Growth Hormone Deficiency as a Consequence of Traumatic Brain Injury

At an industry-sponsored symposium held in conjunction with the American Association of Clinical Endocrinologists’ 2003 Annual Meeting and Clinical Congress, three leaders in the diagnosis and treatment of growth hormone deficiencies in traumatic brain injury (TBI) presented the latest information on detection and correction of hypopituitarism as a result of head trauma. Topics included an introduction to TBI, prevalence of hypopituitarism after TBI and the diagnosis and treatment of hypopituitarism.

This program was supported by an unrestricted educational grant from Pharmacia Corporation.

Introduction to Traumatic Brain Injury

Traumatic brain injury (TBI) is an enormous problem in the United States. Every 15 seconds a traumatic brain injury occurs leading to an estimated 5 million individuals living with the sequelae. Every year 500,000 people are hospitalized following a trauma to the brain with 200,000 of them having residual deficits and 70,000-90,000 being severely disabled (National Center for Injury Prevention.http://www.cdc.gov/ncipc/factsheets/tbi.htm).

TBI strikes the young disproportionally. It is the leading killer and disabler of those under the age of 35. Roughly two-thirds of those injured are male, although there has been a persistent, gradual rise in the number of young females with TBI.

“One of the hardest parts of dealing with TBI is trying to predict outcomes,” said Brent E. Masel, MD, President and Medical Director of the Transitional Learning Center in Galveston, TX. “The family wants to know what is going to happen while the patient is still in the Intensive Care Unit (ICU). We try to categorize the signs and symptoms hoping to get a clue as how things will work out.”

Retrograde amnesia is the period of time preceding the injury for which the patient has no memory. Basically the information does not get moved from the working memory into intermediate memory. Dr. Masel likened it to information never getting from your monitor to the hard drive.

Post-traumatic amnesia constitutes the period of time following emergence from a coma during which the patient cannot create new memories. He said this is analogous to the information never getting to your monitor in the first place, and may be one of the best predictors of outcome.

“There are stages of recovery from TBI,” said Dr. Masel. “Generally, retrograde amnesia improves first following the injury, then post-traumatic amnesia and, hopefully restoration of memory. Keep in mind, it doesn’t always work like this and restoration of memory is often incomplete.”

The Glasgow Coma Scale (GCS) is a 15-point score with up to 4 points given for opening the eyes, one to five for best verbal response and as many as 6 points for best motor response. Glasgow scores less than 9 indicate a poor prognosis.

“TBI complications cover the entire spectrum of medicine,” said Dr. Masel. “Gastrointestinal problems such as bowel incontinence and a need for feeding tubes are not uncommon. Genital and urinary tract infections and bladder concerns are often seen. Respiratory and cardiovascular complications may also arise.”

There is an entity known as mild traumatic brain injury. Those with the disorder would say there is nothing mild about it as they often have anxiety, depression and somatic preoccupation. Although it is sometimes assumed that they are hysterics or malingerers, they are often found to have organic damage if objective neuropsychological tests are completed.

In 1914 Simmonds described the first case of hypopituitarism in a woman with necrosis of the pituitary. This was followed in 1918 by the first case in the literature of hypopituitarism secondary to trauma.

“The number have grown greatly since then largely due to improved survival because of better pre- and in-hospital care,” said Dr. Masel. “Current estimates may be low since the time between trauma and diagnosis is often close to a year. Individuals with brain injuries, and possibly post-traumatic hypopituitarism, are frequently lost to the brain injury medical system by then.”

Prevalence of Hypopituitarism after Traumatic Brain Injury

A summary by Benvenga reported on 367 cases of hypopituitarism published between 1942 and 1998. They found over 95% were diagnosed with hypogonadism, followed by hypothyroidism, adrenal insufficiency, hyperprolactinemia, diabetes insipidus and growth hormone (GH) deficiency (Benvenga S, et al. J Endocrinol Metab. 2000;85:1353) (Table 1).

“I want to point out that this is a summary of case reports in the literature,” said Steven A. Lieberman, MD, Associate Professor at the University of Texas Medical Branch in Galveston. “This is not a true prevalence look at the brain injury problem.”

However, over the last few years three major studies have begun with results of two published. The first from Kelly and associates studied 22 adults with TBI. They collected information in an attempt to identify risk factors for hypopituitarism using dynamic testing of pituitary function.

