A Statistical Model Using Patients Perioperative Clinical Markers for Optimal Predication of Calcium and Vit-D doses in Post-thyroidectomy Hypocalcemia-prospective study.

AS., AlDhahri SF, Mubasher M, AlMuhawas FA, Alessa M, Terkawi RS and Terkawi . 2014

Abstract

Objective: To optimize and individualize post-thyroidectomy hypocalcemia management.
Study Design: A multicenter prospective cohort study.
Setting: Two tertiary care hospitals.
Subjects and Methods: parathyroid hormone (PTH) was measured preoperatively, then at 1 and 6 hours after surgery. The required doses of calcium and vitamin D were defined as those maintaining the patients asymptomatic and their cCa ! 2 mmol/L. They were used as an endpoint in a generalized linear mixed effect model (GLIMMEX) aiming to identify the best predictors of these optimal required doses. Models were evaluated by goodness of fit and Receiver Operating Characteristic (ROC) curves.
Results: One hundred and sixty-eight patients were analyzed; 85.1% were female, 49.3% had BMI . 30, and 64% had vitamin D deficiency. Post-thyroidectomy hypocalcemia was found in 25.6%, of whom 18 (41.9%) were symptomatic and received intravenous calcium. First hour percentage of drop in PTH correlated positively with the severity of hypocalcemia (P \ .0001). The GLIMMIX prediction model for oral calcium requirement was based on first-hour percentage change from preoperative PTH level, preoperative actual PTH, BMI, and thyroid function. The same predictors were identified for vitamin D, except that thyroid function was replaced with vitamin D status. These factors were used to build predictive equations for calcium and vitamin D doses.
Conclusion: Our findings help to optimize management of post-thyroidectomy hypocalcemia by assisting in the early identification of those who are not at risk of hypocalcaemia and by guiding early effective management of those at risk. This may reduce complications and medical cost.