Abstract
Background
People living with human immunodeficiency virus (HIV) may develop endocrine dysfunction, particularly involving the thyroid gland. Although overt thyroid dysfunction in patients living with HIV is similar to that of the general population, subclinical thyroid dysfunction is significantly more prevalent. This study aims to evaluate the prevalence of changes in thyroid function in patients living with HIV on antiretroviral therapy and to determine whether routine thyroid function testing is warranted.
Methods
Retrospective observational study including 249 people living with HIV on antiretroviral therapy and followed in an Internal Medicine outpatient clinic between 2018 and 2023. Patients with a history of thyroid disease before 2018, those receiving treatment with amiodarone, lithium, interferon, and/or tyrosine kinase inhibitors, and those lacking thyroid function assessment or medical records before 2023 were excluded.
Results
Of the 249 patients, 96.4% showed no abnormalities in thyroid function. Thyroid dysfunction was identified in nine patients (3.6%): four (1.6%) had subclinical hypothyroidism, one (0.4%) Hashimoto's thyroiditis, two (0.8%) antibody-negative hypothyroidism, and two (0.8%) subclinical hyperthyroidism.
Conclusion
The overall prevalence of thyroid dysfunction was low, suggesting no benefit in systematic screening of thyroid function among patients living with HIV on antiretroviral therapy.
Introduction
Antiretroviral therapy has changed the clinical evolution of human immunodeficiency virus (HIV) infection, having significantly reduced the morbidity and mortality associated with this disease.1,2
Antiretroviral therapy has been associated with several metabolic adverse effects, including dyslipidemia, insulin resistance, arterial hypertension, lipodystrophy, and decreased bone mineral density. Although the pathophysiological mechanisms underlying these alterations are not completely understood, the pathogenesis seems to be due to complex interactions between effects of chronic HIV infection, antiretroviral therapy, and patient factors, including genetic susceptibility.3–5 Furthermore, individuals infected with HIV may exhibit endocrine system dysfunction, ranging from mild, asymptomatic changes in laboratory values to glandular failure. Regarding thyroid function, the described prevalence of overt thyroid dysfunction in people living with HIV is similar to that of the general population, ranging from 1% to 3%. However, subclinical thyroid dysfunction has been described as significantly more prevalent in people living with HIV.6–16 Several antiretroviral drug classes have been linked to different patterns of thyroid dysfunction, although mechanisms are not completely understood, and evidence remains limited. Protease inhibitors (PIs) regimens, particularly those boosted with ritonavir, have been associated mainly with subclinical or overt hypothyroidism through increased thyroid-stimulating hormone (TSH) levels and interactions with levothyroxine metabolism, whereas non-nucleoside reverse transcriptase inhibitors (NNRTIs), especially efavirenz, have been implicated in hyperthyroidism, and older nucleoside analogs such as stavudine have historically been associated with hypothyroidism.6,8,9,17 Nevertheless, despite the improved tolerability of antiretroviral therapy, thyroid dysfunction remains a relevant clinical issue in people living with HIV, implying a multifactorial etiology.
Diagnosing thyroid dysfunction in people living with HIV can be challenging, as clinical manifestations such as fatigue, weight changes, and mood disturbances are often nonspecific and may overlap symptoms of HIV infection, antiretroviral therapy side effects, or other comorbid conditions, potentially leading to delayed diagnosis and treatment. Importantly, untreated thyroid dysfunction is associated with a broad spectrum of adverse outcomes, including increased cardiovascular risk, metabolic disturbances, neurological complications such as peripheral neuropathy and cognitive impairment, and reduced quality of life. Given that people living with HIV have already a higher risk of cardiovascular and metabolic complications, the presence of thyroid dysfunction represents a modifiable risk factor that warrants early detection and management.8,17–19
Currently, available data are controversial and there is no consensus on the need to monitor changes in thyroid function in people living with HIV on antiretroviral therapy. Some studies report an increased prevalence of subclinical hypothyroidism and other thyroid abnormalities, particularly in patients on antiretroviral therapy and suggest routine monitoring due to potential health risks.8,9,12,13,16 In contrast, other authors advocate that thyroid function testing should be targeted, given that thyroid dysfunction appears similar to the general population and that routine screening may not provide clinical or economic benefit.7,15,20,21
This study aims to evaluate the prevalence of changes in thyroid function in people living with HIV on antiretroviral therapy followed in an Internal Medicine outpatient clinic from 2018 to 2023 and to determine whether routine thyroid function testing is warranted in this population.
