OBJECTIVE Early, safe postoperative hospital discharge has become a priority in the current environment of cost containment and managed care. One determinant for this following operations of the thyroid or parathyroid glands is a stable postoperative calcium level. This study sought to determine whether early postoperative calcium levels could safely predict hypocalcemia following operations of the thyroid or parathyroid glands. METHODS A retrospective chart review of 197 patients undergoing such operations was undertaken. METHODS The study population was divided into patients undergoing parathyroid exploration with removal of one or more glands for parathyroid disease ("parathyroid" group) and patients undergoing bilateral thyroid operations ("nonparathyroid" group). Postoperative calcium levels were plotted as a function of time, and the slope between the first two postoperative calcium levels examined. RESULTS A positive slope predicted normocalcemia in 100% of "nonparathyroid" and 90% of "parathyroid" surgeries. A negative slope was predictive in its magnitude. The nonparathyroid patients who developed postoperative hypocalcemia had an average slope significantly more negative (-0.84% change/h) than patients remaining normocalcemic (-0.49% change/h) (P = .03). Similarly, the parathyroid patients who became hypocalcemic had an average initial slope nearly twice as negative (-1.4% change/h) as patients remaining normocalcemic (-0.78% change/h) (P = .005). CONCLUSIONS These results indicate that an initial upsloping postoperative calcium curve based on two early postoperative calcium measurements is strongly predictive of a stable postoperative calcium level, and a steeply downsloping initial calcium curve is worrisome for eventual hypocalcemia.