A new Vascular Quality Initiative (VQI) data analysis, recently published in the European Journal of Vascular and Endovascular Surgery (EJVES), has found no statistically significant differences between three carotid endarterectomy (CEA) shunting strategies regarding in-hospital stroke and death rate, including in patients with contralateral carotid occlusion or recent stroke.
In their editor’s choice paper, Xavier Hommery-Boucher (Centre Hospitalier de l’Université de Montréal [CHUM], Montreal, Canada) and colleagues outline that the study aimed to evaluate in-hospital outcomes after CEA according to shunt usage, particularly in high-risk groups of patients such as those with contralateral carotid occlusion or recent stroke.
Considering the context for their study, the authors underline a lack of data on the topic of carotid shunting. They note that systematic reviews comparing the three shunting strategies have concluded the evidence is too limited to support one over the others, and that there are no randomised data available on the subject.
In their study, Hommery-Boucher et al set out to perform a registry-based analysis. Specifically, they analysed data from CEAs registered in the VQI database between 2012 and 2020, excluding surgeons with fewer than 10 CEAs registered in the database, concomitant procedures, reinterventions, and incomplete data.
The authors note that participating surgeons were divided into three groups based on their rate of shunt use: non-shunters (<5%); selective shunters (5–95%), and routine shunters (>95%). They analysed primary outcomes of in-hospital stroke, death, and stroke and death rate in both symptomatic and asymptomatic patients.
Hommery-Boucher and colleagues share that, in total, 113,202 patients met the study criteria. Of this total, 31,147 were asymptomatic, while a majority of 82,055 were asymptomatic.
Writing in EJVES, the authors report that 12.1% of the 1,645 surgeons included in the study were non-shunters, while 63.6% were selective and the remaining 24.3% were routine shunters. The number of procedures in each group was 10,557, 71,160, and 31,579, respectively.
Hommery-Boucher et al reveal that, in the symptomatic cohort, in-hospital stroke, death, and the combined stroke and death rate were not statistically different among the three groups, based on univariable analysis. Similarly, they note that the asymptomatic group also did not show a statistically significant difference for any of the three primary outcome measures.
The authors go on to state that a multivariable model did not show a statistically significant difference for the primary outcomes between the three shunting cohorts and that, on subgroup analysis, the stroke and death rates were not statistically significantly different for patients with contralateral carotid occlusion and those presenting with a recent stroke.
“This paper adds new data collected from a large registry regarding postoperative outcomes related to shunt use during [CEA],” Hommery-Boucher and colleagues write in EJVES, remarking that the design of the study enabled comparison between three shunting strategies.
The authors summarise that the results of their study “demonstrated that there was a two-fold increase in the percentage of surgeons using the non-shunting strategy between 2012 and 2020, with no significant difference in outcomes compared with the other two strategies”.
In the discussion of their findings, Hommery-Boucher et al recognise some limitations of their research, including those “intrinsic” to the use of a large database like the VQI.
The researchers conclude that, despite its limitations, their study “could not define a preferential shunting strategy,” leading them to advise that the strategy “should be mainly based on a surgeon’s preference and skillset”. They go on to stress that their study has “provided quality data on the impact of a surgeon’s shunting pattern on postoperative stroke and death rate, particularly for the most at-risk groups”.
