Mary N.Haan, University of Michigan, 109 S. Observatory St, Ann Arbor, MI 48109mnhaan@umich.edu
C. Cramer, S. Galea, K.M. Langa, J. D. Kalbfleisch
Submitted December 14, 2008
We thank Dr. Vos for his comments. We accounted for indication bias by including only cases that occurred after baseline, only statin use that occurred at a visit prior to the diagnosis, and by adjusting for access to health care.
We went beyond a simple cholesterol test and obtained lipids six times during study follow-up. LDL-C was lower in statin users at study baseline and declined over time more rapidly than among non-users. In additional analyses, adjustment for LDL-C as a time dependent co-variate does not influence the association between dementia/CIND and statin use over time. [6]
Dr. Vos compared our study to that of elderly hospitalized patients with heart failure. [7] Our study was population based and included healthy and young people. Horwich et al. showed that sick, elderly people with low cholesterol are more likely to die. [5] The lower LDL reported by Horwich et al. was probably due to underlying pathology, declining weight and/or frailty. The association of lower LDL with higher mortality is expected in this sample and it does not follow that community dwelling elderly would be at similar risk.
Elias et al. [3] averaged change over time in total cholesterol rather than modeling decline and evaluated means or odds ratios for cognition rather than incidence of dementia. [4] Neither study addressed the role of statins. In a statin-free sample, Curb found a quadratic (u-shaped) association between LDL and CHD, attributed the effects of low LDL-C to aging-related metabolic/physiologic changes, and concluded statin treatment as appropriate in healthy elderly. [8] Decline in LDL-C due to underlying disease is not the same as lowering LDL-C by statins or another intervention in relatively healthy, non-frail individuals.
It is inappropriate to compare our study to randomized clinical trials of statin treatment in demented patients with established disease. These trials may have failed because statins cannot slow decline or reverse pathology. The mechanistic roles of statins in neurodegeneration are unclear; our results may be due to the effect statins have on subclinical cerebrovascular disease. For example, the SPARCL trial recently reported that aggressive treatment with atorvastatin reduced stroke risk by 16%. [9]
Cerebrovascular disease is a major contributor to cognitive impairment and dementia so we adjusted for stroke. Cognitive decline and dementia in statin users appear to be less than in non-users. However, this is not a recommendation for treatment for the purpose of preventing dementia. Only a randomized trial in people with normal cognition can determine whether statin treatment will prevent cognitive decline and dementia.
References
6. Haan MN. Use of statins, LDL-C and incidence of cognitive impairment or dementia in a seven year cohort study of older Mexican Americans. Presentation at the International Congress on Alzheimer's disease, August 2008.
7. Vos E. Available at: (http://www.health-heart.org/)
8. Curb JD, Abbott RD, Rodriguez BL, et al. Prospective association between low and high total and low-density lipoprotein cholesterol and coronary heart disease in elderly men. J Am Geriatr Soc 2004;52:1975-1980.
9. Fitchett DH, Goodman SG, Langer A. Ischemic stroke: a cardiovascular risk equivalent? Lessons learned from the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial. Can J Cardiol 2008;24:705-708.
Disclosure: Caryn Cramer was employed by Pfizer Corporation during completion of her doctoral degree during which time this study was conducted. Pfizer did not provide any material support for this study, and did not participate in the design, conduct, management, analysis, interpretation, review, or approval of the study or the manuscript. The other authors have reported no conflicts of interest.
We thank Dr. Vos for his comments. We accounted for indication bias by including only cases that occurred after baseline, only statin use that occurred at a visit prior to the diagnosis, and by adjusting for access to health care.
We went beyond a simple cholesterol test and obtained lipids six times during study follow-up. LDL-C was lower in statin users at study baseline and declined over time more rapidly than among non-users. In additional analyses, adjustment for LDL-C as a time dependent co-variate does not influence the association between dementia/CIND and statin use over time. [6]
Dr. Vos compared our study to that of elderly hospitalized patients with heart failure. [7] Our study was population based and included healthy and young people. Horwich et al. showed that sick, elderly people with low cholesterol are more likely to die. [5] The lower LDL reported by Horwich et al. was probably due to underlying pathology, declining weight and/or frailty. The association of lower LDL with higher mortality is expected in this sample and it does not follow that community dwelling elderly would be at similar risk.
Elias et al. [3] averaged change over time in total cholesterol rather than modeling decline and evaluated means or odds ratios for cognition rather than incidence of dementia. [4] Neither study addressed the role of statins. In a statin-free sample, Curb found a quadratic (u-shaped) association between LDL and CHD, attributed the effects of low LDL-C to aging-related metabolic/physiologic changes, and concluded statin treatment as appropriate in healthy elderly. [8] Decline in LDL-C due to underlying disease is not the same as lowering LDL-C by statins or another intervention in relatively healthy, non-frail individuals.
It is inappropriate to compare our study to randomized clinical trials of statin treatment in demented patients with established disease. These trials may have failed because statins cannot slow decline or reverse pathology. The mechanistic roles of statins in neurodegeneration are unclear; our results may be due to the effect statins have on subclinical cerebrovascular disease. For example, the SPARCL trial recently reported that aggressive treatment with atorvastatin reduced stroke risk by 16%. [9]
Cerebrovascular disease is a major contributor to cognitive impairment and dementia so we adjusted for stroke. Cognitive decline and dementia in statin users appear to be less than in non-users. However, this is not a recommendation for treatment for the purpose of preventing dementia. Only a randomized trial in people with normal cognition can determine whether statin treatment will prevent cognitive decline and dementia.
References
6. Haan MN. Use of statins, LDL-C and incidence of cognitive impairment or dementia in a seven year cohort study of older Mexican Americans. Presentation at the International Congress on Alzheimer's disease, August 2008.
7. Vos E. Available at: (http://www.health-heart.org/)
8. Curb JD, Abbott RD, Rodriguez BL, et al. Prospective association between low and high total and low-density lipoprotein cholesterol and coronary heart disease in elderly men. J Am Geriatr Soc 2004;52:1975-1980.
9. Fitchett DH, Goodman SG, Langer A. Ischemic stroke: a cardiovascular risk equivalent? Lessons learned from the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial. Can J Cardiol 2008;24:705-708.
Disclosure: Caryn Cramer was employed by Pfizer Corporation during completion of her doctoral degree during which time this study was conducted. Pfizer did not provide any material support for this study, and did not participate in the design, conduct, management, analysis, interpretation, review, or approval of the study or the manuscript. The other authors have reported no conflicts of interest.