sign in sign up
<<
Home , Dexamethasone, Isoflurane

http://informahealthcare.com/bij ISSN: 0269-9052 (print), 1362-301X (electronic)

Brain Inj, Early Online: 1–8 ! 2014 Informa UK Ltd. DOI: 10.3109/02699052.2014.943289

ORIGINAL ARTICLE Effects of vs. dexamethasone on brain oedema and inflammatory responses following experimental brain resection

Fei-Fan Xu1,2, Stella Sun1, Amy S. W. Ho1, Derek Lee1, Karrie M. Y. Kiang1, Xiao-Qin Zhang1, Anna M. Wang3,4, Ed X. Wu3,4, Wai-Man Lui1, Bai-Yun Liu1,5,6, & Gilberto K. K. Leung1

1Division of Neurosurgery, Department of , Li Ka Shing Faculty of , The University of Hong Kong, Queen Mary Hospital, Hong Kong, PR China, 2Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, PR China, 3Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong, PR China, 4Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, PR China, 5Beijing Neurosurgical Institute, Beijing, PR China, and 6Department of Neurotrauma, General Hospital of Chinese People’s Armed Police Force, Beijing, PR China

Abstract Keywords Background: Dexamethasone (DEXA) is commonly used to reduce brain swelling during Brain oedema, dexamethasone, , neurosurgical procedures. DEXA, however, has many side-effects that can increase the risks of matrix metalloproteinase 9, progesterone, post-operative complications. In contrast, progesterone (PRO) has fewer side-effects and has surgical brain injury been found to be neuroprotective on (TBI). Whether PRO may be used as an alternative to DEXA during routine procedures has not been fully explored. History Object: To compare the effects of DEXA and PRO on surgical brain injury (SBI). Methods: Seventy-five adult male Sprague Dawley rats were randomized into five groups: (1) SBI Received 17 May 2014 + drug vehicle (peanut oil, 1 ml kg1); (2) SBI + DEXA (1 mg kg1); (3) SBI + low-dose PRO (10 mg Revised 26 June 2014 kg1); (4) SBI + high-dose PRO (20 mg kg1); and (5) sham SBI + drug vehicle. Magnetic Accepted 7 July 2014 resonance imaging study and assessments of brain water content (BWC), blood–brain barrier Published online 5 August 2014 (BBB) permeability, cellular inflammatory responses and matrix metalloproteinase 9 (MMP-9) expression were conducted. Results: This model consistently resulted in increased BWC and BBB disruption. PRO reduced For personal use only. astrocyte and microglia responses and attenuated brain oedema with preservation of BBB. A significant down-regulation of MMP-9 expression occurred in the PRO 20 group. Conclusions: PRO is as effective as DEXA in reducing brain oedema and inflammation following SBI; 10 mg kg1 of PRO was demonstrated to be more effective in relieving acute cellular inflammatory responses.

Introduction constitutes a form of penetrating surgical brain injury (SBI). Although the pathophysiology may differ, the unfavourable

Brain Inj Downloaded from informahealthcare.com by University Of Pittsburgh on 09/05/14 are effective in maintaining blood–brain of DEXA is an argument for seeking and barrier (BBB) integrity and reducing peritumoural oedema using an alternative adjunct during routine brain operations. [1,2]. In daily neurosurgical practice, dexamethasone (DEXA) The ideal agent should have well-defined neuroprotective is the form of glucocorticoids most commonly used during properties and few side-effects, as well as being inexpensive elective brain resection, with the rationale that it may and readily available. facilitate surgery and patient recovery by reducing brain Progesterone (PRO) is a potential candidate. Its neuropro- swelling and inflammatory responses [3,4]. DEXA, however, tective effects have already been demonstrated in many pre- has many side-effects such as glucose intolerance, immuno- clinical studies for a variety of neurologic conditions, suppression, , , gastric erosion and including TBI [6], stroke [7,8], spinal cord injury [9] and neuropsychiatric problems, all of which may cause post- chronic pain [10]. Clinically, two randomized controlled operative morbidities or even mortalities [3]. Furthermore, it studies have independently demonstrated the beneficial has now been established that the use of glucocorticoids is effects of PRO in TBI. PRO resulted in better neurologic detrimental and contraindicated in patients with traumatic recovery, was well tolerated and had minimal side-effects brain injury (TBI) [5]. Elective brain resection essentially [11,12]. This contrasts with DEXA and provides support for studying PRO as an alternative peri-operative adjunct. A Correspondence: Gilberto K. K. Leung, Division of Neurosurgery, previous report has shown that PRO could reduce brain Department of Surgery, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Queen Mary Hospital, Hong Kong, oedema following experimental cortical ablation, but no PR China. E-mail: [email protected] comparison with DEXA was made [13]. The present study 2 F.-F. Xu et al. Brain Inj, Early Online: 1–8

