Pregnancy-associated stroke (PAS) is defined as a stroke occurring during pregnancy, delivery, or within 6 weeks postpartum. It is a rare but catastrophic pregnancy complication that may seriously affect maternal and infant outcomes. Unlike strokes in the general population, which are dominated by ischemic, hemorrhagic stroke (HS) accounts for roughly half of PAS (1, 2). According to previous studies, HS is more highly associated with maternal mortality compared to other types of strokes (3–5).

Hypertensive disorders of pregnancy (HDP) are recognized risk factors of pregnancy-associated HS, affecting 7.3% of all pregnancies (6–9). These disorders consist of the following five categories: (1) chronic hypertension, defined as high blood pressure (BP) predating the pregnancy or discovered before 20 weeks gestation; (2) gestational hypertension, refers to persistent new onset hypertension that develops at or after 20 weeks gestation; (3) pre-eclampsia, refers to diagnosed hypertension between 20 weeks of gestation and puerperium, accompanied by proteinuria or multisystem organ failure; (4) eclampsia, refers to convulsions occur during pregnancy and puerperium, and cannot be attributed to any other pre-existing neurological disease; (5) pre-eclampsia superimposed upon chronic hypertension (9). In addition, there is also a HELLP (H = Hemolysis, EL = Elevated Liver enzymes, LP = Low Platelets) syndrome, which is currently regarded as a variant of severe preeclampsia in some degree (10). Among all types of PAS, HS has the greatest correlation with HDP (11). High-capacity circulation, endothelial dysfunction, and blood-brain barrier disruption resulting from HDP may play a key role in increasing the risk of developing HS during pregnancy and puerperium (6, 12).

Despite HDP having been confirmed as a strong risk factor for stroke in women of childbearing age, there is a dearth of literature examining its impact on PAS outcome, as the majority have conducted prognosis studies in the wider pregnant population (1, 11, 13). Therefore, in the present study, we intended to evaluate the impact of HDP on short-term and long-term outcomes of pregnancy-associated HS to promote a better understanding of their relevance. To our best knowledge, this is the first study considering the long-term functional outcome of PAS.

The study was approved by the Ethics committee of West China Hospital of Sichuan University, Chengdu, China, and conducted according to the Declaration of Helsinki. Written informed consent was obtained from each patient or legal surrogate.

Ninety-four pregnancy-associated HS patients were enrolled, of whom twenty-two were with diagnosed HDP during recent pregnancy. Table 1 summarizes the demographic information and stroke-associated clinical condition. The young women included in this study had almost no chronic diseases other than hypertension (only one nephrotic syndrome in the HDP group and one congenital heart disease in the non-HDP group). Therefore, in the following data analysis process, we neglected the presence of comorbidity. Headache [11 (50%) vs. 49 (68.1%)] was the most common initial symptom in both groups and there was no significant difference in the initial symptom between the two groups except disturbance of consciousness (45.5 vs. 19.4%, p = 0.014). In addition, patients with HS with HDP tended to present worse consciousness levels when admitted (p = 0.003). Table 2 summarizes the general conditions of patients in the HDP group. There were five chronic hypertension, five gestational hypertension, four pre-eclampsia, five eclampsia, and three pre-eclampsia superimposed upon chronic hypertension. Among five patients with eclampsia, two had developed HELLP syndrome, and three of the whole patients (3/22, 13.6%) indicated that they had also suffered from HDP in the previous pregnancy.

Table 3 summarizes the short-term outcomes of included patients. Despite patients diagnosed with HDP during recent pregnancy showed poorer functional outcomes at discharge in univariate analysis (OR = 3.22, 95% CI = 1.07–9.65, p = 0.037), there was no significant difference between the two groups after adjustment (77.3 vs. 51.4%, aOR = 1.43, 95% CI = 0.41–4.96, p = 0.572). Figure 1 summarizes the distribution of mRS scores within each group at discharge, from which we could directly figure out that mRS scores are concentrated at high scores in the HDP cohort but evenly distributed in the non-HDP cohort.

