Parathyroid imaging: the importance of pinhole collimation with both single- and dual-tracer acquisition

Our objective was to rigorously compare pinhole and parallel-hole collimation in an intrapatient, intrastudy design in 2 parathyroid imaging protocols: the first was dual-phase (99m)Tc-sestamibi imaging, and the second was dual-phase (99m)Tc-sestamibi plus dual-tracer ((99m)Tc-sestamibi and (123)I) simultaneous-acquisition subtraction imaging. Thirty-three patients with 37 surgically proven nonectopic parathyroid adenomas were evaluated. Anterior pinhole and parallel-hole images of the neck were available for (99m)Tc-sestamibi at 15 min and 3 h, and for simultaneously acquired (99m)Tc-sestamibi and (123)I subtraction at 15 min, all from a single study. The images were modified so that all had a square border and so that the thyroid filled approximately three quarters of the image. The images were evaluated by 2 experienced nuclear medicine physicians who did not know the surgical results or whether the images were acquired with pinhole or parallel-hole collimation. The observers indicated the location of any identified adenoma and graded the certainty of diagnosis on a 3-point scale. The localization success rate for the 2 observers combined for the single-tracer dual-phase images was 66.2% with pinhole collimation and 43.2% with parallel-hole collimation (P < 0.0001). The localization success rate with the addition of the dual-tracer simultaneous-acquisition subtraction image was 83.8% with pinhole collimation and 62.2% with parallel-hole collimation (P = 0.0018). In addition, the degree of certainty of localization was greater with pinhole collimation with both imaging protocols (P < 0.001 in both cases). In the anterior projection, pinhole collimation is superior to parallel-hole collimation for parathyroid imaging with either dual-phase (99m)Tc-sestamibi or dual-phase (99m)Tc-sestamibi plus dual-tracer ((99m)Tc-sestamibi and (123)I) simultaneous-acquisition subtraction.