Evaluation of BDNF levels in patients hospitalized for physical trauma at an emergency hospital in Porto Alegre, southern Brazil

Objective: To assess the association between brain-derived neurotrophic factor (BDNF) levels and acute stress disorder (ASD) in patients who have suffered physical trauma. Methods: Data were collected at an emergency hospital in Porto Alegre, state of Rio Grande do Sul, southern Brazil. Participants were over 18 years of age, victims of physical trauma, and had been hospitalized for a minimum of 48 hours. A total of 117 hospitalized patients who agreed to participate in the research were grouped according to the shift in which blood was collected (38 subjects from the morning shift and 79 from the afternoon shift), had their BDNF levels measured and responded to other questionnaires. Respondents were further grouped by age into three ranges: 18-30, 31-50 and 51-70 years. Results: We found a significant difference in the distribution of BDNF between the two shifts in which blood samples were collected, with the afternoon group having higher BDNF levels (U = 1906.5, p = 0.018). A difference was observed only between the 18-30 group and the 51-70 group in the afternoon shift (U morning = 1107, p morning = 0.575; U afternoon = 7175, p afternoon = 0.028). Conclusions: The population whose blood samples were collected in the afternoon showed significantly higher values of BDNF compared to those of the morning shift. This same population presented lower BDNF levels when associated with ASD subtypes A1, A2, and A. We hypothesize that the lower values of BDNF measured in the morning shift were due to a response to the circadian cycle of cortisol, whose action inhibits the expression of serum neurotrophins.


Introduction
Acute stress disorder (ASD) is a diagnosis organized under disorders related to trauma and stressors introduced in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV). Since then, a series of works have attempted to evaluate to what extent ASD might be a predictor of posttraumatic stress disorder (PTSD). 1 However, the centrality defined by the DSM-IV says that symptoms need to last for only two days for a diagnosis of ASD to be made. Therefore, for the scope of the present study, the authors deemed it necessary to adopt a more up-to-date version of the DSM. Thus, for ASD only, we used DSM-5 criteria for diagnosis. In the DSM-5, ASD is characterized by the onset of a minimum set of symptoms 3 to 30 days after the person has been exposed to a traumatic event. Avoidance, peritraumatic dissociation, intrusive re-experiencing, hyperexcitation, and disruption of normal social/work functioning are the main symptoms listed. After the 30-day mark, diagnosis should be changed to PTSD. 2 The National Comorbidity Survey Replication 3 estimates that approximately 6.8% of the general population will suffer from PTSD at some point in their lives, with the incidence being as high as 9.2% for individuals in the 45 to 60 age range. Women are more often affected than men, at a ratio of 2:1 when adjusted for the relative frequency of trauma between genders.
There do not seem to be any studies on the prevalence of ASD in the general population, and prospective studies and systematic reviews differ according to the nature of the event and the context in which it is evaluated. 3 Various research centers, primarily American, Israeli, and Australian, are noteworthy for undertaking and publishing multi-pronged investigative projects. They range from diagnostic validation, through the search for biomarkers and relevant neuroendocrine findings, to the proposal and testing of different therapeutic strategies for PTSD. 1,4 Important public health resources may be expended in the post-diagnosis treatment of these individuals, as they often become incapacitated by chronic forms of PTSD or by depression and anxiety. 5,6 Add to it the prejudices brought on by being away from work and/or the loss of productivity due to treatments, and one can see the relevance of the problem.
In our area, studies on the presence of altered brain-derived neurotrophic factor (BDNF) levels in stress disorders have been appearing. 7,8 In light of the epidemiological evidence and the findings of the abovementioned studies, the need to better understand ASD is clear.  13 The original questionnaire forms were bought in the United States to be applied in this study, with permission granted by Pearson Education for the use of assessment products of trauma in childhood.

Mini-International Neuropsychiatric Interview (MINI)
For the evaluation of psychiatric comorbidities, we used During hospitalization, the event continued to be relived (criterion C). The subject presented marked aversion to stimuli that might activate memories of the trauma (criterion D). Finally, the individual had heightened symptoms of elevated anxiety or excitability or negative changes in mood and cognition (criterion E). 14 It is important to note that criterion A was subdivided to enable investigation of the specific types of trauma experienced by ASD patients.

