Intra-rater and inter-rater reliability of ultrasonographic intestinal wall thickness measurements in healthy horses.

BACKGROUND

Abdominal ultrasound is widely used to evaluate the intestinal tract of horses. Despite being a routine examination, there is limited data on the reliability of this diagnostic procedure.

OBJECTIVES

To investigate intra- and inter-rater reliability of ultrasonographic intestinal wall thickness measurements in healthy horses. A second aim was to assess variance within repeated measurements to determine threshold values that distinguish whether differences between repeated examinations are true findings versus solely due to measurement variation.

STUDY

DESIGN

In vivo reliability study.

METHOD: Eight healthy horses (7 Standardbreds, 1 Warmblood) were ultrasonographically examined in six intestinal regions: duodenum, right dorsal colon (RDC), right ventral colon (RVC), caecum, jejunum and left ventral colon (LVC). In each horse, triplicate measurements of intestinal wall thickness were performed by three sonographers on three consecutive days. Intra-class correlation coefficient (ICC) was calculated to assess intra- and inter-rater reliability.

RESULTS

Intra-rater ICC was <0.5 for all regions except the duodenum (0.52). Inter-rater ICC was <0.5 for duodenum, caecum and jejunum and between 0.5 and 0.75 for RDC, RVC and LVC. The standard deviation of repeated measurements was low (0.33-0.45 mm). Across all regions, 95% (±2 SD) of all reported measurements were within a 1 mm range.

MAIN LIMITATIONS: Homogenous study group and use of healthy horses may limit generalisability to clinical populations.

CONCLUSIONS

Ultrasonographic measurements of intestinal wall thickness show limited consistency but low absolute variation. Differences of less than 1 mm fall within expected repeated measurement variability. This study supports the use of repeated ultrasonographic intestinal wall thickness measurements in clinical practice and provides a practical cut-off value of 1 mm to differentiate expected variability within and between observers from true changes in intestinal wall thickness.