Table 4 summarizes the stroke-related clinical complication and life-support therapy received by patients during hospitalization. Except for univariate analysis on intravenous nutrition usage, there was no statistically significant difference in the incidence of stroke-related complications or life-support therapy between the two groups.

Table 5 summarizes the long-term outcomes among the two groups. Follow up averaged 4.7 ± 3.6 years, the total missing rate was 10.64% (10/94, two patients in the HDP group and eight patients in the non-HDP group), and there was still no significant difference in death rate between the two groups at the last follow-up (30.0 vs. 12.5%, aOR = 1.55, 95% CI = 0.40–5.92, P = 0.526). However, patients in HDP groups showed much poorer functional outcomes than those patients in non-HDP groups, and a significant difference was maintained even after adjusting for confounders (70.0 vs. 23.4%, aOR = 4.47, 95% CI = 1.28–15.67, P = 0.019).

Despite that a considerable part of stroke caused by HDP could be prevented through early identification and treatment, the incidence of stroke among women with HDP still increased over time (15, 16). Our study is the first to systematically evaluate the impact of HDP on pregnancy-associated HS prognosis. The main findings of this study are as follows: (1) Among pregnancy-associated HS patients, the long-term functional outcomes of patients diagnosed with HDP were significantly poorer; (2) HDP may not impact the short-term outcomes of pregnancy-associated HS neither increase the occurrence of stroke-related complications or usage for life-support therapy.

For PAS, HDP has always been the focus of attention. Previous studies have established that HDP is an important independent risk factor for HS in women of childbearing age (16–19). In addition, compared with other etiologies of HS such as aneurysms or arterial malformations, HDP possesses the valuable characteristic of “preventable” (4, 18, 20). Inadequate BP control is blamed as the most common mistake for stroke-related maternal death, in other words, adequate BP control may give maximum preventability to stroke-induced poor outcomes in these patients (21). As is known to all, elevated BP is the cornerstone of HDP, which means a considerable proportion of HDP-related stroke deaths can be avoided if timely and correctly treated. Oral labetalol, methyldopa, and nifedipine are the frequently recommended antihypertensive medications during pregnancy, and intravenous hydralazine should be additionally added for acute hypertensive urgency (22). However, despite all guidelines acknowledging the significance of BP control, there is still no consensus on the target BP level for pregnant women due to a lack of high-quality evidence. Further studies on lowering BP therapy among pregnant women with HS are warranted.

Another key element to treat HDP is “placenta delivery” (23). Though the pathophysiology of HDP has yet to be fully elucidated, the theory that the release of antiangiogenic factors caused by placental vascular maldevelopment leads to maternal endothelial dysfunction and resulting hypertension is generally accepted (24). According to the International Society for the Study of Hypertension in Pregnancy guidelines and the American College of Obstetricians and Gynecologists Task Force, in women with preeclampsia with severe features or eclampsia, emergency delivery soon after maternal stabilization is indicated irrespective of gestational age (9, 25). Although our study eliminated the obstetric complication analysis due to its high data missing rate, the roughly estimated preterm birth rate was extremely higher within the HDP group (40.9 vs. 16.9%).

It can be seen that most of our patients (both in the HDP and non-HDP groups) had their onset in the third trimester and puerperium, which is in line with the previous study's findings (11, 26). This may be related to the dramatic increase in plasma volume and cardiac output in the third trimester, which is attributed to the general vasodilatation of the cardiovascular system combined with an increase in water volume due to water retention (27, 28). While in the puerperium, all the above changing parameters rapidly return to normal, leading to an increase in vascular osmotic pressure, a decrease in blood volume, and a relatively vascular constriction, which eventually leads to a mild increase in blood pressure (28). This set of changes may noticeably increase the risk of postpartum intracranial bleedings.

