Implantable Cardioverter–Defibrillators
http://www.100md.com
《新英格兰医药杂志》
To the Editor: Given the results of the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT), reported by Bardy et al. (Jan. 20 issue),1 and other trials, the Centers for Medicare and Medicaid Services has proposed expanding coverage for prophylactic placement of implantable cardioverter–defibrillators (ICDs) to approximately 500,000 beneficiaries, as noted by McClellan and Tunis in their Perspective article.2 At a total cost of approximately $15 billion, this policy will benefit — according to the data presented by Bardy et al. — only 7 of every 100 patients who receive a device (absolute decrease in mortality, 7.2 percentage points over a five-year period), a benefit that translates into a cost of $428,000 per life saved, or $86,000 per year of life saved.
Since Medicare is a zero-sum game, the amount spent on ICDs means that some other services will be withheld. In SCD-HeFT, since nearly four of five of the ICDs implanted never fired appropriately and only 7 of 100 patients received benefit, one must question the cost–benefit ratio. Rather than expanding the population covered, we should consider subgroup analysis or pooling of data, as Hlatky suggests3; such an approach can define more specific groups that might benefit, so that the ratio of devices fired to those implanted can be improved, the absolute mortality benefit increased, and the cost per life saved decreased.
Michael J. Mack, M.D.
Cardiopulmonary Research Science and Technology Institute
Dallas, TX 75230
References
Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med 2005;352:225-237.
McClellan MB, Tunis SR. Medicare coverage of ICDs. N Engl J Med 2005;352:222-224.
Hlatky MA. Evidence-based use of cardiac procedures and devices. N Engl J Med 2004;350:2126-2128.
To the Editor: The median age of the patients in SCD-HeFT was approximately 60 years (interquartile range, 52 to 68). Therefore, a majority were under the age of 65, the age defining Medicare eligibility. According to McClellan and Tunis, Medicare beneficiaries account for 80 percent of persons who die suddenly from cardiac causes. Because of the increasing rate of coexisting conditions associated with aging and other factors among patients 65 years of age or older, the major portion of the clinical benefit noted for ICD implantation in SCD-HeFT may have occurred in patients younger than 65. Extrapolating the clinical significance of SCD-HeFT to the Medicare population, especially the older Medicare population, may not be valid. In concluding his accompanying editorial, Kadish1 writes that "ICDs will prolong life and should be used in most patients with severe left ventricular dysfunction" — a conclusion that is not supported by the clinical evidence presented, since most eligible patients are substantially older than those studied.
Irwin Nash, M.D.
Hospital of St. Raphael
New Haven, CT 06511
inash@srhs.org
References
Kadish A. Prophylactic defibrillator implantation -- toward an evidence-based approach. N Engl J Med 2005;352:285-287.
To the Editor: Bardy et al. find that ICDs reduce mortality from all causes among patients with New York Heart Association (NYHA) class II, but not NYHA class III, congestive heart failure (CHF). The authors had not anticipated this finding, on the basis of other trials of ICD therapy. This is the first time that a significant difference, in a prespecified analysis according to NYHA class, was found in a trial of ICD therapy. However, we believe that the finding is not surprising, since sudden death is a major cause of death among persons with mild CHF, whereas death due to worsening heart failure is more prevalent among those with advanced CHF.1 Recently, we showed that among patients with NYHA class II CHF, sudden death accounted for more than 50 percent of the deaths during 13 years of follow-up2; one may speculate that ICDs might have been beneficial in this patient population. Since an ICD will affect only sudden death, it appears logical that the relative risk reduction is more pronounced among patients with NYHA class II CHF.
Dirk J. van Veldhuisen, M.D., Ph.D.
Adriaan A. Voors, M.D., Ph.D.
University Hospital Groningen
9700 RB Groningen, the Netherlands
a.a.voors@thorax.azg.nl
References
Effect of metoprolol CR/XL in chronic heart failure: Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF). Lancet 1999;353:2001-2007.
Smilde TD, Hillege HL, Voors AA, Dunselman PH, Van Veldhuisen DJ. Prognostic importance of renal function in patients with early heart failure and mild left ventricular dysfunction. Am J Cardiol 2004;94:240-243.