used an animal model mimicking clinical brain resection to the Committee for the Use of Live Animals in Teaching and compare DEXA with two dosages of PRO. It was hypothe- Research of the corresponding authors’ institution. sized that PRO would at least be as effective as DEXA in reducing oedema and inflammation around the resection Surgical procedures cavity. The aim was to provide pre-clinical evidence to support the clinical testing of PRO as a superior alternative to DEXA, The animal model has been described previously [14]. given the former’s more favourable pharmacodynamics. Briefly, following the induction of with 10% chloralhydrate (50 mg/100 g, i.p.), the rats were placed in Materials and methods a prone position on a stereotaxic frame under a surgical operating microscope. A craniotomy was made over the right Study design frontal skull such that the left lower corner of the craniotomy Seventy-five adult male Sprague Dawley rats (250–300 was at the bregma. The dura was opened to expose the right grams) were randomized into five treatment groups: Group frontal lobe. Using a micro-knife, two brain incisions were 1, SBI + drug vehicle (peanut oil, 1 ml kg1, n ¼ 15); Group made leading away from the bregma along the sagittal and 2, SBI + DEXA (1 mg kg1, n ¼ 15); Group 3, SBI + low- coronal planes to sever an area of brain 2 mm lateral to the dose PRO (10 mg kg1, n ¼ 15); Group 4, SBI + high-dose sagittal and 1 mm anterior to the coronal planes. The depth of PRO (20 mg kg, n ¼ 15); and Group 5, sham surgery + drug the incisions extended to the base of the skull, which vehicle (peanut oil, 1 ml kg1, n ¼ 15). Treatments were effectively resulted in a standardized partial frontal lobectomy administrated intraperitoneally (i.p.) at five time points: 12 (Figure 1(A)). Intraoperative surgical packing with gelatin hours before surgery and then at 1, 24, 48 and 72 hours after sponge (GelfoamÕ, Pfizer, New York, NY) was used to secure surgery. The brain samples for western blot (WB) and hemostasis. The dura was closed with 10/0 nylon sutures immunohistochemistry (IHC) were taken from the boundary (Ethicon, Johnson & Johnson, Edinburgh, UK). The skull of lesion site on the same 25 rats with five rats from each window was covered with paraffin film and the skin sutured group (Figure 1(A), WB part (green arrow) and IHC part with 4/0 silk (Ethicon). As the control group, animals in (yellow arrow)). Both PRO and DEXA were obtained from Group 5 received sham surgery that included only the Sigma-Aldrich (St. Louis, MO). The study was approved by craniotomy without any dural opening or brain resection. For personal use only. Brain Inj Downloaded from informahealthcare.com by University Of Pittsburgh on 09/05/14

Figure 1. (A) Intra-operative pictures showing the craniotomy window (left), the resection site (middle) and a harvested brain specimen with frontal lobe resection (right, WB part (green arrow) and IHC part (yellow arrow)). (B) MRI studies taken on post-operative day 3, showing the areas of oedema with T2-hyperintensity (red arrow). (C, left) BWC in different brain regions in the control and sham groups. (C, right) BWC around the operative site in different treatment groups. #p50.05 vs. sham. *p50.05 vs. control. DOI: 10.3109/02699052.2014.943289 Progesterone vs. dexamethasone in brain injury 3

