Neuroendocrine Disturbances One to Five or More Years after Traumatic Brain Injury and Aneurysmal Subarachnoid Hemorrhage: Data from the German Database on Hypopituitarism

Abstract

Neuroendocrine disturbances are common after traumatic brain injury (TBI) and aneurysmal subarachnoid hemorrhage (SAH), but only a few data exist on long-term anterior pituitary deficiencies after brain injury. We present data from the Structured Data Assessment of Hypopituitarism after TBI and SAH, a multi-center study including 1242 patients. We studied a subgroup of 351 patients, who had sustained a TBI (245) or SAH (106) at least 1 year before endocrine assessment (range 1–55 years) in a separate analysis. The highest prevalence of neuroendocrine disorders was observed 1–2 years post-injury, and it decreased over time only to show another maximum in the long-term phase in patients with brain injury occurring ≥5 years prior to assessment. Gonadotropic and somatotropic insufficiencies were most common. In the subgroup from 1 to 2 years after brain injury (n = 126), gonadotropic insufficiency was the most common hormonal disturbance (19%, 12/63 men) followed by somatotropic insufficiency (11.5%, 7/61), corticotropic insufficiency (9.2%, 11/119), and thyrotropic insufficiency (3.3%, 4/122). In patients observed ≥ 5 years after brain injury, the prevalence of somatotropic insufficiency increased over time to 24.1%, whereas corticotropic and thyrotrophic insufficiency became less frequent (2.5% and 0%, respectively). The prevalence differed regarding the diagnostic criteria (laboratory values vs. physician`s diagnosis vs. stimulation tests). Our data showed that neuroendocrine disturbances are frequent even years after TBI or SAH, in a cohort of patients who are still on medical treatment.

Introduction

Hypopituitarism is a common complication after brain injuries such as traumatic brain injury (TBI) and subarachnoid hemorrhage (SAH).^1

–6 In the past years, the prevalence of hypopituitarism after brain injury was investigated in several studies. In a metaanalysis of 13 studies with 911 patients after TBI or SAH, Schneider and coworkers estimated a prevalence of 27.5% in patients with TBI and an even higher prevalence of 47% in patients with SAH.³

Because of the high prevalence rates, it is recommended that a neuroendocrine assessment should be performed “even in patients who had had trauma many years before;” however, little is known about long-term anterior pituitary insufficiency after TBI or SAH.⁷

Although there are considerable data on prevalence rates and risk factors for pituitary insufficiencies after TBI and SAH in general, the observation period in most studies extended only to the immediate post-acute phase after brain injury, which often implies a period of 3 months after trauma.⁸ Several studies investigated patients in the chronic phase, although chronic is defined as >5 months after trauma in these studies. Only a few studies aimed to look at anterior pituitary deficiency 12 months after brain injury.^4

–9 Even fewer studies investigated patients after a longer period of time; that is, after >12 months. In these studies, hypopituitarism seems to be as common in the long-term phase as 12 months after brain injury. Concerning TBI, prevalence rates of hypopituitarism from 24% to 28.4% have been reported from 17 months up to 5 years after brain injury.^10
–12 In these studies, hypogonadism and growth hormone deficiency (GHD) were the most common disturbances, with a prevalence of 17% and 8%, respectively. These first insights into long-term anterior pituitary insufficiency indicated that there might be rather high prevalence rates even years after brain injury. On the other hand, low prevalence rates of GHD have been reported in a nationwide Danish study on 439 patients 1–4 years after TBI.¹³ Only 1% of all patients had GHD when GHD was diagnosed with two different stimulation tests.

Concerning SAH, first insights showed even higher prevalence rates than in TBI. Pituitary insufficiency affecting at least one pituitary axis was diagnosed in 47–55% in patients 12–72 months or 12–24 months after SAH.^14,15 Karaca and coworkers reported neuroendocrine dysfunction 3 years after SAH in 20 patients.¹⁶ In this prospective analysis, the authors describe GHD to be the most common disorder after 1 year (25%), as well as after 3 years (20%). Interestingly, new-onset deficiencies after 3 years were observed. Recently, these high prevalence rates in long-term patients after SAH have been questioned.^17,18

Hence, there is a vigorous debate on the clinical importance of post-traumatic hypopituitarism; for example, Karaca and coworkers concluded that pituitary function had to be assessed 1 year as well as 3 years after SAH because of the high prevalence. On the other hand, Klose and coworkers “question the evidence for newly introduced recommendations for routine pituitary assessment in TBI.”¹³

Therefore, the question arises if patients with a history of brain injury many years ago should be tested endocrinologically, and to what extent. In an effort to contribute to this discussion on the prevalence and clinical importance of pituitary dysfunction years after brain injury, we present data of our cohort of long-term patients after TBI and SAH, the German Database on Hypopituitarism.

Using the Structured Data Assessment of Hypopituitarism after TBI and SAH including 1242 patients,¹⁹ Renner and coworkers already demonstrated that gender did not influence outcome in patients after TBI.²⁰ Neuroendocrine dysfunction in the acute and chronic phases was evaluated by Schneider and coworkers; however, the chronic phase was only defined as at least 5 months after trauma.²¹ The subgroup of chronic patients has not been further divided into subgroups. However, especially in patients ≥5 years after brain injury, data on endocrine dysfunction is very rare. The very few reports point toward a high prevalence of neuroendocrine dysfunction in these patients.^22
–24

In our recent study, we investigated long-term anterior pituitary insufficiencies by analyzing a subgroup of 351 patients at least 1 to ≥5 years (range 1–55 years) after TBI or SAH. The prevalence rates, distribution, and main characteristics of neuroendocrine disturbances were determined for different periods of time, to address the question of their clinical importance and relevance.

Methods

Structured data assessment

Neurological, neurosurgical, and endocrinological centers in Germany (n = 13) and Austria (n = 1) participated from 2005 to 2008 in the Structured Data Assessment of Hypopituitarism after TBI and SAH, including a total number of 1242 patients. Patients received medical treatment in these centers. Data were collected using a structured, internet-based study sheet, involving information on clinical, radiological, and hormonal parameters as well as sociodemographic values. It was allowed to include patients retrospectively or to evaluate hormonal status prospectively during the study period up until November 2008. Patients were entered into the Access-based database online after pseudonymization. All patients or their legal representatives gave written informed consent. The study was approved by the ethics committee of the Bavarian State Chamber of Physicians in Munich, and complied with the Declaration of Helsinki.

Details on methodology were described previously by Kreitschmann-Andermahr and coworkers.¹⁹

Subjects

We selected a subgroup of patients in the Structured Data Assessment of Hypopituitarism after TBI and SAH database whose time interval between brain injury and their endocrine assessment was at least 1 year long. This criterion was met by 351 patients (mean age at injury 40 ± 17 years, range: 3–86 years) after TBI (245) or aneurysmal SAH (106) out of a total number of 1242 patients. Our subgroup of patients was mainly recruited in endocrinological departments (n = 240), followed by neurological departments (n = 64) and neurosurgical departments (n = 47).

Subject characteristics including sex, age, body mass index (BMI), and Glasgow Outcome Scale (GOS) are shown in Table 1. The GOS indicates the outcome after TBI or SAH (I, death; II, persistent vegetative state; III, severe disability (conscious but disabled); IV, moderate disability (disabled but independent); V, good recovery).²⁵

Table 1.

