The Assessment of the Tendon of Insertion of the Iliopsoas Muscle in Dogs Using Diagnostic Ultrasound Has Good Intraobserver Consistency but Lacks Interobserver Consistency

TL;DR

This study finds that while a single veterinarian can reliably assess dog iliopsoas tendons using ultrasound, different veterinarians often reach different conclusions.

Contribution

The study provides quantitative evidence of poor interobserver consistency in ultrasound evaluation of canine iliopsoas tendons.

Findings

Intraobserver consistency was good with coefficients of variation under 20%.

Interobserver consistency was poor with intraclass correlation coefficients below 0.500.

Abnormalities were detected in 15% of presumed normal tendons and 92% of presumed abnormal tendons.

Abstract

The iliopsoas (IPM) is a muscle that is used by dogs to move their back leg forward. Injuries to the tendon at the end of the IPM are common in dogs, particularly active dogs such as those doing agility. In this study, we evaluated 104 IPM tendons from 52 dogs. Some tendons were normal, and other tendons were injured. We evaluated whether a veterinarian who used a portable ultrasound machine to evaluate the tendon of the IPM would find the same changes if they scanned a normal or abnormal tendon twice. We also evaluated differences between the findings of three veterinarians reading scans of these tendons. We found that when a veterinarian evaluated tendons twice, measurements were usually very close to each other, but when different veterinarians read scans of the same tendons, findings often varied among veterinarians. To make findings more consistent, efforts should be made to make sure that all veterinarians interpret scans of tendons of the IPM using the same technique.

The iliopsoas muscle (IPM) is a flexor of the hip joint in dogs that is vulnerable to injury. Intraobserver and interobserver reliability of musculoskeletal ultrasound when evaluating the tendon of insertion of the IPM was studied. The IPM tendons of insertion of a randomly selected cohort were screened by one investigator and recorded. Musculoskeletal ultrasound recordings were separated into two groups of 20 dogs with presumptively normal IPM tendons of insertion and 32 dogs with one or two presumptively abnormal IPM tendons. Recordings were anonymized. The 104 tendons from these 52 dogs were independently reviewed twice by three observers. Abnormalities were detected in 6 of 40 presumptively normal IPM tendons (15%) and 59 of 64 presumptively abnormal IPM tendons (92%). Intraobserver repeatability of measurements of tendon dimensions was good (range, 0.24 to 0.76 mm), intraobserver reliability was good or excellent (range, 0.812 to 0.917), and intraobserver consistency was good (all coefficients of variation <20%). All measurements had poor interobserver consistency (intraclass correlation coefficients <0.500). Measurements of the tendon of insertion of the IPM have acceptable intraobserver repeatability, consistency, and reliability but have poor interobserver consistency, suggesting that efforts should be made to standardize the interpretation methods when using diagnostic ultrasound to evaluate problems affecting the tendon of insertion of the IPM in dogs.

Figures