Measurement (blood analysis)
Four milliliters of blood were drawn from each subject by venipuncture into an anticoagulant-free vacuum tube. The blood was centrifuged at 3000 g for 10 min and serum was kept frozen at -80°C until analysis. BDNF

Ethical considerations
The present study is part of a broader research

Results
A total of 117 patients admitted to the emergency hospital were evaluated. The reasons for their admission were aggression or accident involving physical trauma.
The subjects were grouped according to shift of blood collection: 38 subjects from the morning and 79 from the afternoon shift. Following BDNF measurement and questionnaire response analysis, one significant difference was found in the distribution of BDNF levels between the shifts: patients in the afternoon shift showed higher BDNF levels (U = 1906, p = 0.018) as presented in Table 1.
There were 103 male and 14 female participants.
The distribution of BDNF levels was compared between the ASD criterion and its subtypes. As seen in Table 3, only seven patients presented a positive diagnosis for general ASD (as opposed to one of its subtypes). Regardless of blood collection shift, no statistical differences were observed (U morning = 82, p morning = 0.983; U afternoon = 92, p afternoon = 0.640). Among the negative cases, the mean between the shifts differed by about 10 points. Also, a statistical difference was found between three subtypes of ASD, but only for the afternoon shift. Subjects with ASD subtypes A1 (U afternoon = 892, p afternoon = 0.013), A2 (U afternoon = 1005, p afternoon = 0.025), and A (U afternoon = 581, p afternoon = 0.044) presented a distribution of BDNF lower than that found in subjects without ASD.
Mental disorders were evaluated using the MINI instrument. The most prevalent disorders are presented in Table 4. When BDNF levels were compared according to shift, only substance abuse resulted significant in the afternoon shift (U afternoon = 218, p afternoon = 0.021).
The concentration of BDNF was examined in relation to age, number of ASD symptoms, DSQ domains, and scores

BDNF is a neurotrophin involved in neuronal growth
and survival as well as synaptic plasticity. 15 It is highly expressed in mammal brains, particularly in the hippocampus, whose function is associated with learning and memory. 8,16 In this way, BDNF participates in the formation of long-term memory and other cognitive processes and is an important factor to be evaluated in   In our literature search, no studies with similar findings were encountered, and we hypothesized that this could be due to the influence of circadian endogenous cortisol variations on BDNF serum production. Authors report that cortisol secretion intensity is higher in humans early in the morning and lower at the end of the day, varying from a maximum of approximately 20 μg/dl one hour before awakening in the morning to a minimum of about 5 μg/dl around midnight. 18 A recent study has revealed that glucocorticoid receptors downregulate the expression of BDNF. 19 In that study, the authors demonstrated that exposure to dexamethasone suppressed the expression of the transcription of BDNF in neuronal cells mediated by glucocorticoid receptors.
Thus, based on the hypothesis above, it can be inferred that greater endogenous cortisol serum levels may be associated with diminished BDNF serum expression, leading to the findings of our study (Table 1).
With this in mind, studying BDNF levels in patients who have suffered physical trauma, whether they developed ASD or not, has uncovered a significant difference in the distribution of BDNF levels in blood samples collected in morning vs. afternoon shifts, with higher BDNF levels found in the group tested in the afternoon (U = 1906.5, p = 0.018), as seen in had also been exposed to trauma but did not develop this comorbidity. The final cohort totaled 257 participants (PTSD = 151, no PTSD = 106) and underwent chisquare analysis. 32 Based on earlier studies involving both animal models and human participants, the hypothesis formulated in the current study was that individuals who carry the Met allele would be at greater risk for PTSD than those carrying the Val allele. 33 One study aimed at assessing the association of BDNF levels with PTSD and depression examined a cohort of 2,928 pregnant women (gestational age > 16 weeks) who received treatment at a pre-natal care clinic in Lima, Peru, and found a decrease in BDNF levels in expectant mothers who presented depression and PTSD, but not in those who has PTSD alone. 34 34 In the present study, the following factors were correlated with BDNF levels in 117 patients who were victims of severe physical trauma: collection shift, sociodemographic factors, ASD subtypes, and mental disorders. Seven patients from the sample were also diagnosed with ASD. The group whose blood collection occurred in the afternoon presented significantly greater BDNF values compared to the morning collection. This same group presented lower levels of BDNF when associated with ASD subtypes A, A1, and A2. Regarding mental disorders, only substance abuse was significant in relation to BDNF. Finally, Spearman's correlation showed that rationalization and reactive formation may be associated directly with BDNF levels. It is possible that the low levels of BDNF observed in the morning shift derive from a response to the circadian cortisol cycle, whose action inhibits the expression of serum neurotrophins.

Conclusion
The levels of BDNF in this study were not sufficiently different to distinguish between patients with ASD (only with its subtypes) and without ASD. We consider this a limitation of the study, as only seven patients met ASD criteria. More studies are necessary to obtain a more precise association of BDNF levels with the development of ASD. Perhaps, when this association is clarified, it will be possible to validate the use of this test in clinical practice for posttraumatic stress.