Prevalence of Pituitary Dysfunction After Severe Traumatic Brain Injury in Children and Adolescents: A Large Prospective Study

Transcription

1 ORIGINAL Endocrine ARTICLE Care Prevalence of Pituitary Dysfunction After Severe Traumatic Brain Injury in Children and Adolescents: A Large Prospective Study Claire Personnier, Hélène Crosnier, Philippe Meyer, Mathilde Chevignard, Isabelle Flechtner, Nathalie Boddaert, Sylvain Breton, Caroline Mignot, Yamina Dassa, Jean-Claude Souberbielle, Marie Piketty, Kathleen Laborde, Jean-Philipe Jais, Magali Viaud, Stephanie Puget, Christian Sainte-Rose, and Michel Polak Pediatric Endocrinology, Gynecology, and Diabetology Unit (C.P., H.C., I.F., Y.D., M.V., M.P.), Pediatric Anesthesiology Unit (P.M.), Radiology Unit (N.B., S.B.), Pediatric Neurosurgery Unit (C.M., S.P., C.S.-R.), Functional Explorations Unit (J.-C.S., M.P., K.L.), and Biostatistics Department (J.-P.J.), Assistance Publique-Hôpitaux de Paris, Hôpital Universitaire Necker Enfants-Malades, Paris, France; Rehabilitation Department for Children With Acquired Neurological Injury (M.C.), Hôpitaux de St Maurice, St Maurice, France; Université Paris Descartes (N.B., S.B., S.P., C.S.-R., M.P.), Paris, France; ER6-Université Pierre et Marie Curie, Paris, France (M.C.); and IMAGINE Affiliate (N.B., M.P.), Paris, France Context: Traumatic brain injury (TBI) in childhood is a major public health issue. Objective: We sought to determine the prevalence of pituitary dysfunction in children and adolescents after severe TBI and to identify any potential predictive factors. Design: This was a prospective longitudinal study. Setting: The study was conducted at a university hospital. Patients: Patients, hospitalized for severe accidental or inflicted TBI, were included. The endocrine assessment was performed between 6 and 18 months after the injury. Main Outcome Measures: Basal and dynamic tests of pituitary function were performed in all patients and GH dynamic testing was repeated in patients with low stimulated GH peak ( 7 ng/ml). The diagnosis of proven severe GH deficiency (GHD) was based on the association of two GH peaks less than 5 ng/ml on both occasions of testing and IGF-I levels below 2 SD score. Initial cranial tomography or magnetic resonance imaging was analyzed retrospectively. Results: We studied 87 children and adolescents [60 males, median age 6.7 y (range )] months after the TBI (73 accidental, 14 inflicted). The second GH peak, assessed months after the first evaluation, remained low in 27 children and adolescents. Fifteen patients had a GH peak less than 5 ng/ml (mean IGF-I SD score ) and five (5.7%) strict criteria for severe GHD. Two children had mild central hypothyroidism and one had ACTH deficiency. We did not find any predictive factors associated with existence of GHD (demographic characteristics, growth velocity, trauma severity, and radiological parameters). Conclusion: At 1 year after the severe TBI, pituitary dysfunction was found in 8% of our study sample. We recommend systematic hormonal assessment in children and adolescents 12 months after a severe TBI and prolonged clinical endocrine follow-up. (J Clin Endocrinol Metab 99: , 2014) ISSN Print X ISSN Online Printed in U.S.A. Copyright 2014 by the Endocrine Society Received November 18, Accepted March 4, First Published Online March 17, 2014 Abbreviations: BMI, body mass index; CT, cranial tomography; ft3, free T 3 ; ft4, free T 4 ; GCS, Glasgow Coma scale; GHD, GH deficiency; MRI, magnetic resonance imaging; SBS, shaken baby syndrome; SDS, SD score; TBI, traumatic brain injury jcem.endojournals.org J Clin Endocrinol Metab, June 2014, 99(6): doi: /jc

2 doi: /jc jcem.endojournals.org 2053 Traumatic brain injury (TBI) is a leading cause of childhood mortality and morbidity in industrialized countries and a major public health issue (1); approximately 10% of TBI is moderate or severe (2 4). The severity of brain injury is categorized as mild, moderate, or severe according to the Glasgow Coma scale (GCS) score. A GCS score of 9 12 defines moderate TBI, whereas a GCS score of 8 or less defines severe TBI (5). The causes of TBI in children include falls, pedestrian and road-traffic accidents, and child abuse. Child abuse is significantly more likely to occur in children below the age of 1 year (6 10). Poor developmental and functional outcome persist in 35% of affected children (9, 11, 12), and a recent study reported that 40% of children with moderate or severe TBI have impairments of health-related quality of life and have disabilities (13). In adults, neuroendocrine dysfunction has been recognized as a consequence of moderate or severe TBI, and a large number of studies have reported long-term hypopituitarism in 11% 69% of TBI survivors. Consensus guidelines on screening for pituitary deficiency after a brain injury have therefore been established (14). Retrospective and prospective pediatric studies reported highly variable rates of