Abstract
1450
Objectives: Tc-99m Sestamibi (MIBI) parathyroid scintigraphy has been used for preoperative localization in primary hyperparathyroidism (PHPT) and secondary hyperparathyroidism (SHPT). However, the reported detection rate was not good enough as compared with other imaging studies like 18FCH PET/CT. Therefore, our aim was to evaluate the clinical usefulness of semi-quantitative analysis with standard uptake value (SUV) in Tc-99m MIBI SPECT/CT for diagnosis of diseased parathyroid gland.
Methods: We retrospectively reviewed 129 patients with hyperparathyroidism (70 PHPT and 59 SHPT) who underwent preoperatively Tc-99m MIBI dual-phase scan and delayed SPECT/CT from Jan, 2016 to May, 2018. Maximum standardized uptake value (SUVmax) and the ratio of standardized uptake value (SUVR) were calculated in each parathyroid lesion on SPECT/CT images like PET/CT as follows: SUVmax (g/mL) = (Maximum radioactivity / Volume of voxel) / (Injected radioactivity / Body weight), and SUVR = SUVmax in parathyroid lesion / SUVmean in normal left upper lung, while SUVmean (g/mL) = (Total radioactivity / Volume of VOI) / (Injected radioactivity / Body weight). Demographic information, histopathology result and biomarkers were recorded, and the size of post-operative parathyroid lesion was measured assuming an elliptical shape using the formula: Volume (ml)=4/3π×length/2×width/2×height/2. With regard to descriptive data, analysis of variance (ANOVA) and correlations of variables (r) were carried out. Receiver Operator Characteristics (ROC) curves were generated by the result of pathology in relation to the SUVmax and SUVR levels. The best cut-off values were determined using Youden’s index.
Results: All demographic data were summarized in Table 1. Mean SUVmax and SUVR were 3.56±1.60 and 8.52±4.46 in normal parathyroid glands and significantly higher in adenoma / hyperplasia (5.80±5.04 and 14.96±13.46, p<0.05) for PHPT. Correlation between the size, the weight and serum iPTH of PHPT assessed with Pearson Correlation were r=0.73 for the size and r=0.69 for the weight (p<0.01 for both). The ANOVA and Pearson Correlation for variables were outlined in Table 2. In visual interpretation, the sensitivity, specificity and accuracy for PHPT and SHPT were 71.8%, 27.0%, 56.5% and 78.5%, 52.6%, 76.2%, respectively. In the ROC analysis (Table 3), we found the best threshold values for SUVmax and SUVR levels in predicting adenoma / hyperplasia were 4.07 and 8.24 for PHPT, and 2.32 and 3.34 for SHPT, separately.
Conclusions: Parathyroid adenoma / hyperplasia have higher SUVmax and SUVR levels than normal parathyroid glands. In addition, semi-quantitative analysis showed higher cutoff values in patients with PHPT than SHPT. These findings were consistent with current gold standard of visual interpretation for Tc-99m MIBI SPECT/CT study. The application of semi-quantitative analysis may be a useful tool for preoperative detection and localization of parathyroid adenoma / hyperplasia with optimal diagnostic accuracy.