BACKGROUND: The advantages of preoperative localization in the management of primary hyperparathyroidism have not been clearly demonstrated. The aim of this study was to investigate prospectively the accuracy of three localization techniques in patients with this condition. METHODS: Forty-nine consecutive patients with primary hyperparathyroidism underwent ultrasonography, magnetic resonance imaging (MRI) and technetium-thallium (Tc-Tl) subtraction scanning before surgery, during which an attempt was made to identify all parathyroid glands. A scan was regarded as correct if it identified an enlarged parathyroid gland on the correct side of the neck as subsequently demonstrated at surgery. RESULTS: Ultrasonography had a sensitivity of 38 per cent (18 correct scans in 47 patients) with a positive predictive value of 78 per cent. The sensitivity of MRI was 72 per cent (34 of 47) with a predictive value of 92 per cent. Tc-Tl scanning was 60 per cent sensitive (28 of 47) with a predictive value of 85 per cent. Two patients with negative neck explorations were subsequently found to have mediastinal adenomas. CONCLUSION: Ultrasonography, MRI and Tc-Tl scanning have limited value as localization techniques and the relatively low sensitivity of these investigations means they are of no value before first-time surgery.
BACKGROUND: The advantages of preoperative localization in the management of primary hyperparathyroidism have not been clearly demonstrated. The aim of this study was to investigate prospectively the accuracy of three localization techniques in patients with this condition. METHODS: Forty-nine consecutive patients with primary hyperparathyroidism underwent ultrasonography, magnetic resonance imaging (MRI) and technetium-thallium (Tc-Tl) subtraction scanning before surgery, during which an attempt was made to identify all parathyroid glands. A scan was regarded as correct if it identified an enlarged parathyroid gland on the correct side of the neck as subsequently demonstrated at surgery. RESULTS: Ultrasonography had a sensitivity of 38 per cent (18 correct scans in 47 patients) with a positive predictive value of 78 per cent. The sensitivity of MRI was 72 per cent (34 of 47) with a predictive value of 92 per cent. Tc-Tl scanning was 60 per cent sensitive (28 of 47) with a predictive value of 85 per cent. Two patients with negative neck explorations were subsequently found to have mediastinal adenomas. CONCLUSION: Ultrasonography, MRI and Tc-Tl scanning have limited value as localization techniques and the relatively low sensitivity of these investigations means they are of no value before first-time surgery.