[Skip to Navigation]
April 2011

Methodological Remarks Concerning the Recent Meta-analysis of Carotid Artery Stenting vs Carotid Endarterectomy—Reply

Author Affiliations

Author Affiliations: New York University Medical Center, New York (Dr Bangalore); and Brigham and Women's Hospital and Veterans Affairs Boston Healthcare System, Boston, Massachusetts (Dr Bhatt).

Arch Neurol. 2011;68(4):543-544. doi:10.1001/archneurol.2011.40

In reply

We thank Economopoulos et al for their interest in our article.1 Some of the methodological issues outlined by the authors pertaining to data abstraction are owing to insufficient details in the original articles. While we were able to obtain clarification when some of the authors responded to our query, others did not. For example, the TESCAS-C trial provides data on individual components of the endpoint of death, nonfatal stroke, or nonfatal myocardial infarction and there is limited description of these end points. However, for composite outcomes such as those evaluated in our article, it is not clear from the original publication if the same patient had 2 events, ie, if a patient had periprocedural myocardial infarction and later died during longer-term follow-up. Thus, the composite, as outlined in the letter by Economopoulos et al as a simple addition of individual components, might not be completely accurate. The authors are surprised by the numbers for SPACE and speculate that no periprocedural myocardial infarction occurred in SPACE. In fact, the authors are correct in their speculation. In SPACE, there was no routine measurement of periprocedural enzymes or routine evaluation of the electrocardiogram, and based on the SPACE investigators' data, there were no periprocedural myocardial infarction. This is mentioned in the study limitations.1 In addition, the authors refer to discrepant data abstraction in the Endarterectomy vs Angioplasty in Patients With Symptomatic Severe Carotid Stenosis trial for the outcome of any stroke. These data are in Table 2 of the original publication, and the numbers abstracted are correct. Additionally, in Figure 6A, the Carotid and Vertebral Artery Transluminal Angioplasty Study data refer to death or disabling stroke. Finally, despite the challenges referred to above, we have reanalyzed some of the data based on the numbers presented by the authors, and the results are similar to our published data1 (Table).

Add or change institution