Subject Characteristics Depending Upon the Time Since Injury

	All n = 351	1-2 y n = 126	2-3 y n = 77	3-5 y n = 59	> 5 y n = 89
Sex (male/female)	192/159	65/61	41/36	35/24	51/38
Age^a mean ± SD	39.9 ± 17.2	43.3 ± 17.1	43.7 ± 16.5	43.9 ± 14.3	28.8 ± 14.9
Age^b mean ± SD	45.0 ± 15.8	44.7 ± 17.0	46.1 ± 16.8	47.5 ± 14.4	42.6 ± 13.6
BMI mean ± SD	26.4 ± 5.4	26.4 ± 4.9	25.4 ± 4.2	27.5 ± 4.9	26.3 ± 6.7
	n = 270	n = 97	n = 51	n = 48	n = 74
GOS mean ± SD	4.6 ± 0.7	4.6 ± 0.7	4.6 ± 0.6	4.6 ± 0.7	4.5 ± 0.6
	n = 301	n = 115	n = 70	n = 55	n = 61

At injury.

At hormonal evaluation.

BMI, body mass index; GOS, Glasgow Outcome Scale.

The severity of TBI was classified following the Glasgow Coma Scale (GCS)-based TBI severity classification.^26
–28 Mild TBI was indicated by a GCS 13–15. Patients with a GCS 9–12 were classified as having a moderate TBI whereas patients with a GCS 3–8 had a severe TBI.

In the SAH group, the severity of brain injury was indicated by the Hunt and Hess classification as well as by the radiological Fisher classification.^29
–31 Patients with a Hunt and Hess score I are asymptomatic, or have minimal headache or mild nuchal rigidity. Hunt and Hess score II means moderate or severe headache, nuchal rigidity, or cranial nerve palsy. Hunt and Hess score III is assumed when drowsiness, confusion, or a mild focal deficit occurs. Patients are classified with a Hunt and Hess score IV when they experience stupor, moderate to severe hemiparesis, and vegetative disturbances. Hunt and Hess score V categorizes patients in a deep coma or decerebrate rigidity.

The Fisher score reflects the blood amount in the first cerebral CT (CCT) scan. In detail, Fisher score I indicates the absence of subarachnoid blood. Fisher score II means that there is diffuse blood or a vertical layer <1 mm. Patients with Fisher score III show a dense collection of blood that appears to represent a clot >1 mm in thickness in the vertical plane, or >5 × 3 mm in longitudinal and transverse dimensions in the horizontal plane. The highest Fisher score, IV, implies intracerebral or intraventricular clots in the CCT scan.

Detailed subject characteristics of the TBI and SAH group, respectively, are displayed in Table 2.

Table 2.

Subject Characteristics Depending Upon the Type of Brain Injury

	TBI n (%)	SAH n (%)
All patients	245 (100%)	106 (100%)
Sex (male/ female)	165 (67%)/80 (33%)	27 (25%)/79 (75%)
Traffic accidents	130 (53.1%)
Falls	73 (29.8%)
Direct blows to the head	20 (8.2%)
Different causes	13 (5.3%)
Unknown	9 (3.7%)
Missing	0
Mild TBI	51 (20.8%)
Moderate TBI	6 (2.4%)
Severe TBI	37 (15.1%)
Missing	151 (61.6%)
Coma/loss of consciousness	128 (52.2%)
No coma/loss of consciousness	52 (21.2%)
Unknown	64 (26.1%)
Missing	1 (0.004%)
Diffuse axonal injury	11/104 (10.6%)
Traumatic SAH	61/137 (44.5%)
Basal skull fracture	46/135 (34.1%)
Sella turcica fracture	5/115 (4.3%)
Skullcap fracture	46/131 (35.1%)
Hemorrhage from concussion
Occipital	13/121 (10.7%)
Temporoparietal	47/124 (37.9%)
Frontal	47/131 (35.9%)
ACommA		34 (32.1%)
PCommA		10 (9.4%)
MCA		19 (17.9%)
Basilar artery		6 (5.7%)
ICA		5 (4.7%)
Other		28 (26.4%)
Missing		4 (3.8%)
Neurosurgical intervention		79 (74.5%)
Neuroradiological intervention		37 (34.9%)
Hunt & Hess I		26 (24.5%)
Hunt & Hess II		23 (21.7%)
Hunt & Hess III		25 (23.6%)
Hunt & Hess IV		8 (7.5%)
Hunt & Hess V		10 (9.4%)
Missing data		14 (13.2%)
Fisher I		4 (3.8%)
Fisher II		7 (6.6%)
Fisher III		17 (16.0%)
Fisher IV		27 (25.5%)
Missing data		51 (48.1%)
Brain edema		34/92 (37.0%)
Cerebral vasospasm		35/77 (45.5%)
Vasospasm-related infarction		13/77 (16.9%)
Hydrocephalus		27/88 (30.7%)
Delayed ischemic neurological deficits		8/75 (10.7%)
SIADH		1/48 (2.1%)
(Transient) diabetes insipidus		6/61 (9.8%)

ACommA, anterior communicating artery; ICA, internal carotid artery; MCA, middle cerebral artery; PCommA, posterior communicating artery; TBI, traumatic brain injury; SAH, subarachnoid hemorrhage; SIADH, syndrome of inappropriate antidiuretic hormone secretion.

Hormonal assessments

Hormone levels were measured in the local laboratories of the participating centers and recorded in the study sheets. Reference values differed from center to center depending upon the specific test applied and the local laboratory value. Although it was recommended to investigate all pituitary axes, endocrine assessment was in some cases incomplete. The following definitions of hormone disturbances (deficiencies or excess) were used.

• Corticotropic axis: basal cortisol below or above the local laboratory reference value; for example, <62 μg/L and >194 μg/L, <5 μg/dL and >25 μg/dL, or <171 nmol/L and >536 nmol/L, respectively; the stimulation tests used for testing the integrity of the corticotropic axis were the insulin tolerance test (ITT), the short adrenocorticotrophic hormone (ACTH) test, or the corticotropin-releasing hormone (CRH) test. Corticotropic deficiency was defined as peak cortisol in the stimulation test <500 nmol/L (<180 μg/L). Performing the Metopiron test or the glucagon test was also allowed.

• Gonadotropic axis in men: basal testosterone below or above the local laboratory reference value; for example, <9.9 nmol/L and >27.8 nmol/L, or <2.8 μg/L and >8 μg/L; the gonadotropin releasing hormone (GnRH) test could be used as stimulation test, with an increase of luteinizing hormone (LH) <1.5–2-fold in men indicating gonadotropic insufficiency.

• Gonadotropic axis in women: estradiol in women below or above the local laboratory value (e.g., proliferative stage <13 ng/L and >166 ng/L, or <18.9 pg/mL and >247 pg/mL, respectively; ovulation <86 ng/L and >498 ng/L, or <35.5 pg/mL and >571 pg/mL, respectively; luteal phase <44 ng/L and >211 ng/L, or <22.4 pg/mL and >256 pg/mL, respectively; postmenopause <35 ng/L or <10 pg/mL, and >44.5 pg/mL, respectively, depending upon the local laboratory); the GnRH test could be used as a stimulation test with an increase of LH <20 U/L in women indicating gonadotropic insufficiency.

