Kenneth Nichols PhD, Presenter: Royalties, Syntermed, Inc

Maria Bernadette Tomas MD, Abstract Co-Author: Nothing to Disclose

Gene Gellegani Tronco MD, Abstract Co-Author: Nothing to Disclose

Christopher J. Palestro MD, Abstract Co-Author: Research grant, General Electric Company

To identify factors associated with false negative (FN) MIBI studies in pts with primary hyperparathyroidism (PHP).

This was a retrospective investigation of 659 pts with biochemically confirmed PHP who underwent preoperative parathyroid lesion localization using a dual tracer 99mTc-MIBI/99mTcO4- protocol that included early & late planar pinhole 99mTc-MIBI, pinhole thyroid imaging, image subtraction, & SPECT. All pts subsequently underwent surgery. Pathology reports were reviewed for lesion weights & histology: adenoma (A) or hypercellular, and/or enlarged, and/or fat depleted, but not an adenoma (nonA). One experienced nuclear physician, who had no knowledge of other test results or final diagnoses graded studies on a 5-point scale (0=definitely normal to 4=definitely abnormal) while reading all scintigraphic images simultaneously. Association of false negative results with patient age, sex, lesion weight, lesion location, pre-surgical PTH & Ca levels & cell type (A or nonA) were evaluated by logistic regression.

There were 859 lesions among the 659 pts. 528 pts (80%) had single gland disease (SGD); 131 (20%) had multigland disease (MGD) (74 pts had 2 lesions, 45 had 3, & 12 had 4 lesions). 261 (30%) lesions were A; 598 (70%) were nonA. 45% of SGD lesions were A; 7.5% of MGD lesions were A (p < 0.0001). The factor most strongly associated with a FN result was MGD (Χ2 = 330.7; p<0.0001), followed by lesion weight (Χ2 = 51.3; p<0.0001), nonA histology (Χ2 = 11.5; p=0.0007), & elevated Ca level (Χ2 = 5.8; p=0.02). Age, sex, lesion location & PTH levels were too weakly predictive to permit computing an association. Scintigraphy was significantly more sensitive for SGD than for MGD (96% versus 61%, p<0.0001), with lesions of greater weight (log(weight) = -0.62±1.07 versus -1.55±1.16, p<0.0001) & marginally higher calcium levels (11.2±3.5 versus 10.9±1.0, p=0.042). Scintigraphy was most sensitive for detecting SGD A lesions (FN = 1%), & least sensitive for detecting MGD nonA lesions (FN = 41%).

While several factors are associated with false negative MIBI parathyroid studies in PHP, multigland disease is the single most strongly associated parameter.

Because of lower test sensitivity for MGD, preoperative MIBI imaging alone is insufficient & must be used along with rapid intraoperative parathyroid hormone assay to assure removal of all lesions.

Nichols, K, Tomas, M, Tronco, G, Palestro, C, Factors Associated with False Negative Results in MIBI Parathyroid Scintigraphy.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8004714.html