Why We Should Care About the Sex of Specimens in Cadaveric, Animal, and Cell-Based Orthopaedic Research: Commentary on an article by Jessica Bryant, MD, et al.
Historically, women have been severely underrepresented in human clinical trials, which led to the Revitalization Act of 1993 and the requirement to include women in clinical research funded by the National Institutes of Health (NIH). Under this policy, enrollment rates of women in modern NIH-funded clinical trials are now higher than those of men5. However, little attention has been paid to sex bias among animals or cells utilized in basic science research. In 2014, the NIH leadership unveiled future plans to roll out policies that require balanced use of female and male animals and cells in basic science research6. The rationale for these proposed policies is that elimination of sex bias at the preclinical stage will lead to improved clinical study design and more reliable translation of results to clinical practice.
This proposal has had mixed support in the basic science community. Routine use of male and female cell lines and animals as well as comparison of experimental results by sex may unnecessarily increase experimental costs and variability7; if there is no physiologic basis for sex to play a role in certain experiments, why should it be investigated? However, proponents of the proposal suggest that the extra cost and effort needed to equally include male and female specimens are justified by the downstream costs of failing to do so, namely, failed clinical trials, misdiagnosis and inappropriate therapies for women, and omission of fundamental biological principles8.
In the current review, Bryant et al. investigated the potential for sex bias in orthopaedic basic science research. The authors identified all cadaveric, animal, and cell-based studies shown in 4 of our top journals in 2014 (Journal of Orthopaedic Research, Clinical Orthopaedics and Related Research, The Bone & Joint Journal, and The Journal of Bone & Joint Surgery) and determined that sex is not reported in more than one-third of the included studies. Among studies that did include sex, there was a predilection toward use of male specimens, and authors reported sex-specific results in only 13% of studies that included both sexes. The predilection for male specimens in animal studies, in particular, is based on the unfounded (and debunked)9 assumption that variability in results is intrinsically higher in females because of variability in estrogen cycles.
Clearly, we need to do a better job of consistently reporting sex in basic science research within our field. The current review highlights an important topic and reminds us to, at a minimum, actively consider the sex of specimens when interpreting study results or designing a new basic science experiment. Standardized reporting of the sex of included specimens should be a requirement, as well as an explanation for use of samples from one sex if a balanced sample (of male and female specimens) is not used. This has the potential to increase discovery of previously obscured sex-specific relationships and to improve the clinical applicability of basic science studies within our field.