• Thyrotropic axis: free thyroxine (fT4) below or above the local laboratory value; for example, <12 pmol/L and >22 pmol/L, or <0.93 ng/dL and >1.7 ng/dL; the thyrotropin-releasing hormone (TRH) test could be performed as stimulation test (thyroid stimulating hormone [TSH] increase <2.0 μU/mL indicating thyrotropic deficiency).

• Somatotropic axis: insulin-like growth factor (IGF)-1 level below or above the local age- and sex-specific reference value if no stimulation test was performed; as possible stimulation tests for the growth hormone (GH) axis, we allowed the ITT using a GH cutoff of <3 μg/L (0.14 nmol/L) and the growth hormone releasing hormone (GHRH)-L-arginine test using BMI dependent cutoffs (< 4 μg/L for BMI <25; < 8 μg/L for BMI 25–29.9; < 11.5 μg/L for BMI ≥30, respectively).³² Further, performing the glucagon test to investigate the integrity of the somatotropic axis was allowed.

• Lactotropic axis: prolactin level below or above the local sex-specific reference values.

Criteria for the diagnosis of pituitary insufficiency

In order to compare our results with previous reports, three different criteria were used to assess hypothalamo-pituitary insufficiency.²¹ Lowered basal hormonal values were defined as criterion “A,” analogous to Schneider and coworkers.²¹ Additionally, the treating physicians were asked to classify each pituitary axis as sufficient or deficient, based on basal hormonal values, additional information (e.g., clinical symptoms and signs) and clinical judgment (criterion “B”). A standardized questionnaire for the medical history was not included; physicians were allowed to diagnose pituitary insufficiency themselves according to the criteria previously mentioned. The rationale to establish this criterion was the knowledge that lowered laboratory values do not automatically imply pituitary dysfunction.³³ Lowered laboratory values are only indicative of hypopituitarism, whereas criterion “B” allows a more holistic view in diagnosing hypopituitarism by the treating physician.

Further, stimulation tests were applied for the diagnosis of hypopituitarism (criterion “C”).

Statistical analyses

The Structured Data Assessment of Hypopituitarism after TBI or SAH database and data administration were conducted by ANFOMED^® (Möhrendorf, Germany).

Data were analyzed for the subgroup of patients who had had TBI or SAH >1 year prior to endocrine assessment. This group was further divided into four subgroups, depending upon the time after brain injury: 1) patients with brain injury ≥1 to <2 years previously (“1–2 years”), 2) patients with brain injury ≥2 to <3 years earlier (“2–3 years”), 3) patients with brain injury ≥3 to <5 years earlier (“3–5 years”), and 4) patients with brain injury ≥5 years earlier (“> 5 years”). Data were analyzed separately for each subgroup. Data were analyzed both for the total group and for each subgroup separately. Differences among several subgroups regarding the subjects' characteristics were analyzed using χ² test or analysis of variance (ANOVA). Differences in hormonal disturbances between the TBI group and the SAH group were analyzed using χ² test or Fisher`s exact test. A subgroup analysis was performed in patients after TBI and SAH, regarding the severity of injury assessed by Fisher score and Hunt and Hess score in patients after SAH and the initial GCS (mild, moderate, and severe TBI) in patients after TBI. The correlation between the severity of injury and the number of impaired pituitary axes (0, 1, or ≥2) was analyzed using the Cochran–Mantel–Haenszel test, which is applied for the comparison of two ordinal scaled variables. The relationship between the severity of injury and the incidence of impaired pituitary axes was indicated by the Kendall rank correlation coefficient (Kendall`s τ-b coefficient).

Missing variables were treated as missing at random. All statistical analyses were performed with IBM SPSS Statistics 17 and 22 (IBM Corporation, NY).

Results

Subject characteristics

Patients were divided into four subgroups depending upon their time since injury: 1–2 years (n = 126), 2–3 years (n = 77), 3–5 years (n = 59), and >5 years (n = 89). The mean time between brain injury and endocrine assessment in the group of long-term patients >5 years after brain injury was 13.7 ± 11.5 years with a maximum of 55 years. In most of these patients (60.7%), brain injury had occurred 5–10 years earlier. Further subject characteristics are displayed in Table 1. No statistical differences were observed in sex, age at hormonal evaluation, BMI, or GOS among the four different subgroups. However, the long-term patients in the group >5 years after brain injury were significantly younger at the time of injury (p < 0.0004).

Similar to the distribution of the GCS in all TBI patients in the database,¹⁹ a bimodal GCS distribution was observed for the TBI patients >1 year after trauma, with most patients having a GCS of 3 (n = 17; 7%) on the one hand, or a GCS of 14 (n = 19; 7.8%) or 15 (n = 26; 10.6%) on the other hand. Detailed subject characteristics for the TBI and SAH group are shown in Table 2.

Basal clinical data were available for all patients. Endocrine assessment was performed in 341 patients at a mean age of 43 ± 17 years (range: 12–87 years) in TBI patients and at a mean age of 50 ± 12 years (range: 23–80 years) in SAH patients.

Basal hormonal assessment

Basal hormonal assessment was used to define pituitary insufficiency according to criterion “A.” As described, hormonal testing was missing in 10 cases, hence only 341 patients were included in our analysis. After quality control and plausibility check of the entered data, 318 cortisol levels, 333 fT4 levels, 180 testosterone levels in men, and 94 estradiol levels in women, 314 prolactin levels and 320 IGF-I levels remained. Data were excluded mostly for the following reasons: the local laboratory value was not clearly defined, the ovulation phase in women was not documented, or data were missing because of incomplete hormonal or data assessment, respectively. In general, data were not included in our analysis if a clear classification of the basal hormonal level was not possible.

In 154 men and in 83 women, hormonal levels of all hypothalamo-pituitary axes were available. Regarding these 237 patients with complete data of all hormonal axes, neuroendocrine disturbances in patients >1 year after TBI or SAH were present in 44.7% (106 patients). More than half of all patients (131, 55.3%) showed normal endocrine values. Predominantly, only one pituitary axis was affected (87; 36.7%). Disturbances in two axes were diagnosed in 17 patients (7.2%). Neuroendocrine disturbances in three or more pituitary axes were rarely diagnosed (2, 0.8%).

Considering all patients, including those with incomplete hormonal assessment of the hypothalamo-pituitary axes, most subjects showed lowered testosterone (26/180 men, 14.4%) or lowered estradiol values (10/93 women, 10.8%), respectively, as well as lowered IGF-I values (30/218, 13.8%) possibly indicating that gonadotropic and somatotropic insufficiencies are the most common disturbances in long-term patients. Lowered cortisol values were observed in 7.2% (23/318 patients), lowered fT4 values were observed in 3.3% (11/334 patients). Interestingly, elevated cortisol levels were observed in 11% of all patients. Elevation of other pituitary hormones were less common: elevated prolactin levels were documented in 4.5%, elevated testosterone was documented in 4.4%, elevated fT4 levels were documented in 2.4%, elevated IGF-I was documented in 2.3% and elevated estradiol was documented in 1.1%.

Hormonal disturbances depending upon the time since injury

Depending upon the time since injury, neuroendocrine disturbances were most common in the subgroup from 1 to 2 years after brain injury and in the group of patients whose brain injury occurred >5 years earlier.

