Value of laboratory tests in the differential diagnosis of hypercalcemia

Abstract

Eighteen laboratory tests were compared in the differentiation of primary hyperparathyroidism from hypercalcemia associated with malignancy. Statistical comparisons of the test results were carried out in four patient groups and two control groups. The patient groups evaluated were those with confirmed primary hyperparathyroidism, those with malignancy with hypercalcemia, those with malignancy without hypercalcemia, and those with surgically cured primary hyperparathyroidism. These groups allowed determination of the relative diagnostic values of the tests and a rationale for their value. After exclusion of patients with renal failure from the patient and control groups, these data indicated that the laboratory tests with the greatest differential diagnostic value, in order of efficacy, were: albumin, carboxy-terminal parathyroid hormone, venous pH, cholesterol, chloride, alkaline phosphatase, phosphorus, and the chloride/phosphate ratio. Hemoglobin, hematocrit, and red blood cell count also had some value, particularly in male patients. However, none of these tests individually achieved better than an 81 percent classification accuracy. With application of logistic discriminant analysis, only three tests--albumin, parathyroid hormone, and chloride--were identified as statistically significant in jointly improving the diagnostic separation between these two patient groups. Although the 94.4 percent classification accuracy achieved by use of these three variables in a logistic discriminant function was better than that obtained with any individual variable, incorrect classification was still a significant problem, particularly in the case of patients with malignancy and high concentrations of parathyroid hormone. With the exception of albumin and chloride measurements, the commonly available ancillary laboratory tests proposed to aid this differential diagnosis do not give any more information than the analysis of parathyroid hormone alone and merely add to the increased cost of medical care.