Materials and Methods
The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for observational studies, 22 and the STROBE reporting checklist 23 is included in supplement A.
Study Population
Retrospective observational study including patients living with HIV on antiretroviral therapy followed in an Internal Medicine outpatient clinic between 2018 and 2023. During the study period, 266 patients were actively followed at this center. Although the study focused on evaluating thyroid function in the context of antiretroviral therapy, a small number of patients were not receiving antiretroviral therapy at the time of the most recent evaluation. These individuals had documented prior exposure to antiretroviral treatment but were off therapy due to clinical or adherence-related reasons. They were included in the analysis because previous antiretroviral exposure may have persistent effects on thyroid function and since they reflect real-world clinical scenarios. Thyroid function testing was not performed systematically and there was no standardized protocol guiding its request. Thyroid function assessment was performed at the discretion of the treating physician and was not consistently driven by thyroid-related symptoms.
Patients with a history of thyroid disease before 2018, those receiving treatment with amiodarone, lithium, interferon, and/or tyrosine kinase inhibitors, and those lacking thyroid function assessment or complete medical records before 2023 were excluded. After applying the exclusion criteria, data from 249 individuals were analyzed.
Clinical Assessment
Disease duration was defined as the duration of HIV infection since the first positive HIV test result. CD4 cell count was determined by flow cytometry and HIV-RNA viral load by reverse transcriptase polymerase chain reaction, with a minimum detection limit of 50 copies/mL for HIV type 1 and of 40 copies/mL for HIV type 2. The normal serum levels of TSH, free thyroxine (FT4), and free tri-iodothyronine (FT3) were defined according to the reference values of the central laboratory of the hospital where the patients were followed. The reference values for TSH were 0.35-4.94 mIU/mL between 2018 and 2022 and 0.27-4.20 mIU/mL in 2023. The reference values for FT4 were 0.70-1.48 ng/dL between 2018 and 2022 and 0.93-1.70 ng/dL in 2023, and for FT3 were 1.88-3.18 pg/mL between 2018 and 2022 and 2.00-4.40 pg/mL in 2023. Antithyroid antibody negativity corresponded to an antithyroglobulin antibodies (TgAb) of less than 4.11 IU/mL between 2018 and 2022 or less than 115 IU/mL in 2023, antithyroid peroxidase antibodies (TPOAb) of less than 5.61 IU/mL between 2018 and 2022 or less than 34 IU/L in 2023, and anti-TSH receptor antibodies (TRAbs) of less than 1.0 U/L between 2018 and 2022 or less than 1.22 U/L in 2023.
The presence of thyroid disease was defined based on clinical history, laboratory parameters, and use of thyroid medications. Euthyroidism was defined by normal TSH and FT4 levels. Subclinical hypothyroidism was defined by the presence of TSH greater than 4.94 mIU/mL (2018-2022) or 4.20 mIU/mL (2023) and normal FT4. Clinical hypothyroidism was defined by the presence of TSH greater than 4.94 mIU/mL (2018-2022) or 4.20 mIU/mL (2023) and FT4 lower than 0.70 ng/dL (2018-2022) or 0.93 ng/dL (2023). Subclinical hyperthyroidism was defined by the presence of TSH lower than 0.35 mIU/mL (2018-2022) or 0.27 mIU/mL (2023) and normal FT4. Clinical hyperthyroidism was defined by the presence of TSH lower than 0.35 mIU/mL (2018-2022) or 0.27 mIU/mL (2023) and FT4 higher than 1.48 ng/dL (2018-2022) or 1.70 ng/dL (2023). Graves’ disease was considered by the coexistence of hyperthyroidism and positive TRAbs; hypothyroidism due to Hashimoto's thyroiditis was considered if hypothyroidism and positivity for TgAb and/or TPOAb were present.
Statistical Analysis
Data analysis was performed using IBM SPSS Statistic 28.0.1.0 software (IBM SPSS Inc., NY, USA). Continuous and categorical variables are presented as median (interquartile range) and number and percent number (%), respectively.
Data Source
Demographic, clinical, and laboratory data were collected retrospectively between December 2024 and January 2025 from institutional clinical records. Data were fully anonymized prior to analysis, and the authors had no access to information that could identify individual participants during or after data collection. Data will be securely stored for at least 10 years in accordance with institutional policies.
Results
The demographic and clinical characteristics of patients are shown in
Demographic and Clinical Characteristics of Patients.