In the subgroup from 1 to 2 years (n = 126), lowered testosterone value in men was the most common endocrinological disturbance (12/63, 19%), followed by elevated cortisol levels (21/119, 17.6%), lowered estradiol levels (4/34, 11.8%), lowered IGF-I levels (7/61, 11.5%) and lowered cortisol levels (11/119, 9.2%). Other disturbances of the hormonal axes were less common (see Fig. 1).

FIG. 1.

Long-term anterior pituitary insufficiencies. Impaired pituitary axes with lowered basal hormonal values depending upon time after traumatic brain injury (TBI) or subarachnoid hemorrhage (SAH); in this analysis, patients with single and multiple impaired axes were included. y, years.

After 5 years, lowered IGF-I values were still observed in 24.1% (13/54) patients; 5/21 women (23.8%) showed lowered estradiol values and 8/49 men (16.3%) showed lowered testosterone values, respectively. Figure 1 gives an overview on the distribution of impaired pituitary axes depending upon time after TBI or SAH.

Hormonal disturbances depending on the severity of injury

In patients with SAH, there was no statistically significant correlation between the number of impaired pituitary axes and the severity of injury assessed by the Fisher score (p = 0.141, Kendall`s τ-b = 0.135). Similar results were obtained when the Hunt and Hess score was applied for the severity of injury (p = 0.788, Kendall`s τ-b = 0.823). To assess the severity of TBI, patients were divided into the subgroups of mild, moderate, and severe TBI according to the initial GCS score. However, the severity of TBI did not correlate with the number of impaired pituitary axes in our cohort of patients (p = 0.378, Kendall`s τ-b = 0.298).

Influence of medication

Because medication can affect hormonal values, relevant medication was registered in the study sheets. The treating physician was allowed to initiate a hormone replacement therapy in consultation with an endocrinologist if necessary.

In our study cohort, 47 patients received hydrocortisone replacement therapy. Eight out of 47 patients showed lowered values even under hormone replacement therapy, 5 patients showed elevated cortisol levels, and 34 patients showed normal cortisol levels.

The thyrotropic axis was altered in 36 patients receiving levothyroxine therapy and in 2 patients who were treated with methimazole. Three patients showed lowered fT4 values, in 1 patient elevated fT4 levels were measured, and 34 patients showed normal fT4 levels (including both patients receiving methimazole treatment).

Twenty-two women were treated with estrogens, but it is not apparent from the study sheet if the therapy was in the form of contraception or hormone replacement therapy. Data on testosterone replacement therapy are missing in the study sheet.

Dopaminergic therapy was administered to two patients. One of them showed lowered prolactin levels and the other one showed normal prolactin levels.

Nine patients were treated with growth hormone, and all of them showed normal IGF-I levels.

Regarding the laboratory values, it should be kept in mind that patients receiving hormone replacement therapy often showed normal laboratory values. Nevertheless, the reason for this therapy was an insufficiency. In an additional analysis, patients receiving hormone replacement therapy were classified as “insufficient” even though basal hormonal values were within the reference range. Based on this assumption, higher prevalence rates of neuroendocrine disturbances were apparent, especially in the corticotropic and thyrotropic axes. Disturbances (insufficiency or excess) of hormones of the corticotropic axis were present in 28.9%. Disturbances of the gonadotropic axis were assumed in 18.9% of all men and in 12.8% of all women. The thyrotropic axis was affected in 15.9%. In 14.7%, of patients, the somatotropic axis was disturbed. In 9% of all patients, disturbances of the lactotropic axis were detected.

Physician's diagnosis of hypopituitarism

Changes in basal hormonal levels do not always reflect post-traumatic hypopituitarism. Therefore, physicians had to classify patients additionally as hormone insufficient or not (criterion “B”) based on the hormonal level as well as on further clinical data and clinical symptoms and signs. Corticotropic insufficiency was diagnosed in 25.5%, followed by somatotropic insufficiency in 13.3%, gonadotropic insufficiency in 11.1%, and thyrotropic insufficiency in 7.2%.

Table 3 shows the distribution of anterior pituitary deficiency depending upon the time after brain damage. In the groups from 1 to 3 years after brain injury, hypopituitarism predominantly affected only one axis. Interestingly, there was an increase in multiple hormonal deficiencies in the group who had had TBI or SAH >3 years earlier, pointing toward a high diversity of impaired hormonal axes.

Table 3.

Impaired Single and Multiple Pituitary Axes Based on the Physician`s Diagnosis and Depending Upon Time after TBI or SAH

Hormonal insufficiencies	1-2y n (%)	2-3y n (%)	3-5y n (%)	≥5y n (%)
No disturbances	34 (53.1%)	28 (52.8%)	24 (40%)	44 (52.4%)
Single axis affected	21 (32.8%)	19 (35.8%)	18 (30%)	16 (19.0%)
• Corticotropic	11 (17.2%)	8 (15.1%)	13 (21.7%)	6 (7.1%)
• Gonadotropic	4 (6.3%)	3 (5.7%)	1 (1.7%)	7 (8.3%)
• Thyrotropic	4 (6.3%)	3 (5.7%)	1 (1.7%)	2 (2.4%)
• Somatotropic	2 (3.1%)	5 (9.4%)	3 (5%)	1 (1.2%)
Multiple axes affected	9 (14.1%)	6 (11.3%)	18 (30%)	24 (28.6%)
• Corticotropic + gonadotropic	2 (3.1%)	2 (3.8%)	1 (1.7%)	3 (3.6%)
• Corticotropic + thyrotropic	1 (1.6%)	0	3 (5.0%)	1 (1.2%)
• Corticotropic + somatotropic	3 (4.7%)	4 (7.5%)	5 (8.3%)	4 (4.8%)
• Gonadotropic + thyrotropic	0	0	2 (3.3%)	3 (3.6%)
• Gonadotropic + somatotropic	2 (3.1%)	0	1 (1.7%)	4 (4.8%)
• Thyrotropic + somatotropic	0	0	1 (1.7%)	2 (2.4%)
• Corticotropic + gonadotropic + thyrotropic	0	0	1 (1.7%)	1 (1.2%)
• Corticotropic + gonadotropic + somatotropic	1 (1.6%)	0	1 (1.7%)	2 (2.4%)
• Corticotropic + thyrotropic + somatotropic	0	0	1 (1.7%)	1 (1.2%)
• Gonadotropic + thyrotropic + somatotropic	0	0	1 (1.7%)	2 (2.4%)
• All axes	0	0	1 (1.7%)	1 (1.2%)
Total	64 (100%)	53 (100%)	60 (100%)	84 (100%)

Stimulation tests

In 190 patients, stimulation tests were performed to affirm the diagnosis of hypopituitarism (criterion “C”).

The most frequently used test to confirm hypocortisolism was the ITT (113 patients, 35.5%), followed by the ACTH test (57 patients, 17.9%) and the CRH test (22 patients, 6.9%), respectively. The Metopiron test and the glucagon test had not been applied in this observational study, although the treating physicians were allowed to perform them. The ITT revealed corticotropic insufficiency in 40.7% (46 out of 113 patients). In the ACTH test, corticotropic insufficiency was only diagnosed in 5.3% (3 out of 57 patients). Using the CRH test, corticotropic insufficiency was diagnosed in 45.5% (10 out of 22 patients).