To the Editor: In their report of the results of SCD-HeFT, one of Bardy and colleagues' main conclusions is that "amiodarone does not improve survival among patients with mild-to-moderate systolic CHF." However, a crucial issue is whether amiodarone–digoxin interactions could have confounded the results. The percentage of patients in SCD-HeFT who were taking digoxin at the time of enrollment (about 70 percent) was higher than that in standard European clinical practice (about 50 percent).1 Moreover, the dose of amiodarone used was higher than that used in Europe.2 The amiodarone group in SCD-HeFT also had a smaller increase in the frequency of beta-blocker therapy (increase, 3 percentage points) during the trial than did the other two groups (10 and 13 percentage points). Beta-blockers have repeatedly been shown to be beneficial in patients with heart failure.1 The amiodarone group may have been disadvantaged by this smaller increase in the use of beta-blocker therapy. Given these observations, it may be that the amiodarone group in SCD-HeFT was not entirely comparable to the other treatment groups in this trial.
Ashish Patwala, M.R.C.P.
Cardiothoracic Centre
Liverpool L14 3PE, United Kingdom
Dominik Schlosshan, M.R.C.P.
Airedale Hospital
Keighley BD20 6TD, United Kingdom
Simon G. Williams, M.D.
Wythenshawe Hospital
Manchester M23 9LT, United Kingdom
simongwilliams@hotmail.co.uk
References
The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999;353:9-13.
British National Formulary. No. 48. London: British Medical Association and Royal Pharmaceutical Society of Great Britain, September 2004.
Dr. Bardy and colleagues reply: Dr. Mack raises several important issues. ICD therapy, like coronary bypass surgery, has large up-front costs and, when applied to the correct populations, saves lives. Our analysis shows that ICDs provide good value for money relative to conventional benchmarks.1 Without question, spending on new technologies may divert resources from other high-priority services. We agree with the goal of making ICD therapy more efficient, but until this can be done reliably, it makes little sense to divert resources away from a lifesaving technology.
Dr. Nash suggests that the relatively young age of the patients in SCD-HeFT limits the applicability of the findings to the Medicare population. Approximately 35 percent of the patients in SCD-HeFT were more than 65 years of age, and 9 percent were older than 75. Notably, there was no evidence that the effectiveness of ICD therapy was reduced among older patients. We certainly agree that application of the SCD-HeFT results to an older population will require careful consideration of overall prognosis and competing causes of death related to coexisting conditions. If, however, the primary risk is sudden death and the patient is otherwise healthy, we see no reason to withhold ICD therapy because of age.
Drs. van Veldhuisen and Voors suggest that the finding that ICD therapy offers greater benefit to patients with NYHA class II disease than to those with class III disease is logical. However, we would caution that many medical concepts previously believed to be logical have been disproved. Furthermore, these results were not consistent with findings from the Multicenter Automatic Defibrillator Implantation Trial II (MADIT-II) and the Defibrillators in Non-Ischemic Cardiomyopathy Treatment Evaluation (DEFINITE), in which patients with class III disease benefited at least as much, on a relative scale, as did those with class II disease. This issue remains unresolved.
Dr. Patwala and colleagues wonder whether amiodarone–digoxin interactions diminished survival in the amiodarone group. Although we did not record digoxin levels, the data and safety monitoring board monitored the trial carefully, and there were no adverse-event data to suggest such a problem in the amiodarone group. In addition, the mechanisms of death in the amiodarone group did not include a rise in the incidence of sudden death due to arrhythmia, which one might expect as a consequence of digoxin toxicity.
The final issue raised by these correspondents is that the frequency of beta-blocker use increased to a lesser extent in the amiodarone group than in the other groups during the trial. Although the increase over time was smaller, the rate of beta-blocker use was high in all trial groups. More important, those treated with amiodarone and beta-blockers had a slightly (not significantly) worse outcome than those in the amiodarone group who were not treated with beta-blockers, as shown in Figure 4 of our article. Thus, we maintain that amiodarone confers no survival benefit on patients when used for primary prevention.
Gust H. Bardy, M.D.