Magnetic resonance imaging (MRI) Membranes probed with -actin were then probed with horseradish peroxidase (HRP)-conjugated secondary antibody Seventy-two hours after surgery but before sacrifice for the against rabbit at 1:10 000 dilution (Santa Cruz rest tests, three rats from each group were randomly selected Biotechnology) for 1 hour at room temperature, while those for MRI. The animals were anaesthetized with 1.5% with MMP-9 were probed with Biotin goat-anti-rabbit isoflurane and placed in a plastic stereotactic device. All secondary antibody at 1:5000 dilution (Invitrogen, CA), MRI data were acquired on a 7-Tesla scanner with a followed by Streptavidin HRP Conjugate tertiary antibody maximum gradient of 360 mT m1 (70/16 PharmaScan, (Zymed, Invitrogen) at 1:15 000 dilution for 1 hour at room BrukerBiospin GmbH, Germany). Multi-slice, proton- temperature. Immobilon Western HRP Substrate (Merck weighted scout images were acquired using two-dimensional Millipore, Billerica, MA) was used to detect immunoreactiv- rapid acquisition with refocused echoes (RARE) sequence ity signals. along the coronal, axial and sagittal planes to accurately locate the oedema region and to position the following MRI Immunohistochemistry (IHC) slices. T2-weighted images were acquired with 17 coronal slices, each of 0.5 mm thickness and without inter-slice gap. The right frontal lobes from the remaining animals were They were positioned to cover the rat brain around the harvested 72 hours after surgery (n ¼ 5 for each group). bregma (3.46 mm to 5.04 mm) and were scanned with The sham-operated rats were used as negative controls. the following parameters: RARE, TR/TE ¼ 4800/120 ms, Immunohistochemical staining for GFAP (for astrocytes) FOV ¼ 35 35 mm2, data matrix ¼ 256 256, NEX ¼ 6. and IBA1 (for microglia) was performed on consecutive 6 mm thick paraffin-embedded tissue sections. Briefly, tissue Brain water content (BWC) sections were deparaffinized with and antigens retrieved in 10 mM sodium citrate (pH ¼ 6.0). After perox- Animals (n ¼ 5 for each group) were sacrificed after an over- idase blocking, the specimens were blocked with 5% dose of general anaesthesia 3 days after surgery. The brains normal goat serum (Dako, Glostrup, Denmark) for 1 hour were removed and divided into six portions: frontal ipsilateral at room temperature, followed by incubation with mouse (to the resection), frontal contralateral, parietal ipsilateral, monoclonal anti-GFAP antibody (1:300 dilution; Cell parietal contralateral, brainstem and cerebellum. These parts Signaling Technology, Inc.) and rabbit polyclonal anti- were weighed immediately (wet weight) and again after IBA1 antibody (1:500 dilution; Wako, Inc.) at 4 C drying in an oven at 120 C for 24 hours (dry weight). The overnight. After washing with phosphate-buffered , percentage of water content was calculated as [(wet weight sections were incubated with ready-to-use DAKO En dry weight)/wet weight] 100%. Vision + Kit (Dako). Signals were detected using DAB (Dako) and subsequently counterstained with haematoxylin Blood–brain-barrier permeability (Vector Laboratories, Burlingame, CA). For personal use only. Two per cent Evans-blue (EB) (4 ml kg1, Sigma-Aldrich) in In the GFAP/IBA1-immunostained sections, positive 0.9% saline was injected through the tail vein 3 days after reactivities around the lesion sites were quantified on four surgery (n ¼ 5 for each group). Animals were kept anaes- randomly selected sections per brain. On each section, four thetized for 1 hour after the injection and then perfused randomly selected images were captured along the edge of transcardially with saline to clear the cerebral circulation of lesion boundary zone. These were input to the image analysis EB. The brain was then removed and photographed. To system, Meta Morph Offline and the total area of positive quantify EB extravasation, the operated hemisphere was reactivities was automatically measured. The percentage of homogenized in 3 ml of N,N-dimethylformamide (Sigma- positive reactivities was calculated using the following Aldrich), then incubated at 56 C for 48 hours and formula: Ap/Af 100, where Ap is the total area of positive

Brain Inj Downloaded from informahealthcare.com by University Of Pittsburgh on 09/05/14 centrifuged. The absorption of the supernatant was measured reactivities and Af is area of the whole field. at 620 nm using a spectrophotometer (Thermo Scientific, Waltham, MA). Statistical analysis SPSS (version 18.0) software was used to conduct the Western blot (WB) statistical analysis. The correlation of the investigated Three days after surgery, brain tissue samples (n ¼ 5 for each parameters was analysed by the use of Student t-test or one- group) were taken and homogenized on ice and extracted way ANOVA with a post-hoc Bonferroni test. Results were using the RIPA Buffer (Cell Signaling Technologies, Inc., established as significant if p50.05. Danvers, MA) containing a concentration of Protease Inhibitor Cocktail, according to the manufacturer’s instruc- Results tions (Roche Diagnostics, Mannheim, Germany). Protein Magnetic resonance imaging lysates (50 mg per sample) were separated by 12% SDS-PAGE and transferred onto a PVDF membrane (Bio-Rad Representative T2-weighted images are shown in Figure 1(B). Laboratories, Hercules, CA). After milk-block for 1 hour at These were taken at 2.54 mm anterior to the bregma (slice room temperature, membranes were incubated overnight at thickness ¼ 0.5 mm) on day 3 post-surgery. Control group rats 4 C with the primary antibodies to matrix metalloproteinase showed significant T2-hyperintensity along the upper border (MMP)-9 at 1:700 dilution and -actin as loading control at of the resection site, indicating brain oedema. The oedema- 1:1000 dilution (both from Cell Signaling Technologies). tous cortex swelled and protruded through the bone window. 4 F.-F. Xu et al. Brain Inj, Early Online: 1–8