Hypogonadism was mainly diagnosed by means of basal hormone levels and clinical parameters. Stimulation tests such as the GnRH test were only applied in 24 patients (7.5%). However, the results of the test were not recorded in the study sheet in our subgroup of patients.

Similarly to the gonadotropic axis, stimulation tests were rarely applied for diagnosing hypothyroidism. The TRH test was only performed in 25 patients (8.5%). Unfortunately, in our patient cohort, the results of this stimulation test were not recorded in the study sheet either.

To diagnose GHD, most physicians used stimulation tests. As the ITT is known as the gold standard for diagnosing GHD, it was the most commonly used stimulation test (105 patients) followed by the GHRH-L-arginine test in 67 patients.³⁴ GHD was diagnosed in 15.2% of patients with the ITT (16 out 105) and in 20.9% with the GHRH-L-arginine test (14 out of 67). The glucagon test has not been used in these patients.

Information about stimulation tests was missing in only 33 patients.

Differences in patients with TBI and SAH

For this analysis of differences between injury types, patients with basal hormonal levels outside the reference range as well as patients with normal hormonal levels under hormone replacement therapy were classified as patients with neuroendocrine disturbances. No significant differences in the prevalence of neuroendocrine disturbances were observed in patients after TBI or SAH, respectively.

About 29% TBI patients and 29.1% SAH patients showed disturbances of the corticotropic axis (p = 0.539, n.s.). Free thyroxine values outside the reference range or a hormone replacement therapy were documented in 16.1% patients after TBI and in 15.5% patients after SAH (p = 0.519, n.s.). Changes in prolactin levels did not differ statistically between patients with TBI (8.8%) and those with SAH (9.4%) (p = 0.511, n.s.). Further, gonadotropic insufficiency was indicated in 18.6% of men with TBI and in 21.6% of men with SAH (p = 0.409, n.s.). Although a higher prevalence of gonadotropic disturbances was documented in women after TBI (21.7%) than in women after SAH (12%), statistical significance was not reached (p = 0.139, n.s.). There was a nonsignificant trend (p = 0.061) toward a higher rate of somatotropic insufficiencies in the TBI group (17.3%) than in the SAH group (10%). However, statistical significance was not reached in any hypothalamo-pituitary axis regarding the different types of injury.

Diagnosis of hypopituitarism according to different diagnostic criteria

Diagnosis of pituitary insufficiency differed depending upon the different diagnostic criteria. Table 4 gives an overview of the prevalence of hypopituitarism according to criterion “A” (lowered basal hormonal values), criterion “B” (physician`s diagnosis), and criterion “C” (stimulation tests).

Table 4.

Pituitary Insufficiencies According to the Different Diagnostic Criteria

Criteria	Corticotropic (%)	Gonadotropic (%)	Thyrotropic (%)	Somatotropic (%)
Criterion A	7.3	14.4	3.3	13.8
Criterion B	25.5	11.1	7.2	13.3
Criterion C	25.4	n.a.	n.a.	20

Criterion A: lowered basal hormonal values.

Criterion B: physician`s diagnosis (lowered basal hormonal values, clinical symptoms and signs, and anamnesis).

Criterion C: stimulation test with lowered stimulated hormonal values.

n.a., not applicable, because the results of the stimulation tests for the gonadotropic and thyrotropic results were not recorded.

Discussion

Our data show that neuroendocrine disturbances can be observed in the long-term phase even years after TBI and SAH. Analyzing a group of 351 patients from the Structured Data Assessment of Hypopituitarism after TBI and SAH and their basal hormonal laboratory values, we could demonstrate that neuroendocrine disturbances (i.e., insufficiency or excess) were documented in 44.7% of all patients >1 year after brain injury with complete endocrine assessment (n = 237). In these patients, predominantly one pituitary axis was affected (36.7%). Considering single laboratory values in 341 patients, most patients showed lowered sex hormones (lowered testosterone in 14.4% or lowered estradiol values in 10.8%, respectively) as well as lowered IGF-I values (13.8%), possibly indicating that gonadotropic and somatotropic insufficiencies are the most common disturbances in long-term patients.

Interestingly, the pattern of neuroendocrine insufficiencies changed over time. In the subgroup from 1 to 2 years, gonadotropic insufficiency (19% in men, 11.8% in women) was most commonly followed by somatotropic insufficiency (11.5%). After >5 years, somatotropic insufficiency became the most frequent disorder (24.1%) followed by gonadotropic insufficiency (23.8% in women, 16.3% in men). Corticotropic insufficiency became less frequent, with a decrease from 9.2% to 2.5%. The prevalence of thyrotropic insufficiency fluctuated over time, and was rarely observed (3.3% in all patients). The subgroup of patients 3–5 years after brain injury show a different insufficiency profile compared to the trend over the other groups. This could possibly be the result of the lower number of patients and, therefore, the lower statistical power in this group. Apart from this group, a clear trend of increasing insufficiency is only apparent for the somatotropic axis.

One might assume that pituitary dysfunction caused by brain injury exclusively results in pituitary insufficiency. Therefore, it might be confusing that we reported also on elevated laboratory values. In fact, disturbed pituitary axes with, for example, elevated stress hormone levels have been reported in patients during neurorehabilitation.³⁵ This does not reflect pituitary dysfunction per se, but is mainly a response to chronic stress. Hence, the treating physician should be careful about the diagnosis of pituitary dysfunction based only on laboratory values.

Strictly speaking, changes in basal hormonal levels do not imply post-traumatic hypopituitarism as a matter of course.³³ Therefore, we used three different criteria for the diagnosis of post-traumatic hypopituitarism. In addition to the laboratory value (criterion “A”), the treating physician had to classify patients as hormone insufficient or not (criterion “B”). Applying this criterion, corticotropic insufficiency was the most frequent disturbance (25.5%), followed by somatotropic insufficiency (13.3%), gonadotropic insufficiency (11.1%), and thyrotropic insufficiency (7.2%). At this point, the gap between the laboratory values alone and the diagnosis of hypopituitarism is evident; for example, lowered cortisol levels were documented in 7.3% of all patients, but hypocortisolism was diagnosed by the treating physician in 25.5%. Interestingly, dynamic endocrine testing is recommended for the diagnosis of corticotropic and somatotropic insufficiencies.³⁶ We assume that the treating physicians classified their patients as insufficient, being aware of an impaired test result despite normal basal hormonal values. On the other hand, lowered testosterone values were observed in 14.4% of all men, but the diagnosis of hypogonadism was made in only 11.1%. A possible reason for this discrepancy might be that chronic stress leads to hypercortisolism.^37,38 However, hypercortisolism itself reduces gonadotropin level, thus mimicking post-traumatic hypopituitarism.³⁹ Additional information on dynamic endocrine testing and the more holistic view on all laboratory results might cause the discrepancy between criterion “A” (basal laboratory values) and criterion “B” (physician`s diagnosis).

Another important finding regarding criterion “B” (opinion of the treating physician) is that the number of impaired pituitary axes was higher with increasing time span post-injury. Multiple hormonal deficiencies are more frequent in the group with TBI or SAH occurring >3 years prior to assessment, with the subgroup of patients whose brain injury occurred >5 years before assessment showing a highly diverse set of impaired hormonal axes.