Seattle Institute for Cardiac Research
Seattle, WA 98103
gbardy@sicr.org
Kerry L. Lee, Ph.D.
Daniel B. Mark, M.D., M.P.H.
Duke University
Durham, NC 27708
References
Mark DB, Nelson CL, Anstrom KJ, et al. Cost-effectiveness of ICD therapy in the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT). Presented at the American Heart Association 2004 Scientific Sessions, New Orleans, November 10–17, 2004. abstract.
Dr. Kadish replies: Dr. Nash notes that the median age of the patients in SCD-HeFT was about 60 years and thus suggests that extrapolating data from SCD-HeFT to the Medicare population may not be appropriate. It should be noted that other trials of ICD therapy in patients with coronary disease, including the Multicenter Unsustained Tachycardia Trial1 and MADIT-II,2 involved patients with an average age that was closer to 65 and showed that ICD therapy was associated with a benefit similar to or greater than that seen in SCD-HeFT. In addition, there was no significant difference in the ICD-associated benefit between those 65 or older and those younger than 65 in SCD-HeFT and DEFINITE.3 As the average life expectancy approaches 80 years and as the quality of life among elderly patients increases, it would not be appropriate to withhold lifesaving therapy from patients older than 65 because of a preconceived notion that it will not offer them a benefit.
Dr. Mack performs an estimated cost-effectiveness analysis and concludes that ICD therapy will cost $86,000 per year of life saved and that the amount spent on defibrillators means that other services will be withheld. A preliminary analysis of data from SCD-HeFT shows a much lower cost per life-year saved.4 Overall, Medicare spending is determined annually by Congress, and information on actual health care costs is available for budgeting. No hard ceiling on spending has been established. Finally, subgroup analysis or pooling of data is attractive to generate hypotheses but does not serve as an evidence-based approach to determine therapy. It is true that certain subgroups who benefit less than others from ICD implantation may be identified, and it is hoped that the prospective registry mandated by the Centers for Medicare and Medicaid Services will help provide additional information about which patients will benefit most from ICD therapy. However, before such data are available, the best guide for clinicians with regard to ICD implantation remains the inclusion criteria in clinical trials.
Alan H. Kadish, M.D.
Northwestern Memorial Hospital
Chicago, IL 60611
a-kadish@northwestern.edu
References
Buxton AE, Lee KL, DiCarlo L, et al. Electrophysiologic testing to identify patients with coronary artery disease who are at risk for sudden death. N Engl J Med 2000;342:1937-1945.
Moss AJ, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877-883.
Kadish A, Dyer A, Daubert JP, et al. Prophylactic defibrillator implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med 2004;350:2151-2158.
Kadish A. Prophylactic ICD implantation in nonischemic dilated cardiomyopathy: evidence-based medicine. Presented at Heart Rhythm 2004: 25th annual sessions, San Francisco, May 19–22, 2004.
Drs. McClellan and Tunis reply: Dr. Nash and Dr. Mack both raise the important issue of risk stratification within the broad population of patients for whom a survival benefit has been shown to be associated with prophylactic ICD implantation. Dr. Nash notes that the median age of the patients in the SCD-HeFT trial was 60 years and suggests that the benefit of ICDs might be greater in these younger patients than in older, sicker patients. Although coexisting conditions are an important factor in determining the risk–benefit ratio for ICD implantation, analyses of data from SCD-HeFT and other recent trials of ICD therapy have not shown that age itself is a predictor of ICD-associated benefit. Dr. Mack also comments on the importance of greater attention to risks and benefits in specific patient subgroups and notes that pooling of data from multiple trials might help to determine which patients obtain the greatest benefit. The Centers for Medicare and Medicaid Services has also noted the potential value of pooling data, and efforts in this area are under way.1
We fully agree about the importance of being able to better stratify patients eligible for ICD implantation according to risk, and we believe that systematic collection of prospective, observational data, such as that being planned in the context of the longitudinal ICD registry, should provide additional insights into this important issue. This approach to making technologies available while continuing to gather systematic data on their risks and benefits will generate useful information for physicians, patients, and other decision makers while also ensuring that individual patients are receiving appropriate care.
