J Graves’ Disease: An Analysis of Thyroid Hormone Levels and Hyperthyroid Signs and Symptoms PAULAT. TRZEPACZ, M.D. mt~hrgh, Pennsy/vaniaId~~~ KLEIN, M.D. Manhasset, New ~~~MICHELLE ROBERTS, M.D., JOELGREENHOUSE, Ph.D., GERALD S. LEVEY, M.D. Pittsburgh. Pennsy/vania PURPOSE: Assessmentof disease severity for pa- tients with hyperthyroidism involves clinical evalu- ation and laboratory testing. To determine if there is a correlation between symptoms and thyroid function test results, we prospectively studied hy- perthyroid patients using a standardized symptom rating scale and serum thyroid function parame- ters. PATIENTS AND METElODS We examined 25 patients with untreated, newly diagnosed Graves’ disease using the Hyperthyroid Symptom Scale (HSS) and serum levels of thyroxine (TA), triiodothyronine (Ts) relative insulin area (RIA), and estimates of free thyroxine index (FTI). In addition, we com- pared thyroid hormone levels with standard mea- sures of depression and anxiety in these patients. RJEWLTS: When regression analyses controlling for age were performed, none of these symptom ratings were associated with FTI or T3 RIA. The HSS was correlated with goiter size and anxiety ratings and was inversely correlated with age. CONCLUSION: The present study suggeststhat there is no relationship between the clinical assess- ment of disease severity and serum levels of thy- roid hormone in untreated Graves’ disease. From the Western Psychiatric Institute and Clinic, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania (PlT): Department of Medi- cine, Division of Endocrinology and Metabolism, North Shore University Hospital, Cornell University Medical College, Manhasset, New York (IK); Department of Medicine, Division of Endocrinology and Metabolism, Uni- versity of Pittsburgh School of Medicine., Pittsburgh, Pennsylvania (MR. GSL); and Department of Statistics, Carnegie-Mellon University, Pitts- burgh, Pennsylvania (JG). Research for this work was supported by First Award R29 MH44964-01 (PlT) from the National Institute of Mental Health, and National Institutes of Health General Clinical Research Center Grant 2 MO1 at the Presbvterian Untversitv Hosoital. Universitv of Pitts- burgh School of Medicine.. Requests for reprints’should be addressedto Gerald S. Levey, M.D.. Department of Medicine, University of Pittsburgh School of Medicine, 1218 Scaife Hall, Pittsburgh, Pennsylvania 15213. Manuscript submitted February 16. 1989. and accepted in revised form July21. 1989. T he diagnosis of hyperthyroidism is suspected in the setting of thyrotoxic symptoms and confirmed by standard thyroid function tests [l]. There is a wide variation in the magnitude of symptoms as well as in the level of thyroid hormones. Thyroid function test result8 may be markedly abnormal when the patient is only minimally symptomatic, or conversely, may be modestly elevated at the same time that the patient exhibits classic symptoms [l-4]. Thus, it ha8 been sug- gested that the intensity of thyrotoxic symptoms does not correlate with the degree of elevation of thyroid function tests [2]. This discrepancy may arise if serum levels do not reflect intracellular thyroid hormone concentration8 [2], if there are cellular variations in nuclear thyroid hormone receptor sensitivity [3], or if some hyperthyroid symptoms are adrenergically me- diated [5,6]. In addition, previous studies have not specifically addressed this issue using standardized symptom rating scalesto assess and quantify the se- verity of symptoms. To further clarify this clinically important question, we prospectively studied 25 hyperthyroid patients us- ing a previously described standardized symptom rat- ing scale[7] and serum thyroid function tests to inves- tigate the relationship between symptom severity and the degree of hormone elevation. We have also com- pared the severity of psychiatric symptoms with these parameters. PATIENTS AND METHODS Twenty-five subjects with untreated, newly diag- nosed Graves’ disease were consecutively recruited from all servicesof the University of Pittsburgh-affili- ated hospitals. All gave written informed consent for this project, which had been approved by the Institu- tional Review Board. All were hospitalized for one day at the Clinical Research Unit. Subjects with other sig- nificant medical diseases, pregnancy, substance abuse, or a preexisting major psychiatric disorder were ex- cluded. None of the women were being treated with either oral contraceptives or estrogen. The duration of hyperthyroid symptoms anamnestically reported by the subjects ranged from six to 18 months. None of the subjects were taking o-blocker or antithyroid medica- tions. Procedures A complete medical history and physical examina- tion were performed. Clinical estimates of goiter size in grams were independently made by two endocrinol- ogists. Electrocardiography was used to determine heart rate. Serum thyroid function tests included the total serum thyroxine (TJ, triiodothyronine (Ts) resin uptake, the calculation of the free thyroxine index (FTI), and total serum TZ by radioimmunoassay (RIA). Normal ranges for FTI and TS RIA are 5.0 to 558 November 1999 The American Journal of Medicine Volume 87