# Laxity index measurement on stress radiographs obtained using the Vezzoni-modified Badertscher distension device technique: repeatability and reproducibility in a large cohort of dogs

## TL;DR

This study evaluates the consistency of hip joint laxity measurements in dogs using a specific radiographic technique, finding it reliable within observers but with some variability between them.

## Contribution

The study provides the first evaluation of repeatability and reproducibility of the laxity index in a large cohort using the VMBDD technique.

## Key findings

- Repeatability within observers was excellent with ICCs ranging from 0.94 to 0.99.
- Inter-observer agreement was moderate (ICC = 0.55) with Observer 3 consistently reporting higher laxity index values.
- Differences in measurements were considered clinically negligible despite variability.

## Abstract

The Vezzoni-modified Badertscher distension device (VMBDD) technique is a radiographic method used to assess hip joint laxity, and it is widely used across Europe. While the intra-observer and inter-observer variability of the laxity index (LI) measured on stress radiographs obtained using the VMBDD technique has been reported, it has never been evaluated in a large cohort of patients. The study aims to assess the repeatability and reproducibility of the LI measured on stress radiographs obtained using the VMBDD technique in a large cohort of dogs.

Stress radiographs obtained using the VMBDD method were analyzed for medium to large breed dogs, aged between 4.5 and 6 months and presented between 2021 and 2024 for screening of hip dysplasia. The LI for each hip was blindly measured by three observers with different levels of experience. Significant intra- and inter-observer variability was evaluated to assess the repeatability and reproducibility of the LI, respectively. Statistical testing was performed, and a p-value of <0.05 was considered statistically significant. Inter-observer and intra-observer intraclass correlation coefficients (ICCs) were evaluated.

A total of 195 stress radiographs (390 hip joints) were included. The inter-observer ICC showed moderate agreement (ICC = 0.55, 95% CI 0.50–0.59). Estimated marginal means (EMMeans) indicated that Observer 3 consistently provided higher LI values compared to Observers 1 and 2 across all time points (e.g., at T1: 0.484 vs. 0.410 and 0.438, p < 0.001 for Observer 1 vs. Observer 3). The repeatability within each observer was excellent for all three observers (Observer 1: ICC = 0.94, 95% CI 0.93–0.96; Observer 2: ICC = 0.99, 95% CI 0.99–0.99; Observer 3: ICC = 0.95, 95% CI 0.94–0.96).

In-house evaluation of the LI on stress radiographs obtained using the VMBDD technique showed that it was a highly repeatable procedure but a moderate reproducible measurement due to a systematic upward bias by an observer with less experience. Nevertheless, the mean differences could be considered negligible in a clinical setting due to their low impact on the definitive diagnosis.