Traumatic brain injury (TBI) is a leading cause of disability worldwide, often affecting young individuals and resulting in long-term health consequences. One important but frequently overlooked complication is post-traumatic hypopituitarism. Severe TBI can disrupt the hypothalamic-pituitary-peripheral hormone axes during the acute phase and may lead to chronic pituitary dysfunction [1], [2].

Acute critical illness and its treatments can impair the normal adaptive response of the hypothalamic-pituitary axis through various mechanisms, including alterations in metabolism, hormone binding, and hormone production. In TBI, the risk of direct structural damage to the hypothalamus or pituitary adds an additional layer of complexity, potentially resulting in lasting hormonal dysregulation.

Pituitary hormone abnormalities are commonly observed in the acute phase following TBI; however, diagnosis is challenging, and the clinical significance and therapeutic value of addressing these acute changes remain controversial. Routine assessment of the growth hormone, thyroid, and gonadal axes in critically ill patients is not currently recommended, as there is no clear evidence of benefit from hormonal replacement in this setting. In contrast, adrenal and arginine-vasopressin deficiency (AVP-D) are potentially life-threatening conditions that require prompt recognition and immediate treatment [3].

While a subset of patients may present with profound hormonal deficiencies and benefit from intervention in the longer and non-acute term [4], the majority tend to have subtle or isolated hormonal changes [2]. The impact of hormone replacement therapy in such cases remains uncertain, and until more robust data becomes available, routine anterior pituitary testing and hormone replacement in non-critical cases should be approached with caution.

The mechanisms by which TBI leads to post-traumatic hypopituitarism remain incompletely understood. Several hypotheses have been proposed, including direct injury to the pituitary gland or hypothalamus, vascular compromise, and autoimmune processes. Current evidence suggests a multifactorial aetiology involving both primary mechanical damage and secondary physiological insults.

Autopsy studies provide some insight. Anterior pituitary necrosis has been observed in up to one-third of patients who died from severe head injury [5]. Notably, no infarction was found in the pituitary histology specimens from 12 instantly fatal cases of motorcycle accidents, while 13 of 30 (43 %) persons who survived 3 h to 7 days exhibited acute infarcts of varying extent, ranging from focal lesions to sub-total necrosis involving up to 90 % of the adenohypophysis [6]. The pathogenesis is likely complex, encompassing direct trauma to the pituitary gland, indirect injury through disruption of the hypothalamus, the pituitary stalk, or associated neural pathways (Fig. 1), severe intracranial hypertension and stress inducing hippocampal and pituitary cell apoptosis [7] or secondary insults such as hypotension, hypoxia, anaemia, or cerebral oedema, which may impair blood flow through the hypophyseal portal system [8].

Whether findings from fatal cases can be extrapolated to survivors with long-term hypopituitarism remains unclear. However, neuroimaging studies offer some corroborative evidence. In an observational case–control study including 41 adults with non-fatal TBI, nearly 30 % exhibited acute MRI abnormalities of the pituitary gland, including enlargement, haemorrhage, infarction, signal irregularities, or partial transection of the pituitary stalk [9].

Other imaging data suggest that patients with chronic post-TBI hypopituitarism more frequently display features such as reduced pituitary volume, empty sella, signal heterogeneity, perfusion deficits, and loss of the normal posterior pituitary signal, compared to patients with intact pituitary function [10]. Additional factors such as elevated intracranial pressure, diffuse axonal injury, and basal skull fractures have also been associated with increased risk of long-term endocrine dysfunction [11], [12]. Bavisetty et al. [13] found that acute cranial CT findings were the strongest predictor of long-term hormonal deficiencies. Similarly, Klose et al. [14] reported that a normal head CT was strongly associated with preserved pituitary function. However, these associations are not consistent across studies. For instance, Kleindienst et al. [15] found no significant correlation between acute or delayed CT findings and the eventual development of hypopituitarism, casting doubt on the predictive value of imaging alone.

TBI in patients and animals has been shown to trigger a focal immune response in the brain that is initiated within minutes after the primary injury. This focal brain inflammation involves both brain-resident immune cells and peripheral immune cells with complicated interactions. In addition to this focal inflammatory state, the evidence suggests that post-traumatic brain inflammation disseminates globally both in the acute and chronic stages of injury, sometimes even throughout life. The disseminated brain inflammation may persistently impact brain pathophysiology and is associated with progressive neurodegeneration. There may also be a link to concurrent hypopituitarism. In a 3-year follow-up study of 29 patients with TBI, anti-pituitary antibodies were detected in 13 (44.8 %) patients and were significantly correlated with hypopituitarism [16]. Another study analyzing blood samples from 143 men with moderate-to-severe TBI (1–12 months post-injury) and 39 healthy men found elevated levels of anti-pituitary and anti-glial fibrillary acidic protein antibodies, alongside other inflammatory markers [17]. These immune markers correlated to persistent hypogonadotrophic hypogonadism, yet in different directions, further suggesting involvement of autoimmunity and general inflammation in post-TBI hypopituitarism. The mechanism may be autoimmune destruction by preexisting or elicited autoimmune activity targeting the hypothalamus and/or the pituitary gland [18], but other inflammatory regenerative mechanisms are also possible [17], [18]. Due to the finding of several types of both proinflammatory and other cytokines in the cerebrospinal fluid, a direct effect from these cytokines on pituitary cells is also possible [19].