hypopituitarism after TBI, which may be related to the wide variation of methodology and to the small number of patients with severe TBI (15 19), and strong prospective data to support such guidelines in children are lacking. GH deficiency (GHD) is reported as the most prevalent deficiency. However, in some series any degree of permanent hypopituitarism is rare (20 22), and the question of what indicators of pituitary function to use and what systematic endocrine evaluation to undertake after TBI in children remains controversial. We report a prospective study investigating a large population of French children after severe accidental or inflicted TBI. The primary aim of the study was to determine the prevalence of posttraumatic pituitary dysfunction in this population. The secondary purpose was to identify any characteristics of TBI or brain imaging that were predictive of subsequent endocrine deficiency. Subjects and Methods This study was conducted with patients aged 0 15 years, admitted for TBI between October 2008 and April 2011 to the Department of Neurosurgical resuscitation of the Necker-Enfants Malades Hospital, the only hospital in the Paris area with a designated neurosurgical pediatric intensive care unit, and/or to the Rehabilitation Department of the Saint-Maurice Hospital, after a severe accidental or inflicted TBI. During the admission, the families of all eligible children and adolescents were approached to provide study information and obtain the parents consent for participation. The inclusion criteria were age less than 15 years at injury and severe TBI, defined as initial GCS score of 8 or less for accidental TBI and defined as the presence of an acute subdural hematoma requiring neurosurgical intensive care management for inflicted TBI compatible with shaken baby syndrome (SBS) and supported by social investigation. Infants with inflicted TBI sustain whiplash-shaking injuries, resulting in marked rotational acceleration/deceleration forces, causing widespread parenchymal damage. They often suffer more severe neurological damage than children with accidental injury sustained at the same age; therefore, more severe lesions to the pituitary axis could be expected as well. Exclusion criteria were the presence of known pituitary deficiency before TBI and posttraumatic vegetative state for ethical reasons. The endocrine assessment was performed at the Department of Pediatric Endocrinology of the Necker-Enfants Malades Hospital between 6 and 18 months after the injury. The medical history recorded included characteristics of the trauma, mechanism of injury, the presence of more than one life-threatening lesion (polytrauma), hemodynamic instability, and the use of neurosurgical decompression. Clinical features suggesting pituitary dysfunction, including asthenia, weight progression, appetite, alteration of thirst, hair changes, pubertal and menstrual abnormalities, and mood and school disturbances, were recorded. A full clinical examination was performed by two pediatric endocrinologists; height and weight were measured and plotted over time, and pubertal staging was assessed according to the method of Tanner (23). Height, weight, growth velocity, and body mass index (BMI) SD score were standardized against French norms (24, 25). Baseline fasting serum concentrations of electrolytes, urea, creatinine, urinary and blood osmolalities, basal (8:00 AM) cortisol, IGF-I, free T 4 (ft4), free T 3 (ft3), TSH, and prolactin (two measures 15 min apart) were measured. Plasma/serum concentrations of LH, FSH, and T (male) or estradiol (female) were assessed in girls older than 10 years, boys older than 11 years, and children showing clinical evidence of precocious puberty. Dynamic tests were used to evaluate GH secretion. Children weighing at least 15 kg underwent GH stimulation with betaxolol (2.5 mg per 10 kg orally) at 30 minutes plus glucagon (0.03 mg/kg im, maximum 1 mg) at 0 minutes; blood samples for GH and laboratory glucose were collected at 30, 0, 30, 60, 90, 120, 150, and 180 minutes. Children weighing less than 15 kg or with a history of asthma underwent GH stimulation with glucagon alone (0,1 mg/kg im, maximum 1 mg) at time 0; blood samples for GH and glucose were collected at 0, 15, 30, 45, 60, 90, 120, 150, and 180 minutes. Standard laboratory methods were used to measure electrolytes, urea, creatinine, and urinary and blood osmolalities. GH (SI 98/574, 1 ng/ml 3 IU/mL), prolactin, TSH, ft4, ft3, LH, and FSH were measured by immunoassays performed on an automated platform (Beckman Coulter Access). IGF-I was measured by means of an immunoradiometric assay (IGF-I RIACT; Cis Bio International), cortisol, estradiol, and T concentrations were measured by a RIA (Cis Bio International). A second GH secretion test was performed when first GH peak under stimulation was below 7 ng/ml. In patients weighing at least 15 kg, arginine (0.5 g/kg iv, maximum 25 g) was administered at 0 minutes plus insulin (0.1 U/kg iv) at 90 minutes; GH and glucose blood samples were collected at 0, 15, 30, 45, 60, 90, 105, 120, 135, and 150 minutes. In children weighting less than 15 kg, arginine (0.5 g/kg iv, maximum 25 g) was administered at