HIV, human immunodeficiency virus; INI, integrase inhibitor; NNRTI, non-nucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor.
In 2024, 98.4% of patients were on antiretroviral therapy, with a median treatment duration of 15 years. The antiretroviral regimens included nucleoside reverse transcriptase inhibitors (NRTIs) in 84.5%, integrase inhibitors (INIs) in 65.7%, NNRTIs in 32.7%, and PIs in 18.8%. Among patients undergoing antiretroviral therapy, 49.4% were receiving a combination of NRTI and INI, 21.2% of NNRTI and NRTI, 12.7% of NRTI and PI, 10.2% of NNRTI and INI, 4.5% of INI and PI, 1.2% of NRTI, PI and INI, and 0.8% of NRTI, NNRTI and INI.
Among the 249 patients considered, 96.4% did not present any change in thyroid function, between 2018 and 2023. Among the nine patients (3.6%) with thyroid function changes, four (1.6%) were diagnosed with subclinical hypothyroidism, one (0.4%) was diagnosed with hypothyroidism due to Hashimoto's thyroiditis, two (0.8%) patients were diagnosed with antibody-negative hypothyroidism, and two (0.8%) patients with subclinical hyperthyroidism.
As summarized in
Epidemiologic and Laboratory Features of Patients with Thyroid Function Alterations.
F, Female; HIV, human immunodeficiency virus; INI, integrase inhibitor; M, Male; NA, not available; NNRTI, non-nucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor; s, smoker.
Discussion
The reported prevalence of thyroid dysfunction in people living with HIV on antiretroviral therapy varies widely in the literature, depending on the population studied, the length of follow-up and the diagnostic criteria used.
Before the introduction of antiretroviral therapy, changes in thyroid function occurred due to infiltration of the gland by regional tumors like Kaposi sarcoma, infection of the gland with opportunistic infections (Pneumocystis jirovecii, Cryptococcus neoformans or cytomegalovirus), serious systemic disease or energy deprivation.7,16–19 Additionally, it is also reported as a side effect of the drugs used in the course of the HIV infection, including antiretroviral therapy, anti-tubercular and antifungal drugs, and from direct infection of the gland with HIV.7,15,24–26
The influence of antiretroviral therapy on thyroid function has been widely studied; however, its impact remains unclear.8,17 Several authors have observed that the use of stavudine, a NRTI, is associated with a greater risk of hypothyroidism,6,9,10–12,27 although its use has largely been discontinued in clinical practice due to toxicity concerns. Some studies have reported that hypothyroidism is more frequently observed in patients receiving PIs, including amprenavir, ritonavir and lopinavir,9,12 as well as in those treated with the NNRTI efavirenz. 9 In addition, Hatzl et al found that subclinical hypothyroidism was correlated with antiretroviral therapy duration and NRTI use. 16 Conversely, other studies have found no association between thyroid dysfunction and antiretroviral therapy, including stavudine.8,13,14,20,24,28 In our study, no patients were on stavudine therapy between 2018 and 2023, and thyroid dysfunction occurred across different antiretroviral regimens, with no consistent association with a specific drug class.
In our cohort, the prevalence of changes in thyroid function was 3.6%, which corroborates other studies, such as that of Harsløf et al, which analyzed 826 patients living with HIV on antiretroviral therapy, who presented a prevalence of thyroid dysfunction of 4.6%, and found that the prevalence of hyperthyroidism and hypothyroidism was similar to the general population. 28 Similarly, Madge et al, in a large-scale study with 1565 patients, showed a low prevalence of overt thyroid disease and found no association between thyroid alterations and people living with HIV on antiretroviral therapy, 20 as did Properzi et al, in a study of 6343 patients, reporting only 1.94% of symptomatic thyroid dysfunction. 24 Nonetheless, other studies have reported a significantly higher prevalence of thyroid dysfunction, ranging from 16% to 34.2%,9,25–27 and some studies have related this higher value to the use of stavudine.6,9,27 It is important to note that the low prevalence of thyroid functional abnormalities in our study may be justified by the relatively small sample size and by the fact that stavudine was not used among these patients.