Mark B. McClellan, M.D., Ph.D.
Sean R. Tunis, M.D.
Centers for Medicare and Medicaid Services
Baltimore, MD 21210
References
Centers for Medicare and Medicaid Services. Decision memo for actinic keratoses. (Accessed April 21, 2005, at http://www.cms.hhs.gov/mcd/viewdecisionmemo.asp?id=148.)
Since Medicare is a zero-sum game, the amount spent on ICDs means that some other services will be withheld. In SCD-HeFT, since nearly four of five of the ICDs implanted never fired appropriately and only 7 of 100 patients received benefit, one must question the cost–benefit ratio. Rather than expanding the population covered, we should consider subgroup analysis or pooling of data, as Hlatky suggests3; such an approach can define more specific groups that might benefit, so that the ratio of devices fired to those implanted can be improved, the absolute mortality benefit increased, and the cost per life saved decreased.
Michael J. Mack, M.D.
Cardiopulmonary Research Science and Technology Institute
Dallas, TX 75230
References
Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med 2005;352:225-237.
McClellan MB, Tunis SR. Medicare coverage of ICDs. N Engl J Med 2005;352:222-224.
Hlatky MA. Evidence-based use of cardiac procedures and devices. N Engl J Med 2004;350:2126-2128.
To the Editor: The median age of the patients in SCD-HeFT was approximately 60 years (interquartile range, 52 to 68). Therefore, a majority were under the age of 65, the age defining Medicare eligibility. According to McClellan and Tunis, Medicare beneficiaries account for 80 percent of persons who die suddenly from cardiac causes. Because of the increasing rate of coexisting conditions associated with aging and other factors among patients 65 years of age or older, the major portion of the clinical benefit noted for ICD implantation in SCD-HeFT may have occurred in patients younger than 65. Extrapolating the clinical significance of SCD-HeFT to the Medicare population, especially the older Medicare population, may not be valid. In concluding his accompanying editorial, Kadish1 writes that "ICDs will prolong life and should be used in most patients with severe left ventricular dysfunction" — a conclusion that is not supported by the clinical evidence presented, since most eligible patients are substantially older than those studied.
Irwin Nash, M.D.
Hospital of St. Raphael
New Haven, CT 06511
inash@srhs.org
References
Kadish A. Prophylactic defibrillator implantation -- toward an evidence-based approach. N Engl J Med 2005;352:285-287.
To the Editor: Bardy et al. find that ICDs reduce mortality from all causes among patients with New York Heart Association (NYHA) class II, but not NYHA class III, congestive heart failure (CHF). The authors had not anticipated this finding, on the basis of other trials of ICD therapy. This is the first time that a significant difference, in a prespecified analysis according to NYHA class, was found in a trial of ICD therapy. However, we believe that the finding is not surprising, since sudden death is a major cause of death among persons with mild CHF, whereas death due to worsening heart failure is more prevalent among those with advanced CHF.1 Recently, we showed that among patients with NYHA class II CHF, sudden death accounted for more than 50 percent of the deaths during 13 years of follow-up2; one may speculate that ICDs might have been beneficial in this patient population. Since an ICD will affect only sudden death, it appears logical that the relative risk reduction is more pronounced among patients with NYHA class II CHF.
Dirk J. van Veldhuisen, M.D., Ph.D.
Adriaan A. Voors, M.D., Ph.D.
University Hospital Groningen
9700 RB Groningen, the Netherlands
a.a.voors@thorax.azg.nl
References
Effect of metoprolol CR/XL in chronic heart failure: Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF). Lancet 1999;353:2001-2007.
Smilde TD, Hillege HL, Voors AA, Dunselman PH, Van Veldhuisen DJ. Prognostic importance of renal function in patients with early heart failure and mild left ventricular dysfunction. Am J Cardiol 2004;94:240-243.