3 2054 Personnier et al Pituitary Dysfunction in Children After Severe TBI J Clin Endocrinol Metab, June 2014, 99(6): minutes, and blood samples were collected at 0, 15, 30, 45, 60, and 90 minutes. Patients with insufficient 8:00 AM basal cortisol level (compared with normal ranges provided by Necker-Enfants Malades Hospital Laboratory Services) underwent a Synacthen test (0.25 and mg older than and younger than 2 years of age, respectively, as an iv push) and cortisol measured at 0 and 60 minutes. A gonadotropin stimulation test (GnRH, 100 g/m 2 iv at time 0, maximum 100 g; LH and FSH measured at 0, 10, 20, 30, 60, 90, and 120 min) was performed in patients showing abnormal sex hormone levels or clinical evidence of precocious or delayed puberty. TSH and prolactin response to a TRH stimulation test (TRH, 7 g/kg iv at 0 minutes, maximum 200 g; TSH and prolactin measured at 0, 10, 20, 40, 60, and 120 minutes) were performed in patients with abnormal basal thyroid hormones or prolactin levels (26). A fluid deprivation test was planned for patients showing clinical or biological signs of central diabetes insipidus (27). Baseline hormone values and IGF-I values were compared with age, sex, and pubertal normal reference ranges provided by Necker-Enfants Malades Hospital Laboratory Service according to age, sex, and pubertal stage. GHD was defined as stimulated GH peak below 7 ng/ml at two tests (partial GHD with peak between 5 and 7 ng/ml and severe GHD with peak below 5 ng/ml at two tests). Low levels (below 2 SDS in relation to age and sex) of IGF-I supported the diagnosis of GHD (28). We did not use any sex hormone priming in peripubertal subjects prior to GH stimulation test. ACTH deficiency was confirmed in case of peak stimulated cortisol below 550 nmol/l at time 60 minutes. A central diabetes insipidus was excluded if urinary osmolality was above 800 mosmol/kg. Precocious puberty was defined in girls as Tanner stage 2 breast development at younger than 8 years of age and in boys as testicular length of 25 mm or more before the age of 9 years. Delayed puberty was defined in girls as Tanner stage 1 breast development after 13 years of age and in boys as testicular length of 25 mm or less after 14 years of age. An evaluation of bone age based on an X-ray of the left hand and wrist was performed for all patients aged at least 3 years. All patients had cranial cranial tomography (CT) at the acute phase. In addition, initial and postoperative head and intracranial injuries, as documented by CT and (or) magnetic resonance imaging (MRI), were retrospectively analyzed, at the time of the study, by the same two experienced neuroradiologists, blinded to the results of the hormonal measurements. Also recorded were fractures of the basal skull, sella, cranial vault, and facial bones; pituitary and pituitary stalk lesions; cerebral edema; sub/epidural and traumatic subarachnoid hemorrhage; intracerebral parenchymal hematoma; and intracerebral (supra- or infratentorial) petechial hemorrhages. Patients data are expressed as mean SD if Gaussian or median (ranges) if non-gaussian distribution. Differences in patient-reported data were tested with a Fisher s exact test or a Wilcoxon-Mann-Whitney test; correlations were tested with a Pearson test. A value of P.05 was considered to be significant. This study was approved by the French Research Ethics Committee (Comité de protection des personnes), and written informed consent was obtained from parents of all participants. Results A total of 103 children and adolescents, 81 patients with accidental TBI and 22 with inflicted TBI, met the inclusion criteria and were recruited. Five of the 103 families could not be contacted and one child had died from late TBI aftereffects, leaving 97 families that were contacted, of which 10 declined participation in the study. The characteristics of the 87 patients (73 accidental TBI, 14 inflicted TBI) who were assessed are summarized in Table 1. All patients were admitted to the neurosurgical intensive care unit for a mean duration of days. All patients, except three with inflicted TBI, were mechanically ventilated in the acute phase for a mean duration of days. Auxological and pubertal evaluation For 85 patients in whom preinjury measurements were available, mean growth velocity was within the normal range. One child had clinical evidence of short stature (height SDS of 2.3 for target parental height SDS of 1) related to intrauterine growth retardation. Six patients were overweight. A total of sixty-four children (73.6%) were Tanner stage 1, six adolescents (6.9%) were Tanner stage 2, eight (9.2%) Tanner stage 3 4, and nine (10.3%) Tanner stage 5. All children who sustained inflicted TBI were Tanner stage 1. Among