The most frequent alteration in our study was subclinical hypothyroidism, present in 1.6% of patients. Subclinical hypothyroidism affects 3-10% of the population, with the highest prevalence among women and elderly individuals.19,29,30 Several authors have found that this is the most prevalent thyroid function alteration in people living with HIV, with a prevalence that reaches up to 14.4%.6,8–12,14–15,21,27 Bongiovanni et al observed that treatment-naïve patients starting antiretroviral therapy, exhibited a higher prevalence of subclinical hypothyroidism compared to those who had been on antiretroviral therapy, for at least 1 year prior to the study. 13 This can be related to a potential acute effect of therapy on thyroid function, mediated by immune mechanisms or other drug-related factors. However, it is also important to emphasize that aging is a key determinant that influences thyroid function and the interpretation of TSH levels. Large population-based studies have consistently shown a rise in TSH levels in healthy elderly people as a physiological consequence of aging.19,31 In this context, the finding that patients with subclinical hypothyroidism were relatively older in our cohort may partly reflect age-related changes in thyroid homeostasis rather than a direct effect of HIV infection or antiretroviral therapy.
Despite the high prevalence of subclinical hypothyroidism in people living with HIV on antiretroviral therapy, progression to clinical hypothyroidism is rare and, in most cases, does not require treatment.8,15,26 In accordance with guidelines for the general population, in patients under 65 years, replacement therapy with L-thyroxine is recommended only if TSH≥10 mIU/L, regardless of symptoms, and should be considered if there are symptoms suggestive of hypothyroidism and TSH<10 mIU/L. In patients over 65 years, replacement therapy should be considered if TSH≥10 mIU/L with clear symptoms of hypothyroidism or high vascular risk. 30 In our study, none of the patients with subclinical hypothyroidism had a TSH higher than 10 mIU/L, and there was no need to start treatment with levothyroxine.
In populations without iodine deficiency, clinical hypothyroidism has an estimated prevalence of 0.3-2%, 29 with this value being similar in people living with HIV.6,8–10,20 However, unlike the general population, most cases of clinical hypothyroidism are not associated with autoimmunity.6,7,9,20,26 Likewise, in the sample studied, we observed only one case of Hashimoto's thyroiditis and two cases of hypothyroidism without positivity for thyroid antibodies.
Innumerous studies have explored the association between thyroid dysfunction and HIV-related factors, particularly immune status. Lower CD4 cell counts have been associated with higher TSH levels, suggesting a link between hypothyroidism and HIV disease progression. Additionally, individuals with subclinical hypothyroidism have been reported to have significantly lower CD4 counts compared to those with normal thyroid function.5,6,8,9 A longitudinal study of 182 patients living with HIV on antiretroviral therapy, followed for 18 to 24 months, observed that TSH level correlated positively with the duration of antiretroviral therapy, and negatively with the CD4 count nadir. 25 The impact of HIV viral load on thyroid dysfunction has been less extensively investigated and most studies do not include detailed analyses of the influence of HIV-RNA levels on thyroid function. 8 While some reports suggest a higher prevalence of thyroid dysfunction in individuals with virological non-suppression, 8 other studies, including Ji et al, have found no significant association between HIV viral load and thyroid dysfunction. 25 In our cohort, interpretation of the association between thyroid dysfunction and HIV-related factors, namely CD4 cell count and HIV viral load was limited by the small number of patients with thyroid dysfunction and incomplete availability of CD4 cell count and HIV viral load data. Patients with overt hypothyroidism exhibited CD4 cell counts above 400 cells/μL both at the time of HIV diagnosis and during follow-up between 2018 and 2023. Among patients with subclinical hypothyroidism, two patients had low CD4 nadir values (27 and 36 cells/μL) and baseline CD4 cell count data were unavailable in other two cases; however, during follow-up, all patients maintained CD4 counts above 200 cells/μL. Most patients with thyroid dysfunction were virologically suppressed during follow-up between 2018 and 2023 and overall, thyroid dysfunction occurred mainly in the context of viral suppression. Additionally, the duration of HIV infection among patients with thyroid dysfunction varied widely, ranging from 4 to 29 years. Overall, no consistent pattern linking CD4 cell count, viral load or disease duration to thyroid dysfunction was observed.
