Efficacy of intraoperative parathyroid hormone monitoring to predict success of parathyroidectomy for secondary hyperparathyroidism.
- Author:
Woo Young KIM
1
;
Jae Bok LEE
;
Hoon Yub KIM
Author Information
- Publication Type:Original Article
- Keywords: Hyperparathyroidism; Secondary; Parathyroidectomy
- MeSH: Follow-Up Studies; Humans; Hyperparathyroidism; Hyperparathyroidism, Secondary; Parathyroid Hormone; Parathyroidectomy; Retrospective Studies; ROC Curve; Sensitivity and Specificity
- From:Journal of the Korean Surgical Society 2012;83(1):1-6
- CountryRepublic of Korea
- Language:English
- Abstract: PURPOSE: In secondary hyperparathyroidism (sHPT), it is unclear which criterion for intraoperative parathyroid hormone (ioPTH) decline performs best and whether ioPTH monitoring is sufficiently efficacious in predicting postoperative surgical cure by retrospective review. METHODS: In 80 parathyroidectomies for sHPT, patients with ioPTH monitoring and follow-up PTH assay were categorized into physiologically accepted and failed groups. Receiver operating characteristic (ROC) curves were used to identify the criterion with the best performance and to determine the regression equation for prediction of surgical success. RESULTS: There was a statistically significant difference in the percentage reduction of ioPTHs between the two groups (P < 0.05). With our criterion, cure was predicted with a sensitivity of 86% and specificity that could be improved to 60% using an 85% ioPTH decrease as cut-off level when this value was compared to the Miami Criterion. There was a slightly negative correlation between the natural logarithm of percentage reduction of ioPTH and percentage reduction of PTH at follow-up (R2 = 0.109, P = 0.003). CONCLUSION: ioPTH measurements are a useful tool for improvement of the cure rate of operations for sHPT. Because this study aimed at 40 minutes (ioPTH3) as an optimal process in the operation, we recommend using a criterion of more than 85% ioPTH decline at 40 minutes by use of the ROC curve, and the use of the presently calculated regression equation may enable prediction of success.
