Author + information
- Masoor Kamalesh, MD⁎ ( and )
- Thomas Sharp, MD
- ↵⁎Krannert Institute of Cardiology, 1481 West 10th Street, Indianapolis, Indiana 46202
In their recent letter about our paper (1), Drs. Farooq and Serruys make the assertion that the VA CARDS (Veterans Affairs Coronary Artery Revascularization in Diabetes) study is not applicable to contemporary coronary revascularization based on: 1) the angiographic inclusion criteria being too strict; 2) the small percentage of screened patients who were enrolled; and 3) that our study was underpowered to evaluate SYNTAX (Synergy Between PCI With Taxus and Cardiac Surgery) scores. The letter completely misses important aspects of our study.
Our angiographic criteria were based on subsets of patients known to have better survival with surgery than with medical treatment. We deliberately excluded patients when the primary role of revascularization would be symptom relief. These patients were extensively studied in COURAGE (Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation) and BARI 2D (Bypass versus Angioplasty Revascularization Intervention- 2 Diabetes) (2,3). Including these subsets would have increased enrollment but also would have diluted the power of the study to find survival differences.
Our screened patients included all patients with diabetes referred for a diagnostic angiogram for any reason. To compare VA-CARDS to SYNTAX, we need to know the total number of patients having diagnostic angiography at the 85 sites over their 2-year enrollment (4). An average of 500 diagnostic angiograms per year per SYNTAX site would yield a total of 85,000 diagnostic coronary angiograms. The 1,800 patients enrolled in SYNTAX would then represent 2.1% of this total, which is lower than our study.
Our study was not designed or powered to examine SYNTAX score subgroups. The SYNTAX scores in our study merely show that there was no systematic bias in the distribution of scores to explain the observed survival difference. If anything, low SYNTAX scores were more frequent for PCI than surgery. It is important to note in this discussion that the SYNTAX trial itself was not powered to compare small subgroups based on SYNTAX terciles. There is no SYNTAX score that leads to an absolute improvement in outcome for percutaneous coronary intervention over surgery among patients with 3-vessel coronary disease. The failure to find a significant p value in the subanalysis of 352 patients with low SYNTAX scores and 3-vessel coronary artery disease is likely to represent a type II error (5). The SYNTAX investigators need to report a power analysis of each of the subgroups that they analyze. The assumption that the failure to find a difference means that there is no difference is misleading.
- American College of Cardiology Foundation
- Kamalesh M.,
- Sharp T.G.,
- Tang X.C.,
- et al.
- Mohr F.W.,
- Morice M.C.,
- Kappetein A.P.,
- et al.