Inter- and intraobserver variability of (99m)Tc-DMSA renal scintigraphy: impact of oblique views.

UNLABELLED (99m)Tc-Dimercaptosuccinic acid (DMSA) scintigraphy is a frequently used diagnostic test to assess the presence and severity of cortical damage. The aim of this study is to investigate the variability in the interpretation of (99m)Tc-DMSA scans, evaluate the usefulness of oblique images, and assess their impact on scan interpretation. METHODS Two experienced nuclear medicine physicians independently interpreted 100 (99m)Tc-DMSA scans (197 kidneys) 4 times. Interpretation was twice based on posterior projection images and twice based on posterior and posterior oblique projection images. For each kidney, the observers had to choose between the following results: normal, abnormal, and indeterminate. The indices of variability used were the percentage of agreement, kappa-statistic, and marginal homogeneity. RESULTS Intraobserver and interobserver reproducibility (kappa-values) varied between 0.683 and 0.708 and between 0.609 and 0.671, respectively, for posterior images. Disagreement in (99m)Tc-DMSA scan interpretation occurred in 18% of kidneys within observers and in 21% of kidneys between observers when only posterior images were used. Oblique views changed the interpretation in 14% and 11.5% of kidneys for the first and second observers, respectively. The use of oblique views increased the agreement rate within and between observers (kappa-values, 0.725-0.812 and 0.768-0.732, respectively; mean agreement, 86.5 and 87.25, respectively). CONCLUSION Oblique views were found useful in approximately 13% of kidneys and affected inter- and intraobserver variability. Our results suggest that oblique views should be used routinely in children with clinically suspected urinary tract infection to reliably interpret images.