Presurgical localization of abnormal parathyroid glands using a single injection of Tc-99m sestamibi: comparison of high-resolution parallel-hole and pinhole collimators, and interobserver and intraobserver variation.

PURPOSE

Scintigraphy using Tc-99m sestamibi is sensitive in localizing abnormal parathyroid glands in patients with hyperparathyroid disease (HPT). Various methods have been described to increase the sensitivity of the single-tracer technique, such as SPECT, factor analysis of dynamic structures, and use of a pinhole collimator, but often the gain in sensitivity is accompanied by a loss of specificity.

PURPOSE

In this study, the authors compared the sensitivity and specificity of side localization of the diseased gland(s) using the double-phase single-tracer method performed with and without the addition of a pinhole collimator in the early and late phases of imaging. The combined high-resolution parallel-hole and pinhole collimator imaging protocol was further validated by investigation of interobserver and intraobserver variation.

MATERIALS AND METHODS

Forty-seven patients with primary HPT and 16 patients with secondary HPT examined from 1996 to 1999 with the Tc-99m sestamibi double-phase technique and who had subsequent surgery formed the basis of the study. Their histologic and follow-up data were also factored into this analysis. Tc-99m sestamibi (750 to 900 MBq; 18.9 to 24.3 mCi) was injected. Ten-minute neck and mediastinum images acquiring 1,000 K counts were obtained with the high-resolution parallel-hole collimator, and a neck image containing 500 K counts was obtained with the pinhole collimator. Two to three hours later, the same parallel-hole and pinhole collimator images were obtained that had the same acquisition time as the early images. Two observers who were nuclear medicine specialists independently viewed all the parallel-hole scintigrams and afterward all parallel-hole and pinhole scintigrams two times.

RESULTS

Thirty-eight (81%) of the patients with primary HPT had a single adenoma. The sensitivity and specificity for the correct side of localization were 54% and 89%, respectively, using the high-resolution parallel-hole collimator, and 88% and 77%, respectively, with the addition of the pinhole collimator in all patients with primary HPT. In patients with secondary HPT, the sensitivity and specificity for localization of the correct side were 58% and 100%, respectively, using the high-resolution parallel-hole collimator, and 85% and 100%, respectively, with the addition of the pinhole collimator. The interobserver agreement was acceptable, with an overall agreement of 84% and a kappa value of 0.67. The intraobserver agreement was even better, with an overall agreement of 88% and 90% and kappa values of 0.76 and 0.79 for the two observers.

CONCLUSIONS

Sensitivity is increased considerably when the pinhole collimator is added to the imaging protocol of parathyroid scintigraphy in patients with primary or secondary HPT. A loss of specificity occurred only in patients with primary HPT. The precision of the combined approach is very high.