Severe hypothyroidism as a complication of anticonvulsant treatment.

Abstract

The clinical relevance of the interaction of enzyme-inducing anticonvulsants (AEDs) with thyroxine-replacement therapy has sometimes been underestimated. We report here a case of severe hypothyroidism precipitated by AEDs in the intensive care setting (ICU). A 72-year-old woman with arterial hypertension and primary hypothyroidism due to Hashimoto thyroiditis and stabilized on 100-μg levothyroxine-replacement therapy was admitted to our neurological ICU for an acute ischemic stroke, complicated by epileptic convulsions. Phenytoin (PHT) was given intravenously for 4 days (250-mg bolus, followed by an infusion of 1,000 mg/24 h on the first day, and then tapered off: 375 mg on day 2, 250 mg on day 3, and gradually discontinued on day 4), and was substituted thereafter with carbamazepine (CBZ), 400 mg orally on day 4 and 800 mg daily there after. On day 3, sodium valproate (VPA) was added and increased gradually from 500 to 1,000 mg daily until day 23, when it was discontinued. Convulsions were rapidly suppressed by the PHT bolus, and did not recur. The patient's level of consciousness, however, worsened progressively (day 1: somnolent but oriented and communicating in short sentences; day 6: single words; day 18: no verbal response, just eye opening to loud voice). Consistent with the computed tomography (CT) scan finding of large ischemic lesions in the right hemisphere, left-sided hemiplegia occurred with left facial nerve palsy. Although circulatory and respiratory parameters were initially stable, the patient died on day 30 of septicemia secondary to a urinary tract infection. On admission, her thyroid function tests were normal: thyrotrophin (TSH), 1.07 mIU/L, total thyroxine (tT4) 104.6 nM, and total triiodothyronine (tT3), 1.39 nM. On day 14, however, despite continuation of 100 μg levothyroxine daily, her TSH increased to 43.5 mIU/L, whereas free thyroxine (fT4) decreased to 6.29 pM, and free triiodothyronine (fT3) decreased to undetectable values (<1.0 pM), suggesting severe primary hypothyroidism. Although on day 18, levothyroxine was increased to 200 μg daily, on day 20, TSH increased to 91.4 mIU/L, fT4 was 3.73 pM, and fT3 was undetectable. Clinical signs of hypothyroidism included lethargy, constipation, and eyelid edema and were possibly confounded by the ischemic brain damage, infection, and drugs. No bradyarrhythmias or hypothermia was found. However, during progression of her infectious complication in the last 2 weeks, her heart rate was slower (80 to 90 beats/min), and her temperature did not reach 38.0°C. A decrease in thyroid hormone levels caused by PHT was first described in 1961 (1). Further studies have shown complex interactions between some AEDs, including PHT and/or CBZ, and thyroid hormones at multiple levels (2-4). These interactions may include decreased plasma protein binding (4) and, more important, increased metabolism of thyroid hormones (3,5). Patients with intact pituitary/thyroid axis respond to decreased availability of thyroid hormones by feedback increase in their production and stay euthyroid (6). In primary hypothyroidism, however, induction of metabolism of exogenous T4 cannot be compensated for by an increased synthesis of the hormone, and the patient becomes hypothyroid (7, 8). In these situations, an increase in replacement T4 dose is necessary. At times, even doubling the dose might not be sufficient, even though in hypothyroid patients, it may take ≥1 month to normalize the thyroid levels after a dose change, and our patient did not survive for a period sufficient to assess fully the effects of the change in hormone-replacement therapy. When enzyme-inducing AEDs are administered, replacement hormone therapy should be carefully adjusted, based on repeated measurement of TSH, which is the most sensitive (and most rapidly changing) marker of thyroid status. Alternatively, an AED without enzyme-inducing capacity, such as VPA (9), may be used. In the intensive care setting, as in our case, the evaluation could further be obscured by adaptive changes in the pituitary/thyroid axis, leading to decreased TSH, T4, and T3 in critically ill patients (10). Although suppressed conversion of T4 to T3 by this mechanism might explain the massive decrease of fT3 to undetectable levels in our patient, the fast elevation of TSH clearly documented severe hypothyroidism precipitated by PHT and CBZ.