We read the report published by Badheka et al regarding the impact of symptoms, gender, co-morbidities, and operator volume on the outcome of carotid artery stenting. The investigators convincingly demonstrated that operator volume is an important predictor of postprocedural outcomes and resource utilization. There have been concerns regarding the accuracy of Nationwide Inpatient Sample in studying carotid stenting and carotid enderarterectomy, raised by several investigators in the past. These issues might be related to inadequate documentation of preprocedural symptoms and periprocedural strokes in the medical records, leading to subsequent coding errors in the hospital discharge abstracts from which Nationwide Inpatient Sample data are extracted.

A close look at the study reveals a few inconsistencies surrounding the codes used for analysis, especially in the differentiation of codes for stroke into preprocedure versus postprocedure. In the Supplementary Table 2 (Deyo modification of Charlson co-morbidity index), it has been stated that cerebrovascular disease includes International Classification of Diseases, Ninth Revision , codes from 430 to 438. A close review of Supplementary Table 1, detailing the postprocedural complications, reveals that several codes of the form 438.xx were included in the definition of postoperative stroke and/or TIA. These appear to be clearly duplicated in the preprocedural and postprocedural codes for stroke. The certainty of inclusion of these codes into the definition of postoperative outcomes remains unclear. In addition, this discrepancy would likely render the classification of the study population into symptomatic and asymptomatic cohorts a little questionable.