Optimization of MDCT of the wrist to achieve diagnostic image quality with minimum radiation exposure

OBJECTIVE. This study tests various acquisition and reconstruction protocols for MDCT of the wrist to determine the optimal protocol for obtaining diagnostic image quality with minimal radiation exposure. MATERIALS AND METHODS. Thirty anatomic specimens were examined with an MDCT collimation of 4.0 x 1.0 mm and 2.0 x 0.5 mm (80,120, and 140 kV; 80,100,130,160, and 200 mA; rotation time, 0.5 0.75, 1.0 see; pitch, 1.0, 1.3, 1.5, and 2.0). Coronal images were reconstructed using a slice thickness of 0.5, 1.0, and 2.0 min with 60% overlap. Three observers evaluated all images independently for gross and fine anatomic detail. Diagnostic confidence was tested using Shrout-Fleiss intraclass correlation coefficients. Interobserver agreement was assessed by Kappa statistics and the Kruskal-Wallis test. RESULTS. Fine anatomic detail was best presented in 0.5-min or 1.00-mm reconstructions based on a 2.0 x 0.5 mm acquisition. A rotation time of >= 0.75 sec resulted in fewer artifacts; a significant dose reduction was achieved with 80 kV and 100 MA at the expense of somewhat increased noise, but without significant loss of anatomic detail in bone presentation. Artifacts were tolerable with a pitch of 1.5 or less. CONCLUSION. MDCT at the described optimal settings allows significant dosage reduction without sacrificing image quality. An acquisition and reconstruction thickness of 0.5 mm results in the best depiction of anatomic detail. A reconstruction thickness of 1.0 mm with a reconstruction interval of 0.5 min represents a good trade-off between noise and resolution when using low-dose protocols.