Purpose The purpose of this systematic review and meta-analysis was to investigate the variability in femoral geometric ratios among knees of diferent sexes and races, and to appreciate whether the observed variability is accommodated by commonly implanted total knee arthroplasty (TKA) components. The hypothesis was that the anthropometric studies report considerable variability of femoral geometric ratios among sexes and races. Methods This systematic review and meta-analysis was performed according to the Preferred Reporting Items for Systematic reviews and Meta-analyses (PRISMA) criteria. Two authors independently conducted an electronic search using MEDLINE ® and Embase ® on 6 July 2020 for clinical studies reporting on femoral geometric ratios among sexes and races. Femoral geometric ratios were graphically represented as means and 2 standard deviations, and compared to those of 13 commonly implanted femoral components. Results A total of 15 studies were identiied that reported on a combined total of 2627 knees; all reported the aspect ratio, and 2 also reported the asymmetry ratio and trapezoidicity ratio. Men had wider knees than women, and compared to Caucasian knees, Arabian and Indian knees were wider, while East Asian were narrower. There were no diferences in asymmetry ratio between men and women, nor among Caucasian and East Asian knees. Men had more trapezoidal knees than women, and East Asian knees were more trapezoidal than Caucasian knees. The commonly implanted femoral components accommodated less than a quarter of the geometric variability observed among sexes and races. Conclusion Anthropometric studies reported considerable sexual dimorphism and racial diversity of femoral geometric ratios. Since a surgeon generally only uses one or a few TKA brands, bone-implant mismatch remains unavoidable in a large proportion of knees. These indings support the drive towards personalized medicine, and accurate bone-implant it may only be achievable through customisation of implants, though the clinical beneits of custom TKA remain to be conirmed. Level of evidence IV.