chronic conditions affecting growth, asthma was observed in seven patients (three treated with low doses of inhaled steroids); pre-tbi growth velocity of these seven children was normal (mean height SDS). Endocrine evaluation (the results are summarized in Figure 1) GH secretion and GH-IGF-I axis Peak GH response to the first stimulation test was below 7 ng/ml in 35 patients (36.7%), 29 who suffered accidental TBI and six inflicted TBI, at a mean age of years. Two families of children with accidental TBI declined further investigation, and 33 patients underwent a second provocative GH test month after the first evaluation. GH peak remained below 7 ng/ml in 27 patients (31%) whose characteristics are given in Table 2. Fifteen patients (17.2%), 13 accidental TBI (study numbers 1 3, 5, 6, 8 10,12 15, and 18) and two inflicted brain trauma (study numbers 23 and 27), had severe GHD. Their mean IGF-I SDS was ; 12 of them had an IGF-I below 1 SDS and five (study numbers 9, 10, 15, 18, and 27) below 2 SDS. Six (study numbers 9, 10, 14, 18, 23, and 27) of these patients exhibited poor growth velocity, with a decrease in their height SDS of more than 0.5 between the two assessments. Two adolescents (study numbers 1 and 12), Tanner stage 3, had T levels at 3.4 and 3 ng/ml, respectively, and an IGF-I in the low part of the normal ranges. In the remaining 12 patients with partial GHD, all but one (study number 25) had normal growth. Among pa-

4 doi: /jc jcem.endojournals.org 2055 Table 1. Baseline Characteristics of the Evaluated Children (n 87; 73 Accidental TBI and 14 Inflicted Trauma) Total Accidental TBI SBS Gender (M/F) 60/27 48/25 12/2 Median age at TBI, y 5.9 ( ) 7.3 ( ) 0.5 ( ) Cause of injury Fall 21 (24.1%) RA 50 (57.5%) Pedestrian 31 (35.6%) Cyclist 8 (9.2%) MVA 11 (12.6%) Other a 2 (2.3%) SBS 14 (16.1%) Initial CT findings Fracture Basal scull 19 (21.8%) 19 (26.0%) 0 (0.0%) Cranial vault 37 (42.5%) 35 (47.9%) 2 (14.3%) Hematoma 51 (58.6%) 37 (50.7%) 14 (100%) Surgical intervention b 29 (33.3%) 16 (21.9%) 13 (92.8%) Polytrauma 56 (64.4%) 53 (72.6%) 3 (21.4%) Age at assessment, y 6.7 ( ) 8 ( ) 1.1 ( ) Time since TBI, mo Height SDS at assessment Weight SDS at assessment BMI SDS at assessment Height SDS-MPH SDS Abbreviations: F, female; M, male; MPH, mean parental height; MVA, motor vehicle accident; RA, road accident. Data are median (range), mean SD, n (percent), or ratio. a Other: horse riding, skateboard. b Neurosurgical decompressive intervention tients with a weak GH between 5 and 7 ng/ml, mean IGF-I SDS was , significantly higher (P.003) than mean IGF-I in severe GHD. Excluding infants younger than 6 months of age at TBI, the mean height velocity SDS before TBI in normal GH patients and GHD patients were similar ( and , respectively; P.9) Mean BMI SDS of GHD patients and severe GHD patients were not significantly different from that of patients with normal GH secretion, ( , P.7, and , P.93, respectively, vs ). In overweight children with low GH response, one child (study number 18) had low growth velocity and IGF-I less than 2 SDS, and two other children (study numbers 5 and 13) had an IGF-I below 1 SDS. TSH axis Three patients with accidental TBI and three patients with inflicted head trauma had low ft4 levels for age at initial assessment (mean ft pmol/l; normal range pmol/l) associated with normal TSH and ft3 levels. The ft4 level remained subnormal in two patients with accidental TBI on subsequent verification. One young girl, treated with carbamazepine, showed appropriate TSH response to the TRH stimulation test (1.2, 16.5, and 4.6 miu/l for basal, peak, and 120 minute TSH levels); ft4 remained low after completion of carbamazepine treatment. In the other child, permanent low ft4 was associated with mild elevated TSH (5.38 miu/l; normal range miu/l/l) and with GHD. ACTH axis The mean basal cortisol level was nmol/l (normal range nmol/l). One male patient, in the group of children and adolescents with accidental TBI, had a low basal cortisol level (44.7 nmol/l) and suboptimal cortisol response (350.4 nmol/l); he reported a history of fainting on a few occasions. His other pituitary functions were normal. Other pituitary functions One girl, aged 5.4 years, with accidental TBI exhibited, during the first evaluation 6 months after the injury, signs of posterior pituitary dysfunction (polyuria, polydipsia, low urinary osmolality). Her fluid deprivation test was in the normal range so that the diagnosis of diabetes insipidus was excluded. No patient experienced precocious puberty, and basal levels of estradiol, T, LH, and FSH were appropriate in the tested subjects. No patient exhibited delayed puberty, and secondary amenorrhea was not observed in adolescents. At initial assessment, the basal prolactin level was high in seven patients ( ng/ml; normal range 2 20 ng/ml). In these patients, the second measure, 15 minutes