In contrast, DEXA- and PRO-treated animals displayed less three hormone-treated groups showed significantly reduced significant hyper-intense areas and less swelling. EB extravasation surrounding the operative site when compared with the control group (p50.05, Bonferroni test). Brain water content (BWC) However, there was no difference among these three groups. Comparing between the sham and control groups, signifi- Cellular inflammatory responses cantly greater BWC was found in the latter around the operative site in the residual right frontal lobe (Figure 1(C), GFAP (for astrocyte) and IBA1 (for microglia) expressions left, p50.001, Student t-test). There was no difference were studied as markers of acute cellular inflammatory between these two groups in other regions of the brain, responses. The expressions of both markers were significantly indicating that the experimental model could consistently increased in the control group (p50.05, Bonferroni test) induce localized brain oedema. All surgical groups (1–4) also when compared with sham treatment (Figure 2(B)). With showed higher BWC than the sham group over the right hormone treatments, both markers showed significantly lower frontal lobe (p50.05, Bonferroni test). When compared with levels of expression than the control group (Figure 2(B)). the control group, both DEXA and PRO treatments reduced Furthermore, low-dose PRO resulted in significantly less BWC to significantly lower levels (Figure 1(C), right, cellular inflammatory responses than DEXA and high-dose p50.05, Bonferroni test). No significant difference was PRO (Figure 2(B), p50.05, Bonferroni test). found between the three hormone-treated groups. MMP-9 expression Blood–brain barrier disruption This study also investigated the potential mechanism under- EB extravasation in the control group was markedly increased lying the effect of PRO by studying the expression of MMP-9. when compared with the sham group (Figure 2(A), p ¼ 0.000, In comparison with the sham group, this SBI model induced a Bonferroni test), indicating that the SBI model could significantly higher level of expression of MMP-9 in the induce a significant and localized BBB breakdown. All control group (Figure 3). Except the PRO 10 group, both For personal use only. Brain Inj Downloaded from informahealthcare.com by University Of Pittsburgh on 09/05/14

Figure 2. (A) The upper panel shows representative pictures of EB dye extravasation in different groups. The lower panel shows the quantitative analysis of EB extravasation. #p50.05 vs. sham. *p50.05 vs. control. (B) Representative pictures of immunohistochemical staining of GFAP and IBA1. The bar charts show the amount of positively stained area around the operative sites in each group. #p50.05 vs. sham. *p50.05 vs. control. **p50.05 vs. DEXA and PRO 20. DOI: 10.3109/02699052.2014.943289 Progesterone vs. dexamethasone in brain injury 5

DEXA has many unwanted side-effects [5] and there is now definitive evidence that it should not be used in TBI patients. This somewhat poses a conceptual problem with its use during routine brain surgery, except for its ability to reduce peritumoural oedema. In contrast, PRO has fewer side-effects and is well tolerated and effective in various forms of CNS insults [23]. Moreover, PRO can enter the CNS readily with a wide therapeutic window. Given its safety, low cost and accessibility, it is potentially a superior alternative to DEXA in the management of SBI.

Progesterone in SBI Most reported studies on PRO and TBI used animal models Figure 3. The upper panel shows a representative western blot of MMP- that simulate blunt head injuries [24]. PRO was found to 9 expression. The lower panel shows the results of quantitative analysis reduce oedema formation after frontal cortical contusion [25] of protein expression. #p50.05 vs. sham. *p50.05 vs. control. and impact-acceleration diffuse TBI [26]. An inverse correl- ation between serum PRO level and the degree of oedema has been described [27]. To date, only a handful of reports DEXA and PRO 20 groups caused substantial down- focused on cortical resection injuries such as ours. Attella regulation of MMP-9 (p50.05, Bonferroni test) when et al. [28] found a lesser degree of oedema in pseudopregnan- compared with the control group. trats (with higher PRO level) than in normal cycling ones after prefrontal cortex lesion. However, the authors suggested Discussion that the difference was due to different levels of vasopressin. Surgical brain injury Asbury et al. [13] also reported decreased neuronal death and better functional outcomes after bilateral medial pre-frontal Neurosurgical procedures, whether performed in an emer- cortex ablation in animals given PRO. Other investigators, gency or elective setting, may result in additional trauma due however, did not find any benefit from PRO after sensori- to cortical incision, parenchymal resection, brain retraction, motor cortex ablations in rats [29]. Previous work showed that haemorrhages and thermal injuries from electrocautery. These systemically administered PRO could ameliorate brain injury may cause serious post-operative complications such as brain induced by electrocautery [30]. The present study was swelling, ischaemia, raised intracranial pressure, functional conducted to further investigate the effects of PRO following deficits or even deaths. Various medications and methods are