As expected, stimulation tests (criterion “C”) were most often used for the corticotropic and somatotropic axes. GHD was more often diagnosed with the GHRH-L-arginine test (20.9%) than with the ITT (15.2%). Interestingly, the ITT revealed corticotropic insufficiency in ∼40.7%. This shows that stimulation tests are needed in particular for the corticotropic and somatotropic axes in order to diagnose hypocortisolism or GHD, respectively. Another interesting finding is that stimulation tests such as the Metopiron test and the glucagon test had not been applied in our patient sample.

Although previous reports assume higher prevalence rates of post-traumatic hypopituitarism in SAH than the prevalence rates reported in TBI, no statistical difference between these injury types was observed in our study.^3,14
–16

It might be somewhat surprising that there was no correlation between the severity of injury and the number of impaired pituitary axes in our analyses. Previous data report on the association of post-traumatic hypopituitarism with the severity of brain injury.^3,23,40 This assumption has been generally accepted; therefore, most recommendations on the screening of patients after brain trauma rely on the fact that patients with moderate to severe TBI are particularly at risk for neuroendocrine disorders.⁴¹ However, similar to our results, other studies could not find that the severity of TBI is predictive for the presence of hypopituitarism.^10,42,43 It should be emphasized that there was a preselection of patients, because only patients in medical treatment even years after brain trauma were included into the study. The parameters indicating the severity of injury (Fisher score, Hunt and Hess score, initial GCS score) were assessed on the 1st day of the patient`s injury; however, a subsequent worsening might affect patients even more. Our patients were in medical treatment even years after their injury. This indicates that they were still severely affected, independent of their initial score.

Our study cohort is part of the cohort of Schneider and coworkers, who investigated patients after >5 months following brain injury.²¹ However, we focus on long-term pituitary insufficiencies and give a more detailed overview on the pattern of endocrine disturbances and their development over time in the present study.

There are several limitations of our study. First, the prevalence rates might be overestimated, because only patients in medical treatment were included in the data assessment, and patients without symptoms were probably under-represented. Because of the character of the database and the study, the prevalence rates evaluated with a longitudinal observational study design investigating all patients after a defined period of time after brain injury might have been probably lower. Therefore, these results cannot be directly compared with a screening program of all patients after TBI or SAH. Most patients were recruited from endocrinological departments, which might provide a selection bias, because only neurological patients suspected of hormonal disturbances were referred to an endocrinologist. Nevertheless, our data reflect the daily routine of the treating physicians and are, therefore, representative for clinical routine. In fact, neuroendocrine disturbances have been observed not only after TBI or SAH, but also after other types of brain damage such as stroke.⁴⁴ In the present study, only patients who had had TBI and SAH were included. Unfortunately, the severity of TBI was not known in the majority of patients. Further, endocrine diagnostic assessment was performed at the local laboratory of each center. Because of different cutoffs and different methods, the test results might differ from center to center. Several data had to be excluded during the quality check process because, for example, local laboratory values were not clearly defined, or the menstrual cycle in women was not documented. Although recommended, not all pituitary axes had been checked in every patient. Nevertheless, we investigated 341 patients with partial hormonal assessment and 237 patients with complete hormonal assessment.

The pathophysiology of chronic hypopituitarism after brain injury is not completely understood. Dusick and coworkers discussed, in addition to the primary brain injury itself, secondary insults, the stress of critical illness, and medication effects as possible reasons for post-traumatic hypopituitarism.⁴⁵ Because the pituitary gland of TBI patients is enlarged in the acute phase and shows a loss of volume or perfusion deficits in the chronic phase especially in patients with clinical hypopituitarism,^46,47 a degenerative process might occur and enhance late-onset hypopituitarism after brain injury.

Recently, the question of diagnostic criteria, especially the necessity of confirmation tests, was raised. Two studies reported significantly lower prevalence rates in patients after TBI (< 1%; 1/107) and SAH (0%; 0/62).^17,48 Both studies required a stimulation test to confirm the diagnosis of hypopituitarism after brain injury. Further, it was shown that the prevalence of GHD decreased significantly after repeated stimulation testing.¹³ In our study, repeated stimulation testing was not envisaged. But it is evident from our study that stimulation tests are necessary for the diagnosis of hypocortisolism or GHD. Our study protocol did not imply a general screening for hypopituitarism after brain injury, but reflects the clinical routine.

Conclusion

Patients who are still in need of medical treatment even years after TBI or SAH are at risk for neuroendocrine disturbances. All pituitary axes should be assessed, with special regard to gonadotropic and somatotropic axes, which were the most frequently disturbed. Stimulation tests should be performed in patients with suspicion of GHD or hypocortisolism. It is important to highlight that clinical signs of neuroendocrine dysfunction in combination with a history of brain injury should lead to hormonal assessment of the hypothalamo-pituitary axes, even many years after brain injury.

Footnotes

Acknowledgments

The authors are grateful to all patients who contributed to the Structured Data Assessment of Hypopituitarism after TBI and SAH and to all centers that included patients in the registry. G.K.S., I.K.A., E.K., M.B., and E.U. were part of the advisory board of the database. The following investigators recruited patients to our subgroups of the database: M.F., University Hospital Cologne (n = 65); A.K. and M.S., Schön Klinik Bad Aibling (n = 53); H.J.S. and G.K.S., Max-Planck-Institute of Psychiatry (n = 52); Dieter-Karsten Böker, University Hospital Giessen (n = 48); Ulrich Tuschy, Helios Hospital Erfurt (n = 42); I.K.A., University Hospital Aachen (n = 41); C.B., University of Essen (n = 17); H.W., University Hospital Greifswald (n = 16); C.R., NRZ Neurological Rehabilitation Center, University of Leipzig (n = 7); E.U., Hospital Klagenfurt (n = 6); Diethard Steube, Neurological Clinic Bad Neustadt/Saale (n = 3); and Dr. Ralf Sudhoff, Fachklinik Enzensberg (n = 1). We thank Kristin Lucia for proofreading.

Author Disclosure Statement

A.K. received a travel grant from Ipsen. H.J.S. received research and travel grants and speaker fees from Pfizer, and travel grants from Lilly and Sandoz. I.K.A. is a member of the German PATRO^® board. C.K., M.S., M.B., M.F., C.B., H.W., E.U., E.K., C.R., M.J., and G.K.S. have nothing to disclose. The Structured Data Assessment of Hypopituitarism after TBI and SAH was supported financially by an independent investigator grant from Pfizer GmbH Germany to GKS, Clinical Neuroendocrinology Group, Max Planck Institute (MPI) of Psychiatry, Munich, the details of which are outlined in a contract between the MPI Munich and the pharmaceutical company. The grant covers the setup of the database. Pfizer has no access to the data included in the database and does not have the right to analyze the data independently.

References

Schneider

H.J.

, Aimaretti

, Kreitschmann-Andermahr

, Stalla

G.K.

, and Ghigo

(2007). Hypopituitarism. Lancet, 369, 1461–1470.

Berg

, Oeffner

, Schumm–Draeger

P.M.

, Badorrek

, Brabant

, Gerbert

, Bornstein

, Zimmermann

, Weber

, Broecker–Preuss

, Mann

, and Herrmann

B.L.

