | 3.2 (2.2–4.6) | < 0.001 |
| DVT | 64 (1.4) | 28 (0.6) | 2.1 (1.3–3.3) | 0.001 |
| PE | 55 (1.2) | 14 (0.3) | 3.6 (2.0–6.6) | < 0.001 |
| Other venous thrombosis* | 11 (0.2) | 0 | -- | 0.001 |
| Arterial thrombotic outcomes | ||||
| Arterial events | 28 (0.6) | 22 (0.4) | 1.1 (0.6–2.0) | 0.544 |
| Stroke | 21 (0.4) | 19 (0.4) | 1.0 (0.5–1.9) | 0.946 |
| Myocardial infarction | < 10 | < 10 | -- | -- |
*Other venous thrombosis includes hepatic venous thrombosis, portal venous thrombosis, or cerebral venous sinus thrombosis. MPN myeloproliferative neoplasm, HR hazard ratio, HTN hypertension, IUGR intrauterine growth retardation, PPH post partum hemorrhage, VTE venous thromboembolism, DVT deep venous thrombosis, PE pulmonary embolism
Thrombotic outcomes
Thrombotic outcomes were divided into venous and arterial thrombotic outcomes. VTE events were significantly higher in the MPN cohort as compared to the non-MPN cohort (2.7% vs. 0.7%, HR: 3.2 (2.2–4.6)). Similarly, the individual DVT (1.4% vs. 0.6%, HR: 2.1 (1.3–3.3)) and PE (1.2% vs. 0.3%, HR: 3.6 (2.0–6.6)) were also significantly higher in the MPN cohort. The incidence of atypical site thrombosis (like HVT, PVT, and CVST) was also higher in the MPN cohort (n = 11 (0.2%) vs. 0, p < 0.001). In contrast, there was no significant difference in the arterial thrombotic events (0.6% vs. 0.4%, HR:1.1 (0.6–2.0)) (Table 2).
Subgroup analysis
Further, we checked whether having a history of a prior VTE event (either DVT or PE) or atrial fibrillation has any bearing on these outcomes. We found that all-cause mortality is significantly higher in those having a prior history of VTE or atrial fibrillation (10.3% vs. 0.6%, p < 0.001; HR: 6.6 (3.9–10.9)). The subsequent VTE rates were also higher in those with a prior history of VTE (25.2% vs. 0.9%, p < 0.001; HR:28.5 (19.5–41.4)). However, obstetric outcomes like gestational hypertension, preeclampsia, eclampsia, preterm labor, IUGR, and PPH were not significantly different between the subgroups (Table 3).
Secondly, we identified the MPN cohort on IFN therapy (n = 88, 1.9%) and analyzed whether IFN treatment affected these outcomes. The number of VTE events occurred in zero patients on IFN vs. 3.9% (n = 178) in those not on IFN. The remaining obstetric outcomes could not be evaluated due to a low number of events (Table 4).
Table 4.
Subgroup analysis based on whether the patient is on IFN*%
| Number of patients with outcome (n = 4563) | ||||
|---|---|---|---|---|
| Outcomes | Pregnancy with the MPN on IFN (n = 88) | Pregnancy with MPN not on IFN (n = 4475) | HR (95% CI) | Log-rank p-value |
| Obstetric outcomes | ||||
| Gestational HTN | 20 (22.7) | 1122 (25.0) | 1.3 (0.8–2.1) | 0.135 |
| Thrombotic outcomes | ||||
| VTE | 0 | 178 (3.9) | -- | -- |
*Arterial thrombotic events, Abruptio placenta, Oligohydramnios, Polyhydramnios; IUGR, PPH; individual VTE events could not be reported as their number is less than 10. MPN myeloproliferative neoplasm, HR hazard ratio, HTN hypertension, IUGR intrauterine growth retardation, PPH postpartum hemorrhage, VTE venous thromboembolism, DVT deep venous thrombosis, PE pulmonary embolism
Discussion
In this large multicenter retrospective propensity-matched study, we demonstrate that pregnancies complicated by BCR: ABL1-negative MPNs are associated with significantly increased risks of both adverse obstetric and thrombotic outcomes. After robust PSM, the presence of MPN remained independently associated with higher all-cause mortality (1.1%), hypertensive disorders of pregnancy (gestational HTN, preeclampsia, and eclampsia) (21.5%), preterm labor (6.1%), IUGR (7.3%), PPH (5.5%), and VTE (2.7%). These findings provide contemporary, real-world evidence that reinforces and extends prior observational data in this high-risk population [14].
