Henrik Falhammar

Clinical endocrinology • March 17, 2025

Hadis Mirzaei, Jekaterina Patrova, Buster Mannheimer, Jonatan Lindh, Henrik Falhammar

We express our gratitude for the insightful commentary by Dr Notarianni on our study. As accurately pointed out, in all patients with non-overtly functional adrenal tumours (NOFATs) diagnosed in Sweden between 2005 and 2019 (n = 20,390) the prevalence of unspecified dementia, Alzheimer's disease (AD) and vascular dementia (VD) were reduced compared to controls without diagnosed adrenal tumours (n = 125,392) [1]. This is probably because individuals diagnosed with dementia tend to undergo imaging less frequently. Moreover, the incidence of new unspecified dementia, AD and VD was similar as controls during the follow-up period up (median 4.9 years, IQR 2.2–8.2). However, it is important to note that these were diagnosed cases of dementia, and many cases of mild dementia may go undiagnosed. We appreciate the thorough analysis and the opportunity to address some of the important points raised by Dr Notarianni regarding the potential role of the apolipoprotein E (APOE) genotype as a confounding factor. The APOE ε4 allele is a well-established risk factor for dementia, particularly AD and VD. APOE ε4 has also been associated with an atherogenic lipid profile, which increases the risk of cerebrovascular disease, a key contributor to vascular dementia [2]. Recent research further supports that APOE variants influence various forms of vascular disease. For instance, a study by Rasmussen et al. demonstrated that APOE carriers exhibit increased risk of ischaemic cerebrovascular disease, highlighting the complex interplay between APOE genotype and cerebrovascular pathology [3]. APOE polymorphisms were found to influence lipid metabolism, with the ε4 allele being associated with higher levels of triglycerides (TG) and low-density lipoprotein cholesterol (LDL-C) while reducing high-density lipoprotein cholesterol (HDL-C). Furthermore, ε4 acted as an independent risk factor for both cancer and cardiovascular disease [4]. These findings align with previous research suggesting that the APOE ε4 genotype contributes to both metabolic and vascular risk factors that may influence dementia development. However, in our register-based study, we did not have access to genetic data for either cases or controls, limiting our ability to directly analyse APOE allele distributions. We assumed similar APOE distributions in both groups, but we acknowledge that if there were significant differences, this could have influenced our findings. To mitigate this, we accounted for key cardiometabolic confounders, such as hyperlipidaemia and atherosclerotic heart disease, in our statistical models. Given the well-documented links between APOE and lipid metabolism, this approach may have helped address some of the potential confounding effects. Additionally, studies have linked APOE ε4 to an increased risk of tumorigenesis and cardiovascular disease [4]. While the relationship between APOE and adrenal tumours has not been explicitly studied, there is evidence suggesting that adrenal tumours, particularly NOFATs, also carry an increased baseline risk of tumorigenesis. Patrova et al. reported that patients with NOFATs exhibited increased overall mortality, including from cardiovascular disease and cancer, with a more pronounced effect in younger individuals [5]. Future studies exploring this intersection may provide valuable insights into shared pathways linking APOE, adrenal function, and dementia risk. In conclusion, while our study found no increased risk of dementia in patients with NOFATs, we recognise that APOE genotype could be an important variable. We contend that prospective studies incorporating genomic data would prove invaluable in further delineating these associations. We reiterate our appreciation for the thoughtful feedback and for contributing to this important discussion.

Hormones And Behavior • April 21, 2025

Anna Strandqvist, Martin Asperholm, Henrik Falhammar, Angelica Hirschberg, Anna Nordenström, Agneta Herlitz

Psychological outcomes in people with congenital adrenal hyperplasia (CAH) and complete androgen insensitivity syndrome (CAIS) may provide information contributing to the understanding of development of behaviors that typically show sex differences. In this study, we investigated gender identity, friendship quality and occupational choices. Participants were women with 46,XX classic CAH (C-CAH; n = 29), non-classic CAH (NC-CAH; n = 13), women with 46, XY CAIS (n = 11), male controls (n = 147) and female controls (n = 142). Participants completed an online survey with questions on gender identity, friendship, (sex of friends in childhood, adolescence, adulthood), friendship style, and occupation. Results showed that (1) female and male controls differed on most outcomes. (2) Women with CAIS and women with NC-CAH responded in a pattern not different from female controls on most questions regarding gendered behavior. (3) Women with C-CAH and women with CAIS responded more similarly to male controls than female controls on the friendship questionnaire. (4) Women with C-CAH worked in occupations with a male sex distribution whereas females with CAIS worked in occupations that were not different from those of female or male controls. (5) More severe forms of CAH were associated with a response pattern more in line with that of male controls, whereas the opposite was true for females with less severe forms of CAH.

