Higher-Quality Data Collection Is Critical to Establish the Safety and Efficacy of Pediatric Mechanical Thrombectomy

• Published in: Stroke (1970) Vol. 52; no. 4; pp. 1213 - 1221
• Main Authors: Barry, Megan, Barry, Dwight, Kansagra, Akash P, Hallam, Danial, Abraham, Michael, Amlie-Lefond, Catherine
• Format: Journal Article
• Language: English
• Published: United States 01.04.2021
• Subjects: publication bias; odds ratio; morbidity; thrombectomy; child
• ISSN: 0039-2499; 1524-4628
• DOI: 10.1161/STROKEAHA.120.032009

Because children often have lifelong morbidity after stroke, there is considerable enthusiasm to pursue mechanical thrombectomy in childhood stroke based on literature reports. However, current published data may reflect inconsistent reporting and publication bias, which limit the ability to assess safety and efficacy of mechanical thrombectomy in childhood stroke. This retrospective cohort study compared reporting quality and clinical outcomes for mechanical thrombectomy between a trial-derived cohort of 42 children treated with mechanical thrombectomy for acute stroke at study sites and 133 patients reported in the literature. National Institutes of Health Stroke Scale at baseline, 24 hours after mechanical thrombectomy, and at discharge were compared between study site patients and literature patients. Odds ratios (ORs) were used to compare reporting frequencies. Proportional odds logistic regression was used to compare outcomes. Premechanical thrombectomy National Institutes of Health Stroke Scale was available in 93% of study patients compared with 74% of patients in the literature (OR, 4.42 [95% CI, 1.47-19.89]). Postmechanical thrombectomy National Institutes of Health Stroke Scale was available in 69% of study patients compared with 29% of literature cases at 24 hours (OR, 5.48 [95% CI, 2.62-12.06]), and 64% of study patients compared with 32% of cases at discharge (OR, 3.85 [95% CI, 1.87-8.19]). For study sites, median scores were 12 at baseline, 9 at 24 hours, and 5 at discharge. Median scores in case reports were 15 at baseline, 4 at 24 hours, and 3 at discharge. ORs for differences in outcomes between groups were 5.97 (95% CI, 2.28-15.59) at 24 hours and 3.68 (95% CI, 1.45-9.34) at discharge. Study site patients had higher rates of National Institutes of Health Stroke Scale reporting and worse short-term outcomes compared with literature reports. Rigorous data collection is needed before treatment guidelines for pediatric mechanical thrombectomy can be developed.