We observed significantly increased incidence of hypertensive disorders of pregnancy, including gestational hypertension and preeclampsia/eclampsia, in the MPN cohort. These findings are consistent with prior studies suggesting that placental vascular dysfunction plays a central role in adverse pregnancy outcomes in MPN [15, 16]. Mechanistically, constitutive activation of the JAK-STAT pathway, most commonly due to the JAK2 mutation, promotes a pro-inflammatory and prothrombotic endothelial phenotype, which may impair placental perfusion and lead to ischemic placental disease [16]. This is further supported by the increased rates of IUGR and preterm labor observed in our study, both of which are well-recognized consequences of placental insufficiency [17]. Collectively, these adverse obstetric and thrombotic complications may contribute to the higher all-cause mortality observed in the MPN cohort (1.1%) as compared to the 2–3 deaths per 100,000 live births in a normal risk singleton pregnancy [18]. Overall, gestational hypertension (11.1%) is the most common maternal complication in our study, followed by preeclampsia and eclampsia (10.4%). In contrast, a meta-analysis of 1210 MPN pregnancies found that pre-eclampsia is the most common complication, with a pooled incidence rate of 3.1% [19]. Additionally, PPH risk is also increased after balancing for antiplatelet and anticoagulant usage, underscoring the complex hemostatic balance among patients with MPN. Given that ET constituted most of the cohort, qualitative platelet dysfunction, acquired von Willebrand syndrome, might have predisposed to bleeding complications [20]. Another study by Landtblom et al. showed a trend toward increased bleeding complications in MPN pregnancies compared to age-matched controls (14% vs. 9%, P = 0.11), and similarly, increased bleeding complications were reported in an analysis of 129 PV pregnancies [4, 21].
VTE incidence in patients with MPN in our index cohort is 2.7% (n = 123). Though we could not analyze the timing of VTE in the study (antepartum versus postpartum), in a meta-analysis of 756 MPN pregnancies, the absolute VTE risk in the antepartum and postpartum settings was 2.5% and 4.4%, respectively, and Maze et al. reported pooled incidences of VTE to be 1.5% [19]. In comparison, VTE complicates 0.1%−0.2% of pregnancies in the general population [22]. Routine thromboprophylaxis using low molecular weight heparin (LMWH) across the risk spectrum is recommended in the postpartum period as per current guidelines [23]. Additionally, arterial thrombotic events, including stroke and myocardial infarction, were not significantly increased in our study in the MPN group. This distinction is noteworthy, as arterial thrombosis is a recognized complication of MPN in the general population. The relative absence of increased arterial events in our cohort may reflect the younger age of the study population and the lower prevalence of atherosclerotic risk factors.
MPN not only affects maternal outcomes, but it also affects fetal outcomes. A large-scale population-based study of 342 pregnancies from Sweden found that preterm birth and low birth weight were significantly higher in the MPN cohort [4]. Our findings are also consistent with these observations, with a significantly higher incidence of preterm labor and IUGR as compared to the matched controls. Although the risk of thrombosis is increased among MPN patients in general, it is unclear if mutational status also predicts thrombosis or other maternal and fetal outcomes, with some studies finding associations of the JAK2 mutation with early and late pregnancy loss and other studies finding no such associations. One of the limitations of our study is the lack of information regarding the mutational impact [24–26]. Also, given the conflicting evidence, at this time, there are no changes in pregnancy management of MPN patients based on driver mutation status [14].