Growth hormone stimulation (GHS) was assessed using the insulin tolerance test (ITT) and subnormal responses were seen in 18% of the patients. Gonadotropin-releasing hormone (GnRH) stimulation found abnormal luteinizing hormone (LH) and follicle-stimulating hormone (FSH) responses in just fewer than 25% of the cohort. More than one deficiency was seen in nearly 15% of those enrolled (Kelly DF, et al. J Neurosurg. 2000;93:743).

The researchers looked at other measures. They identified GCS scores of <10 and diffuse brain swelling on magnetic resonance imaging (MRI) as significant predictors of hypopituitarism. Hypotensive or hypoxic episodes were seen in all patients with GH deficiency, but did not reach significance for hypopituitarism as a whole.

The other published study was by Dr. Lieberman and colleagues who enrolled 70 adults with TBI. Again they used a variety of pituitary function tests. The GH stimulation test in this case was the glucagon stimulation test (GST) with GH deficiencies found in 15% of patients. There were low insulin-like growth factor 1 (IGF-1) levels seen in 19% of patients. Frankly abnormal thyroid function occurred in 22% of those involved with low morning cortisol levels in 45%. There were no cases seen of diabetes insipidus, no hypogonadism in either gender or hyperprolactinemia (Lieberman SA, et al. J Clin Endocrinol Metab.2001;86:2752).

“Put all of these together and you have a single abnormal result in about half the patients,” said Dr. Lieberman. “There were dual abnormalities in a substantial minority.”

When they looked closer at the corticotropic axis, 45% of measured serum cortisol values were below the laboratory’s lower limit; however, their response to the cosyntropin stimulation test was largely normal. Five out of 70 patients fell below the cutoff point of 18. To see if steroid therapy after head injury might have contributed to this, they looked at basal and stimulated cortisol values in three different time periods. If there was a significant effect of the treatment, there should have been lower levels in the earlier groups and there were none.

The researchers next turned to the thyroid axis. They found a number of patients with low thyroid stimulating hormone (TSH), free T4 or both. When they plotted free T4 levels against TSH values, there was also a shift toward the lower range instead of the predicted mid-normal ranges.

“This is something that definitely piqued our interest,” said Dr. Lieberman. “We need to look into this in more detail before we can attach any clinical significance to this finding.”

They also did thyrotropin-releasing hormone (TRH) stimulation testing in a subset of patients. There was evidence of tertiary hypothyroidism in none of the patients and a flat response in 4 of 27.

Lieberman’s group compared patients who were GH deficient with those who were GH sufficient. There were no differences with respect to height, age, weight, GCS score, time since injury or age at the time of injury.

Even though height and weight were not significant, there were statistically significant differences seen in body mass index (BMI). The group with the higher BMI was more likely to be GH deficient. When they looked at hormonal studies in these two groups, the IGF-1 levels were significantly lower in the GH deficient group. In addition, the percentage of patients who had subnormal IGF-1 was significantly higher in the deficient group.

They then turned their attention to abnormal results for the glucagon stimulation (15%), L-dopa (close to 20%) and IGF-1 (20%) test. When put together the affected axes were GH, thyroid and cortisol. Gonadatropins, prolactin, and ADH secretion were unaffected or relatively spared.

There are some treatment options that should be considered. Those with clear-cut abnormalities in adrenal and thyroid function should be given replacement therapy. Growth hormone deficiency is an approved indication for prescribing growth hormone.

A third, as yet unpublished study on TBI comes from Italy. This uses a third test for GH secretion, GH-releasing hormone (GHRH) and arginine. As before, GH deficiency was found in 20.4% patients (unpublished data).

“We have seen three recent studies looking at the prevalence of hypopituitarism in brain injury patients,” said Dr. Lieberman “They show a remarkably consistent prevalence of 15-20% of GH deficiency in these patients using three different tests.”

Diagnosis and Treatment of Hypopituitarism

Currently the experts are unsure of the pathophysiology behind pituitary injury in TBI. The shearing forces that accompany TBI could disrupt blood flow to the pituitary.

Most studies have focused on severe TBI. However, there is also a group with mild brain trauma who do not go into long-term rehabilitation and may be missed.

“One of the questions that remains is the appropriateness of screening these less severely injured patients,” said Randall J. Urban, MD, Distinguished Chair in Internal Medicine at the University of Texas Medical Branch. “If we screen these patients will we see hypopituitarism to the same extent we have already seen in severely injured patients? While we don’t have any hard figures yet, if someone comes to me with vague symptoms and a history consistent with TBI, I will screen them for pituitary deficiencies.”

