# Consistently high accuracy of digital 2D templating in total knee arthroplasty across different levels of surgical training

## TL;DR

Digital 2D templating in knee replacement surgery is highly accurate and not affected by the surgeon's training level.

## Contribution

Demonstrates that 2D digital templating accuracy in TKA is consistent across different surgical training levels.

## Key findings

- 92.7% of femoral and 88.7% of tibial components were within ±1 size of the implanted size.
- No significant differences in templating accuracy were found between junior residents, senior residents, and board-certified surgeons.

## Abstract

To evaluate the accuracy of two-dimensional (2D) digital templating in primary total knee arthroplasty (TKA) and assess whether surgical training level affects templating accuracy.

A total of 424 patients who underwent primary TKA with preoperative 2D digital templating using the Attune system were retrospectively analyzed. Templating was performed in TraumaCad (Brainlab AG) by junior residents (< 3 years of training), senior residents (≥ 3 years), or board-certified orthopaedic surgeons. Planned and implanted component sizes were compared, and accuracy was assessed as exact matches and deviations of ± 1, ± 2, and ± 3 sizes. Pearson correlation analysis was used to assess the association between planned and implanted sizes. One-way ANOVA was used to compare mean absolute deviation across training levels. Additionally, the proportion of cases with a deviation greater than ± 1 size was calculated for both components across experience levels and compared using chi-square tests.

A total of 424 patients (61% female) were included. The median planned component sizes were 6 (IQR, 5–7) for the femoral and 6 (IQR, 5–7) for the tibial component; the median implanted sizes were 6 (IQR, 5–7) and 6 (IQR, 4–7), respectively. Planned and implanted sizes were very strongly correlated for both femoral (r = 0.864; P < 0.001) and tibial components (r = 0.841; P < 0.001). Templating accuracy was high, with 92.7% of femoral and 88.7% of tibial components within ± 1 size. No significant differences in correlation strength or mean absolute deviation were observed across training levels (P > 0.05). The proportion of cases with > ± 1 size deviation was low across all groups and did not differ significantly between training levels for either component (femoral: P = 0.874; tibial: P = 0.791).

2D digital templating for primary TKA demonstrated high accuracy, with reliable prediction within a ± 1 size range and no significant influence of surgical training level. These findings support its continued clinical use and confirm that templating can be reliably performed by residents at all stages of training.

Video Abstract

Video Abstract

Level III, diagnostic study.

The online version contains supplementary material available at 10.1186/s42836-025-00356-5.