5 2056 Personnier et al Pituitary Dysfunction in Children After Severe TBI J Clin Endocrinol Metab, June 2014, 99(6): Figure 1. Summary of endocrine evaluation in children with severe TBI. a, One child has two GH peak under stimulation less than 7 ng/ml with low IGF-I. later, showed that prolactin levels were normal in three patients, strongly decreased in two (46 to 27 ng/ml and 85 to 38 ng/ml) and remained unchanged in two (28 to 24 ng/ml and 40 to 42 ng/ml). In these four latter cases, high prolactin levels were attributed to stress, and no further investigations were performed; the other pituitary axes were normal in three patients and low peak GH was observed in the fourth. Predictive factors As summarized in Table 3, no demographic and trauma characteristics was significantly associated with the presence or absence of GHD. In patients with GHD and severe GHD, growth velocity between first and second assessments were correlated with neither first GH stimulated peak (r 0.19, P.42; r 0.37, P.22; respectively) nor second GH stimulated peak (r 0.19, P.42; r 0.82, P.074, respectively). All 87 patients had cranial CT at the acute phase, immediately after the TBI. For 82 of these, CT or MRI was reported by experienced neuroradiologists [61 of initial CT, nine CT, and five MRI performed in early postoperative period ( 15 d); and seven CT performed in the late postoperative period]. No radiological parameter, type of lesions or location of lesions, was significantly associated with existence or not of GHD. Discussion We prospectively assessed 87 children and adolescents after severe TBI, and this cohort is, to our knowledge, the largest sample of children and adolescents with severe accidental TBI studied. Twelve months after the severe TBI, we found prevalence rates of 17% with suspected severe GHD and, applying strict criteria including two stimulated peaks GH response less than 5 ng/ml and IGF-I 2 SDS, 6% with proven severe GHD in this cohort. Multiple pituitary deficiencies or other isolated pituitary deficiencies (thyroid and adrenal axis) were very rare, and we have so far not observed precocious or delayed puberty. Although many patients reported increase of fluid intakes without polyuria, probably within the framework of posttraumatic hyperphagia, we did not diagnose any case of diabetes insipidus. Using stimulated GH tests alone (GH 7 ng/ml), up to 31% of our study population might have been diagnosed with GHD. Recent studies conducted in adults after a TBI and healthy adults (29, 30) showed a high risk of bias in pituitary testing and the risk of overdiagnosis of GHD. In a pediatric population, the lack of recent normal age-related reference values for GH dynamic tests and of stringent reliability of these tests requires a combination of auxological and biochemical information (IGF-I levels) to establish the diagnosis of GHD (28, 31, 32). In our study,

6 doi: /jc jcem.endojournals.org 2057 Table 2. Auxological Data and Hormonal Status of 27 Patients With Two Stimulated GH Peak Below 7 ng/ml First Evaluation 1 Second Evaluation 2 Patient Number Gender Age, y Cause of Injury poor growth velocity between the two assessments was found only in six patients with severe GHD, but for all but one patient, the interval between two measurements was too short to calculate growth velocity accurately and a longer follow-up period is justified. Low levels of IGF1 in the patients with GH peaks less than 5 ng/ml support the diagnosis of proven GHD (28); however, normal levels (IGF-I 1 SDS) do not. TS Height BMI IGF-I GH, ng/ml Height BMI IGF-I GH, ng/ml 1 M 13.5 ATBI F 4.5 ATBI F 7.7 ATBI F 5.1 ATBI M 11.4 ATBI M 5.9 ATBI M 4.8 ATBI F 8.8 ATBI M 4.7 ATBI M 1.3 ATBI M 1.2 ATBI M 12.8 ATBI F 10.6 ATBI M 12.9 ATBI M 3.1 ATBI F 5.5 ATBI M 5.1 ATBI M 6.1 ATBI M 7.5 ATBI M 0.7 ATBI F 7.3 ATBI M 6.0 ATBI M 0.2 ITBI M 0.2 ITBI M 1.2 ITBI M 0.2 ITBI M 0.5 ITBI Abbreviations: age, age at injury; ATBI, accidental TBI; F, female; ITBI, inflicted TBI; M, male; TS, Tanner stage. Table 3. Interval Between 2 and 1, mo Other recent studies reported low GH response to repeated stimulation test in smaller cohorts of children 1 year after the TBI; one of these studies reported 5% of GHD (19) and the other reported 34% of low GH response in children with normal IGF-I and increasing BMI (33). These results contrast with a previous study, which detected no basal pituitary dysfunction 6.8 years after the TBI; however, the sample size was limited and injury se- Demographics and Injury Characteristics of Study Patients According to Stimulated Peak GH Response Normal GHD Severe GHD P Value a P Value b Age at injury, y Gender (F/M) 20/40 7/20 4/ ATBI/ITBI 51/9 22/5 13/ GCS score Hemodynamic instability, % Length of intubation, d PICU admission, d Cranial lesions, % Polytrauma, % Neurological disabilities, % Abbreviations: ATBI, accidental traumatic brain injury; F, female; ITBI, inflicted TBI; M, male; PICU, pediatric intensive care unit. Data are mean SD, percentage, or ratio. a GHD vs normal. b severe GHD vs normal.

