Letter to the editor regarding "Clinical outcomes of modified partial eversion carotid endarterectomy: 27-month follow-up"
Abdullah Burak Balci
DOI: 10.9739/tjvs.2026.09.063 · 1 Views · 0 Downloads · 0 Citations
Abstract
DEAR EDITOR
We read with interest the article by Pinar et al. reporting the mid-term clinical outcomes of modified partial eversion carotid endarterectomy [1]. The authors present a neat single-center cohort undergoing an interesting technical variation performed by a single surgical team with low perioperative complication rates and an average clamp time of 16 minutes.
While these early results are encouraging, several methodological and analytical points warrant further discussion.
First, the primary endpoint was restricted to overall survival. While mortality is a major benchmark, the surgical durability of carotid endarterectomy is primarily reflected in stroke prevention, freedom from transient ischemic attack, ipsilateral restenosis, and reintervention rates, aligning with major European Society for Vascular Surgery (ESVS) guidelines [2]. Because these stroke- and carotid-specific surveillance endpoints were not systematically tracked, the current data establish short-term procedural safety rather than the long-term anatomical or neurological efficacy of MPE-CEA. Furthermore, given that all arteriotomies were closed primarily without patch angioplasty, systematic postoperative duplex ultrasound data would have been essential to evaluate early and late restenosis risks.
Second, chronic kidney disease (CKD) was identified as an independent predictor of overall mortality. Although this association is clinically plausible and consistent with broader vascular surgical literature [3], only five patients had CKD in this cohort, four of whom died. This extremely small event subset yielded a remarkably wide confidence interval (HR: 10.446; 95% CI: 1.257–86.777), indicating substantial uncertainty around the true effect size. In addition, the manuscript lacks specific baseline definitions for CKD, such as eGFR cutoffs, disease staging, or dialysis status.
Third, the multivariable Cox proportional hazards model included three covariates despite recording only seven death events during follow-up. Incorporating three predictors for seven events raises a clear risk of model overfitting and parameter instability, as statistical guidance generally recommends a minimum of 10 to 15 events per variable [4]. Variable selection based solely on univariate significance further exacerbates this risk. Clarifying whether proportional hazards assumptions were evaluated or if model overfitting was guarded against would assist in interpreting these findings.
Fourth, the reported 5-year Kaplan–Meier survival estimate requires careful reading. With a mean follow-up of 27 months, the number of patients remaining at risk beyond 3 to 4 years was likely minimal. Providing the number of patients at risk beneath the survival curves and reporting median follow-up times with interquartile ranges would give readers a clearer picture of late survival reliability. Lastly, a minor numerical discrepancy appears in Table 1 [1]: the survival group lists 33 symptomatic patients as 48%, which does not mathematically correspond to 33 out of 48 patients (68.8%).
In conclusion, Pinar et al. demonstrate that MPE-CEA can be performed safely with low immediate morbidity. However, the absence of a control group, lack of routine duplex surveillance, and limited neurological follow-up restrict firm conclusions regarding comparative superiority or long-term durability. Future prospective studies with standardized renal staging and objective duplex-based vascular outcomes will better clarify the exact place of this technical modification in carotid surgery.