Comparative Effectiveness of Carotid Endarterectomy vs Initial Medical Therapy in Patients With Asymptomatic Carotid Stenosis

Salomeh Keyhani, E. M. Cheng, K. J. Hoggatt, P. C. Austin, E. Madden, P. L. Hebert, E. A. Halm, A. Naseri, J. M. Johanning, D. Mowery, W.. Chapman, D. M. Bravata

JAMA Neurol. doi:10.1001/jamaneurol.2020.1427

Selezionato dal Lettore: Pietro Falco Università degli Studi di Roma La Sapienza

Motivation: In this study, the authors analyze the efficacy of early carotid endarterectomy (CEA) compared to initial medical therapy in clinical practice for fatal and non-fatal strokes prevention, among patients with asymptomatic carotid stenosis. The authors’ initial hypothesis is that CEA superiority, demonstrated by old randomized clinical trials (RCTs), is currently no longer reproducible in clinical practice for two main reasons: worse surgical outcome outside the selective RCTs inclusion criteria and new pharmacological advances, with more effective control of atherosclerotic risk factors.

The authors retrospectively selected 5˙221 patients from a registry of more than 170˙000 American war veterans, aged 65 years or older, who showed asymptomatic carotid stenoses ≥70% at carotid imaging and received early carotid endarterectomy (CEA) or initial medical therapy. From this population, they obtained two groups: one RCT-like cohort, with the strict inclusion criteria of the most recent RCT in the field, and a “pragmatic” cohort, similar to the general population, with non-selected patients. After mimicking a retrospective randomization to two treatments, the five-year cumulative incidence of stroke was analyzed. The results indicate, in the pragmatic cohort, a five-year risk of fatal and non-fatal stroke of 5,6% among patients randomized to CEA, compared with 7,8% in patients randomized to initial medical therapy. The risk difference of 2,3% was half lower than that showed in the most recent RCT, whose data are based on a population studied twenty-five years ago. Furthermore, accounting for the competing risk of death associated with non-stroke causes, there was a non-significant risk difference of 0.8%. Surprisingly, similar results were found in the RCT-like cohort too.

This study suggests for the first time that medical therapy may be an equally acceptable treatment strategy, respect to surgical approach, for patients with asymptomatic carotid stenosis in clinical practice. Both new pharmacological advances for atherosclerotic risk factors control and surgical outcome variability (poor patients’ selection, perioperative complications in low-volume hospitals) may contribute to this result. The large sample size, though only constituted by elderly males, and the accurate statistical analysis reinforce such conclusions, making this study of extreme usefulness in neurological clinical practice.