Thyrotoxic periodic paralysis (TPP) is a rare complication of hyperthyroidism TPP has been reported in male patients of Asian descent, as well as being reported in individuals with some degree of Native American genetic admixture (Native American Indians, Mexican Americans, Mexicans). Given the increased susceptibility for TPP in some ethnic groups, and considering the autoimmune basis of Graves’ disease, finding an immunogenetic marker that would identify the populations at risk seems attractive. The present report describes the clinical, biochemical and HLA-serological characteristics of a group of Mexican mestizo patients with TPP.

Methods

The diagnosis of hyperthyroid Graves’ disease was established between 1990 and 1997 by the presence of signs and symptoms of thyrotoxicosis, accompanied by a diffuse goiter, a diffusely increased radioiodine uptake (>40% at 24 h) and biochemical evidence of hyperthyroidism. TPP was diagnosed based on one or more acute episodes of flaccid paralysis accompanied by a low serum K⁺ occurring in the context of biochemical and clinical evidence of hyperthyroidism. The ethnic background of the patients was sought going back three generations, and a family history of similar episodes was carefully investigated. Individuals taking corticosteroids or diuretics were excluded from the study.

Eight TPP cases were available for serologic HLA typing. As a control population, 10 Graves’ disease patients were HLA-serotyped without TPP. The results were compared to a large HLA database of 100 healthy Mexican mestizo adults.

TSH, total and free T4 and T3 were measured by commercially available RIA and IRMA assays. HLA-A, B, DR and DQ typing was performed in glass-bead defibrinized venous blood, using the two-step microcytotoxicity method on precharged plates that included 12 HLA-A, 21 HLA-B, 12 HLA-DR and 3 HLA-DQ specificities (Pel-Freeze, Rogers, AK, USA). HLA-A and HLA-B antigens were determined in peripheral blood lymphocytes separated by Ficoll-Hypaque gradient (Sigma Chemical Corp., St. Louis, MO, USA). HLA-DR and DQ typing was carried out on B lymphocytes obtained by either the lympho kwik B (One Lambda, Los Angeles, CA, USA), or the nylon wool column methods.

Data analysis was carried out employing the ?² test using 2 × 2 contingency tables, and Yates’ correction for continuity. P values were obtained by Fisher’s exact P test and were corrected (PC), multiplying by the number of specificities studied.

Results

Fourteen patients with TPP were identified. Ethnically, they were all Mexican mestizos. Their mean age at presentation was 27.8 ± 7.3 years (range 21–50 years), and all were male. Hyperthyroidism was documented in all cases as described in Methods. In six cases, acute episodes of hypokalemic paralysis had been occurring for an average of 4.8 months before the diagnosis of hyperthyroidism was established. Five patients had been diagnosed with hyperthyroidism, a mean of 26 months before the development of paralysis. In three cases, hyperthyroidism and hypokalemic paralysis were diagnosed simultaneously. Of the 14 cases, 9 presented with quadriplegia, the remaining had lower limb paraplegia. Only one patient had respiratory muscle involvement and required mechanical ventilation. All patients had hypokalemia at presentation, the mean level was 2.2 ± 0.4 mEq/l (range 1.3–3.3 mEq/l). Neither the severity nor the number of episodes of hypokalemic paralysis correlated with the potassium level or the degree of thyrotoxicosis. All patients were initially treated with KCl replacement with improvement of the neuromuscular symptoms. Paralysis episodes did not recur once patients were rendered euthyroid by means of radioactive iodine and/or methymazol treatment (mean follow-up of 1 year).

Eight patients with Graves’ disease and TPP were available for HLA testing, and were compared to a group of 10 Graves’ patients without paralysis. Graves’ patients without TPP had an increased frequency of HLA-DR3 compared to the general Mexican mestizo population (60 vs. 9.4%, RR 6.4, PC <0.01). In contrast, patients with TPP had a frequency of HLA-DR3 comparable to the general Mexican mestizo population (25 vs. 9.4%, NS). No associations with other class I or II antigens were found in the Graves’ patients in this study, regardless of whether or not they had TPP.