For personal use only. brain resection. Brain swelling after brain injury may be being used to minimize these including osmotic agents, vasogenic and/or cytotoxic in nature. The former commonly glucocorticoids, ‘neuroprotective’ anaesthetic drugs and occurs within a few hours post-injury [31,32], followed by the induced hypothermia [15]. Given the large number of latter that may persist for weeks [31]. Brain oedema usually neurosurgical procedures performed on a daily basis and the peaks on 3–7 days after surgery and then decreases gradually. potential negative impact of SBI on patient outcome, further In this study, considering the available number of the animals research effort in finding an effective perioperative adjunct is in total for one time point has reached 75, it was decided to of major clinical importance [16]. The present study took into mainly focus on investigating the early stage of brain oedema considerations recent pre-clinical and clinical findings on instead of the late stage [33,34]. Although the observed PRO and investigated its use in modulating SBI. differences in BWC between groups in this study may appear Brain Inj Downloaded from informahealthcare.com by University Of Pittsburgh on 09/05/14 to be small (ranging from 82.4 ± 1.5 to 85.1 ± 1.2%), it must Progesterone as a be emphasized that even a small increase in BWC can lead to PRO is commonly known as a gonadal hormone. It is, a significant increase in tissue volume and intracranial however, also synthesized by oligodendrocytes and neurons pressure [35]. within the central nervous system (CNS), where it serves The mechanisms underlying the observed effects of PRO many important functions [17]. During , PRO are not fully understood. The occurrence of brain oedema is protects the foetal brain from oxidative stress and immune- closely related to the integrity of the BBB [33]. In general, the inflammatory reactions around the time of neuronal develop- BBB becomes maximally permeable from 4–6 hours after ment. It is currently thought that healing after CNS injury and injury, before regaining integrity throughout a 7-days period degeneration recapitulates many of these pre-natal events [18] [36,37]. PRO is thought to maintain BBB integrity during the and that exogenous PRO may improve the survival, differen- early phase by modulating the expression of many important tiation and functioning of post-natal neurons and glial cell mediators such as matrix metalloproteinases (MMPs) [38,39], [19]. There is now a growing body of evidence that PRO may p-glycoprotein efflux pump [40] and aquaporin-4 [25]. be a safe and effective treatment for TBI and other neurologic MMPs, in particular, are capable of degrading extracellular disorders in humans [8,20–22]. matrix proteins such as tight junction proteins and the This study compared PRO with DEXA because the latter neurovascular basal lamina of the BBB. Indeed, dysregulated has been in use for decades to reduce brain swelling and MMPs expressions are commonly observed with BBB inflammatory responses around the time of surgery [3]. breakdown after TBI, neurodegeneration and stroke [41–44]. 6 F.-F. Xu et al. Brain Inj, Early Online: 1–8

Amongst different MMPs, MMP-9 is the most up-regulated additional impact of pre-existing peritumoural oedema was one after TBI [45] and is involved in tissue remodelling unknown. Blood PRO level, known to correlate with brain following many pathologic conditions [33]. MMP-9 gene oedema, was not measured [27]. Furthermore, only pathologic knock-down, on the other hand, may protect against TBI and outcomes at one early post-operative time-point were studied. ischaemia due to reduced BBB breakdown, brain oedema How PRO may affect later events such as apoptosis, and inflammatory responses [46,47]. The present study found cavitation, glial scarring and functional recovery require that MMP-9 down-regulation was also likely to play a role in further investigations. Since MMP-9 expression has been reducing post-operative oedema. proved to have a positive relationship with the severity of The activation and migration of inflammatory cells and brain oedema or BBB breakdown, the guess is that it may also mediators can initiate a cascade of damaging inflammatory change after PRO 10 treatment in a different time compared reactions. Cellular responses such as reactive astrogliosis with control. Future studies should focus on these as well as and microglial activation contribute significantly towards different treatment regimens and the effects of combinatorial secondary injuries seen in TBI [48]. It is now thought that therapy. For instance, PRO may be used together with DEXA PRO and its active metabolites may attenuate post-traumatic to assess the former’s glucocorticoids-sparing effect or with inflammation by inhibiting cytokine release [49,50] as well vitamin D, which is known to act synergistically with PRO as immune cell activation and migration [23]. It was found [53]. Whether PRO would exert similar effects in animals that PRO could reduce astrocyte and microglia development (and patients) of different genders, age groups and endogen- in response to SBI, although between-group differences did ous PRO levels also need to be carefully evaluated in order to not correlate fully with those of BWC levels, suggesting that inform future clinical trials and applications. other mechanisms may be at work. Conclusions Timing and dosage Brain resection can cause significant post-operative morbid- The dosing regimen of PRO requires careful considerations. ities in daily neurosurgical practice. This pilot study demon- Roof et al. [51] showed that earlier treatments resulted in strates that, like DEXA, PRO can reduce brain oedema, BBB greater effects, although treatment given as late as 24 hours leakage and inflammatory responses surrounding a resection post-injury may still be beneficial. For obvious reasons, only cavity. Down-regulation of MMP-9 expression may play a post-injury treatment is feasible for accidental TBI. However, role in reducing oedema after brain resection. A lower dose as an adjunct for elective operations, both pre- and post- (10 mg kg1) of PRO was demonstrated to be more effective operative treatments are possible, if not desirable. In a than a higher dose (20 mg kg1) in relieving acute cellular previous study on electrocautery-induced injury, just a single inflammatory responses. Given the safety, availability and pre-operative dose of PRO was already effective in reducing more favourable pharmacodynamics of PRO, its use for the astrocytic hypertrophy and macrophage infiltration [30]. The protection against brain resection and other forms of SBI For personal use only. present study used both pre- and post-operative treatments to deserves further investigation. mimic the common regimen of DEXA in order to enhance clinical relevance and found it to be beneficial. Regarding the Acknowledgements optimal dosage, Goss et al. [52] reported that the efficacy of PRO followed a U-shaped curve, with the most effective We thank Dr C. F. Fung for his personal donation to the dosages being between 8–16 mg kg1; lower (4 mg kg1) and Department of Surgery, Li Ka Shing Faculty of Medicine, The higher (32 mg kg1) dosages were less effective. It was found University of Hong Kong, in support of academic research in that, in terms of reducing acute cellular inflammatory neurosurgery. responses, PRO was more effective at 10 mg kg1 than at a

