Ericka Teleg, MD
This study begins with emphasizing the lack of evidence in the management of pregnant patients or those in the early postpartum period confronted with acute ischemic stroke within the time window for endovascular therapy. In the advent of the landmark clinical trials on the benefit of endovascular therapy for acute ischemic stroke, it is the authors’ hypothesis that this particular group will also show a favorable clinical course and short-term outcomes likened to those found in the general population. Pregnant and postpartum women were systematically excluded from the clinical trials in acute reperfusion therapies for acute ischemic stroke. Acute reperfusion therapy with endovascular thrombectomy in the setting of pregnancy and the postpartum period is an important area to navigate. Pathophysiology of stroke among this population includes a hypercoagulable physiological state. It is important that this study answers this need in terms of benefit, complications, and outcomes, as stroke physicians are bound to encounter these complex cases in their lifetime.
This is a large cross-sectional analysis study that utilized the National Inpatient Sample (NIS) from year 2012 to 2018 and that is maintained by the Healthcare Cost and Utilization Project (HCUP), a large inpatient care database in the United States that is accessible to the public. Since a large, unweighted data is made available, approximately 7,000,000 patients, this allows for a robust data analysis. Patients selected were identified, include definition of pregnant and early postpartum patients, which are less than six weeks following delivery. Baseline demographic and clinical characteristics included age, race, other variables, acute stroke severity and manifestations. Clinical endpoints include functional outcome, all-cause in-hospital mortality, length of hospital stay and, importantly, discharge disposition. Neurological complications of relevance to treatment of acute ischemic stroke with mechanical thrombectomy included intracranial hemorrhage and need for decompressive hemicraniectomy. Other medical complications included thromboembolic events, deep venous thrombosis, pulmonary embolism, urinary tract infection, sepsis, myocardial infarction and acute kidney injury. More importantly, tertiary endpoints included pregnancy-related complications and puerperium complications.
The methodology included the use of descriptive statistics, multivariable logistic regression models, a 1:1 propensity-score matching protocol and chi-square analysis and t-test. The article justifies the use of these statistical methods and explains the need for each. It is important that comparison of the frequency in utilization of thrombectomy before years 2012 to 2014 and after years 2015 to 2018 in which the era of mechanical thrombectomy begun. This is necessary to gauge whether the thrombectomy usage among pregnant and postpartum patients paralleled that of the general ischemic stroke population. More importantly, in dealing with a large amount of data with multiple variables, their statistical methods more so address confounding and evaluate adjusted comparisons in clinical endpoints between this particular subgroup of patients treated with mechanical thrombectomy and those managed medically.
The results of this retrospective study identified 52,825 female patients, and among which 4,590 (8.7%) were pregnant or postpartum during the seven-year period. One hundred eighty pregnant and postpartum patients underwent mechanical thrombectomy for treatment of acute ischemic stroke.
Points highlighted in their results include:
1. In comparison with non-pregnant female patients treated with mechanical thrombectomy, they experienced lower rates of both intracranial hemorrhage and a poor functional outcome at discharge. The multivariable analysis adjusting for age, illness severity and stroke severity demonstrated that the pregnant and postpartum patients were associated with lower likelihood of developing intracranial hemorrhage. This is in contrast with the previous multicenter retrospective study that showed increased symptomatic intracranial hemorrhage following treatment with intravenous thrombolysis. They explained that their present analysis cannot distinguish between symptomatic and asymptomatic hemorrhage due to inability of billing codes, a younger cohort, and the presence of lesser burden of comorbid conditions.
2. Severe acute ischemic stroke patients experience progressive neurological deterioration due to progression of stroke and/or complications of reperfusion therapy. Decompressive hemicraniectomy is often performed in the setting of mechanical thrombectomy failure. Thus, their analysis demonstrated that there was a concurrent decrease in hemicraniectomy. Their finding paralleled national trends in year 2019 study that evaluated thrombectomy and decompressive hemicraniectomy for acute ischemic stroke patients in which upon the increased utilization of mechanical thrombectomy, there was a decrease in hemicraniectomy rates.
3. Pregnant and postpartum women treated with mechanical thrombectomy are already hypercoagulable, and general immobility complicates discharge and hospital state. Additional vigilance is necessary to keep a monitor for thromboembolic complications.
Additionally, limitations of the study identified by the authors include: the reliance on a database, billing codes rather than clinical data, and the inability to identify the patients who were ineligible for thrombectomy at presentation or the rate at which eligible patients were offered endovascular therapy. Also, there may be inconsistency with the use of these. There is an inability of such a study to demonstrate causation with regard to treatment modality and outcome.
Despite these limitations, as involvement of this particular subgroup of patients is a challenge to include in clinical trials, these results suggest that endovascular thrombectomy is viable and safe as first-line therapy in the setting of pregnancy and the postpartum period.
