Birmingham formula is not a reliable instrument for preoperative prediction of thyroid malignancy [prospective study]

Thyroid nodules are very common in population all around the world. Preopoerative anticipation of malignancy in a thyroid nodule is crucial for optimum management. There are limitations for all available tests especially in follicular tumors. The Birmingham formula was described to predict the malignancy risk in each individual with a thyroid nodule depending on phenotyping association. We tried to explore prospectively the predictivity of this formula in comparison with the postoperative pathological report. Data for 512 patients with a thyroid nodular disease, planned for surgery, were collected 399 of which were included in the study. Malignancy risk was calculated for each patient and compared to postoperative pathology report. Statistical analysis showed a low sensitivity and a positive predictive value of around 50%, the performance of the formula did not improve when we excluded microcarcinoma. Data did not support the efficacy of Birmingham formula as a reliable instrument in predicting malignancy in thyroid nodules

Performance of chloride/phosphate test in patients with primary hyperparathyroidism. Is it related to calcium level?

To examine the sensitivity of the chloride/phosphate [Cl/PO[4]] ratio with a cut-off point of >33 as a diagnostic test for primary hyperparathyroidism [pHPT] in surgically proven patients, and its performance at different calcium levels. This is a retrospective medical records based study. Data of 120 patients diagnosed with pHPT, already operated in the Department of Surgery, Cisanello Hospital, Pisa, Italy between March 2010 and June 2011 were reviewed. They were divided into 4 subgroups according to their calcium levels. The Cl/PO[4] ratio was measured for each patient, with a cut-off point of 33, sensitivity of Cl/PO[4] test was measured. Test sensitivity was calculated for each subgroup, and a correlation with the parathyroid hormone [PTH] level was investigated. Performance of the equation was tested for the normocalcemic patients with a suitable control group. The sensitivity of Cl/PO[4] ratio for the whole group was 0.883 [0.809-0.932]. The sensitivity was 0.9340 [0.857-0.973] for patients with serum calcium above normal levels. A similar result of 0.933 [0.830-0.978] was demonstrated for the subgroup with hypercalcemia <1 [mg/dL] above normal level. Normocalcemic patients constituted 24%; for this subgroup, the sensitivity test was 0.724 [0.562-0.887], specificity was 0.763 [0.628-0.898], positive predictive value was 0.700 [0.536-0.864], and negative predictive value was 0.784 [0.651-0.916]. No correlation was identified between the performance of formula and serum PTH level. The Cl/PO[4] test seems to be a good tool to anticipate pHPT and showed a fair performance in normocalcemic patients