LAUSR

TO THE EDITOR: We read with interest, concern, and ultimately outrage the argument made by Rind in Smetana and colleagues' grand rounds discussion ( 1) that not only is screening for diabetes a waste of time and resources but that diabetes (his use of quotation marks) is not itself a disease state. His contention that diabetes is not a disease worth screening for and treating is an astonishing position to take in the 21st century. As recently as the 1980s, some clinicians believed that high glucose levels only needed to be treated sufficiently to maintain day-to-day wellness and that the vascular complications associated with diabetes had little or nothing to do with glucose levels per se. But the publication of the landmark DCCT (Diabetes Control and Complications Trial) ( 2) and UKPDS (United Kingdom Prospective Diabetes Study) ( 3) in the 1990s provided irrefutable evidence that intensifying blood glucose control reduced the rate of microvascular complications. Further, long-term follow-up in those studies clearly established the relation between diabetes control and reduction in macrovascular disease and death ( 4). Rind acknowledges some value in screening for diabetes if it is symptomatic, although he does not explain what that means. A heart attack, overt retinopathy, or nephropathy? Or does he mean hyperglycemia severe enough to lead to a hyperosmolar hyperglycemic state or diabetic ketoacidosis? The medical profession has not waited for diabetes to become symptomatic to initiate treatment for at least 30 yearsor, really, since the discovery of insulin almost a century ago. We cannot wait for symptoms to begin screening for and diagnosing diabetes. Fully 36% of participants enrolled in the UKPDS already had microvascular disease upon diagnosis of diabetes ( 3). The Euro Heart Survey found that 22% of patients presenting with acute myocardial infarction had undiagnosed diabetes at the time of their cardiac event ( 5). Vascular disease of all types is strongly associated with diabetes, and early detection and management of diabetes have been proved to reduce long-term risk, with earlier intervention leading to better outcomes. Vascular complications do not appear overnight in persons with diabetes. The DCCT and UKPDS have shown that, in intervention versus control groups, it can take 9 to 17 years for differences in rates of microvascular disease to appear and 27 to 30 years for those of macrovascular disease and death to do so. Yet, when those differences do appear, they are striking and substantial. The overwhelming consensus in the medical professionand from the Centers for Disease Control and Prevention, National Institutes of Health, and Centers for Medicare & Medicaid Servicesis that diabetes detection is highly worthwhile. However, Rind is prepared to bide his patients' time by waiting for symptoms to appear before screening and treating them for this condition. For these patients, given the time that it takes for vascular complications to arise, Rind may be retired before his patients experience the full effect of his delay in initiating effective management.