HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Correspondence to:

ARTICLES: C. M. Tanner, S. M. Goldman, D. A. Aston, R. Ottman, J. Ellenberg, R. Mayeux, and J. W. Langston Smoking and Parkinson’s disease in twins Neurology 2002; 58: 581-588 [Abstract] [Full text] [PDF]

Correspondence: Submit a response to this article

Correspondence published:

Reply to Letter to the Editor

Caroline M. Tanner, Samuel A. Goldman, Diana Aston, Ruth Ottman, Jonas Ellenberg, Richard Mayeux, and J. William Langston   (17 July 2002)

Smoking and Parkinson’s disease in twins

Mohamed Farouk Allam   (17 July 2002)

Reply to Letter to the Editor 17 July 2002
Caroline M. Tanner Parkinson's Institute Sunnyvale CA, Samuel A. Goldman, Diana Aston, Ruth Ottman, Jonas Ellenberg, Richard Mayeux, and J. William Langston Send Correspondence to journal: Re: Reply to Letter to the Editor ctanner{at}parkinsonsinstitute.org Caroline M. Tanner, et al.	We are surprised that Dr Allam overlooked the care we took to address the concern that some aspect of the disease process of PD may have caused the parkinsonian twin to smoke fewer cigarettes than his nonparkinsonian twin brother. We addressed this concern directly by including a comparison of smoking behavior not only at the time of PD in the affected twin, but also at 10 years and 20 years before the onset of PD in the affected twin in our analysis. These data are shown in Table 2 of our manuscript and were addressed in detail in the discussion. We explored not only Dr. Allam’s concern that problems with motor control would affect the prevalence of smoking in parkinsonian patients, but also the possibility that more subtle deficits could have affected outcome. Our results (as clearly pointed out in our original discussion) provide little support for Dr Allam's assertion that the difference we observe is the result of a disease-associated change in smoking behavior, as the difference between brothers IN SMOKING BEHAVIOR is similar 20 years before PD onset. We would also point out that this observation is particularly remarkable in view of the very strong correlation for smoking behavior that we observed in the twins. We do not understand Dr Allam’s concern that the difference in smoking behavior between twins with PD and their nonparkinsonian brothers is only significant when both MZ and DZ twins are analyzed together. As he correctly states, statistical power was less in these strata as sample size was halved. However, the same pattern of difference is observed when twins are analyzed within strata defined by zygosity, in our minds supporting the direction observed in the combined analysis. We regard the "p value" as just one of several tools to assist in inference, and found the consistency of our results to be more compelling than the p values for this secondary analysis. We agree that attribution of cause and effect cannot be inferred with confidence from a cross-sectional retrospective design. Importantly, in prospective studies cigarette smoking has been associated with a lower risk of PD, and the effect was dose-dependent. {1, 2} We would urge Dr Allam to consider these very important studies, and the added insights they provide before drawing conclusions about the relationship between cigarette smoking and PD. References: 1) Ross G, Abbott R, Petrovitch H et al. Association of coffee and caffeine intake with the risk of Parkinson Disease. JAMA 2000;283:2674- 2679. 2) Ascherio A, Zhang S, Herman M et al. Prospective study of caffeine consumption and risk of Parkinson’s disease in men and women. Ann Neurol 2001;50:56-63.
Smoking and Parkinson’s disease in twins 17 July 2002
Mohamed Farouk Allam University of Cordoba Spain Send Correspondence to journal: Re: Smoking and Parkinson’s disease in twins fm2faahm{at}lucano.uco.es Mohamed Farouk Allam	The negative association between smoking and the risk of PD has been a controversial finding over the last forty years. Tanner et al. examined the protective effect of cigarette smoking against PD in male twin pairs identified from the National Academy of Sciences–National Research Council World War II Veteran Twins Cohort [1]. The authors after identifying smoking habit in pack-years until the time of PD onset in the affected twin or until the time of death for the unaffected twin, whichever came first, concluded that there is a true biologic protective effect of smoking. I have deep concerns regarding the study methodology and results, which do not match with the authors’ conclusions. First, neither the 33 discordant monozygotic twin pairs (p value 0.077) nor the 39 discordant dizygotic twin pairs (p value 0.17) included had differences regarding the amount of cigarettes. The significant negative association was found only after combining both discordant monozygotic and dizygotic twin pairs results; most probably because of the small sample size of this retrospective cohort study. Second, the authors compared the smoking habit of the discordant twin pairs according to their pack-years. Patients with PD may stop smoking during the long prodromal symptoms of the disease or even after its onset because of physiologic or psychologic dislike, and in turn affects the truthfulness of retrospective studies’ results. PD is a chronic neurodegenerative disorder associated with incomplete motor control and comparing these patients with healthy subjects gives an overestimation of smoking in the more physical and motor control subjects. I have recently reviewed all retrospective studies that evaluated the association between PD risk and smoking habit until December 2000 [2]. The pooled analysis of current versus never smokers demonstrated one-third risk of PD in current smokers (odds ratio 0.37 (95% confidence interval 0.30 to 0.46)). Meanwhile, former versus never smokers had pooled odds ratio of 0.87 with 95% confidence interval nearly overlapping unity (95% confidence interval 0.79 to 0.96). Thus, the reverse causation is a more probable explanation on seeing the pooled analyses of current and former smokers, and this is against a real protective effect of tobacco smoking. It is quite possible that the movement disorders of PD protect against smoking. This systematic bias could explain the inverse epidemiological association, which until now has no definite biochemical basis. References: 1. Tanner CM, Goldman SM, Aston DA, Ottman R, Ellenberg J, Mayeux R, Langston JW. Smoking and Parkinson’s disease in twins. Neurology 2002;58:581-588. 2. Allam MF. Meta-analysis of risk factors in Parkinson’s disease. European PhD Thesis. Faculty of Medicine, Cordoba University, Spain, 2002.