To the Editor:

We read with interest the report of the cohort study by Haugen et al on the association of finger length pattern, osteoarthritis (OA), and knee injury (Haugen IK, Niu J, Aliabadi P, Felson DT, Englund M. The associations between finger length pattern, osteoarthritis, and knee injury: data from the Framingham community cohort. Arthritis Rheum 2011;63:2284–8). The authors concluded that in the general population, low index finger–to–ring finger (2D:4D) phalangeal ratios in men were associated with knee injury, and they found no significant association with knee OA. While we believe this was a well-conducted study exploring a possible unique and interesting association, and a good addition to OA research, some questions still remain.

The 2D:4D length ratio is interesting, but we do not believe its relevance in knee OA is clearly explained in this report. If a positive association between low 2D:4D length ratio and knee OA is established, are there opportunities for preventative care? Why were the variables knee injury and meniscal lesions included in this study? Neither question is clearly answered in the article, in our opinion.

Furthermore, the authors used logistic regression to confirm whether low 2D:4D length ratios are associated with radiographic knee OA. Reading the article, we assumed that the 2D:4D length ratio was considered the dependent variable and knee OA was considered the independent variable. Shouldn't knee OA have been the dependent variable in this study? This would allow determination of whether 2D:4D length ratio is a risk factor in knee OA, and would be especially relevant in regard to covariates included in the analysis.

The authors mention that the Framingham community cohort was used in this study, but it is not clear to us what kind of cohort this is. A brief description of the population and the purpose of the cohort, in addition to a reference, would be helpful in assessing the generalizability of the conclusion to the general population. Additionally, they note that the 2D:4D length ratio of the right hand has been suggested to be a better indicator of prenatal androgenization than that of the left hand. Is this in any way related to the dominant hand? Haugen and colleagues' study included 33 in whom the left hand was measured. Why were these participants not excluded? Doing so would not have had a great influence on the power.

Finally, why were groups divided into tertiles instead of into predefined categories with fixed-value boundaries that were similar for both sexes? In our opinion, the findings of this study may be too extensive to be sufficiently addressed in a brief report. We welcome the authors' thoughts on these comments and questions.

F. A. Vijfvinkel*, D. Schiphof MSc*, A. P. Verhagen PhD*, * Erasmus MC University Medical Centre Rotterdam, Rotterdam, The Netherlands.