Interobserver and Intraobserver Variability in Crural Angiography.

Objective: The gold standard imaging modality for assessing crural arterial disease is digital subtraction angiography (DSA). Using DSA, the operator can estimate the diameter and degree of stenosis and determine the diameter of the balloon. Since these measures are estimates, it allows for interobserver and intraobserver variability. This study aimed to determine the interobserver and intraobserver variability in the assessment of crural angiography.

Method: A total of 15 angiographies with corresponding patient characteristics were presented to 7 assessors, including 3 vascular surgery residents and 4 vascular surgeons on 3 separate occasions. A series of questions was presented, including vessel identification, estimated diameter, estimated degree of stenosis, treatment decision, and which type of balloon or stent would be used including diameter. The primary outcomes were interobserver and intraobserver variability in the estimated diameter of the artery, the estimated degree of stenosis and treatment decision. Secondary outcomes were vessel identification and how the lesion would be treated. Interobserver and intraobserver agreement was calculated using intraclass correlation coefficients (ICCs) and Hubert's kappa, interpreted as poor (<.50), moderate (.50-.75), good (.75-.90), and excellent (>.90) agreement.

Results: Interobserver agreement was moderate (.62) for estimated vessel diameter, moderate (.72) for degree of stenosis and moderate (.53) for whether the physician would treat the stenosis. The interobserver agreement was poor regarding balloon diameter (.21). The intraobserver agreement ranged from moderate to good (.62-.80) for estimated vessel diameter and was good for estimated degree of stenosis (.76-.87). The intraobserver agreement regarding treatment decision ranged from moderate to excellent (.76-.87) and ranged from poor to moderate (.07-.57) for balloon diameter.

Conclusion: This study shows moderate interobserver agreement and moderate to good intraobserver agreement regarding estimated vessel diameter. In addition, moderate to excellent intraobserver agreement and moderate interobserver agreement regarding treatment decision was found. This study also showed poor to moderate agreement, both interobserver and intraobserver, regarding balloon sizing. There is a need for more objective diagnostic modalities to assess the diameter and morphology of crural lesions.

Clinical impact: Digital subtraction angiography (DSA) is currently the golden standard imaging modality guiding below-the-knee percutaneous transluminal angioplasty (PTA). Due to the lack of objective measurements with DSA, physicians estimate vessel diameter and percentage of stenosis. The present study shows that there is inter- and intraobserver variability in the assessment of DSA, indicating there is a need for objective imaging modalities guiding below-the-knee PTA.