Clinical Endocrinology

Which patient requires neuroendocrine assessment following traumatic brain injury, when and how?

Authors


Correspondence: Dr Amar Agha, Academic Department of Endocrinology, Beaumont Hospital, Beaumont Road, Dublin 9, Ireland. Tel.: + 353 1 8093000; Fax: +353 1 8572979; E-mail: amaragha@beaumont.ie

Summary

Traumatic brain injury (TBI) is an important public health problem, particularly among young adults in industrialized countries. Hypopituitarism is a common occurrence among survivors of TBI and may contribute to the associated morbidity seen in the acute and chronic phases following injury. The available data suggest that survivors of moderate to severe TBI should undergo screening for hypopituitarism particularly in the first year after injury. This requires a close liaison between endocrinologists, neurosurgeons, neuropsychologists, intensive care and rehabilitation physicians. Patients who suffer milder forms of TBI should also be considered for endocrine evaluation if they exhibit any clinical features of pituitary hormone deficiencies.

Introduction

Traumatic brain injury (TBI) is the leading cause of death and disability among young adults living in industrialized countries.[1] Survivors frequently suffer multiple physical and neuropsychiatric complications with an estimated 2% of the US population requiring assistance with activities of daily living following head injury.[2] Damage to the hypothalamic–pituitary axis in the setting of head trauma has been recognized for several decades, with post-mortem data demonstrating pituitary gland infarction in up to one-third of patients after fatal TBI.[3] Recently, several clinical studies have demonstrated a high frequency of hypothalamic–pituitary hormone deficiencies among adult TBI survivors.

Notwithstanding the methodological differences between various studies, there is a broad agreement that anterior hypopituitarism is a common finding after moderate and severe head injury with an estimated prevalence of approximately 25% among adult long-term survivors.[4] There is also considerable cross-over between the chronic sequelae of TBI (i.e. fatigue, memory impairment, emotional lability) and clinical features of hypopituitarism, suggesting that post-traumatic hypopituitarism (PTHP) may be a contributing factor.

Owing to the high incidence of TBI and limited health care resources, it would not be feasible to evaluate all these patients for PTHP. A rational approach to pituitary screening, which will identify those most likely to suffer hypopituitarism and consequently those who may benefit from treatment, is necessary. In this article, we discuss the appropriate selection of patients for endocrine evaluation, the potential sequelae of PTHP and the appropriate timing for assessment.

How common is pituitary dysfunction after TBI and what is the natural history?

The annual incidence of head trauma is approximately 100–300 cases/100 000 of the population.[5] Approximately one million patients attend the emergency department, and nearly 300 000 are hospitalized for TBI in the United States annually.[6]

Several retrospective and prospective studies have reported a variable prevalence of anterior pituitary dysfunction following head injury in adults ranging from 15% to 68%.[7-12] Much of this variability can be explained by differences in patient selection, timing of testing in relation to the injury and, importantly, the methodological differences with respect to the type of dynamic test used and whether abnormal results were confirmed or otherwise using a second test. A systematic review that examined the heterogeneity between various studies reported a pooled prevalence of 27·5% following TBI.[4]

In addition, research also shows that anterior pituitary dysfunction following TBI is a dynamic process. Acute deficiencies may be transient and, conversely, new hormone deficiencies can manifest in the post-acute phase.[13] Therefore, endocrine evaluation of patients post-TBI involves a process from the acute post-injury period through to the chronic phase, that is, 1-year post-injury.

Disturbance of neurohypophyseal function is also common following moderate to severe head injury. Diabetes insipidus can occur in up to 26% of TBI victims in the acute phase.[14] The general trend is towards improvement with a prevalence of 7% in long-term survivors in a large series.[15] Hyponatraemia owing to the syndrome of antidiuresis (SIAD) can also occur with a reported prevalence ranging from 2·3% to 33% post-TBI.[16-18] This is typically an early and transient phenomenon.

What is the clinical significance of PTHP?

ACTH deficiency may be life-threatening particularly in the acute phase following head injury. The present authors and colleagues have previously identified acute ACTH deficiency among 16% of a cohort of patients (n = 50) following moderate to severe TBI using the glucagon stimulation test.[14] Complications such as hypoglycaemia, hyponatraemia or unexplained hypotension in the hours or days after injury should alert the clinician to the possibility of acute hypopituitarism. Other investigators have demonstrated a lower blood pressure and higher vasopressor requirements in patients with hypocortisolaemia in the acute phase of TBI.[19] Dramatic improvement in the clinical condition of such patients following glucocorticoid replacement has been reported.[20] A positive association between acute phase hypocortisolaemia and mortality has also been shown although it is unclear whether a casual relationship exists.[21]

The chronic recovery phase following moderate to severe TBI is often complicated by behavioural disturbance, cognitive impairment, poor motivation and lethargy.[22] Recent evidence suggests that during this rehabilitation phase, patients with PTHP have a worse phenotype than patients without hypopituitarism but a similar degree of injury. TBI survivors with PTHP have altered body composition, a less favourable lipid profile, reduced exercise capacity and a poorer quality of life compared to eupituitary TBI patients.[23, 24] Therefore, unrecognized and untreated hypopituitarism may impair recovery or limit rehabilitation following moderate to severe TBI.

Who should be investigated for post-traumatic hypopituitarism?

It is clearly not feasible to test all TBI patients for post-traumatic hypopituitarism without consuming an enormous amount of health care resources. Recent research efforts have tried to identify predictors of PTHP, including patient and injury-specific risk factors.