The Journal Of Clinical Endocrinology And Metabolism • March 21, 2025

Fredric Hedberg, Jonatan Lindh, Buster Mannheimer, Tereza Planck, Jakob Skov, Mikael Lehtihet, Henrik Falhammar, Jan Calissendorff

Background: Autoimmune hypothyroidism is a common endocrine disorder affecting 1-2% of the population in iodine sufficient areas. While levothyroxine is standard treatment, a substantial number of patients report persistent symptoms despite adequate treatment. The use of liothyronine as an adjunct to levothyroxine therapy has increased. The psychiatric characteristics of patients receiving liothyronine remain largely unknown. This study examines the association between preexisting psychiatric morbidity and subsequent liothyronine use in autoimmune hypothyroidism. Methods: This nationwide retrospective cohort study includes all adults in Sweden with autoimmune hypothyroidism and initiated on treatment with thyroid hormones between 2006 and 2020. Data were obtained from the National Patient Register and the National Prescribed Drug Register. Psychiatric morbidity prior to diagnosis was identified using ICD-10 codes and ATC-codes for psychiatric medications. Logistic models estimated associations, adjusting for sex, age, and region. Results: Among 353,708 patients, 44.8% had a history of psychiatric morbidity. These patients were more likely to receive liothyronine (adjusted odds ratio (aOR) 1.90, 95% confidence interval (95% CI) 1.83-1.97, p<0.001) compared to those without a psychiatric history. This was most evident among individuals with affective or anxiety morbidity (aOR 1.91, 95% CI 1.84-1.98, p<0.001). No association was found for psychotic morbidity (aOR 1.08, 95% CI 0.98-1.19, p=0.11). Conclusions: Patients with a psychiatric history before autoimmune hypothyroidism were more likely to receive liothyronine, especially among those with affective or anxiety morbidity. This may reflect persistent symptoms and affect subsequent decisions in the treatment of hypothyroidism.

European Journal Of Internal Medicine • February 28, 2025

Objective: Over the last decades, advances in understanding of previously described associations have important implications for diagnosis and workup of hyponatremia. In addition, new drug groups potentially affecting sodium balance and water homeostasis have evolved. The aim of this review is to summarize current evidence on drug-induced hyponatremia in clinical care. Methods: We searched PubMed using the string "Inappropriate ADH Syndrome/chemically induced"[Mesh] OR "Inappropriate ADH Syndrome/diagnosis"[Mesh]) OR ("Hyponatremia/chemically induced"[Mesh] OR "Hyponatremia/diagnosis"[Mesh]), January 1st, 2008, to September 2nd 2024. In total 2003 articles were found and reviewed. Relevant articles referenced herein were subsequently traced backwards and also reviewed. Results: Drugs associated with hyponatremia, including selective serotonin reuptake inhibitors, antipsychotics, antiepileptic drugs and proton pump inhibitors, typically cause hyponatremia shortly after initiation of treatment. For thiazide diuretics, the number one culprit in drug-induced hyponatremia, the risk for hyponatremia is highest the first weeks after initiation and then gradually decreases to a stable but still increased level after around 3 months. Several drugs that promote a negative water balance such as loop diuretics, lithium and of sodium-glucose cotransporter-2 inhibitors appear to decrease the risk for hyponatremia. Treatment with immune checkpoint inhibitors is associated with an increased risk of hypophysitis and adrenalitis resulting in hyponatremia due to secondary and primary cortisol deficiency. Conclusions: For most drugs associated with hyponatremia, including thiazides, the cause-effect relationship is tightly linked to newly initiated treatment. Further research is warranted to characterize the association between hyponatremia and newly developed drugs such as sodium-glucose cotransporter-2 inhibitors and immune checkpoint inhibitors.

Clinical Endocrinology • February 16, 2025

Ana Carasel, Jan Calissendorff, C Juhlin, Henrik Falhammar

Objective: The incidence of adrenal tumours has increased in the last decades, mainly due to increased use of imaging. The diagnostic evaluation of adrenal masses can be complex and, in some cases, necessitates cytological evaluation. However, concerns remain regarding the potential complications associated with adrenal gland biopsy. Design: We conducted a retrospective cohort study to evaluate the safety and diagnostic effectiveness of cytology in patients who underwent fine-needle aspiration (FNA) of adrenal glands at Karolinska University Hospital in Stockholm, Sweden, between 2000 and 2022. The aim was to evaluate the accuracy of the sample and the complication rate. Patients and measurements: A total of 241 patients and 251 FNAs were included, with 10 patients undergoing two FNAs each. Data on clinical, radiological and laboratory presentation was collected and corelated with cytological findings and outcomes. Results: Diagnostic FNA was obtained in 90% of patients (n = 217) with endoscopic ultrasound technique being most successful (95.8%), followed by CT (88.7%) and transabdominal ultrasound technique (86.7%). The sensitivity and the specificity were 93.8% respectively 96.7%. More than half of the FNA samples (52.7%) indicated a diagnosis consistent with metastases to the adrenal gland. The complication rate was 7.9% (n = 20). Based on the FNA results, adrenalectomy was performed on 13.6%, while 52.8% of the patients with benign findings were managed conservatively. Chemotherapy was started for 78.7% of patients with malignant findings. Conclusion: FNA of the adrenal glands is a safe, minimally invasive diagnostic procedure that can be useful in the assessment of adrenal lesions.