Impacts of Pinhole Images on Image Quality and Interpretation of Pediatric (superscript 99m)Tc-DMSA Renal Scintigraphy

Abstract

The purpose of this study was to compare the image qualities and interpretations of pediatric (superscript 99m)Tc-DMSA renal scintigraphy between parallel-hole and pinhole images. Methods: Between Jan 2007 and Dec 2007, 202 pediatric patients (99 boys and 103 girls; age range from 1 month to 14 years old) with urinary tract infection underwent (superscript 99m)Tc-DMSA renal scintigraphy for diagnosis of acute pyelonephritis (APN). Both of parallel-hole and pinhole images were acquired in all patients. A total of 808 images were assessed by two nuclear medicine specialists. The qualities of images were classified into three grades: poor (blurred renal contour, score 1), fair (clear renal contour but internal architecture not clearly visible, score 2), and good (renal contour and internal architecture clearly visible, score 3). Three categories of image interpretation for APN (negative, equivocal, and positive) were employed. Results: Good agreements within and between the two observers were noted in image quality assessments of parallel-hole images. Excellent and good agreements within and between the two observers, respectively, were noted in image quality assessments of pinhole images. In both of interpretations for parallel-hole and pinhole images, the intraobserver and interobserver agreements by the two observers were excellent and good, respectively. The mean quality scores of parallelhole and pinhole images were 2.10±0.55, 2.40±0.60 by observer A and 2.08±0.50, 2.32±0.58 by observer B, respectively (Wilcoxon test, both P ＜0.001). Pinhole collimator improved the qualities of DMSA images in 126/404 (31.2%) kidneys (23 poor to fair, 4 poor to good, and 99 fair to good) by observer A and in 111/404 (27.5%) kidneys (21 poor to fair, 3 poor to good, and 87 fair to good) by observer B. With pinhole images, interpretations were changed in 45/404 (11.1%) kidneys (20 negative to equivocal, 18 equivocal to positive, and 7 equivocal to negative) by observer A and in 50/404 (12.4%) kidneys (17 negative to equivocal, 9 equivocal to positive, 21 equivocal to negative, and 3 positive to equivocal) by observer B. Conclusion: (superscript 99m)Tc-DMSA scintigraphy using pinhole collimator in pediatric patients can improve image qualities in comparison with parallel-hole collimator. The pinhole images also have impacts on the interpretation for APN. Accordingly, we suggest using pinhole collimator in (superscript 99m)Tc-DMSA scintigraphy for all pediatric patients routinely.