Subclinical hyperthyroidism has been reported in people living with HIV, although with reduced prevalence.8,15,16,20,28 The progression of subclinical hyperthyroidism in people living with HIV is highly variable, leading to controversy regarding the benefits of treatment with thionamides. It may precede overt hyperthyroidism, remain stable over time or revert to normal thyroid functioning. However, given its potentially life-threatening cardiovascular risks, heightened monitoring of people living with HIV with diagnosis of hyperthyroidism is warranted.7,18
Clinical hyperthyroidism is uncommon in people living with HIV, and there does not appear to be a significant increase in its prevalence in patients on antiretroviral therapy compared to the general population.8,28 The main etiology is Graves’ disease and it may arise mainly in the context of immune reconstitution inflammatory syndrome after initiation of antiretroviral therapy, especially in patients who have very low CD4 cell counts before starting treatment.5,7,8,18,24,26,32
The need for screening for thyroid dysfunction in asymptomatic people living with HIV is controversial, as there is insufficient scientific evidence to prove its benefit. Although some authors recommend screening in all patients on antiretroviral therapy,8,9,12,13,16 others argue that this is unnecessary.7,15,20,21 However, some of these studies suggest that thyroid function analysis should be performed in patients with suggestive symptoms or risk factors (prolonged use of antiretroviral therapy, co-infection with hepatitis C virus and/or presence of thyroid autoantibodies).7,8 Beltran et al and Madeddu et al also recommend screening in patients undergoing antiretroviral therapy, especially if the treatment includes stavudine, and in patients with low CD4 cell counts.6,9 As our study implies that the prevalence of overt thyroid disease is likely to be similar to that observed in HIV-negative populations, routine screening of this specific population does not appear to be beneficial.
The present study has some limitations. Firstly, some patients did not have available information regarding CD4 counts and viral load values at the onset of the disease. Secondly, although the total duration of antiretroviral therapy was available, detailed historical data regarding the specific antiretroviral drug classes used since treatment initiation was not consistently accessible, because there have been cases involving transfers from other health services or countries. Consequently, cumulative lifetime exposure to each antiretroviral class could not be accurately determined. Thirdly, there was no comparison of these results with a control group not infected with HIV. Finally, since the number of individuals who presented alterations in thyroid function was small, it was not possible to apply a statistical test to assess the association between risk factors and thyroid functional abnormalities.
Therefore, we believe that larger studies with comparison with uninfected individuals are necessary to gain a deeper understanding of changes in thyroid function in people living with HIV, especially in the presence of severe immunosuppression. Dynamic and prospective studies with prolonged follow-up could better clarify the impact of HIV infection and antiretroviral therapy on thyroid function, contributing to the definition of more precise guidelines for screening and treatment in these patients.
Conclusion
In the observed population of patients living with HIV under antiretroviral therapy, the overall prevalence of thyroid dysfunction was similar to the estimated prevalence in the general population. For this reason, there appears to be no benefit in systematic screening of thyroid function in asymptomatic patients living with HIV on antiretroviral therapy.
Supplemental Material
sj-docx-1-jia-10.1177_23259582261439968 - Supplemental material for Thyroid Function Screening in People Living With Human Immunodeficiency Virus on Antiretroviral Therapy
Supplemental material, sj-docx-1-jia-10.1177_23259582261439968 for Thyroid Function Screening in People Living With Human Immunodeficiency Virus on Antiretroviral Therapy by Ana Filipa Bolas, MD, Umbelina Caixas, MD, Paula Bogalho, MD, and José Silva-Nunes, MD, PhD in Journal of the International Association of Providers of AIDS Care (JIAPAC)
Footnotes
Acknowledgments
The authors wish to acknowledge Sandra Cruz for support with statistical analysis.
Ethical Considerations
This study was conducted in accordance with the ethical principles outlined in the Belmont Report and the Declaration of Helsinki and was approved by the Ethics Committee of Unidade Local de Saúde de São José (protocol code 1611/2024, approved on December 11, 2024).
Consent to Participate
The Ethics Committee of Unidade Local de Saúde de São José waived the need to obtain consent for the collection, analysis, and publication of the retrospectively obtained and anonymized data for this non-interventional study, since it is a non-interventional, retrospective study, which guarantees the confidentiality and anonymity of patients.
Author Contributions
Conceptualization was done by AFB and UC; methodology was done by AFB; validation was done by UC, PB, and JSN; formal analysis was done by AFB; investigation by AFB, UC, PB, and JSN; resources by AFB and UC; data curation by AFB; writing—original draft preparation by AFB; writing—review and editing by AFB, UC, PB, and JSN; visualization by AFB, UC, PB, and JSN; supervision by JSN; project administration by JSN. All authors have read and agreed to the published version of the manuscript.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The data presented in this study are available on request from the corresponding author; however, they are not publicly accessible due to privacy constraints.
Supplemental material
Supplemental material for this article is available online.
