Special Feature Cardiovascular Death in Dialysis Patients: Lessons We Can Learn from AURORA Allan D. Sniderman,* Amirreza Solhpour,* Ahsan Alam, † Ken Williams, ‡ and James A. Sloand § *Mike Rosenbloom Laboratory for Cardiovascular Research and † Division of Nephrology, McGill University Health Centre, Montreal, Quebec, Canada; ‡ KenAnCo Biostatistics, San Antonio, Texas; § Baxter Healthcare Corporation, Renal Division, McGaw Park, Illinois; and  University of Rochester School of Medicine and Dentistry, Renal Division, Rochester, New York Cardiovascular events are the dominant cause of death in patients with ESRD. Until recently, plaque rupture due to atherogenic dyslipoproteinemias was presumed to be a major mechanism of cardiovascular events in dialysis patients. But how reasonable was that hypothesis and was it entirely discredited by the results of 4D and AURORA? This article places the conventional lipids— cholesterol and triglyceride—within the more physiologic framework of the apoB lipoproteins. Viewed from the perspective of atherogenic particle number, the failure of statins to lower cardiovascular mortality in hemodialysis patients versus the continuing potential for success in peritoneal dialysis patients becomes comprehensible. In the former, apoB is characteristically not elevated and therefore apoB-lowering therapy can have only limited effect; in the latter, apoB is characteristically high and therefore apoB-lowering therapy might have considerable clinical benefit. Nevertheless, plaque rupture is only one mechanism leading to cardiac death. In addition to those previously noted, a new mechanism is suggested for consideration—recurrent reperfusion injury. The coronaries of dialysis patients are often narrowed, the microcirculation underdeveloped, and the left ventricle hypertrophied—all of these plus transient hypotension could produce severe ischemia followed by reperfusion necrosis. The minor but common elevations of troponin that are so well known yet widely disregarded may be markers of an adverse sequence of events that could each trigger a fatal arrhythmia and tend to reduce left ventricular function. Thus sudden death due to arrhythmia and slow progressive death due to heart failure could be manifestations of reperfusion injury. Clin J Am Soc Nephrol 5: 335–340, 2010. doi: 10.2215/CJN.06300909 I t all seemed so simple, so straightforward, so certain. Cardiovascular disease is the major cause of death in dialysis patients (1,2), dyslipidemia is common in dialysis patients, and statins reduce cardiovascular event rates because they treat atherogenic dyslipidemia; therefore, statin therapy will reduce the abysmally high death rate in dialysis patients. The 4D study and AURORA have brutally contradicted this line of logic (3,4). In neither of these well conducted studies did statin therapy reduce cardiac deaths or major cardiac events. Yet does this mean that statins have no place in the therapy of patients with renal disease? And what does it say about the mechanisms of cardiac death in dialysis patients? Our purpose is to demonstrate why the outcomes in 4D and AURORA should not have been surprising; to outline where we think lipid-lowering therapy now stands in patients with renal disease; and to identify mechanisms other than plaque rupture, mechanisms not primarily driven by plasma lipids, that could contribute to cardiac death, particularly in dialysis patients. We will start by examining the issue of the relations of VLDL and LDL to the risk of vascular disease. Until recently, disorders of lipoprotein metabolism were characterized only in terms of the major plasma lipids—tri- glycerides and cholesterol. Most triglyceride is contained in VLDL particles, whereas most cholesterol is present in LDL particles (5). Elevated LDL cholesterol (LDL C) is unquestion- ably a potent risk factor for vascular disease, and statins are unquestionably a potent therapy to reduce LDL C and cardio- vascular risk. On the other hand, hypertriglyceridemia is much more common than hypercholesterolemia in patients with vas- cular disease (6,7). Hypertriglyceridemia is also a more fre- quent finding than hypercholesterolemia in patients with ESRD, particularly in those treated with hemodialysis (HD) (8). Nevertheless, most studies fail to demonstrate that triglycerides are an independent risk factor for vascular disease and, to date, clinical trials have not demonstrated any clear benefit related to lowering of plasma triglycerides. Thus one abnormality— hy- pertriglyceridemia—appears common, but not necessarily im- portant, whereas the other— hypercholesterolemia—is less common, but indisputably important. A very different understanding emerges if one examines plasma lipids as lipoprotein particles. This can be done in different ways. The approach we have developed, which can be Published online ahead of print. Publication date available at www.cjasn.org. Correspondence: Dr. Allan Sniderman, Mike Rosenbloom Laboratory for Cardiovas- cular Research, McGill University Health Centre, Room H7.22, Royal Victoria Hos- pital, 687 Pine Avenue West, Montreal, Quebec H3A 1A1, Canada. Phone: 514-934- 1934 extension 34637; Fax: 514-843-2843; E-mail: allansniderman@hotmail.com Copyright © 2010 by the American Society of Nephrology ISSN: 1555-9041/502–0335