To the Editor: In their report of the results of SCD-HeFT, one of Bardy and colleagues' main conclusions is that "amiodarone does not improve survival among patients with mild-to-moderate systolic CHF." However, a crucial issue is whether amiodarone–digoxin interactions could have confounded the results. The percentage of patients in SCD-HeFT who were taking digoxin at the time of enrollment (about 70 percent) was higher than that in standard European clinical practice (about 50 percent).1 Moreover, the dose of amiodarone used was higher than that used in Europe.2 The amiodarone group in SCD-HeFT also had a smaller increase in the frequency of beta-blocker therapy (increase, 3 percentage points) during the trial than did the other two groups (10 and 13 percentage points). Beta-blockers have repeatedly been shown to be beneficial in patients with heart failure.1 The amiodarone group may have been disadvantaged by this smaller increase in the use of beta-blocker therapy. Given these observations, it may be that the amiodarone group in SCD-HeFT was not entirely comparable to the other treatment groups in this trial.
Ashish Patwala, M.R.C.P.
Cardiothoracic Centre
Liverpool L14 3PE, United Kingdom
Dominik Schlosshan, M.R.C.P.
Airedale Hospital
Keighley BD20 6TD, United Kingdom
Simon G. Williams, M.D.
Wythenshawe Hospital
Manchester M23 9LT, United Kingdom
simongwilliams@hotmail.co.uk
References
The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999;353:9-13.
British National Formulary. No. 48. London: British Medical Association and Royal Pharmaceutical Society of Great Britain, September 2004.
Dr. Bardy and colleagues reply: Dr. Mack raises several important issues. ICD therapy, like coronary bypass surgery, has large up-front costs and, when applied to the correct populations, saves lives. Our analysis shows that ICDs provide good value for money relative to conventional benchmarks.1 Without question, spending on new technologies may divert resources from other high-priority services. We agree with the goal of making ICD therapy more efficient, but until this can be done reliably, it makes little sense to divert resources away from a lifesaving technology.
Dr. Nash suggests that the relatively young age of the patients in SCD-HeFT limits the applicability of the findings to the Medicare population. Approximately 35 percent of the patients in SCD-HeFT were more than 65 years of age, and 9 percent were older than 75. Notably, there was no evidence that the effectiveness of ICD therapy was reduced among older patients. We certainly agree that application of the SCD-HeFT results to an older population will require careful consideration of overall prognosis and competing causes of death related to coexisting conditions. If, however, the primary risk is sudden death and the patient is otherwise healthy, we see no reason to withhold ICD therapy because of age.
Drs. van Veldhuisen and Voors suggest that the finding that ICD therapy offers greater benefit to patients with NYHA class II disease than to those with class III disease is logical. However, we would caution that many medical concepts previously believed to be logical have been disproved. Furthermore, these results were not consistent with findings from the Multicenter Automatic Defibrillator Implantation Trial II (MADIT-II) and the Defibrillators in Non-Ischemic Cardiomyopathy Treatment Evaluation (DEFINITE), in which patients with class III disease benefited at least as much, on a relative scale, as did those with class II disease. This issue remains unresolved.
Dr. Patwala and colleagues wonder whether amiodarone–digoxin interactions diminished survival in the amiodarone group. Although we did not record digoxin levels, the data and safety monitoring board monitored the trial carefully, and there were no adverse-event data to suggest such a problem in the amiodarone group. In addition, the mechanisms of death in the amiodarone group did not include a rise in the incidence of sudden death due to arrhythmia, which one might expect as a consequence of digoxin toxicity.
The final issue raised by these correspondents is that the frequency of beta-blocker use increased to a lesser extent in the amiodarone group than in the other groups during the trial. Although the increase over time was smaller, the rate of beta-blocker use was high in all trial groups. More important, those treated with amiodarone and beta-blockers had a slightly (not significantly) worse outcome than those in the amiodarone group who were not treated with beta-blockers, as shown in Figure 4 of our article. Thus, we maintain that amiodarone confers no survival benefit on patients when used for primary prevention.
Gust H. Bardy, M.D.
Seattle Institute for Cardiac Research
Seattle, WA 98103
gbardy@sicr.org
Kerry L. Lee, Ph.D.
Daniel B. Mark, M.D., M.P.H.
Duke University
Durham, NC 27708
References
Mark DB, Nelson CL, Anstrom KJ, et al. Cost-effectiveness of ICD therapy in the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT). Presented at the American Heart Association 2004 Scientific Sessions, New Orleans, November 10–17, 2004. abstract.
