Response to Letter Regarding Article, “Long-Term Survival and Repeat Coronary Revascularization in Dialysis Patients After Surgical and Percutaneous Coronary Revascularization With Drug-Eluting and Bare Metal Stents in the United States”
We appreciate the interest of Dr Magalhaes and colleagues in our recent article.1 Our responses to their points are as follows.
(1) We agree that factoring in the acuity of the clinical presentation during revascularization is pertinent because it may influence revascularization decisions. We performed additional analyses pertaining to this issue, which we will present at the upcoming 2013 American Heart Association Scientific Sessions in Dallas, Texas; we hope to see Drs Magalhaes, Waksman, and Barbash there to review our (currently embargoed) findings. (2) To address concerns about right censoring, we provide the survival probability (upper and lower confidence intervals with the standard error [SE]) for each cohort at 6 years: coronary bypass group, 22.3% (20.6%, 24%; 0.9% SE); internal mammary graft group, 23.9% (21.7%, 26.2%; 1.2% SE [21% shown in the article was a typographic error]); non-internal mammary graft group, 18.5% (16.0%, 21.1%; 1.3% SE); drug-eluting stent group, 18.5% (17.4%, 19.7%; 0.6% SE); and bare-metal stent group, 14.6% (12.6%, 16.6%; 1.0% SE). Based on the SE estimates, we believe that the degree of imprecision even at 6 years was limited. Furthermore, 218 patients (0.95%) were censored from the study because of renal transplant; these small numbers are also unlikely to have influenced the overall results. (3) Our study focused on the outcomes of dialysis patients undergoing coronary revascularization; by nature of the study design, dialysis patients with coronary artery disease not undergoing revascularization were not included. (4) We agree that the survival improvement in the study population over time could be related to factors other than revascularization, such as improvements in medical care. (5) As acknowledged in our article, lack of clinical variables, such as coronary angiography, left ventricular ejection fraction, etc, is an inherent limitation of administrative databases; the advantage of the United States Renal Data System database is that it is truly representative of US dialysis patients. (6) Staged procedures may indeed be performed in certain scenarios; however, one important clinical reason for performing a staged procedure is prevention of contrast nephropathy, which may not be as applicable to patients already receiving dialysis. Because staged procedures very likely occur within a few weeks of the index procedure, we further analyzed the data and determined that repeat revascularization occurred in only 1009 patients (4.5%) within 1 month of the index revascularization procedure. Thus, we conclude that staged procedures did not contribute substantially to repeat revascularization procedures. Finally, we made a concerted effort in our observational study to refrain from any direct comparisons between different revascularization modalities because we recognize that choice of revascularization procedure may involve selection bias. We encourage clinicians to adopt an individualized approach to management of dialysis patients undergoing coronary revascularization, with an enhanced understanding of the pros and cons of each revascularization modality. These observational data highlight the clinical tradeoffs in the complex decisions regarding revascularization in these high-risk patients, which should factor in short-term and long-term mortality, the ability to use an internal mammary graft, and the probability of repeat revascularization.
Gautam R. Shroff, MBBS
Division of Cardiology
Department of Medicine
Hennepin County Medical Center and University of Minnesota
Craig A. Solid, PhD
Charles A. Herzog, MD
Cardiovascular Special Studies Center
United States Renal Data System
Minneapolis Medical Research Foundation
- © 2013 American Heart Association, Inc.