Further, the subgroup analysis based on the prior history of VTE or atrial fibrillation is noteworthy. These populations were included, assuming these patients are on therapeutic anticoagulation. We found that all-cause mortality is significantly higher in those with a prior history of VTE or atrial fibrillation (10.3% vs. 0.6%, p < 0.001). In addition, this subgroup demonstrated higher rates of recurrent VTE events (25.2% vs. 0.9%, p < 0.001) and a higher incidence of preeclampsia/eclampsia (14.1% vs. 10.1%, p = 0.037). Collectively, these findings may suggest that bad disease biology and higher-risk disease leading to higher thrombotic complications and obstetric complications contribute substantially to mortality, emphasizing the need for specialist-level multidisciplinary team care and tailored anticoagulation strategies in this high-risk cohort.
The exploratory analysis of IFN therapy, although limited by a small sample size (n = 88), is hypothesis-generating. There were no VTE events in the IFN-treated group compared with a 3.9% incidence in patients not on IFN. Interferon is increasingly used in younger patients with MPN due to its disease-modifying properties and safety profile in pregnancy [27]. Although no VTE events were observed among IFN-treated patients, the small sample size precludes inference regarding treatment efficacy and its impact on pregnancy outcomes.
Our study has several strengths. This was a large, population-based study that included a diverse cohort of patients. We used PSM to reduce selection or confounding bias commonly associated with observational studies. However, this study has several limitations. First, reliance on ICD-10 coding may have led to misclassification of exposures and outcomes. We lacked detailed clinical data, including laboratory parameters (e.g., hematocrit, platelet count), mutation status for most patients, disease duration, and treatment adherence. Although we adjusted for antiplatelet and anticoagulant use, data on dosing, timing, and adherence, particularly for low-molecular-weight heparin, were not available. The small number of patients receiving interferon limits the power of subgroup analyses. Additionally, the definition of pregnancy onset and timing of outcomes may be imprecise in real-world databases. Because the cohorts taking or not taking IFN were not analyzed, given a low sample size, the protective benefit of IFN could not be explored. Pregnancy-related factors might also have contributed to adverse pregnancy outcomes. We attempted to minimize these differences through propensity score matching, incorporating key variables that might influence the outcome. However, the possibility of unmeasured or residual confounders that could influence the association between MPN and adverse outcomes, including immortal time bias, cannot be completely excluded. We relied on ICD codes to identify obstetric and VTE outcomes, as we did not have access to individual electronic health records. Based on a systematic review, ICD codes have moderate-to-high sensitivity and specificity for the identification of VTE in electronic databases [28]. While the accuracy of VTE diagnosis could be improved by combining ICD codes with the use of anticoagulation, we were unable to do so given the constraints of the TriNetX platform. This introduces the possibility of misclassification of VTE outcomes. However, such misclassification would likely affect both cohorts similarly and therefore reduce their potential impact on the risk estimates [29]. It is also important to note that data from electronic health records do not capture medication adherence and, therefore, we were unable to assess or account for medication adherence in the analyses. Early pregnancy loss, including early miscarriage and later fetal demise, could not be reliably characterized in this dataset. Because pregnancy loss terminates subsequent exposure to many later obstetric outcomes, the inability to censor follow-up at pregnancy loss or account for it as a competing event may have influenced estimates of complications occurring later in pregnancy. Accordingly, these outcomes should be interpreted as coded events occurring within the 350-day follow-up window rather than as risks conditional on continuation of pregnancy. Further, the database did not permit reliable ascertainment of the delivery date or termination date for each pregnancy episode, precluding classification of thrombotic events as antepartum or postpartum. Therefore, thrombotic outcomes should be interpreted as events occurring within the prespecified follow-up period after the pregnancy index event rather than as pregnancy- or postpartum-specific events.
