Factors Contributing to Negative Parathyroid Localization: An Analysis of 1000 patients


    loading  Checking for direct PDF access through Ovid

Abstract

Background.Localizing studies are the key for determining the optimal surgical strategy in patients with primary hyperparathyroidism (HP). Most of the data in the literature are retrospective in nature and from analysis on a per patient basis. This is a prospective study looking at the characteristics of the patient and the gland that determine the likelihood of an abnormal parathyroid to be detected by ultrasonography (US) and sestamibi scan (MIBI).Methods.This is a prospective analysis of 1000 consecutive patients with HP who underwent parathyroidectomy at a tertiary care center. The study group included HP with single gland disease (63%), double adenoma (15%), as well as hyperplasia (15%), familial HP (2%), and secondary/tertiary HP (6%). All patients underwent surgeon-performed neck US followed by MIBI scan. Univariate logistic regression and multivariate analyses were performed on pre- and intraoperative variables.Results.A total of 1845 abnormal glands were analyzed. Overall, US was superior to MIBI for the detection of abnormal glands in all subgroups. On multivariate analysis, body mass index (BMI), gland size, and gland volume were the statistically significant independent factors predicting detection by both US and MIBI in primary HP. The sensitivity of US was better for single gland disease than for multigland disease in primary HP, but the sensitivity of MIBI was similar in both groups. For a given size, hyperplastic glands in primary HP imaged less well with US and MIBI than in familial or secondary/tertiary HP.Conclusion.This prospective study demonstrates that BMI and gland size independently predict accurate detection of abnormal parathyroid glands by US and MIBI in sporadic primary HP. Understanding the factors that affect the accuracy of parathyroid localization tests will allow the surgeon to develop a successful surgical strategy in a given patient.

    loading  Loading Related Articles