There is limited literature on the clinical outcomes of pregnant women with HS, especially on long-term outcomes. In a study compared the short-term outcomes (in-hospital mortality and home-discharge rate) between pregnant and non-pregnant HS patients, they found that pregnant SAH patients had lower in-hospital mortality, higher home discharge and independent ambulation rate than their non-pregnant counterparts, and pregnant ICH patients also had lower odds of in-hospital mortality than non-pregnant ICH patients (13). The reported overall (ICH+SAH) in-hospital mortality of pregnant-associated HS in this study was 6.7%, which is similar to our results (6.4%). A previous study using the 1994–2011 Nationwide Inpatient Sample examined the stroke-related complication and showed that among stroke hospitalizations, patients with HDP had higher rates of complications than those without it (16). However, this study did not conduct further analysis based on stroke subtypes, and worse outcomes in the HDP group may be associated with a higher proportion of HS which generally induces more severe clinical conditions. Race is a proven independent risk factor for stroke occurrence whether in women of whole childbearing age or gravida with HDP, and it is also an influencing factor related to maternal outcomes of PAS (7, 29, 30). Some Asian population-based studies have shown that Asian women are prominently more likely to develop HS during pregnancy or puerperium compared to western countries (18, 19). Therefore, it is crucial to identify the impact of HDP on the prognosis of pregnancy-associated HS based on Asian women.

Consistent with the characteristics of “hypertensive” of HDP, the admission BP is significantly higher in the HDP group in our study. Some studies argued that the pathophysiology of stroke concerning HDP is similar to hypertensive encephalopathy in that the increase of BP results in the automatic regulation disturbance of cerebral blood flow, cerebral hyper-perfusion, blood-brain barrier disruption, brain edema, and hypertension-induced vessel wall damage (15, 31, 32). However, many women who develop stroke in the setting of preeclampsia or eclampsia were exactly at BP within the normal range of cerebral autoregulation which is considerably lower than those reported for hypertensive encephalopathy (33). Thus, elevated BP is not the only causative factor for this condition, and other mechanisms, such as endothelial dysfunction, increased sympathetic activity, and hyperreflexia that occurs in pre-eclampsia may also be involved (33–35). Further research on the pathophysiology of stroke related to HDP is warranted.

There are several limitations to be noted. First, this is a single-center retrospective study. The number of included patients is small, which restricted us from performing subgroup analysis within each HDP subtype. In addition, some data may be missing due to its retrospective nature, such as maternal birth-related complications. Second, it was reported that subjects with anemia, heart disease, or gestational diabetes mellitus, which factors are also correlated with stroke outcomes, have a higher risk of HDP occurrence when compared to their corresponding controls (36, 37). However, we did not include these confounders in our multivariable logistic regression due to the limited sample size. Third, due to the retrospective design and the large confidence interval of follow-up, the collection of specific data regarding potential confounders such as rehabilitation activities, complications, and readmission status of patients after discharge was insufficient. Fourth, the number of fetuses and the time of pregnancies are associated with the stroke risk (38), and the possibility of their impact on outcomes cannot be ruled out. Due to the limited availability of relevant clinical data, our study omitted this part's analysis, which may bias the results. Our study for the first time determines the impact of HDP on short-term and long-term outcomes of pregnancy-associated HS by systematically comparing in-hospital death, discharge, and post-hospital mRS score, stroke-related complications, and life-support therapies. We hope this study could lay the groundwork for future research on this topic.

In this retrospective study, women with hypertension disorders of pregnancy did not show worse short-term outcomes of pregnancy-associated hemorrhagic stroke compared to those without but had poorer long-term functional outcomes. This underlines the importance of prevention, recognition, and treatment of hypertension disorders in these women.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by Ethics Committee of West China Hospital of Sichuan University, Chengdu, China. The patients/participants provided their written informed consent to participate in this study.

Conceptualization and writing—original draft preparation: MF. Validation: WX and CT. Investigation: JW and ZW. Writing—review and editing: FX. Visualization: MF, YC, and FX. Supervision: LM and CY. Project administration: XH. All authors approved the final version of the manuscript.