Brain Inj Downloaded from informahealthcare.com by University Of Pittsburgh on 09/05/14 higher dosage. Conversely, the result showed that there was Declaration of interest no statistical difference between low dose and high dose PRO on reducing brain oedema and protecting the integrity of the The authors report no conflicts of interest. The BBB. However, within the two groups, the p-values of these authors alone are responsible for the content and writing of two parameters (p ¼ 0.075 for BWC analysis and p ¼ 0.072 the paper. for EB extravasation, ANOVA) were very close to a significantly statistical level (p50.05). Whether or not low References dose PRO has a better effect on reducing oedema and BBB 1. Barnes PJ. Molecular mechanisms and cellular effects of gluco- leakage than high dose needs to be investigated with more . Immunology & Clinics of North America animal samples. 2005;25:451–468. 2. Miller JD, Leech P. Effects of mannitol and therapy on intracranial volume-pressure relationships in patients. Journal of Limitations and future studies Neurosurgery 1975;42:274–281. 3. Bebawy JF. Perioperative for peritumoral intracranial This model only studied brain resection without testing : a review of mechanisms, efficacy, and side effects. Journal other common causes of SBI such as brain retraction. The of Neurosurgical Anesthesiology 2012;24:173–177. establishment of the model strictly followed Yamaguchi 4. Vecht CJ, Hovestadt A, Verbiest HB, van Vliet JJ, van et al.’s [14] method and, since they had already proved its Putten WL. Dose-effect relationship of dexamethasone on Karnofsky performance in metastatic brain tumors: a stabilization and reproducibility, the variation between groups randomized study of doses of 4, 8, and 16 mg per day. was not quantified. Without using a tumour implant, the Neurology 1994;44:675–680. DOI: 10.3109/02699052.2014.943289 Progesterone vs. dexamethasone in brain injury 7