In conclusion, this large, real-world analysis demonstrates that pregnancy in patients with MPN is associated with a significantly increased risk of adverse maternal and thrombotic outcomes, particularly venous thromboembolism and hypertensive complications. These findings underscore the need for multidisciplinary management, risk stratification, and individualized prophylactic strategies. Future prospective studies and collaborative registries are needed to refine therapeutic approaches and improve outcomes in this complex and growing patient population.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We thank all the co-authors for their support in preparing the manuscript. An abstract version of this manuscript is presented as a poster at the Society of Hematologic Oncology (SOHO) annual meeting 2025.
Author contributions
NV: Conceptualization, Drafting initial manuscript, Statistical analysis, Critical Revision, Final Manuscript. SRS: Statistical analysis, Critical Revision, Final Manuscript. SN: Critical Revision, Final Manuscript. MVK: Critical Revision, Final Manuscript. RP: Critical Revision, Final Manuscript. CA: Critical Revision, Final Manuscript. GK: Critical Revision, Final Manuscript. VS: Conceptualization, Critical Revision, Final Manuscript. All authors reviewed the final version of the manuscript.
Funding
None.
Data availability
Data used for this analysis is found on the TrinetX research platform which is selectively available through institutional access.
Declarations
Ethical statement
Not applicable.
Patient consent
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Spivak JL (2017) Myeloproliferative Neoplasms. N Engl J Med 376(22):2168–2181 [DOI] [PubMed] [Google Scholar]
- 2.Barbui T, Thiele J, Vannucchi AM, Tefferi A (2016) Myeloproliferative neoplasms: Morphology and clinical practice. Am J Hematol 91(4):430–433 [DOI] [PubMed] [Google Scholar]
- 3.Alimam S, Bewley S, Chappell LC, Knight M, Seed P, Gray G et al (2016) Pregnancy outcomes in myeloproliferative neoplasms: UK prospective cohort study. Br J Haematol 175(1):31–36 [DOI] [PubMed] [Google Scholar]
- 4.Landtblom AR, Andersson TM, Johansson ALV, Wendel SB, Lundberg FE, Samuelsson J et al (2022) Pregnancy and childbirth outcomes in women with myeloproliferative neoplasms-a nationwide population-based study of 342 pregnancies in Sweden. Leukemia 36(10):2461–2467 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Williams N, Lee J, Mitchell E, Moore L, Baxter EJ, Hewinson J et al (2022) Life histories of myeloproliferative neoplasms inferred from phylogenies. Nature 602(7895):162–8 [DOI] [PubMed] [Google Scholar]
- 6.How J, Garcia JS, Mullally A (2023) Biology and therapeutic targeting of molecular mechanisms in MPNs. Blood 141(16):1922–1933 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gangat N, Tefferi A (2021) Myeloproliferative neoplasms and pregnancy: Overview and practice recommendations. Am J Hematol 96(3):354–366 [DOI] [PubMed] [Google Scholar]
- 8.Troiano NH (2018) Physiologic and Hemodynamic Changes During Pregnancy. AACN Adv Crit Care 29(3):273–283 [DOI] [PubMed] [Google Scholar]
- 9.Lapoirie J, Contis A, Guy A, Lifermann F, Viallard JF, Sentilhes L et al (2020) Management and outcomes of 27 pregnancies in women with myeloproliferative neoplasms. J Matern Fetal Neonatal Med 33(1):49–56 [DOI] [PubMed] [Google Scholar]
- 10.Robinson S, Ragheb M, Harrison C (2024) How I treat myeloproliferative neoplasms in pregnancy. Blood 143(9):777–85 [DOI] [PubMed] [Google Scholar]
- 11.Robinson SE, Harrison CN (2020) How we manage Philadelphia-negative myeloproliferative neoplasms in pregnancy. Br J Haematol 189(4):625–634 [DOI] [PubMed] [Google Scholar]
