Dear Editor:We recently had the privilege of reading the article by Salim et al. [1], which integrates data from 37 centers across North America, Asia, and Europe to explore the management of acute ischemic stroke (AIS) due to distal medium vessel occlusion (DMVO). Specifically, the study compares intravenous thrombolysis (IVT) plus mechanical thrombectomy (MT) versus IVT alone to determine which approach is superior. The article mentions that there was no significant difference between the groups in achieving good functional recovery, and mortality rates at 90 days were similar between the two groups. However, while the incidence of symptomatic intracerebral hemorrhage was comparable, any type of intracranial hemorrhage was significantly higher in the MT-IVT group. These results provided us with further insights into the advantages and disadvantages of treatment strategies for distal medium vessel occlusion, which can be applied in clinical practice to improve patient outcomes. After reviewing the article, we believe there are several aspects worthy of further consideration.Firstly, we would like to address some statistical issues raised in the study. The authors mention that due to the rarity of DMVO in clinical practice, data collection was challenging. According to Table 3 and Supplementary Table 4 in the article, the IVT-only group showed better 90-day mRS 0-1 outcomes and lower mortality, although these differences did not reach statistical significance. Such results might be attributed to insufficient statistical power, specifically due to the smaller sample size. Since there are many more individuals in the unexposed group than in the exposed group, it may be worth considering a higher matching ratio in propensity score matching (PSM) to enhance statistical power [2].Secondly, the authors included baseline comorbidities, medication usage, and baseline mRS as covariates in the PSM to improve the matching of patient characteristics between the two groups. We believe this approach may be insufficient in matching patient characteristics. For example, prior studies have highlighted the impact of chronic kidney disease [3] and previous stroke history [4] on stroke outcomes, yet these factors were not considered in the present study. Additionally, other established cardiovascular diseases [5,6], such as coronary artery disease and peripheral artery disease, were not explored in this study. While the unaccounted-for confounders may or may not introduce bias, we suggest that these factors should be incorporated as covariates in the PSM to obtain more accurate results. Similarly, in the subgroup analysis presented in Figure 3, the treatment approach appears to influence the outcomes. We recommend including this variable in the PSM covariates as well to avoid bias resulting from differences in treatment.Furthermore, the study utilized missing data imputation to address the issue of missing data, increasing the available sample size and enhancing statistical power, which we consider to be a commendable approach. However, we have concerns regarding the assumption that all missing data were either completely missing at random or randomly missing. Upon examining Supplementary Figure 1, we observe missing data in the imaging category. This may be due to the fact that these patients had worse outcomes, leading to missing imaging data (i.e., patients may have died prior to imaging). In this case, the missing data could be considered ”missing not at random” (MNAR), and thus, imputation methods may not be valid for these cases [7]. Additionally, in Supplementary Table 4, the sensitivity analysis indicated that although the outcomes before and after imputation were largely similar, the hemorrhagic infarction type 2 became statistically significant after imputation, which necessitates more rigorous handling of missing data, as it could indeed affect the study’s results.Lastly, we have some concerns regarding the numbers and expression in certain tables. In Table 3 (ICH by type), both HI1 and SAH outcomes showed higher values in the MT-IVT group, but the numbers in parentheses were lower. We believe there was a typographical error, where 14 was mistakenly presented as 1.4, and 11 as 1.1. Although this is a minor error, it could potentially affect the reader’s understanding of the data.Overall, we are honored to have read the article by Salim et al. [1] and gained a deeper understanding of the management of DMVO. The suggestions provided here in no way detract from the value of the study, but we believe they can serve as important considerations for future research and development in this field.References:Salim HA, Yedavalli V, Musmar B, et al. Mechanical Thrombectomy Versus Intravenous Thrombolysis in Distal Medium Vessel Acute Ischemic Stroke: A Multinational Multicenter Propensity Score-Matched Study. J Stroke . 2024;26(3):434-445.Austin PC. Informing power and sample size calculations when using inverse probability of treatment weighting using the propensity score. Stat Med. 2021;40(27):6150-6163.Ghoshal S, Freedman BI. Mechanisms of Stroke in Patients with Chronic Kidney Disease. Am J Nephrol. 2019;50(4):229-239.Chen Y, Wright N, Guo Y, et al. Mortality and recurrent vascular events after first incident stroke: a 9-year community-based study of 0·5 million Chinese adults. Lancet Glob Health. 2020;8(4):e580-e590.Müller MD, Jongen LM, Altinbas A, et al. Silent Intracerebral Hemorrhage in Patients Randomized to Stenting or Endarterectomy for Symptomatic Carotid Stenosis. J Stroke. 2019;21(1):116-119.Lincoff AM, Brown-Frandsen K, Colhoun HM, et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes. N Engl J Med. 2023;389(24):2221-2232.Heymans MW, Twisk JWR. Handling missing data in clinical research. J Clin Epidemiol. 2022;151:185-188.