Dr. Kadish replies: Dr. Nash notes that the median age of the patients in SCD-HeFT was about 60 years and thus suggests that extrapolating data from SCD-HeFT to the Medicare population may not be appropriate. It should be noted that other trials of ICD therapy in patients with coronary disease, including the Multicenter Unsustained Tachycardia Trial1 and MADIT-II,2 involved patients with an average age that was closer to 65 and showed that ICD therapy was associated with a benefit similar to or greater than that seen in SCD-HeFT. In addition, there was no significant difference in the ICD-associated benefit between those 65 or older and those younger than 65 in SCD-HeFT and DEFINITE.3 As the average life expectancy approaches 80 years and as the quality of life among elderly patients increases, it would not be appropriate to withhold lifesaving therapy from patients older than 65 because of a preconceived notion that it will not offer them a benefit.
Dr. Mack performs an estimated cost-effectiveness analysis and concludes that ICD therapy will cost $86,000 per year of life saved and that the amount spent on defibrillators means that other services will be withheld. A preliminary analysis of data from SCD-HeFT shows a much lower cost per life-year saved.4 Overall, Medicare spending is determined annually by Congress, and information on actual health care costs is available for budgeting. No hard ceiling on spending has been established. Finally, subgroup analysis or pooling of data is attractive to generate hypotheses but does not serve as an evidence-based approach to determine therapy. It is true that certain subgroups who benefit less than others from ICD implantation may be identified, and it is hoped that the prospective registry mandated by the Centers for Medicare and Medicaid Services will help provide additional information about which patients will benefit most from ICD therapy. However, before such data are available, the best guide for clinicians with regard to ICD implantation remains the inclusion criteria in clinical trials.
Alan H. Kadish, M.D.
Northwestern Memorial Hospital
Chicago, IL 60611
a-kadish@northwestern.edu
References
Buxton AE, Lee KL, DiCarlo L, et al. Electrophysiologic testing to identify patients with coronary artery disease who are at risk for sudden death. N Engl J Med 2000;342:1937-1945.
Moss AJ, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877-883.
Kadish A, Dyer A, Daubert JP, et al. Prophylactic defibrillator implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med 2004;350:2151-2158.
Kadish A. Prophylactic ICD implantation in nonischemic dilated cardiomyopathy: evidence-based medicine. Presented at Heart Rhythm 2004: 25th annual sessions, San Francisco, May 19–22, 2004.
Drs. McClellan and Tunis reply: Dr. Nash and Dr. Mack both raise the important issue of risk stratification within the broad population of patients for whom a survival benefit has been shown to be associated with prophylactic ICD implantation. Dr. Nash notes that the median age of the patients in the SCD-HeFT trial was 60 years and suggests that the benefit of ICDs might be greater in these younger patients than in older, sicker patients. Although coexisting conditions are an important factor in determining the risk–benefit ratio for ICD implantation, analyses of data from SCD-HeFT and other recent trials of ICD therapy have not shown that age itself is a predictor of ICD-associated benefit. Dr. Mack also comments on the importance of greater attention to risks and benefits in specific patient subgroups and notes that pooling of data from multiple trials might help to determine which patients obtain the greatest benefit. The Centers for Medicare and Medicaid Services has also noted the potential value of pooling data, and efforts in this area are under way.1
We fully agree about the importance of being able to better stratify patients eligible for ICD implantation according to risk, and we believe that systematic collection of prospective, observational data, such as that being planned in the context of the longitudinal ICD registry, should provide additional insights into this important issue. This approach to making technologies available while continuing to gather systematic data on their risks and benefits will generate useful information for physicians, patients, and other decision makers while also ensuring that individual patients are receiving appropriate care.
Mark B. McClellan, M.D., Ph.D.
Sean R. Tunis, M.D.
Centers for Medicare and Medicaid Services
Baltimore, MD 21210
References
Centers for Medicare and Medicaid Services. Decision memo for actinic keratoses. (Accessed April 21, 2005, at http://www.cms.hhs.gov/mcd/viewdecisionmemo.asp?id=148.)