De-Gendering the Knee: Overemphasizing Sex Differences as a Problem

The Challenge

In 2007, an estimated 500,000 total knee arthroplasty (TKA) procedures were performed worldwide—about two-thirds in women (Kurtz et al., 2011; Blunt et al., 2008). In the 1990s, with an increased attention to women’s health research, manufacturers began producing “gender-specific” knees, and marketed them directly to women. Does this lead to better healthcare quality?

Overemphasizing Sex Differences as a Problem

Overall, there is a lack of evidence that female-specific prostheses improve women’s TKA outcomes (Jacobs et al., 2007). Overemphasizing sex differences is a problem, especially when companies market female-specific knees directly to women without evidence of clinical advantages. Such overemphasis could result in over-reliance on sex as a variable in choosing a knee implant for a given patient when in reality height is a better predictor of morphology than is sex. Further, because knee morphology differs within a sex, the “female” knee may be a poor fit for some women and a good fit for some men (Blaha et al., 2009).

Gendered Innovations:

1. Examining Sex within the Context of other Variables: While sex-specific prosthesis design remains controversial, analyzing how sex intersects with other key variables (such as height, ethnicity, and body composition) represents a gendered innovation (Bellemans et al., 2010). Such research raises awareness of differences and questions the “neutrality” of a white male standard model of medicine. Analyzing sex in relation to other key variables helps ensure research quality and patient safety.

The Challenge

Total knee arthroplasty (TKA) is a common procedure—with women a majority of patients (Kurtz et al., 2011). Prior to 2006, prostheses used in knee replacement were unisex, modeled off the knees of both women and men, although women were typically underrepresented in these studies (42-62%) compared to their representation among TKA recipients overall (66%) (Mahfouz et al., 2007; Mahfouz et al., 2006; Hitt et al., 2003; Chin et al., 2002). Unisex TKA implants were not “one size fits all”; surgeons had access to implants of varying sizes and femoral shapes, and fit TKA candidates primarily based on measurements of the anterior-posterior dimension of the knee (Fricka et al., 2009).

Zimmer, Inc., performed a meta-analysis of sex differences in knee anatomy and released a report concluding that women have, on average, greater Q-angles (Figure 1), a less prominent anterior condyle, and a reduced medial-lateral to anterior-posterior aspect ratio (Figures 2-3) (Conley et al., 2007).

Studying Sex Differences in the Knee

A number of differences in female and male knee anatomy have been identified, but does designing for these differences provide the best clinical outcomes for women and men?

Gendered Innovation 1: Examining Sex within the Context of other Variables

Studying sex differences in the knee marks a gendered innovation because it raises questions about diversity and challenges medical models which are often male by default. The relevance of sex differences in knee anatomy, however, must be explored in the context of other variables, especially height and ethnicity (see Method).

Method: Intersectional Approaches

It is important to analyze sex differences before ruling them out. Many additional factors also influence outcomes in TKA, including age, body composition, comorbidities, diagnosis, preoperative knee mobility, ethnicity, sex, and surgeon or hospital volume. Problems can emerge if sex is inaccurately identified as the key difference, without taking intersections between sex and other variables into consideration.

View General Method

 

Height
Sex may appear to be the most important variable in implant choice until height is considered. Specifically, research shows that that two of the anatomical sex differences identified above (greater Q-angle and lesser anterior condylar height in women) disappear when corrected for standing height (Grelsamer et al., 2005). The third anatomical difference (reduced medial-lateral: anterior-posterior aspect ratio) has not been shown to be clinically significant (Merchant et al., 2008). An anatomical study of 60 cadavers revealed that the male linear knee joint dimensions are on average greater than female dimensions, but that sex differences disappear when female and male cadavers are matched by femur length, which predicts knee anatomy more accurately than sex (Dargel et al., 2010). This suggests that height is more important than sex in determining the size of knee implant a patient should receive (see Term: Overemphasizing Sex Differences as a Problem).

Race and Ethnicity
Most U.S. FDA-approved knee implants were modeled on research done in Caucasians (van den Heever et al., 2011). This raises several questions: Do other populations have specific biological traits or cultural behaviors that need to be taken into consideration when designing knee implants (MacDonald et al., 2008)? Would creating race- or ethnicity-specific knees improve patient outcomes? What effect would the proliferation of knee implant models have on the cost of care, and would race/ethnicity-specific knees create the risk that prosthetics companies, clinicians, or patients would overemphasize race/ethnicity as has occurred with regard to sex?

It is important to keep in mind that differentiating the effects of race and ethnicity from the effects of socioeconomic status, cultural and physical environment, geographic location, access to healthcare, etc. may be difficult or impossible. Further, race and ethnicity are often used inconsistently.

Most reported ethnic differences in knee morphology relate to Asians versus Caucasians. Cadaveric studies have suggested that Chinese, Koreans, and Singaporeans have smaller knees on average than Caucasians, and an intra-operative study in Taiwan also found that Asian knees tend to be smaller than Caucasian knees. Researchers have reported differences in knee shape (rather than overall size) which remain when patient height is controlled for (Chin et al., 2011; Ho et al., 2006).

Cultural considerations may also lead to important innovations in prosthesis design: for example, high flexion may be important to TKA patients in Asian countries, where getting up from the floor or sitting cross-legged may be more common than in Western countries (Bin et al., 2007). However, knee designs which allow higher flexion have not been shown to produce superior TKA outcomes in Asian populations (Kim et al., 2010)

Conclusions

Design must be done with diverse populations in mind. Analyzing sex in knee anatomy, prosthesis design, and outcomes research is important—but sex is not the only variable that should be considered. Critical evaluation of different populations and their needs is important –if research demonstrates that better prosthetic fit results in better clinical outcomes, then it might be appropriate to bring to market a spectrum of knees based on sex, ethnicity, height, and other significant “fit” factors. Because of the complex relationships between knee replacement and lifestyle, environmental, and health factors, such as race and ethnicity, nationality, socioeconomic status, diet, exercise, etc., selecting the best prosthesis for a patient requires careful attention.

Next Steps

More basic research needs to be done in this area. Key questions include:

1. Do female-specific implants improve women's outcomes relative to a unisex knee? One prospective study with 146 women showed no difference (Johnson et al., 2011). Many studies designed to evaluate outcomes have been performed retrospectively, and the American Academy of Orthopedic Surgeons has described how this limits the level of evidence provided (Jacobs et al., 2007).

2. How do the outcomes of TKA with unisex prostheses compare in women and men? A 2003 report commissioned by the U.S. National Institutes of Health concluded that no evidence pointed to sex as a strong predictor of outcome, but that this conclusion was "tempered by the limitations of many of the designs of the studies included in the analysis" (Kane et al., 2003).

3. What are the financial relationships between prosthesis developers, researchers, consultants, physicians, and universities, and how might these affect the objectivity of outcome studies (Gelberman et al., 2010)? The International Committee of Medical Journal Editors requires researchers to declare any associations with "commercial entities that provided support for the work reported" as well as with "entities that could be viewed as having an interest in the general area of the submitted manuscript." However, such policies are not universally or uniformly implemented (Drazen et al., 2010).