Dr. Urban suggests testing for gonadotropin deficiency by a menstrual history in women, total testosterone and FSH in men and prolactin levels for both genders. Free T4 and TSH should be checked for thyroid function while an a.m. cortisol and cosyntropin-stimulating hormones assesses cortisol deficiencies. Finally GH deficiencies can be assessed measuring serum IGF-1 or performing a GHS test of choice.

“I can say that if you screen someone and find a true deficiency, then replacement therapy is warranted,” said Dr. Urban. “Where we are struggling is deciding appropriate interventions for the subtle differences we find in many patients. Are they a subset of hypopituitarism patients that we need to be aware of?”

Treatment of GHD in hypopituitarism in patients without TBI has major impacts on important indices of survival. Rosen and Bengtsson studied 333 consecutive patients with hypopituitarism diagnosed between 1956 and 1987 that had been given routine replacement therapy. The overall mortality was higher than in an age- and sex- matched population. Mortality risk was raised irrespective of whether hypopituitarism was due to pituitary adenoma or secondary to other diseases (Rosen T, Bengtsson BA. Lancet. 1990;336:285).

Pfeifer and others noted that hypopituitary patients have increased mortality from vascular disease, and in these patients, early markers of atherosclerosis such as increased carotid artery intima-media thickness (IMT) and reduced distensibility are more prevalent. Before treatment, there were significant differences between those with GH deficiency and controls.

Replacement treatment had normalized the IMT of the common carotid artery by 6 months and of the carotid bifurcation within 90 days. Flow-mediated endothelium-dependent dilation of the brachial artery was significantly improved at three months. They concluded that GH replacement reversed these early changes (Pfeifer M, et al. J Clin Endocrinol Metab. 1999;84:453).

GH deficiency has been linked to changes in bone mineral density (BMD) leading to osteopenia. Colao and others found a significant reduction of BMD associated with abnormalities of bone turnover with very severe or severe growth hormone deficiencies. Normal BMD values were found in non-GH deficient hypopituitary patients. These abnormalities were consistently present in all patients regardless of the presence of additional hormone deficits, suggesting that this deficiency plays a central role in the development of osteopenia in hypopituitary patients (Colao A. J Clin Endocrinol Metab. 1999;84:1919).

GH deficiencies are treatable using replacement medications. However the changes, especially in performance and strength are very subtle and occur over long periods of time. Data from Jorgenson indicates that about 36 months is required before they begin to respond as measured by exercise capacity (Jorgenson JOL, et al. Eur J Endocrinol. 1994;130;224) (Figure 1).

GH replacement can affect quality of life. A study by Gibney and others showed a significant increase in energy and a positive change in emotional reactions (Gibney J, et al. J Clin Endocrinol Metab. 1999;84;2596).

Similar results were found in a study of the opinions of partners, parents and other caregivers. Significant differences were found in the patient becoming more alert, more active and having greater endurance according to the perceptions of the partners (Burman P, et al. J Clin Endocrinol Metab. 1995;80:3585).

“This is important since many of our TBI patients are not going to be able to tell us if they feel better,” said Dr. Urban. “It is their parent, partner or caregiver who are going to be primary sources of this information.”

Dr. Urban and others are beginning a study that will attempt to answer some of the treatment questions associated with TBI. They are planning on enrolling chronic, long-term patients at least one year post-TBI and will look at replacing thyroid, cortisol and growth hormone.

After screening, if a person is normal in all parameters, they will not be tested further. There will be separate cortisol, growth hormone and growth deficiency arms depending on what the screening finds. Those with frank findings will be replaced on an open-label basis. If they have results in the lower limits of normal, they will be given the appropriate medications in a double-blind, placebo-controlled fashion for six months. Following that they will be tested for neuropsychiatric response and those on placebo will be allowed to go on open-label replacement therapy.

The growth hormone arm will be a little more involved. If the tests results are in a gray area, the patient will be placed in a double-blind placebo-controlled replacement medication arm. The medication will be increased up to 600 mg a day using IGF-1 levels to determine GH dose. At one year, those on placebo will be given the option to go on GH for a year while those already in the treatment arm will continue for another 12 months. In addition to the psychiatric testing they also plan to reproduce the earlier studies on muscle strength, body composition and single fiber muscle contraction studies.

“We are trying this with the GH patients because we’ve anecdotally seen some beneficial effects,” said Dr. Urban. “One of the focuses of this study is to see if these benefits are still seen under more rigorous conditions.”