Contrast-enhanced triggered harmonic sonography for assessment of periarticular hemodynamic changes in experimental arthritis

• Published in: Pediatric radiology 2006-12, Vol.36 (12), p.1242-1251
• Main Authors: Doria, Andrea S, Karshafian, Raffi, Moineddin, Rahim, Mohanta, Arun, Zhong, Anguo, Mendes, Maria, Pritzker, Kenneth, Jong, Roland, Burns, Peter
• Format: Article
• Language: English
• Subjects: Animals; Area Under Curve; Arthritis, Experimental - diagnosis; Blood Flow Velocity; Body Weights and Measures - methods; Contrast Media - administration & dosage; Disease Models, Animal; Fluorocarbons; Hindlimb - blood supply; Hindlimb - diagnostic imaging; Image Enhancement - methods; Image Processing, Computer-Assisted - methods; Injections, Intra-Articular; Joints - blood supply; Joints - diagnostic imaging; Male; Microbubbles; Rabbits; ROC Curve; Sensitivity and Specificity; Serum Albumin - administration & dosage; Serum Amyloid A Protein; Sodium Chloride - administration & dosage; Synovial Membrane - ultrastructure; Time Factors; Ultrasonography
• ISSN: 0301-0449; 1432-1998
• DOI: 10.1007/s00247-006-0300-5

Objective quantification is critical for assessment of functional sonography in inflammatory arthritis. To create a microbubble contrast-enhanced image of vessels that lie below the resolution of a standard US system, a technique is required that detects preferentially the contrast agent echo, rejecting that from background tissue: harmonic imaging. To investigate the ability of contrast-enhanced triggered harmonic sonography (CETHS) to evaluate periarticular hemodynamic changes over the course of experimental arthritis and to discriminate presence and absence of arthritis based on measurement values obtained at specific time-points. Arthritis was induced in rabbits knees by intra-articular injection of serum bovine albumin, which acted as an antigen. A total of 11 rabbits (8 with unilateral arthritis and 3 control animals) were imaged at 0, 1, 7, 14, 21 and 28 days of antigen-induced arthritis and euthanized at 28 days. A continuous infusion protocol was performed (triggering times 30.0, 20.0, 10.0, 5.0, 2.0, 1.0, and 0.5 s). Hemodynamic indices of synovial microvasculature (vascular volume, mean velocity and flow rate) were obtained and compared with clinical, laboratory, and histological surrogate markers. Although interval CETHS changes were noted for flow rate (P=0.007) and vascular volume (P=0.003) ratios in albumin-injected knees, no significant differences in ratios were identified over time between albumin-injected and non-injected knees for flow rate (P=0.52), vascular volume (P=0.23) and mean velocity (P=0.19). Flow rate most accurately differentiated between presence and absence of arthritis according to clinical measurements in early (day 1) arthritis, and mean velocity in mid-term arthritis (day 14; both P=0.02). Although the measurement properties of CETHS indices were poor in the evaluation of hemodynamic differences over time in albumin-injected knees compared with non-injected knees, they enabled discrimination between presence and absence of arthritis at specific time-points in different stages.