7 2058 Personnier et al Pituitary Dysfunction in Children After Severe TBI J Clin Endocrinol Metab, June 2014, 99(6): verity was variable from mild to severe (20). Moreover, a recent study (22), with the same strict criteria of GHD, found no case of GHD in children 6.5 years after TBI in early childhood (1.7 y). The most notable difference compared with our study is the delay of the assessments after the TBI. Previous studies in adults (34) have demonstrated that GH dysfunction present 12 months after the TBI may not be persistent. In children, no prospective study has described the natural history of pituitary dysfunction occurring after TBI so that it is not possible to determine whether early GH deficiency may remit (22). The hormonal status of the 22 children and adolescents with low GH response to repeated provocative tests and normal IGF1 (25.3%) ( 2 SDS) remains unclear. Some children were in the peripubertal period, and in this study we did not prime peripubertal children with exogenous sex steroid prior to GH stimulation testing, so that temporarily blunting of GH secretion might be attributable to prepuberty (35). A previous study (21) reported the same difficulties with the interpretation of low GH response without evidence of growth impairment in adolescents a few years after TBI. However, at present, there is no consensus regarding the range of normal for priming GH test. In two adolescents in our study with high T levels, appropriate to their pubertal stage, a GHD diagnosis was strongly suspected because the GH secretion was low and the IGF-I was less than 1 SDS. Moreover, as it has been commonly reported in patients after a TBI (36), 22% of patients with normal GH secretion and 27% of patients with severe GHD became overweight. Low GH responses to stimulation with normal IGF-I levels are common in overweight children (37). Therefore the overweight child with low GH secretion and IGF-I less than 2 SDS was diagnosed as proven GHD; the diagnosis remained uncertain in two other overweight children with IGF-I less than 1 SDS. The children with GHD 1 year after a severe TBI had normal height, and only one child and one adolescent were treated by recombinant human GH by the end of the study. For the other children and adolescents with low stimulated GH peak, the assessment at 1 year after the TBI may be too soon to observe a decrease in growth velocity: patients need to be clinically monitored over longer periods of time to survey auxological parameters and establish accurate height velocity. Irrespective of the growth velocity, the treatment of GHD in this population is a clinical challenge. It may be of clinical importance to diagnose GHD before the decrease of height velocity in those children because a positive impact of recombinant human GH on height velocity and subsequently on muscle strength is to be expected (38), and it has been reported that pituitary dysfunction may interfere with cognitive and physical rehabilitation of children with severe brain injury (39). In adults with GHD after TBI, a few studies have reported an improvement of cognition and quality of life on GH therapy (40, 41), and a recent study reports the same result in children (42). It seems important for us to assess the possibility of GHD in children affected by a severe TBI earlier than is the usual practice. Further studies are necessary to confirm or to refute the possible benefit of GH treatment. Our findings on the prevalence of thyroid dysfunction are lower than the incidence of 5% 12% at 1 year after the injury described in previous pediatric studies (16 18). Some antiseizure medications are known to reduce peripheral thyroid hormone levels among 25% 50% of treated patients. The mechanism of decrease is still debated and may be related to a liver enzyme induction, a competition on thyroid binding globulin binding sites, or an interference with the hypothalamic regulation (43). One of our patients with abnormal thyroid evaluation was taking carbamazepine when assessed. Because thyroid abnormalities have initially been considered a consequence of treatment, persistence of thyroid dysfunction after completion of carbamazepine treatment made a diagnosis of TSH deficiency more plausible. The second patient with thyroid abnormalities presented a mild elevation of TSH level consistent with hypothalamic dysfunction, which has already been described in children after a brain injury (19). In contrast to two recent pediatric studies (20, 23) in which precocious puberty was the most common pituitary abnormality found, we did not observe any case of precocious puberty in our sample at initial assessment. However the study was performed relatively early after a TBI, and a longer follow-up period is necessary to determine the prevalence of pubertal abnormalities, including delayed puberty. In our study, initial and follow-up imaging after a TBI failed to reveal any consistent relationship between pituitary dysfunction and the site or type of cranial and cerebral lesions. Specifically, CT revealed no abnormalities of the region of the sella turcica and hypothalamus; although microhemorrhages correlated with diffuse axonal injury, their frequency was not increased in children with pituitary dysfunction. In a prospective study of 78 adults, diffuse axonal injury and basal skull fracture were associated with a higher frequency of posttraumatic pituitary insufficiency 1 year after a TBI (44). MRI is more sensitive than CT in the identification of diffuse axonal injury (45), but unfortunately, only a few children of our study had MRI available after the TBI. Therefore, predictors of post-tbi pituitary dysfunction have not been identified. In conclusion, severe accidental or inflicted childhood TBI appears to be a cause of pituitary dysfunction in this large population of children 1 year after the injury. How-