- 12.Palchuk MB, London JW, Perez-Rey D, Drebert ZJ, Winer-Jones JP, Thompson CN et al (2023) A global federated real-world data and analytics platform for research. JAMIA Open 6(2):ooad035 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lin TL, Chen YJ, Wu CY (2026) Choosing real-world data for clinical and epidemiological research: methodological lessons from NHIRD and TriNetX-A narrative review. Ann Med 58(1):2616549 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.How J, Hobbs G (2024) Pregnancy: MPN management before, during, and after pregnancy. Hematology Am Soc Hematol Educ Program 2024(1):541–6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lavi N, Brenner B, Avivi I (2013) Management of pregnant women with myeloproliferative neoplasms. Thromb Res 131(Suppl 1):S11–S13 [DOI] [PubMed] [Google Scholar]
- 16.Venugopal S, Mascarenhas J (2020) Novel therapeutics in myeloproliferative neoplasms. J Hematol Oncol 13(1):162 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kiladjian JJ, Chomienne C, Fenaux P (2008) Interferon-alpha therapy in bcr-abl-negative myeloproliferative neoplasms. Leukemia 22(11):1990–1998 [DOI] [PubMed] [Google Scholar]
- 18.Xu J, Murphy SL, Kochanek KD, Arias E (2022) Mortality in the United States, 2021. NCHS Data Brief. (456):1–8 [PubMed]
- 19.Maze D, Kazi S, Gupta V, Malinowski AK, Fazelzad R, Shah PS et al (2019) Association of treatments for myeloproliferative neoplasms during pregnancy with birth rates and maternal outcomes: a systematic review and meta-analysis. JAMA Netw Open 2(10):e1912666 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Tefferi A, Gangat N, Loscocco GG, Guglielmelli P, Szuber N, Pardanani A et al (2025) Essential thrombocythemia: a review. JAMA 333(8):701–14 [DOI] [PubMed] [Google Scholar]
- 21.Wille K, Brouka M, Bernhardt J, Rüfer A, Niculescu-Mizil E, Gotic M et al (2023) Outcome of 129 pregnancies in polycythemia vera patients: a report of the European LeukemiaNET. Hemasphere 7(5):e882 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Snow V, Qaseem A, Barry P, Hornbake ER, Rodnick JE, Tobolic T et al (2007) Management of venous thromboembolism: a clinical practice guideline from the American College of Physicians and the American Academy of Family Physicians. Ann Fam Med 5(1):74–80 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Gerds AT, Gotlib J, Ali H, Bose P, Dunbar A, Elshoury A et al (2022) Myeloproliferative Neoplasms, Version 3.2022, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 20(9):1033–1062 [DOI] [PubMed] [Google Scholar]
- 24.Rumi E, Bertozzi I, Casetti IC, Roncoroni E, Cavalloni C, Bellini M et al (2015) Impact of mutational status on pregnancy outcome in patients with essential thrombocytemia. Haematologica 100(11):e443–e445 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Passamonti F, Randi ML, Rumi E, Pungolino E, Elena C, Pietra D et al (2007) Increased risk of pregnancy complications in patients with essential thrombocythemia carrying the JAK2 (617V > F) mutation. Blood 110(2):485–489 [DOI] [PubMed] [Google Scholar]
- 26.How J, Leiva O, Bogue T, Fell GG, Bustoros MW, Connell NT et al (2020) Pregnancy outcomes, risk factors, and cell count trends in pregnant women with essential thrombocythemia. Leuk Res 98:106459 [DOI] [PubMed] [Google Scholar]
- 27.Edahiro Y (2022) Treatment options and pregnancy management for patients with PV and ET. Int J Hematol 115(5):659–671 [DOI] [PubMed] [Google Scholar]
- 28.Liu B, Hadzi-Tosev M, Eisa K, Liu Y, Lucier KJ, Garg A et al (2024) Accuracy of venous thromboembolism ICD-10 codes: a systematic review and meta-analysis. Thromb Updat 14:100154 [Google Scholar]
- 29.Baumgartner C, Go AS, Fan D, Sung SH, Witt DM, Schmelzer JR et al (2020) Administrative codes inaccurately identify recurrent venous thromboembolism: the CVRN VTE study. Thromb Res 189:112–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data used for this analysis is found on the TrinetX research platform which is selectively available through institutional access.