5. Alderson P, Roberts I. Corticosteroids for acute traumatic brain 28. Attella MJ, Nattinville A, Stein DG. Hormonal state affects injury. Cochrane Database System Review 2005;CD000196. recovery from frontal cortex lesions in adult female rats. 6. Gibson CL, Gray LJ, Bath PM, Murphy SP. Progesterone for the Behavioral & Neural Biology 1987;48:352–367. treatment of experimental brain injury; a systematic review. Brain 29. Goldstein LB, Bullman S. Age but not sex affects motor recovery 2008;131:318–328. after unilateral sensorimotor cortex suction-ablation in the rat. 7. Jiang C, Wang J, Li X, Liu C, Chen N, Hao Y. Progesterone exerts Restorative Neurology & Neuroscience 1999;15:39–43. neuroprotective effects by inhibiting inflammatory response after 30. Un KC, Wang YC, Wu W, Leung GK. Systemic progesterone for stroke. Inflammation Research 2009;58:619–624. modulating electrocautery-induced secondary brain injury. Journal 8. Sayeed I, Stein DG. Progesterone as a neuroprotective factor in of Clinical Neuroscience 2013;20:1329–1330. traumatic and ischemic brain injury. Progress in Brain Research 31. Barzo P, Marmarou A, Fatouros P, Hayasaki K, Corwin F. 2009;175:219–237. Contribution of vasogenic and cellular edema to traumatic brain 9. Garay L, Deniselle MC, Lima A, Roig P, De Nicola AF. Effects of swelling measured by diffusion-weighted imaging. Journal of progesterone in the spinal cord of a mouse model of multiple Neurosurgery 1997;87:900–907. sclerosis. Journal of Steroid Biochemistry & Molecular Biology. 32. Hanstock CC, Faden AI, Bendall MR, Vink R. Diffusion-weighted 2007;107:228–237. imaging differentiates ischemic tissue from traumatized tissue. 10. Kilts JD, Tupler LA, Keefe FJ, Payne VM, Hamer RM, Naylor JC, Stroke 1994;25:843–848. Calnaido RP, Morey RA, Strauss JL, Parke G, et al. Neurosteroids 33. Donkin JJ, Vink R. Mechanisms of in traumatic and self-reported pain in veterans who served in the U.S. Military brain injury: therapeutic developments. Current Opinion in after September 11, 2001. Pain Medicine 2010;11:1469–1476. Neurology 2010;23:293–299. 11. Wright DW, Kellermann AL, Hertzberg VS, Clark PL, Frankel M, 34. Marmarou A. Pathophysiology of traumatic brain edema: current Goldstein FC, Salomone JP, Dent LL, Harris OA, Ander DS, et al. concepts. Acta Neurochirurgica Supplement 2003;86:7–10. ProTECT: a randomized of progesterone for acute 35. Marmarou A, Fatouros PP, Barzo´ P, Portella G, Yoshihara M, traumatic brain injury. Annals of Emergency Medicine 2007;49: Tsuji O, Yamamoto T, Laine F, Signoretti S, Ward JD, et al. 391–402. Contribution of edema and cerebral blood volume to traumatic 12. Xiao G, Wei J, Yan W, Wang W, Lu Z. Improved outcomes from brain swelling in head-injured patients. Journal of Neurosurgery the administration of progesterone for patients with acute severe 2000;93:183–193. traumatic brain injury: a randomized controlled trial. Critical Care 36. Barzo P, Marmarou A, Fatouros P, Corwin F, Dunbar J. Magnetic 2008;12:R61. resonance imaging-monitored acute blood-brain barrier changes in 13. Asbury ET, Fritts ME, Horton JE, Isaac WL. Progesterone experimental traumatic brain injury. Journal of Neurosurgery 1996; facilitates the acquisition of avoidance learning and protects against 85:1113–1121. subcortical neuronal death following prefrontal cortex ablation in 37. Habgood MD, Bye N, Dziegielewska KM, Ek CJ, Lane MA, Potter the rat. Behavioural Brain Research 1998;97:99–106. A, Morganti-Kossmann C, Saunders NR. Changes in blood-brain 14. Yamaguchi M, Jadhav V, Obenaus A, Colohan A, Zhang JH. Matrix barrier permeability to large and small following metalloproteinase inhibition attenuates brain edema in an in vivo traumatic brain injury in mice. European Journal of Neuroscience model of surgically-induced brain injury. Neurosurgery 2007;61: 2007;25:231–238. 1067–1075. 38. Ishrat T, Sayeed I, Atif F, Hua F, Stein DG. Progesterone and 15. Baughman VL. Brain protection during neurosurgery. attenuate blood-brain barrier dysfunction follow- Anesthesiology Clinics of North America 2002;20:315–327. ing permanent focal ischemia by regulating the expression of 16. Jadhav V, Solaroglu I, Obenaus A, Zhang JH. Neuroprotection matrix metalloproteinases. Experimental Neurology 2010;226: against surgically induced brain injury. Surgical Neurology 2007; 183–190. 67:15–20. 39. Wang J, Jiang C, Li X, Liu C, Cheng N, Hao Y. The protective For personal use only. 17. Wojtal K, Trojnar MK, Czuczwar SJ. Endogenous neuroprotec- mechanism of progesterone on blood-brain barrier in cerebral tive factors: neurosteroids. Pharmacological Reports 2006;58: ischemia in rats. Brain Research Bulletin 2009;79:426–430. 335–340. 40. Cutler SM, Cekic M, Miller DM, Wali B, VanLandingham JW, 18. Stein DG. Is progesterone a worthy candidate as a novel therapy for Stein DG. Progesterone improves acute recovery after traumatic traumatic brain injury? Dialogues in Clinical Neuroscience 2011; brain injury in the aged rat. Journal of Neurotrauma 2007;24: 13:352–359. 1475–1486. 19. Schumacher M, Robel P, Baulieu EE. Development and regener- 41. Candelario-Jalil E, Yang Y, Rosenberg GA. Diverse roles of matrix ation of the nervous system: a role for neurosteroids. metalloproteinases and tissue inhibitors of metalloproteinases in Developmental Neuroscience 1996;18:6–21. neuroinflammation and cerebral ischemia. Neuroscience 2009;158: 20. Singh M, Su C. Progesterone and neuroprotection. Hormones & 983–994. Behavior 2013;63:284–290. 42. Ding JY, Kreipke CW, Schafer P, Schafer S, Speirs SL, Rafols JA.