8 doi: /jc jcem.endojournals.org 2059 ever, our results did not enable us to identify predictive factors of subsequent pituitary dysfunction. The possibility that GHD has the potential to impair health-related quality of life, physical rehabilitation, and functional outcome of children affected by severe TBI leads us to recommend systematic hormonal assessment in children 12 months after severe TBI as well as prolonged clinical endocrine follow-up of this population. A coordinated approach by pediatric endocrinologists, neurosurgeons, and specialists of pediatric neurorehabilitation units is fully recommended to provide adequate care to these children. Acknowledgments We thank the nursing staff under the direction of Myriam Faivre in the Endocrinology, Gynecology, and Diabetology Ward at Necker Enfants Malades Hospital. We also thank our colleagues, Dr Hanna Touré (Rehabilitation Department for Children With Acquired Neurological Injury, Hôpitaux de Saint-Maurice) for her help in collection of clinical data and Professor Gérard Friedlander (Functional Explorations Unit, Hôpital Universitaire Necker Enfants-Malades) for his help in the supervision of biochemical data. We also thank Professor Colin Kennedy (University Hospital and University of Southampton) for his help in correcting the English in the manuscript. Address all correspondence and requests for reprints to: Hélène Crosnier, MD, or Michel Polak, MD, PhD, Hôpital Universitaire Necker-Enfants Malades, Service d Endocrinologie, Gynécologie, et Diabétologie Pédiatriques, 149 Rue de Sèvres, Paris, France. h.savajol-crosnier@orange.fr; or michel.polak@nck.aphp.fr. This work was supported by through an educational grant from Pfizer SAS. The sponsor was not involved in the data collection, manuscript preparation, or publishing decisions. Disclosure Summary: The authors have nothing to declare. References 1. World Health Organization. Neurological Disorders: Public Health Challenges. Geneva: WHO Press; Parslow C, Morris KP, Tasker RC, et al. Epidemiology of traumatic brain injury in children receiving intensive care in the UK. Arch Dis Child. 2005;90: Faul M, Xu L, Wald MM, Coronado VG. Traumatic Brain Injury in the United States: Emergency Department Visits, Hospitalizations and Deaths, Atlanta: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; McKinlay A, Grace RC, Horwood LJ, Fergusson DM, Ridder EM, MacFarlane MR. Prevalence of traumatic brain injury among children, adolescents and young adults: prospective evidence from a birth cohort. Brain Injury. 2008;22: Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical scale. Lancet. 1974;2: Masson F, Maurette P, Salmi LR, et al. Characteristics of head trauma in children: epidemiology and a 5-year follow-up. Arch Pediatr. 1996;3: Hiu Lam W, Mackersie A. Paediatric head injury: incidence, aetiology, and management. Paediatr Anaesth. 1999;9: Barlow KM, Minns RA. Annual incidence of shaken impact syndrome in young children. Lancet. 2000;356: Keenan HT, Bratton SL. Epidemiology and outcomes of pediatric traumatic brain injury. Dev Neurosci. 2006;28: Shao J, Zhu H, Yao H, et al. Characteristics and trends of pediatric traumatic brain injuries treated at a large pediatric medical center in China, PLoS One. 2013;7:e Michaud LJ, Rivara FP, Grady MS, Reay DT. Predictors of survival and severity of disability after severe brain injury in children. Neurosurgery. 1992;31: Ducrocq SC, Meyer PG, Orliaguet GA, et al. Epidemiology and early predictive factors of mortality and outcome in children with traumatic severe brain injury: experience of a French pediatric trauma center. Pediatr Crit Care Med. 2006;7: McCarthy ML, MacKenzie EJ, Durbin DR, et al. Health-related quality of life during the first year after traumatic brain injury. Arch Pediatr Adolesc Med. 2006;160: Ghigo E, Masel B, Aimaretti G, et al. Consensus guidelines on screening for hypopituitarism following traumatic brain injury. Brain Injury. 2005;19: Aimaretti G, Ambrosio MR, Di Somma C, et al. Traumatic brain injury and subarachnoid haemorrhage are conditions at high risk for hypopituitarism: screening study at 3 months after the brain injury. Clin Endocrinol (Oxf). 2004;61: Einaudi S, Matarazzo P, Peretta P, et al. Hypothalamo-hypopysial dysfunction after traumatic brain injury in children and adolescents: a preliminary retrospective and prospective study. J Pediatr Endocrinol Metab. 2006;19: Niederland T, Makovi H, Gal V, Andreka B, Abraham CS, Kovacs J. Abnormalities of pituitary function after traumatic brain injury in children. J Neurotrauma. 