Shoe size alone can predict implant sizes for total knee arthroplasty preoperative planning with over 92% accuracy.
The insights provided in this article are derived from a comprehensive scientific investigation by Prof. Cavaignac and his peers titled “Shoes size can predict implant sizes for primary total knee arthroplasty in a quick, reliable and costless manner.” This study, published in the Journal of Experimental Orthopaedics in 2025, explores how simple morphological (physical body structure) data can revolutionize the way surgeons prepare for joint replacement.
By understanding the relationship between a patient’s shoe size and their internal bone dimensions, medical teams can significantly streamline surgical workflows.
Why Preoperative Planning in Total Knee Arthroplasty Matters?
Preoperative planning in total knee arthroplasty (or TKA) refers to the process of anticipating which implant sizes a patient will need before the surgery actually takes place. This might sound like a minor logistical detail, but its impact on the entire surgical workflow is significant.
Without planning, operating rooms must stock every available size of every implant component: femoral (the part that replaces the thigh bone surface), tibial (the shin bone surface) and patellar (the kneecap component). This requires large storage areas, frequent expiry-date checks by nursing staff and last-minute trips to supply rooms during surgery, all of which cost time and money.
As TKA procedures increase globally, many now performed as same-day outpatient surgeries, and as supply chain disruptions continue to affect medical device availability, the pressure to optimize logistics has never been greater. Knowing implant sizes in advance allows hospitals to order only what they need, reduce stock, improve OR (operating room) workflow and even shorten the surgical learning curve for less experienced surgeons.
Traditional approaches to preoperative planning for primary total knee arthroplasty rely on radiographic templating, measuring standardized X-rays or CT scans to estimate implant dimensions. While effective, these methods require specialized software (which carries a cost), standardized imaging protocols that not all facilities can implement and significant surgeon time that is difficult to delegate. A simpler, cheaper and equally reliable method has long been sought.
How Shoe Size Enables Reliable Preoperative Planning for Knee Arthroplasty
Research on preoperative planning for knee arthroplasty has long explored whether easily available patient data (height, weight, age, sex) could predict implant sizes accurately. Shoe size has emerged as a particularly promising predictor, since foot length correlates strongly with overall skeletal dimensions.
This study tested whether the European shoe sizing system (which differs from the UK and US systems) could reliably predict femoral, tibial and patellar component sizes in 227 consecutive TKA patients operated on with the Score II implant (AMPLITUDE®) between April 2022 and July 2023. Shoe size was self-reported by patients before surgery, and the data was then compared to the actual implant sizes used in the operating room.
The methodology used Spearman correlation coefficients: a statistical measure of how strongly two variables are related, on a scale from 0 (no correlation) to 1 (perfect correlation). The results were striking:
- Femoral component: rho = 0.8095 (very strong correlation)
- Tibial component: rho = 0.8400 (very strong correlation)
- Patellar component: rho = 0.6393 (moderate-to-strong correlation)
Crucially, the model requires only two pieces of information from the patient: their shoe size and their sex. Using a simple chart provided in the study, any member of the healthcare team can look up the predicted implant sizes instantly, with no imaging, no software and no additional cost.
The Results: Accuracy That Rivals Radiographic Templating
The core question is: how often does the prediction land within one size of the actual implant used? A margin of ±1 size is considered clinically acceptable, since surgeons can always fine-tune intraoperatively.
After adjusting for sex, the shoe size model achieved:
- Femoral component: correct within ±1 size in 92% of cases (210/227)
- Tibial component: correct within ±1 size in 94% of cases (213/227)
- Patellar component: correct within ±1 size in 97% of cases (220/227), after additionally adjusting for BMI (body mass index)
These figures are comparable, and in some cases superior, to published results from radiographic templating methods, which report ±1 size accuracy ranging from 90% to 98.5% depending on the study and implant system. The shoe size model performs on par with these gold-standard techniques, while being dramatically simpler to implement.
One important caveat: the model does not reliably predict the thickness of the polyethylene (PE) insert, the plastic spacer placed between the metal components. The correlation here was weak (rho = 0.1532), a finding consistent with previous research. PE insert thickness is more closely tied to the degree of ligament correction needed intraoperatively and remains a decision made in the OR.
Limitations and Future Directions for Preoperative Planning in Knee Arthroplasty
No study is without limitations, and the authors are transparent about these. First, the model was validated on a single implant brand (Score II by AMPLITUDE®) in a single center with a single surgeon. This minimizes variability, which strengthens the internal validity of the findings, but it also means the charts cannot be directly transposed to other TKA systems without re-validation.
Second, while foot length tends to remain stable throughout adult life, foot width can change with aging, oedema (fluid retention caused by swelling), or prior surgery. However, the study notes that such changes do not appear to significantly affect prediction accuracy, consistent with findings from comparable research on unicompartmental knee arthroplasty, a partial knee replacement that only addresses one compartment of the joint.
Third, a safety stock of all implant sizes must still be maintained, since the model is not 100% accurate. However, this stock will be considerably smaller than if no preoperative planning were done at all and for patients requiring extreme sizes, the model helps flag the need to order those components from the manufacturer well in advance.
Future research should focus on external validation across different centers, patient populations and implant brands. The authors also suggest measuring the actual improvement in OR efficiency when the model is routinely applied, a practical and compelling next step for the field of preoperative planning.
Conclusion & Findings
Shoe size, combined with the patient’s sex, is a simple, free and highly accurate tool for preoperative planning in total knee arthroplasty. With prediction accuracy exceeding 92% for both femoral and tibial components, it matches the performance of far more complex and costly methods. It can be implemented immediately, delegated to nursing staff and applied across both high-volume arthroplasty centers and facilities that perform TKA only occasionally.
If you found this research valuable, we encourage you to explore the other published studies by Professor Etienne Cavaignac, whose work consistently advances evidence-based orthopedic practice.
For patients and referring physicians alike, Professor Cavaignac‘s expertise in knee surgery and his commitment to practical, patient-centered innovation make him an outstanding choice for the management of complex knee conditions.