Brain Inj Downloaded from informahealthcare.com by University Of Pittsburgh on 09/05/14 21. Singh M, Sumien N, Kyser C, Simpkins JW. and Synapse loss regulated by matrix metalloproteinases in traumatic progesterone as neuroprotectants: what animal models teach us. brain injury is associated with hypoxia inducible factor-1alpha Frontiers in Bioscience 2008;13:1083–1089. expression. Brain Research 2009;1268:125–134. 22. Stein DG. Progesterone exerts neuroprotective effects after brain 43. Rosenberg GA. Matrix metalloproteinases and their multiple injury. Brain Research Review 2008;57:386–397. roles in neurodegenerative diseases. Lancet Neurology 2009;8: 23. Stein DG, Wright DW, Kellermann AL. Does progesterone have 205–216. neuroprotective properties? Annals of Emergency Medicine 2008; 44. Rosenberg GA, Yang Y. Vasogenic edema due to tight junction 51:164–172. disruption by matrix metalloproteinases in cerebral ischemia. 24. Deutsch ER, Espinoza TR, Atif F, Woodall E, Kaylor J, Wright Neurosurgical Focus 2007;22:1–9. DW. Progesterone’s role in neuroprotection, a review of the 45. Hayashi T, Kaneko Y, Yu S, Bae E, Stahl CE, Kawase T, van evidence. Brain Research 2013;1530:82–105. Loveren H, Sanberg PR, Borlongan CV. Quantitative analyses of 25. Guo Q, Sayeed I, Baronne LM, Hoffman SW, Guennoun R, Stein matrix metalloproteinase activity after traumatic brain injury in DG. Progesterone administration modulates AQP4 expression and adult rats. Brain Research 2009;1280:172–177. edema after traumatic brain injury in male rats. Experimental 46. Asahi M, Wang X, Mori T, Sumii T, Jung JC, Moskowitz MA, Fini Neurology 2006;198:469–478. ME, Lo EH. Effects of matrix metalloproteinase-9 gene knock-out 26. Maghool F, Khaksari M, Siahposht Khachki A. Differences in brain on the proteolysis of blood-brain barrier and white matter edema and intracranial pressure following traumatic brain injury components after cerebral ischemia. Journal of Neuroscience across the estrous cycle: involvement of female sex steroid 2001;21:7724–7732. hormones. Brain Research 2013;1497:61–72. 47. Wang X, Jung J, Asahi M, Chwang W, Russo L, Moskowitz MA, 27. Wright DW, Bauer ME, Hoffman SW, Stein DG. Serum proges- Dixon CE, Fini ME, Lo EH. Effects of matrix metalloproteinase-9 terone levels correlate with decreased cerebral edema after gene knock-out on morphological and motor outcomes after traumatic brain injury in male rats. Journal of Neurotrauma 2001; traumatic brain injury. Journal of Neuroscience 2000;20: 18:901–909. 7037–7042. 8 F.-F. Xu et al. Brain Inj, Early Online: 1–8

48. Das M, Mohapatra S, Mohapatra SS. New perspectives on central 51. Roof RL, Duvdevani R, Heyburn JW, Stein DG. Progesterone and peripheral immune responses to acute traumatic brain injury. rapidly decreases brain edema: treatment delayed up to 24 hours is Journal of Neuroinflammation 2012;9:236. still effective. Experimental Neurology 1996;138:246–251. 49. Anderson GD, Farin FM, Bammler TK, Beyer RP, Swan AA, 52. Goss CW, Hoffman SW, Stein DG. Behavioral effects and anatomic Wilkerson HW, Kantor ED, Hoane MR. The effect of progesterone correlates after brain injury: a progesterone dose-response study. dose on gene expression after traumatic brain injury. Journal of Pharmacology Biochemistry & Behavior 2003;76:231–242. Neurotrauma 2011;28:1827–1843. 53. Tang H, Hua F, Wang J, Sayeed I, Wang X, Chen Z, Yousuf S, Atif 50. Grossman KJ, Goss CW, Stein DG. Effects of progesterone on the F, Stein DG. Progesterone and vitamin D: improvement after inflammatory response to brain injury in the rat. Brain Research traumatic brain injury in middle-aged rats. Hormones & Behavior 2004;1008:29–39. 2013;64:527–538. For personal use only. Brain Inj Downloaded from informahealthcare.com by University Of Pittsburgh on 09/05/14