The severity of head trauma is a useful screening tool. The Glasgow Coma Scale (GCS), which assesses the level of consciousness of the patient, is the most commonly used method for assessing the severity of brain injury. Scores between 13 and 15 following injury are considered mild TBI, 9–12 is deemed moderate, while ≤8 is severe. Some studies have found a correlation between severity of injury and likelihood of hypopituitarism, and this is supported by the results of recent meta-analysis.[4] TBI severity can also be graded according to the degree of abnormality on cross-sectional imaging of the head, which may be useful when initial GCS score is not available or inconsistent.[25] As most of the available data on PTHP concerns patients admitted to hospital with moderate or severe TBI, it is reasonable to focus pituitary testing on this cohort (GCS scores of between 3 and 12). Furthermore, this particular group is often left with significant post-traumatic physical and neuropsychiatric morbidity and is therefore more likely to benefit from pituitary replacement. While regional practice regarding in-patient care of patients with TBI (in particular, those with mild to moderate injury) may vary, it is estimated that 10–20% of those presenting to hospital could ultimately be considered for screening.[6] Other identified risk factors for PTHP include raised intra-cranial pressure, long admission to the intensive care unit (ICU), diffuse axonal injury on brain imaging and base of skull fracture.[10, 26] These additional factors have only moderate sensitivity but may be used to risk-stratify patients for pituitary screening in a resource-constrained environment.

Recent research supports the contention that patients with less severe TBI presenting to the emergency department have a low prevalence of endocrine dysfunction when strict diagnostic criteria are used and perhaps should be considered for endocrine evaluation only if they develop clinical features suggestive of pituitary deficiency.[27]

When should survivors of TBI undergo endocrine evaluation?

Acute phase

Assessment of pituitary function in the acute phase following injury is fraught with difficulties. While acute deficiency may be life-threatening in the case of ACTH, other hormonal alterations may simply represent physiological responses to critical illness. For example, low sex-steroid concentration is present in over 80% of patients after severe or moderate TBI but recovers in the majority later on.[13] Provocative stimulation of pituitary function in the immediate post-injury phase is either not practical (as in the case of the insulin tolerance test) or inappropriate (as in the case of the short synacthen test). Currently, there is no evidence that replacement of growth hormone, sex steroids or thyroid hormone in the acute period is of benefit. Therefore, the focus during the acute phase should be on detecting glucocorticoid deficiency.

It has been our practice to monitor early morning serum cortisol for the first 7 days following TBI in hospitalized patients. The appropriate serum cortisol concentration in head injury victims is difficult to define as it depends on the severity of the injury (or injuries), presence or absence of sepsis and other factors. We use a clinical cut-off of 300 nm (11 μg/dl) based on data from non-TBI surgical ITU patients.[21] Therefore, we regard values <300 nm to be highly suggestive of ACTH deficiency. Therapeutic glucocorticoid replacement is indicated in such cases until re-assessment can take place in the recovery phase. Morning cortisol levels of 300–500 nm should be interpreted within the clinical context and glucocorticoid replacement is indicated if the patient displays any clinical features of glucocorticoid deficiency.

Cranial diabetes insipidus should be easily recognizable in the acute phase with hypernatraemia associated with hypotonic polyuria. The treatment is to administer desmopressin (subcutaneously or intramuscularly) and hypotonic fluids guided by the urine output and the plasma sodium. Hyponatraemia is usually mild and mainly because of SIAD.[17] The evaluation and management of hyponatraemia in neurosurgical patients is beyond the scope of this article. However, a major caveat is that hyponatraemia can occasionally be a feature of acute hypocortisolaemia.

Chronic phase

Figure 1 outlines a suggested algorithm for the assessment of patients after TBI. Following the acute phase evaluation outlined above, we recommend that patients selected for screening undergo assessment of the adrenal, thyroid and gonadal axes between three and 6 months. Baseline thyroid function tests, gonadotrophins and sex-steroid concentrations (menstrual history in premenopausal women) should be checked in all patients selected for screening. The choice of a dynamic test for assessing the HPA axis will depend on the preference and experience of individual endocrine units but the short synacthen test is a practical choice owing to its simplicity and usefulness in excluding clinically significant adrenal insufficiency.[28] This chronic phase assessment should be offered to all survivors of moderate to severe head injury in collaboration with the multidisciplinary acquired brain injury team, which includes the neurology, neurosurgery, neuropsychiatry/psychology and rehabilitation services.

Figure 1.

Algorithm for assessment of patients following TBI.

If pituitary hormone deficiencies are detected, then appropriate replacement should be instituted and a repeat assessment at 1 year should be considered as recovery may occur.[13] Assessment of growth hormone (GH) reserve in patients who would be candidates for GH replacement (those with poor recovery, neuropsychiatric symptoms and other features, which may overlap with the syndrome of adult GH deficiency) should perhaps be delayed until 1 year as acute GH deficiency may recover. We typically use the insulin tolerance test (ITT) but the glucagon stimulation or the arginine + GHRH tests (using appropriate cut-off values) are suitable alternatives to the ITT, which may be contraindicated in some survivors of TBI due to a coexisting seizure disorder.

Conclusion

Hypopituitarism is a common complication among adult survivors of TBI and may contribute to poor quality of life and impair rehabilitation following injury. Selecting appropriate candidates for screening requires consideration of patient and injury-related factors. Severity of head injury, as measured by GCS, currently provides a good guide as to who requires pituitary evaluation after TBI. Other head injury survivors should be considered for screening if they display any clinical features of hypopituitarism. A rational approach to endocrine testing begins with assessment of ACTH reserve in the acute phase and is followed by a more comprehensive assessment of pituitary function during the chronic recovery phase.

Conflict of interests

Nothing to declare.

Ancillary