One possible explanation for the difficulty in identifying reliable early predictors lies in the potential regenerative capacity of pituitary tissue. Experimental studies have demonstrated that pituitary stem cells—and even some differentiated hormone-producing cells—may retain the ability to regenerate and restore function [20].

Animal models have contributed to our understanding of hypothalamus-pituitary dysfunction following TBI. The review by Vennekens & Vankelecom [21] summarized the limited number of studies exploring hypothalamus-pituitary dysfunction in experimental animal TBI models. They summarized the studies in a table with respect to hypothalamus pituitary endocrine functions after an experimentally induced trauma by either of many cranial trauma models (14 heterogenously designed experimental protocols), particularly with respect to specific hormonal axes. The studies were rarely covering the broader structural and functional consequences for the heterogeneous phenotype of the pituitary gland, so even these investigations were mainly observational, and did mostly not look into the possible mechanisms causing disturbance of the hypothalamus-pituitary functions. In fact, most were not designed to elucidate the underlying pathophysiological mechanisms, limiting their relevance for explaining the full spectrum of post-traumatic endocrine dysfunction. Thus, more research efforts are needed to uncover development of pituitary hormone dysregulation following TBI, at the same time asking for more reliable study models. Moreover, given the spontaneous restoration in some patients, it would be worthwhile to explore the underlying reparative mechanisms, and in particular the occurrence of tissue regeneration including the reaction of the local pituitary stem cells to TBI. Understanding this regenerative process may open new treatment options for hypopituitarism.

Diagnosing hypothalamic–pituitary hormone axis dysfunction is complex, even for experienced neuroendocrinologists. The challenge is especially pronounced in patients with non-classic causes of hypopituitarism with no visible hypothalamic-pituitary or intracranial abnormalities on imaging. Except for deficiency of adrenocorticotrophin (ACTH) from the anterior pituitary and AVP from the posterior pituitary lobe, — which require immediate evaluation and treatment —most pituitary hormone deficits present with subtle, nonspecific symptoms. These are generally milder than those seen in corresponding primary endocrine disorders, making clinical detection difficult.

Moreover, the lack of reliable biomarkers that reflect end-organ hormone action, combined with the interdependence of hormone axes, further complicates both diagnosis and management. Still, a meticulous clinical history and physical examination may offer important diagnostic clues, in patients where non-classical causes of hypopituitarism are suspected, including patients considered for testing due to suspected post-TBI hypothalamic or pituitary injury.

In classic cases of hypopituitarism caused by mass lesions in the sella turcica or post-radiation effects, growth hormone (GH) and gonadotropin deficiencies typically appear before ACTH and thyroid-stimulating hormone (TSH) deficits. However, it remains unclear whether this progression is similarly observed following TBI. The presence of regular menstrual cycles in premenopausal women and intact sexual function in men make central hypothyroidism and adrenal insufficiency unlikely [21].

A major obstacle in diagnosing post-traumatic hypopituitarism is the low pretest probability of pituitary hormone deficiency in this population, especially in the absence of overt clinical signs. Applying Bayes’ Theorem [22], the likelihood of a true positive result for any given hormonal axis is likely overestimated. Diagnostic accuracy is also heavily influenced by pre-analytical and analytical variables *[23], [24].

Historically, pituitary function testing was largely restricted to tertiary care settings and to patients with clear hypothalamic–pituitary lesions or multiple hormonal deficiencies, where the probability of true hypopituitarism was high. Despite limited robust evidence, patients with TBI—often presenting with isolated deficiencies and lacking confirmatory testing—have more recently been included among high-risk groups considered for pituitary evaluation. This practice increases the risk of false positives, particularly due to borderline ("grey zone") test results. Therefore, hormone testing should always be interpreted within the context of clinical findings and pretest probability, with particular caution when evaluating GH and gonadal axes, which are especially prone to diagnostic confounding (discussed in Table 1) *[23], [25].

