Using Stereology Methods in Weightbearing CT 3D Volume Measurement of the Ankle Syndesmosis; is There a More Accessible and Reliable Method?

Research Type: Level 3 - Retrospective cohort study, Case-control study, Meta-analysis of Level 3 studies Introduction/Purpose: Accurate assessment of ankle syndesmosis is critical for diagnosing and managing syndesmotic injuries. Weight-bearing CT (WBCT) provides a three-dimensional (3D) evaluation of the syndesmosis; however, the current measurement techniques are time-consuming, have inter- and intra-observer variabilities, and need specific software tools to conduct. This study aimed to compare the reliability and reproducibility of two WBCT measurement methods: conventional measurement method for syndesmotic area and volume using specific software with 3D measurement tools and Stereology methods using a program made in-house that can be used on any computers and smart devices for free. We hypothesized that Stereology methods would demonstrate non-inferior intra- and inter-observer reliability compared to the conventional method, making it a more accessible and faster approach for clinical and research applications. Methods: A total of 42 subjects underwent WBCT to assess syndesmotic 2D area and 3D volume. The conventional method was based on a study by Ashkani-Esfahani et al. (2021). The “point-counting volume density” method was used for Stereological assessment as shown in Figure 1 (Ashkani-Esfahani et al. 2016). Two independent observers performed two separate measurements (with a 1-week interval) using both conventional and stereology methods (four measurements per observer). Intra-observer reliability (repeatability) was assessed by repeated measurements of the same observer, and inter-observer reliability was evaluated by comparing measurements between two different observers. Intraclass Correlation Coefficients (ICCs) were calculated to assess agreement, with ICC >0.90 considered excellent, 0.75–0.90 good, and < 0.50 poor. Wilcoxon Signed-Rank tests compared values obtained from each measurement method. Additionally, percentage differences between affected and normal sides were analyzed to determine method agreement and consistency. Outcomes were exhibited as median (interquartile range), and p-value < 0.05 was considered significant. Results: The absolute area measurements using conventional and Stereology were 1.3 cm2 and 1.22 cm2, respectively (P=0.06), and volumes were 9.3 cm3 and 10.2 cm3, respectively (P=0.05). The conventional method demonstrated slightly superior intra-observer reliability for volume measurement (ICC=0.90) and inter-observer reliability (ICC=0.84), whereas Stereology exhibited good intra- and inter-observer reliability (ICC=0.84). Nevertheless, the time consumed to conduct Stereology measurement was one-third of the conventional method (Avg: 5 minutes vs. 15 minutes, respectively). Conclusion: Although the conventional measurement method demonstrated slightly superior reliability and reproducibility for WBCT-based syndesmotic 3D volume assessment, the Stereology method was shown to have good inter- and intra-observer reliability and faster measurement. Multiple area measurements in volume assessments may introduce observer biases. Despite easier and free access, Stereology has a learning curve for practical adoption. We suggest further studies to introduce this method to clinicians and to assess whether it is worth adopting it or we should allocate more resources towards automation of WBCT image interpretation, particularly with the rapid growth of AI.