45.07 Predictors of Operative Failure in Parathyroidectomy for Primary Hyperparathyroidism

D. C. Cron1, S. Kapeles1, S. Kwon1, P. Kirk1, E. Andraska1, B. McNeish1, D. T. Hughes1  1University Of Michigan,Surgery,Ann Arbor, MI, USA

Introduction:  Cure rates for hyperparathyroidism with parathyroidectomy are typically 95%; however, studies analyzing risk factors for operative failure and persistent hyperparathyroidism are limited. This study sought to identify predictors of operative failure in parathyroidectomy for primary hyperparathyroidism.

Methods:  This is a retrospective review of 2084 adult patients with primary hyperparathyroidism who underwent initial parathyroidectomy at a single center from 1999-2012.  Parathyroidectomy included both directed-approach (59%) and bilateral explorations (41%). The following data were collected: age, sex, perioperative lab values, imaging results, thyroid pathology, intraoperative data including parathyroid hormone (IOPTH) levels, and number of excised glands. Biochemical cure was defined as both an IOPTH decrease of ≥50% from the baseline and a level in the normal range (<75pg/ml). Operative failure (persistent disease) was defined by two or more postop serum calcium measurements ≥10.2mg/dL or a single calcium ≥11.0mg/dL when no other data available. Univariate logistic regression and backwards-stepwise multivariate logistic regression were used to identify significant predictors of operative failure. 

Results: Operative failure was identified in 65 patients (3%), with 35% having reoperative parathyroidectomy as a result. Significant predictors of failure on univariate analysis included: failure to meet IOPTH criteria, lower excised gland weight (< median of 600mg), multiple glands excised, bilateral exploration, lower preop PTH, higher preop calcium, and higher final IOPTH. A correct pre-op sestamibi was protective (Fig. A). Preop ultrasound localization was not associated with operative success. Predictors retained in the multivariate model included: IOPTH criteria met (OR=0.2, CI 0.1-0.3, P<0.01), low gland weight (OR=4.2, CI 2.0-9.0, P<0.01), high preop calcium (OR=1.9, CI 1.3-2.8), and low preop PTH (OR=0.99 CI 0.99-1.0, P=0.03). The C-statistic for this model was 0.80 (Fig. B). 

Conclusion: Operative failure of parathyroidectomy for primary hyperparathyroidism is rare; however, failure to meet IOPTH criteria, low preoperative PTH levels, high preoperative calcium levels, and low gland weight are risk factors for persistent disease.