The utility of IGF-1 was shown of limited value as a screening tool also in TBI [26], and as GH secretion is pulsatile, assessment of GH deficiency requires dynamic stimulation testing for accurate diagnosis in most cases [27]. Obesity is a well-known confounder that blunts GH response and can falsely suggest deficiency; this effect is reversible with weight loss [28], [29]. As such, body mass index (BMI)-adjusted cut-off values are essential when interpreting GH stimulation tests, including commonly used protocols such as the arginine GH-releasing hormone and glucagon tests [30], *[31], [32].

Early studies assessing GH deficiency post-TBI did not apply BMI-based thresholds, largely because this confounder was not yet widely recognized, nor were BMI-adjusted reference values available at the time. As a result, these studies likely overestimated the prevalence of GH deficiency.

A newer diagnostic tool—macimorelin, an orally active GH secretagogue—has emerged as a promising alternative. Unlike the insulin tolerance test (ITT), macimorelin does not induce hypoglycaemia or cause significant patient discomfort. It has shown good overall concordance with the ITT and appears to be unaffected by variables such as BMI, age, or sex. Given these advantages, macimorelin may become the preferred GH stimulation test in future clinical practice [33], [34].

In the acute phase following TBI, conventional adrenal stimulation tests—such as the Cosyntropin (ACTH) test—are unreliable. This is due to the abrupt cessation of endogenous ACTH secretion; despite reduced ACTH drive, the adrenal glands can still mount a normal response to supraphysiological exogenous ACTH. As a result, a normal Cosyntropin test in the acute setting does not exclude central adrenal insufficiency.

Instead, baseline plasma cortisol measurements can provide initial guidance. As a rule of thumb, a baseline morning cortisol level below 100 nmol/L strongly suggests adrenal insufficiency. However, the utility of baseline cortisol in diagnosing adrenal insufficiency in acutely ill patients remain limited, as there is no definitive diagnostic threshold. In such cases, the diagnosis is primarily based on clinical assessment. Nevertheless, a low baseline cortisol level can reinforce the clinical suspicion of adrenal insufficiency. However, normal or mildly elevated levels do not reliably rule out the diagnosis.

In contrast, after the acute phase has resolved, dynamic adrenal testing becomes crucial to determine whether the clinical suspicion of adrenal insufficiency can be confirmed or ruled out. Since over 95 % of serum cortisol is bound to corticosteroid-binding globulin (CBG) and albumin, any condition that alters binding protein levels—such as the use of oral contraceptives, severe illness, sepsis or rare CBG deficient cases—can confound test interpretation. Reference cut-offs for these altered physiological states are not established and biomarkers awaited.

Diagnosing central hypothyroidism remains challenging. Serum TSH concentrations are unreliable indicators in central hypothyroidism. TSH may be within the normal range or even elevated due to abnormal glycosylation, resulting in bioinactive TSH forms that are still detectable by current immunoassays [35].

Therefore, diagnosis hinges on identifying an inappropriately low or normal TSH level in the presence of a low or low-normal free thyroxine (fT4) level—a biochemical profile that may resemble non-thyroidal illness [36]. Adding to the difficulty, fT4 measurements themselves often suffer from limited analytical precision and reproducibility [37], *[38].

Diagnosing central androgen or oestrogen deficiency following TBI is complicated by various confounding conditions. Obesity, type 2 diabetes, depression, and opioid use can all alter hypothalamic–pituitary–gonadal function. In obesity and diabetes in men, this is largely due to reduced sex hormone-binding globulin (SHBG), which lowers total testosterone levels. Opioids can suppress the hypothalamus–pituitary–gonadal axis at multiple levels, while commonly prescribed anticonvulsants may accelerate hepatic testosterone metabolism.

In women of reproductive age, the presence of regular menstrual cycles typically excludes significant gonadal dysfunction. In postmenopausal women, low gonadotropin levels are often inappropriate. As with TSH, measurable gonadotropins may be bioinactive, leading to falsely reassuring assay results [39]. The diagnosis, therefore, relies on finding a low serum sex hormone level in combination with an inappropriately low gonadotropin concentration, taking into account potential assay interference.

Assessing AVP disturbances, such as syndrome of inappropriate antidiuretic hormone secretion (SIADH) or diabetes insipidus, nowadays referred to as AVP deficiency (AVP-D) [40], are particularly challenging due to their dependence on the patient’s fluid and electrolyte status. The diagnostic approach differs depending on which condition is suspected.

In the acute phase post-TBI, careful monitoring is essential in case AVP dysregulation is suspected. Urinary output should be assessed hourly, with regular measurements of serum and urine sodium and osmolality.

SIADH should be suspected in cases of hyponatraemia with low serum osmolality, elevated urine osmolality, and reduced urine output with a very high sodium concentration [41]. Management typically involves fluid restriction or administration of hypertonic saline, in accordance with institutional guidelines [42].

