The Effect of Glucocorticoid Therapy on a Falsely Raised Thyrotropin Due to Heterophilic Antibodies

Dear Editor:

Glucocorticoids can suppress the level of thyrotropin (1). Here, we present a patient in whom glucocorticoid therapy revealed the presence of a falsely raised thyrotropin (TSH) level.

A 41-year-old female patient presented with periorbital edema and infected conjunctivae. She complained of gritty eyes and excess tearing. She had started taking propranolol, 20 mg three times a day, for migraine headaches 6 weeks before. Her mother and sister were both being treated for hypothyroidism. Her vital signs were normal, with a regular pulse of 88 beats per minute. She had a firm thyroid gland with moderate thyromegaly, but had no symptoms or signs of thyroid dysfunction. Serum thyroid hormone levels were normal (free triiodothyronine [fT3], 5.5 pmol/L, normal 3.7–6.5; free thyroxine [fT4], 19 pmol/L, normal 10.3–21.9), but her TSH level, measured by an automated chemiluminescence system, was elevated (13.9 mIU/L; normal 0.34–4.25). Levels of anti–thyroid peroxidase and antithyroglobulin antibodies (TPOAb and TgAb, respectively) were both elevated (48.6 IU/mL, normal 1–16, and 271.4 IU/mL, normal 5–100, respectively). Sonography revealed a moderately enlarged thyroid gland with decreased echogenicity. On axial and coronal computed tomography imaging of the orbits, there was swelling of extraocular muscles. She was thought to have mild Hashimoto's hypothyroidism with the rare association of thyroid ophthalmopathy. Prednisone, 75 mg daily, was prescribed for her ophthalmopathy. Six weeks later, her eye symptoms and signs had improved significantly. Her fT4, although slightly higher than before, was still in the normal range (fT4, 20.5 pmol/mL; fT3, 3.9 pmol/mL). However, her TSH level had dropped to a subnormal value (0.12 mU/L). A test for thyroid receptor antibodies (TRAb) was positive (13.18 IU/L; normal 0–1), and TSH did not increase with thyrotropin-releasing hormone (TRH) stimulation (baseline, 0.15 mIU/L; 30 and 60 min levels after 400 μg TRH IV, 0.11 and 0.14 mIU/L respectively). Therefore, Graves' disease was suspected. Prednisone was tapered and finally stopped over a 3 month period. Her TSH level had now risen again to 8.90 mU/L, and it could not be increased any further with TRH stimulation (baseline, 9.09 mIU/L; 30 and 60 min levels after TRH, 8.72 and 9.06 mU/L respectively). Moreover, treatment including T3 suppression with a total dose of 100 μg daily for 10 days did not suppress the TSH (8.02 and 8.09 mU/L). A pituitary magnetic resonance image was normal. The presence of human anti-mouse antibodies (HAMA) was therefore suspected, and the patient's serum tested positive (220 ng/mL; normal <40 ng/mL). With the use of HAMA-blocking reagents, a TSH level of 0.006 mIU/L was obtained. As part of the reevaluation, a retrospective analysis of the serum obtained at initial presentation disclosed a HAMA concentration of 344 ng/mL with a TSH level of 0.005 mU/L after the application of HAMA-blocking reagents. In the same sample, the TSH level measured by another laboratory, using a different assay, was also very low (0.004 mU/L; normal 0.4–4.5). Notably, the HAMA reactivity was shown to be <40 ng/mL in the serum saved after the first 6 weeks of prednisone therapy. When propranolol had to be withdrawn because of a developing asthma, the patient experienced nervousness and irritability, her pulse rose to 120 beats per minute, and she had tremors of her hands. Treatment with methimazole was required to relieve her symptoms and signs of thyrotoxicosis.

In hindsight, this patient was in fact hyperthyroid at the initial presentation. She presented with a falsely raised TSH level and normal peripheral thyroid hormone levels. With prednisone therapy, the TSH was found to be suppressed but became again elevated after the discontinuation of the glucocorticoid therapy. This prompted the measurement of HAMA levels, which were positive. The literature has so far been short of information regarding the effect of glucocorticoids on falsely raised TSH levels due to heterophilic antibodies. Our patient indicates that a spuriously elevated thyrotropin due to HAMA (2 –4) has been revealed after glucocorticoid therapy that led to a reduction of the heterophilic antibodies. At the initial presentation, propranolol had masked the symptoms and signs of hyperthyroidism, and the heterophile antibody had falsely raised the TSH level into a range suggestive for hypothyroidism and, together with positive TPOAb and TgAb, leading to the erroneous diagnosis of Hashimoto's hypothyroidism instead of Graves' disease.