Short Communication: Hyperthyroidism in Human Immunodeficiency Virus Patients on Combined Antiretroviral Therapy: Case Series and Literature Review

With the increasing use of combined antiretroviral therapy (cART) for human immunodeficiency virus (HIV) infection, the incidence of immune reconstitution syndromes has also increased. ¹ Immune reconstitution syndromes are classically associated with the unmasking or paradoxical worsening of concomitant infectious diseases. In general, a large increase in absolute CD4⁺ T-cell count leads to enhanced antigen-specific immune responses and inflammatory mediator production, which sometimes results in autoantibody formation and aberrant tissue infiltration by activated lymphocytes. When this response becomes pathological and is directed toward infectious agents, it is termed immune reconstitution inflammatory syndrome (IRIS). However, it can also present as other autoimmune phenomena—termed immune restoration disease (IRD)—involving a wide range of organ systems. Autoimmune thyroid disease (AITD) is a recently recognized potential complication of cART that may manifest months after initiation of ART. ² Here we describe a case of immune restorative hyperthyroidism 2 years after initiation of cART and further examine our cohort of similar cases.

A 38-year-old black male with untreated HIV infection presented for initiation of care 7 years after initial diagnosis of HIV. At baseline, laboratory testing was notable for CD4 count, 3 cells/μl (1% CD4 cells); HIV viral load, 391,000 copies/ml; and thyroid-stimulating hormone (TSH) level, 1.7 mIU/liter (normal 0.5–5). He was started on a fixed-dose combination of efavirenz, emtricitabine, and tenofovir. His CD4 counts rose steadily at 1-, 6-, 12-, and 18-month follow-ups.

Twenty-three months after initiating cART, he developed heat intolerance, palpitations, tremors, and breast tenderness. Physical examination demonstrated a diffusely enlarged thyroid without nodularity, bilateral gynecomastia, and a fine resting tremor. His CD4 count was 486 cells/μl (24% CD4 cells) and his HIV viral load was undetectable (<40 copies/ml). Thyroid function testing revealed TSH, 0.016 mIU/liter; free thyroxine (fT4), 3.87 ng/dl (normal 0.6–1.6); and triiodothyronine (T3), 425 ng/dl (normal 87–178). There was no evidence of pituitary dysfunction. His symptoms were managed acutely with a long-acting beta-blocker, and he was treated with methimazole 20 mg daily; his cART was continued. After 6 months of this therapy his clinical symptoms had resolved and his serum fT4 had normalized; his viral load remained undetectable and his CD4 count was 745 cells/μl (30% CD4 cells).

We present four additional cases encountered in 1 year at our clinic, all without a previous history or risk factors for thyroid disease, who developed signs and symptoms of hyperthyroidism after initiation of cART (Table 1). Similar to the case patient, at the time of cART initiation, these patients had low baseline CD4 counts (median baseline CD4 of 6.5 [range 3–40] cells/μl) and high viral loads (median 4.19 [range 3.36–5.38] log₁₀ copies/ml), and all had been on cART for several years (median interval of 38.5 [range 23–93] months) before the development of hyperthyroidism. At the time of diagnosis of hyperthyroidism, all of the patients had immunologic recovery (median CD4 count was 486 [range 170–578] cells/μl) and had achieved virologic suppression (≥2 log₁₀ copies/ml decline in their viral load). The clinical diagnosis of primary hyperthyroidism was supported by thyroid function testing (suppressed TSH and elevated fT4), with a median TSH level at the time of diagnosis of 0.026 mIU/liter (0.016–0.063), and confirmed with diffuse I₁₂₃ uptake in radioactive iodine uptake testing and/or the presence of thyrotropin receptor antibodies. After symptomatic therapy, four of the five patients underwent radioiodine ablation with complete resolution of hyperthyroidism after the procedure.

