. Serum iron levels and the risk of Parkinson disease: a mendelian randomization study. PLoS Med. 2013 Jun;10(6):e1001462. PubMed.


Please login to recommend the paper.


  1. My overall impression of this paper is that it adds an interesting piece of evidence, but that the story of iron is not a simple one. Iron regulation is complex and, like most critical biological pathways, has multiple redundancies and contributing factors. The crucial assumption of this paper is that the three single nucleotide polymorphisms (SNPs) the authors selected are critical determinants of serum iron levels. However, the data for this are not published, so that assumption has not been peer reviewed. But, assuming it is correct, while I am not an expert by any means, I worry that it is a gross oversimplification to assume that a single gene variant (two did not have statistically significant associations) that regulates serum iron levels would necessarily be key to explaining iron metabolism in the central nervous system.

    We have looked at features potentially related to peripheral iron metabolism and PD risk in the period before the development of the disease (see Abbott et al., 2010). The authors cite two other studies (Logroscino et al., 2008, and Savica et al., 2009) that indirectly support the idea put forth in this paper that higher serum iron protects against PD. Our finding runs somewhat counter, as we found higher normal hemoglobin in elderly men was associated with a higher incidence of PD. In these studies the relationship between peripheral iron and central iron homeostasis, and how the latter might relate to PD pathogenesis, remains an open question. But because these three papers include information collected before PD diagnosis, they avoid the confounding due to disease or treatment that makes other studies so hard to interpret.

    Regarding the analytic approach of Mendelian randomization, I think it is an interesting one and adds an additional tool to our scientific armamentarium, but there are also significant limitations, namely the necessary assumptions regarding the effect of the SNPs, which in most cases will not be met, since biologic processes generally have more than one determinant. Using this approach when the assumptions can be met with confidence would be very helpful in understanding mechanisms for disease.


    . Late-life hemoglobin and the incidence of Parkinson's disease. Neurobiol Aging. 2012 May;33(5):914-20. PubMed.

    . Dietary iron intake and risk of Parkinson's disease. Am J Epidemiol. 2008 Dec 15;168(12):1381-8. PubMed.

    . Anemia or low hemoglobin levels preceding Parkinson disease: a case-control study. Neurology. 2009 Oct 27;73(17):1381-7. PubMed.

    View all comments by Caroline M. Tanner

Make a Comment

To make a comment you must login or register.

This paper appears in the following:


  1. Genotype-Based Meta-Analysis Suggests Serum Iron Protects Against PD