Possible Involvement of Hypothalamic Dysfunction in Long COVID Patients Characterized by Delayed Response to Gonadotropin-Releasing Hormone.

Long COVID (LC) may involve endocrine dysfunction; however, the underlying mechanism remains unclear. To examine hypothalamic-pituitary responses in patients with LC, we conducted a single-center retrospective study of patients with refractory LC referred to our University Hospital who underwent anterior pituitary stimulation tests. Between February 2021 and November 2025, 1251 patients with long COVID were evaluated, of whom 207 (19%) had relatively low random ACTH or cortisol levels. Ultimately, 16 underwent anterior pituitary stimulation tests and were included. All tests were performed in an inpatient setting without exogenous steroids. Fifteen patients (six women, mean age 35.6 years) underwent corticotropin-releasing hormone (CRH), thyrotropin-releasing hormone (TRH), and gonadotropin-releasing hormone (GnRH) tests. All patients had mild acute COVID-19, eight had ≥2 vaccinations, and the mean interval from infection was 343 days. Frequent symptoms included fatigue (100%), insomnia (66.7%), headache (60.0%), anorexia/nausea (40.0%), and brain fog (40.0%). Mean early-morning cortisol and 24 h urinary free cortisol were 7.5 μg/dL and 41.0 μg/day, respectively. MRI showed an empty sella in one case. Peak hormonal responses were preserved (ΔACTH 247%, ΔTSH 918%, ΔPRL 820%, ΔFSH 187%, ΔLH 1150%); however, peaks were delayed beyond 60 min in ACTH (13%), LH (33%), and FSH (87%). Notably, significantly delayed elevations remained at 120 min in the responses of TSH (4.1-fold), PRL (1.8-fold), LH (9.3-fold), and FSH (2.8-fold), suggesting possible hypothalamic involvement, particularly in the gonadotropin responses. Additionally, serum IGF-I was lowered (-0.70 SD), while GH response (mean peak 35.5 ng/mL) was preserved by growth hormone-releasing peptide (GHRP)-2 stimulation. Low-dose hydrocortisone and testosterone were initiated for three patients. Although direct viral effects and secondary suppression have been proposed, our findings may suggest that, at least in part, the observed response characteristics are consistent with functional secondary hypothalamic dysfunction rather than irreversible primary injury. These findings highlight the need for objective endocrine evaluation before initiating hormone replacements.