There has been increasing recognition of a broad spectrum of disorders associated with chronic inflammation, immune activation, and immune system recovery in HIV-infected individuals. This immune dysregulation is thought to underlie many of the noninfectious end-organ complications of chronic HIV infection, such as cardiovascular disease. ³ IRIS typically occurs in the context of cART initiation, virologic suppression, and recovery of immune function, and usually manifests as a pathological response directed toward pre-existing infectious agents or microbial antigens. However, these exuberant inflammatory responses may also target autoantigens—these IRDs can manifest as a variety of clinical syndromes whose pathophysiology is heterogeneous and poorly understood.

Thyroid conditions are one form of IRD, and may lead to both hypothyroidism and hyperthyroidism. ⁴ In Graves' disease, the majority of cases are idiopathic, but immune restoration after cART has been identified as a cause, now called Graves' immune restoration disease (G-IRD). ⁵ The pathogenesis is thought to be similar to that of Graves' disease, with autoantibodies binding to and activating the TSH receptor. Clinical and laboratory diagnostic criteria for G-IRD are unchanged compared with Graves' disease. To date, no clear risk factors for G-IRD have been identified. Classic Graves' disease more commonly affects women than men, and it is presumed that G-IRD has the same predilection. Whether or not certain ART medications are more likely to result in immune dysregulation is unclear, but general consensus is that G-IRD is due to overall physiology of immune restoration rather than a certain brand or class of medication. However, we believe more studies are needed, such as a prospective longitudinal study with systematic screening of thyroid disease in HIV patients before initiation of cART, particularly with newer agents, to ascertain the predictors of G-IRDs.

Our series of patients with G-IRD indicates the need for heightened awareness of AITD as a potential long-term complication of cART. Prevalence estimates for AITD among HIV-infected patients are difficult to establish because there are no evidence-based guidelines for routine screening of thyroid dysfunction in this population. In several recent cross-sectional studies, abnormalities in thyroid function tests were present in up to 76% of HIV-infected individuals, ⁶ but there was a substantially lower prevalence (<5%) of AITD, as defined by the presence of thyroid autoantibodies. ^{6 –8} One recent analysis in the United Kingdom reported 11 cases of Graves' disease out of a cohort of 1,186 HIV-infected patients. ⁹ Notably, in our series the median time interval between initiation of cART and the diagnosis of G-IRD was 38.5 months, substantially greater than the interval described in the first reported cases of G-IRD (mean of 22 months). ¹⁰ In a more recent report, two patients developed G-IRD after being on cART for 12 and 28 months. ¹¹ Similar to previous cases, ^2,5,12 in our series most of the patients with G-IRD had advanced immunosuppression (low CD4⁺ T-cell count) before the initiation of cART, and experienced rapid recovery of CD4⁺ T-cell counts and virologic suppression well before the detection of G-IRD.

Recovery of the immune system after cART initiation occurs in a biphasic pattern: an initial phase involving recovery of the CD4⁺ and CD8⁺ memory T-cell populations that plateaus within the first month after the start of cART, followed by a later phase with continuous slow production of naive T cells. ^13,14 IRIS, which usually occurs within 6 months after initiation of cART, is thought to be associated primarily with CD45RO⁺ (memory) T-cell recovery, resulting in an exaggerated response of a recovering immune system to a foreign antigen. In G-IRD, there is a relative dearth of inflammatory cytokines—instead the process appears to be mediated through the Th2 immune response through expansion of naive CD4⁺ cells. Patients with G-IRD have increased proportions of naive CD4⁺ T cells and nonthyroid-related autoantibody titers. ¹⁵ This putative delayed migration of naive CD4 cells could account for the later presentation of G-IRD when compared with IRIS.

In conclusion, clinicians should be aware of the wide range of noninfectious immune restoration phenomena that may occur after the initiation of cART, particularly among individuals who initiate therapy late in the course of HIV infection. Our series of patients with G-IRD also highlights the need for providers to remain vigilant for these conditions many years after cART initiation.