(2010). Prevalence of anterior pituitary dysfunction in patients following traumatic brain injury in a German multi-centre screening program. Exp. Clin. Endocrinol. Diabetes, 118, 139–144.

Schneider

H.J.

, Kreitschmann–Andermahr

, Ghigo

, Stalla

G.K.

, and Agha

(2007). Hypothalamopituitary dysfunction following traumatic brain injury and aneurysmal subarachnoid hemorrhage: a systematic review. JAMA, 298, 1429–1438.

Schneider

H.J.

, Schneider

, Saller

, Petersenn

, Uhr

, Husemann

, von Rosen

, and Stalla

G.K.

(2006). Prevalence of anterior pituitary insufficiency 3 and 12 months after traumatic brain injury. Eur. J. Endocrinol., 154, 259–265.

Tanriverdi

, Senyurek

, Unluhizarci

, Selcuklu

, Casanueva

F.F.

, and Kelestimur

(2006). High risk of hypopituitarism after traumatic brain injury: a prospective investigation of anterior pituitary function in the acute phase and 12 months after trauma. J. Clin. Endocrinol. Metab., 91, 2105–2011.

Fernandez–Rodriguez

, Bernabeu

, Castro

A.I.

, and Casanueva

F.F.

(2015). Hypopituitarism after traumatic brain injury. Endocrinol. Metab. Clin. North Am., 44, 151–159.

Gasco

, Prodam

, Pagano

, Grottoli

, Belcastro

, Marzullo

, Beccuti

, Ghigo

, and Aimaretti

(2012). Hypopituitarism following brain injury: when does it occur and how best to test?. Pituitary, 15, 20–24.

Kokshoorn

N.E.

, Wassenaar

M.J.

, Biermasz

N.R.

, Roelfsema

, Smit

J.W.

, Romijn

J.A.

, and Pereira

A.M.

(2010). Hypopituitarism following traumatic brain injury: prevalence is affected by the use of different dynamic tests and different normal values. Eur. J. Endocrinol., 162, 11–18.

Giordano

, Aimaretti

, and Ghigo

(2005). Variations of pituitary function over time after brain injuries: the lesson from a prospective study. Pituitary, 8, 227–231.

10.

Agha

, Rogers

, Sherlock

, O'Kelly

, Tormey

, Phillips

, and Thompson

C.J.

(2004). Anterior pituitary dysfunction in survivors of traumatic brain injury. J. Clin. Endocrinol. Metab., 89, 4929–4936.

11.

Herrmann

B.L.

, Rehder

, Kahlke

, Wiedemayer

, Doerfler

, Ischebeck

, Laumer

, Forsting

, Stolke

, and Mann

(2006) Hypopituitarism following severe traumatic brain injury. Exp. Clin. Endocrinol. Diabetes, 114, 316–321.

12.

Leal–Cerro

, Flores

J.M.

, Rincon

, Murillo

, Pujol

, Garcia–Pesquera

, Dieguez

, and Casanueva

F.F.

(2005). Prevalence of hypopituitarism and growth hormone deficiency in adults long-term after severe traumatic brain injury. Clin Endocrinol (Oxf), 62, 525–532.

13.

Klose

, Stochholm

, Janukonyté

, Lehman Christensen

, Frystyk

, Andersen

, Laurberg

, Christiansen

J.S.

, and Feldt–Rasmussen

(2014). Prevalence of posttraumatic growth hormone deficiency is highly dependent on the diagnostic set-up: results from The Danish National Study on Posttraumatic Hypopituitarism. J. Clin. Endocrinol. Metab., 99, 101–110.

14.

Kreitschmann–Andermahr

, Hoff

, Saller

, Niggemeier

, Pruemper

, Hutter

B.O.

, Rohde

, Gressner

, Matern

, and Gilsbach

J.M.

(2004). Prevalence of pituitary deficiency in patients after aneurysmal subarachnoid hemorrhage. J. Clin. Endocrinol. Metab., 89, 4986–4992.

15.

Dimopoulou

, Kouyialis

A.T.

, Tzanella

, Armaganidis

, Thalassinos

, Sakas

D.E.

, and Tsagarakis

(2004). High incidence of neuroendocrine dysfunction in long-term survivors of aneurysmal subarachnoid hemorrhage. Stroke, 35, 2884–2889.

16.

Karaca

, Tanriverdi

, Dagli

A.T.

, Selcuklu

, Casanueva

F.F.

, Unluhizarci

, and Kelestimur

(2013). Three years prospective investigation of pituitary functions following subarachnoid haemorrhage. Pituitary, 16, 76–82.

17.

Klose

, Brennum

, Poulsgaard

, Kosteljanetz

, Wagner

, and Feldt–Rasmussen

(2010). Hypopituitarism is uncommon after aneurysmal subarachnoid haemorrhage. Clin. Endocrinol. (Oxf). 73, 95–101.

18.

Hannon

M.J.

, Behan

L.A.

, O'Brien

M.M.

, Tormey

, Javadpour

, Sherlock

, and Thompson

C.J.

(2015). Chronic hypopituitarism is uncommon in survivors of aneurysmal subarachnoid haemorrhage. Clin. Endocrinol. (Oxf). 82, 115–121.

19.

Kreitschmann–Andermahr

, Hartmann

, Poll

, Schneider

H.J.

, Buchfelder

, and Stalla

G.K.

(2011). The German Database on Hypopituitarism after Traumatic Brain Injury and Aneurysmal Subarachnoid Hemorrhage – Description, Objectives and Design. 119, 15–20.

20.

Renner

, Hummelsheim

, Kopczak

, Steube

, Schneider

H.J.

, Schneider

, Kreitschmann–Andermahr

, Jordan

, Uhl

, and Stalla

G.K.

(2012). The influence of gender on the injury severity, course and outcome of traumatic brain injury. Brain Inj. 26, 1360–1371.

21.

Schneider

H.J.

, Schneider

, Kreitschmann–Andermahr

, Tuschy

, Wallaschofski

, Fleck

, Faust

, Renner

C.I.

, Kopczak

, Saller

, Buchfelder

, Jordan

, and Stalla

G.K.

(2011). Structured assessment of hypopituitarism following traumatic brain injury and aneurysmal hemorrhage in 1242 patients – the German interdisciplinary database. J. Neurotrauma, 28, 1693–1698.

22.

Ulfarsson

, Arnar Gudnason

, Rosén

, Blomstrand

, Sunnerhagen

K.S.

, Lundgren–Nilsson

, and Nilsson

(2013). Pituitary function and functional outcome in adults after severe traumatic brain injury: the long-term perspective. 30, 271–280.

23.

Nemes

, Kovacs

, Czeiter

, Kenyeres

, Tarjanyi

, Bajnok

, Buki

, Doczi

, and Mezosi

(2015). Predictors of post-traumatic pituitary failure during long-term follow-up. Hormones (Athens), 14, 383–391.

24.

Tanriverdi

, De Bellis

, Ulutabanca

, Bizzarro

, Sinisi

A.A.

, Bellastella

, Amoresano Paglionico

, Dalla Mora

, Selcuklu

, Unluhizarci

, Casanueva

F.F.

, and Kelestimur

(2013). A five year prospective investigation of anterior pituitary function after traumatic brain injury: is hypopituitarism long-term after head trauma associated with autoimmunity?. J. Neurotrauma, 30, 1426–1433.