2007;24: Phoomtavorn P, Maixner W, Zacharin M. Pituitary function in paediatric survivors of severe traumatic brain injury. Arch Dis Child. 2008;93: Kaulfers AM, Backeljauw PF, Reifschneider K, et al. Endocrine dysfunction following traumatic brain injury in children. J Pediatr 2010;157: Moon RJ, Wilson P, Kirkham FJ, Davies JH. Growth monitoring following traumatic brain injury. Arch Dis Child. 2009;94: Khadr SN, Crofton PM, Jones PA, et al. Evaluation of pituitary function after traumatic brain injury in Childhood. Clin Endocrinol (Oxf). 2010;73: Heather NL, Jefferies C, Hofman PL, et al. Permanent hypopituitarism is rare after structural traumatic brain injury in early childhood. J Clin Endocrinol Metab. 2012;97: Tanner JM, Whitehouse RH. Clinical longitudinal standards for height, weight, height velocity, weight velocity, and stages of puberty. Arch Dis Child. 1976;51: Sempé M, Pedron G, Roy MP. Auxologie, Méthodes et Séquences. Paris: Theraplix; Rolland-Cachera MF, Cole TJ, Sempe M, Tichet J, Rossignol C, Charraud A. Body mass index variations: centiles from birth to 87 years. Eur J Clin Nutr. 1991;45: Foley TP Jr, J Owings J, Hayford JT, Blizzard RM. Serum thyrotropin responses to synthetic thyrotropin-releasing hormone in normal children and hypopituitary patients. A new test to distinguish primary releasing hormone deficiency from primary pituitary hormone deficiency. J Clin Invest. 1972;51: Czernichow P, Polak M. Testing water regulation. In: Ranke MP, Mullis PE, eds. Diagnosis of Endocrine Function in Children and Adolescents. Basel: Karger; GH Research Society. Consensus guidelines for the diagnosis and treatment of growth hormone (GH) deficiency in childhood and

9 2060 Personnier et al Pituitary Dysfunction in Children After Severe TBI J Clin Endocrinol Metab, June 2014, 99(6): adolescence: summary statement of the GH Research Society. J Clin Endocrinol Metab. 2000;85: Kokshoorn NE, Wassenaar JE, Biermasz NR, Roelfsema F, Smit JWA. Hypopituitarism following traumatic brain injury: prevalence is affected by the use of different dynamic tests and different normal values. Eur J Endocrinol. 2010;162: Klose M, Stochholm K, Janukonyté J, et al. 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. 2014;99: Carel JC, Tesca J-P, Letrait M, et al. GH values: growth hormone testing for the diagnosis of growth hormone deficiency in childhood: a population register-based study. J Clin Endocrinol Metab. 1997; 82: Ranke MB. Growth hormone deficiency: diagnostic principles and practices. In: Ranke MP, Mullis PE, eds. Diagnostic of Endocrine Function in Children and Adolescents. Basel: Karger; 2011: Casano-Sancho P, Suarez L, Ibanez L, Garcia-Fructuoso G, Medina J, Febrer A. Pituitary dysfunction after traumatic brain injury in children: is there a need for ongoing endocrine assessment? Clin Endocrinol (Oxf) /cen.12237/abstract. Accessed June 27, Tanriverdi F, Ulutabanca H, Unluhizarci K, Selcukli A, Casanueva FF, Kelestimur F. Three years prospective investigation of anterior pituitary function after traumatic brain injury: a pilot study. Clin Endocrinol (Oxf). 2008;68: Albertsson-Wikland K, Rosberg S, Karlberg J, Groth T. Analysis of 24-hour growth hormone profiles in healthy boys and girls of normal stature: relation to puberty. J Clin Endocrinol Metab. 1994; 78: Jourdan C, Brugel D, Hubeaux K, Touré H, Laurent-Vannier A, Chevignard M. Weight gain after childhood traumatic brain injury: a matter of concern. Dev Med Child Neurol. 2012;54: Loche S, Cappa M, Borrelli P, et al. Reduced growth hormone response to growth hormone-releasing hormone in children with simple obesity: evidence for somatomedin C mediated inhibition. Clin Endocrinol (Oxf). 1987;27: Polak M. Growth hormone and muscle strength in children. J Clin Endocrinol Metab. 2013;98: Wamstad JB, Horwood KW, Rogol AD, et al. Neuropsychological recovery and quality-of-life in children and adolescents with growth hormone deficiency following TBI: a preliminary study. Brain Inj. 2013;27: Reimunde P, Quintana A, Castanon B, et al. Effects of growth hormone (GH) replacement and cognitive rehabilitation in patients with cognitive disorders after traumatic brain injury. Brain Injury. 2011;25: Moreau OK, Cortet-Rudelli C, Yollin E, Merlen E, Daveluy W, Rousseaux M. Growth hormone replacement therapy in patients with traumatic brain injury. J Neurotrauma. 2013;30: Devesa J, Alonso B, Casteleiro N, et al. Effects of recombinant growth hormone (GH) replacement and psychomotor and cognitive stimulation in the neurodevelopment of GH-deficient (GHD) children with cerebral palsy: a pilot study. Ther Clin Risk Manag. 2011; 7: Cansu A, Sersaroglu A, Camurdan O, Hirfanoglu T, Bideci A, Gücüyener K. The evaluation of thyroid function, thyroid antibodies, and thyroid volumes in children with epilepsy during short-term administration of oxcarbazepine and valproate. Epilepsia. 2006; 47: Schneider M, Schneider HJ, Yassouridis A, Saller B, von Rosen F, Stalla GK. Predictors of anterior pituitary insufficiency after traumatic brain injury. Clin Endocrinol (Oxf). 2008;68: Azouvi P. Neuroimaging correlates of cognitive and functional outcome after traumatic brain injury. Curr Opin Neurol. 2000;13: