Computed tomographic arthrography (CTA) has been described as a method for detecting articular cartilage defects in equine carpal joints; however, published studies on the effects of contrast volume for lesion detection are currently lacking. The purpose of this prospective, experimental, pilot study was to determine a threshold volume of iodinated contrast for CTA of the antebrachiocarpal (ABC) and middle carpal (MC) joints for detection of articular cartilage surface defects. Articular cartilage defects were iatrogenically created in the surfaces of the ABC and MC joints of 20 equine cadaver limbs using arthroscopy. Unaltered articular surfaces within some joints acted as controls. Joints were imaged precontrast using multidetector CT. The ABC and MC joints were injected with a 150 mg iodine/ml nonionic contrast medium, in 5 ml increments from 5 to 50 ml per joint with CT performed subsequent to each increment. Cartilage defects were measured grossly using a caliper. Detection (qualitative) and measurement (quantitative) of the defects were independently performed by two board-certified radiologists using medical imaging software. At each volume of contrast, the interrater reliability for gross examination and the two observers in the detection of a defect was calculated (Gwet's AC1). Logistic mixed-effects models of selected volumes, 0, 5, 10, 15, and 50 ml, demonstrated that at 10 ml contrast and above, no statistically significant difference between either observer and gross examination for defect detection was identified for either joint. Findings supported using a dose of 10 ml for 150 mg iodine/ml concentration contrast media when performing CTA of equine carpal joints.