25.

Jennett

, and Bond

(1975). Assessment of outcome after severe brain damage. Lancet, 1, 480–484.

26.

Narayan

R.K.

, Michel

M.E.

, Ansell

, Baethmann

, Biegon

, Bracken

M.B.

, Bullock

M.R.

, Choi

S.C.

, Clifton

G.L.

, Contant

C.F.

, Coplin

W.M.

, Dietrich

W.D.

, Ghajar

, Grady

S.M.

, Grossman

R.G.

, Hall

E.D.

, Heetderks

, Hovda

D.A.

, Jallo

, Katz

R.L.

, Knoller

, Kochanek

P.M.

, Maas

A.I.

, Majde

, Marion

D.W.

, Marmarou

, Marshall

L.F.

, McIntosh

T.K.

, Miller

, Mohberg

, Muizelaar

J.P.

, Pitts

L.H.

, Quinn

, Riesenfeld

, Robertson

C.S.

, Strauss

K.I.

, Teasdale

, Temkin

, Tuma

, Wade

, Walker

M.D.

, Weinrich

, Whyte

, Wilberger

, Young

A.B.

, and Yurkewicz

(2002). Clinical trials in head injury. J. Neurotrauma, 19, 503–557.

27.

Saatman

K.E.

, Duhaime

A.C.

, Bullock

, Maas

A.I.

, Valadka

, and Manley

G.T.

(2008). Classification of traumatic brain injury for targeted therapies. J. Neurotrauma, 25, 719–738.

28.

Teasdale

, and Jennett

(1974). Assessment of coma and impaired consciousness. A practical scale. Lancet, 2, 81–84.

29.

Hunt

W.E.

, and Hess

R.M.

(1968). Surgical risk as related to time of intervention in the repair of intracranial aneurysms. J. Neurosurg., 28, 14–20.

30.

Fisher

C.M.

, Kistler

J.P.

, and Davis

J.M.

(1980). Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery, 6, 1–9.

31.

Rosen

D.S.

, and Macdonald

R.L.

(2005). Subarachnoid hemorrhage grading scales: a systematic review. Neurocrit. Care, 2, 110–118.

32.

Brabant

, von zur Muhlen

, Wuster

, Ranke

M.B.

, Kratzsch

, Kiess

, Ketelslegers

J.M.

, Wilhelmsen

, Hulthen

, Saller

, Mattsson

, Wilde

, Schemer

, and Kann

(2003). Serum insulin-like growth factor I reference values for an automated chemiluminescence immunoassay system: results from a multicenter study. Horm. Res., 60, 53–60.

33.

Kopczak

, Kilimann

, von Rosen

, Krewer

, Schneider

H.J.

, Stalla

G.K.

, and Schneider

(2014). Screening for hypopituitarism in 509 patients with traumatic brain injury or subarachnoid hemorrhage. J. Neurotrauma, 31, 99–107.

34.

Hoffman

D.M.

, Nguyen

T.V.

, O'Sullivan

A.J.

, Baxter

R.C.

, and Ho

K.K.

(1994). Diagnosis of growth hormone deficiency in adults. Lancet, 344, 482–483.

35.

Marina

, Klose

, Nordenbo

, Liebach

, and Feldt–Rasmussen

(2015). Early endocrine alterations reflect prolonged stress and relate to 1-year functional outcome in patients with severe brain injury. Eur. J. Endocrinol., 172, 813–822.

36.

Tanriverdi

, Schneider

H.J.

, Aimaretti

, Masel

B.E.

, Casanueva

F.F.

, and Kelestimur

(2015). Pituitary dysfunction after traumatic brain injury: a clinical and pathophysiological approach. Endocr. Rev., 36, 305–342.

37.

Boonen

, and Van den Berghe

(2014) Cortisol metabolism in critical illness: implications for clinical care. Curr. Opin. Endocrinol. Diabetes Obes., 21, 185–192.

38.

Poll

E.M.

, Gilsbach

J.M.

, Hans

F.J.

, and Kreitschmann–Andermahr

(2013) Blunted serum and enhanced salivary free cortisol concentrations in the chronic phase after aneurysmal subarachnoid haemorrhage—is stress the culprit? Stress. 16, 153–162.

39.

Kalantaridou

S.N.

, Makrigiannakis

, Zoumakis

, Chrousos

G.P

. Stress and the female reproductive system. J. Reprod. Immunol., 2004, 62, 61–68.

40.

Kelly

D.F.

, Gonzalo

I.T.

, Cohan

, Berman

, Swerdloff

, and Wang

(2000). Hypopituitarism following traumatic brain injury and aneurysmal subarachnoid hemorrhage: a preliminary report. J. Neurosurg., 93, 743–752.

41.

Glynn

, and Agha

(2013). Which patient requires neuroendocrine assessment following traumatic brain injury, when and how?. Clin. Endocrinol. (Oxf). 78, 17–20.

42.

Schneider

, Schneider

H.J.

, Yassouridis

, Saller

, von Rosen

, and Stalla

G.K.

(2008). Predictors of anterior pituitary insufficiency after traumatic brain injury. Clin. Endocrinol. (Oxf). 68, 206–212.

43.

Aimaretti

, Ambrosio

M.R.

, Di Somma

, Gasperi

, Cannavò

, Scaroni

, Fusco

, Del Monte

, De Menis

, Faustini–Fustini

, Grimaldi

, Logoluso

, Razzore

, Rovere

, Benvenga

, Degli Uberti

E.C.

, De Marinis

, Lombardi

, Mantero

, Martino

, Giordano

, and Ghigo

(2005) Residual pituitary function after brain injury-induced hypopituitarism: a prospective 12-month study. J. Clin. Endocrinol. Metab., 90, 6085–6092.

44.

Bondanelli

, Ambrosio

M.R.

, Zatelli

M.C.

, Basaglia

, and Degli Uberti

E.C.

(2008). Prevalence of hypopituitarism in patients with cerebrovascular diseases. J. Endocrinol. Invest., 31, 16–20.

45.

Dusick

J.R.

, Wang

, Cohan

, Swerdloff

, and Kelly

D.F.

(2012). Pathophysiology of hypopituitarism in the setting of brain injury. Pituitary, 15, 2–9.

46.

Maiya

, Newcombe

, Nortje

, Bradley

, Bernard

, Chatfield

, Outtrim

, Hutchinson

, Matta

, Antoun

, and Menon

(2008). Magnetic resonance imaging changes in the pituitary gland following acute traumatic brain injury. Intensive Care Med. 34, 468–475.

47.

Schneider

H.J.

, Samann

P.G.

, Schneider

, Croce

C.G.

, Corneli

, Sievers

, Ghigo

, Stalla

G.K.

, and Aimaretti

(2007). Pituitary imaging abnormalities in patients with and without hypopituitarism after traumatic brain injury. J Endocrinol Invest, 30, RC9-RC12.

48.

van der Eerden

A.W.

, Twickler

M.T.

, Sweep

F.C.

, Beems

, Hendricks

H.T.

, Hermus

A.R.

, and Vos

P.E.

(2010). Should anterior pituitary function be tested during follow-up of all patients presenting at the emergency department because of traumatic brain injury?. Eur. J. Endocrinol., 162, 19–28.