AVP-D, conversely, presents with hypernatraemia, high plasma osmolality, low urine osmolality, and polyuria. This condition requires prompt treatment with vasopressin, either titrated to response or administered in fixed doses. Close monitoring is critical to avoid overtreatment, which can lead to iatrogenic SIADH. Notably, some patients may exhibit a transient shift between AVP-D and SIADH in the early unstable phase of pituitary injury [43].

In the chronic phase, the standard diagnostic tool for AVP-D remains the water deprivation test [44]. However, newer approaches—such as measuring copeptin (a stable surrogate marker of vasopressin) following hypertonic saline infusion or arginine stimulation—are showing promise and may eventually replace traditional testing, pending further validation [44], *[45]. These tests show stronger validity in ruling out AVP-D than ruling it in.

Studies assessing hypopituitarism after TBI have used diverse hormone assays, diagnostic panels, and cut-off limits, contributing to significant variability in reported prevalence rates [2]. A key step toward improving diagnostic precision is the establishment of local assay-specific diagnostic thresholds. However, such context-specific cut-off limits are often lacking.

Efforts have been made to standardize and evaluate hormone assays, particularly for GH, but assay variability remains a challenge across all endocrine axes. A notable example is the introduction of automated cortisol immunoassays using monoclonal antibodies and mass spectrometry-based standardization. These advances have led to lower diagnostic thresholds for adrenal insufficiency following the 250 µg Cosyntrophin test—from the traditional 500–550 nmol/L (18.1–20.0 µg/dL) down to 350–420 nmol/L (12.7–15.2 µg/dL) [46], [47], [48].

Many TBI studies have not included confirmatory re-testing of hormone deficiencies, even though such re-evaluation is critical—particularly for isolated hormonal deficits—due to intra-individual variability and the transient effects of stress or illness. Although practical constraints often limit adherence to this practice, guidelines increasingly advocate for re-testing. For instance, the American Association of Clinical Endocrinologists and American College of Endocrinology guidelines recommend re-evaluating the GH axis in patients with isolated GH deficiency and avoiding testing in those with low pre-test probability [27].

The need for re-testing is not limited to GH. For testosterone, up to 30 % of initially abnormal results normalize upon repeat testing [49]. A similar variability and spontaneous normalisation are also seen in the thyroid function tests [50]

Inter-hormonal interactions further complicate endocrine evaluation. Ignoring these can lead to significant over- or underestimation of deficiencies across pituitary axes [51], *[52], [53] (Table 2).

Hormonal alterations are common in the acute phase post-TBI, but patterns vary across endocrine axes. While adaptive increases in stress hormones such as cortisol, GH, prolactin, and AVP are typical, suppression of the gonadal and thyroid axes may mimic central hypogonadism or hypothyroidism—though these are often non-specific responses to acute illness rather than true pituitary failure [54], *[55], [56] (Fig. 2).

These hormonal shifts are more pronounced with greater illness severity and are associated with increased morbidity and mortality [57], [58], [59]. Importantly, these alterations are not exclusive to TBI but reflect generalized endocrine adaptations to critical illness, driven in part by inflammatory cytokines.

Not least, acute and critical illness also affects hormone-binding proteins and their metabolism. Many ICU medications alter hormone levels or binding protein concentrations, adding to diagnostic uncertainty. As such, neither total nor free hormone levels provide a reliable basis for diagnosing anterior pituitary hormone deficiencies in critically ill patients.

Among the hormonal systems, the HPA axis demands particular attention during the acute phase. Untreated adrenal insufficiency can be life-threatening. Critical illness induces complex changes ensuring systemic cortisol availability, including early activation of the HPA axis, reduced plasma binding by CBG and albumin with elevated free cortisol, and decreased cortisol clearance [60], *[61]. These changes obscure traditional biochemical indicators of adrenal insufficiency, and appropriate thresholds for initiating glucocorticoid replacement remain undefined.

Cortisol synthesis may also be impaired by specific drugs—such as etomidate and ketoconazole—or suppressed due to exogenous corticosteroid or opioid use, or increased hepatic metabolism (e.g., with phenytoin). These effects are pharmacologic and not directly linked to TBI-induced adrenal dysfunction.

Currently, routine anterior pituitary hormone screening is not recommended in all critically ill patients due to the diagnostic complexities outlined above and lack of evidence supporting the benefit of routine hormonal replacement. For example, pharmacologic GH therapy in critical illness has been linked to increased morbidity and mortality [62], [63], while the role of thyroid hormone therapy remains controversial [59], [64]. Similarly, androgen replacement has not demonstrated clear benefits in prolonged critical illness [54].