| Traumatic brain injury (TBI) is a leading cause of death and disabilityworldwide. While experimental models have provided valuable insight intothe pathophysiology of traumatic brain injury, most commonly used modelsrepresent significant departures from the conditions and forces encountered byhuman patients. Head fixation, opening of the skull and prolonged anesthesiathat are typically employed in experimental TBI models likely alter the courseof TBI development, reducing secondary injury and preventing the monitoringof intracranial pressure (ICP). We have developed a murine model of TBI,termed ‘Hit&Run’, which produces consistent brain contusion in thetemporal lobe without prolonged anesthesia, head fixation or skull openingthat allows chronic monitoring of ICP. The severity of injury is tightlycontrolled by a pneumatic impactor, and differing pressure settings cangenerate highly consistent mild, moderate or severe TBI. In characterizing thehistopathology and behavioral deficits at these injury grades we observed thatmild TBI-treated animals exhibited widespread cortical and subcorticalreactive gliosis. While they did not show evidence of frank cortical disruption,blood brain barrier opening, changes in ICP or any detectable cognitive deficit,mild TBI animals exhibited transient white matter atrophy, diffuse axonalinjury and neuronal loss comparable to moderate and severe TBI animals. Thisdiffuse neuronal injury was accompanied by chronic sensorimotor deficits inmild TBI animals. These results highlight the ‘Hit&Run’ TBI model as animportant advance compared to current models of TBI in which thepathophysiology of mild to severe TBI can be evaluated. We further detailcomplex changes in post-traumatic AQP4expression and localization inreactive astrocytes that may make important contributions to edema formationand resolution after traumatic injury. Most TBI cause cerebrospinal fluid (CSF) leak, at the same time, we studythe frontal sinus drainage pathway (FSDP) and evelop a novel method treatwith CSF rhinorrhea which make basic research transfer to clinicalachievement. We use percutaneous injection of fibrin glue manage frontalsinus CSF rhinorrhea which caused by trauma brain injury (TBI) set aexample for translational medicine, could be used more aggressively in a widearray of clinical scenarios.Part1White matter degeneration, neuronal loss and behavior deficit inthe modified CCI model of traumatic brain injury (TBI)Objectives: To develop a new experimental murine model that we call‘Hit&Run’ traumatic brain injury which is highly reproducible and can bemodulated to generate mild, moderate or severe grades of injury.Methods: Male C57Bl/6mice, aged10-12weeks were used in ourexperiments. The device modified from the commercially available ControlledCortical Impact Device (CCI). Measurement of infarct volume and bloodbrain barrier disruption, Histology and image analysis, Evaluation of cerebraledema and intracranial pressure; Neurobehavioral tests include neuroscore,open field test, rotarod, novel object recognition test, Barnes maze test.Results:1Optimization of the Hit&Run traumatic brain injury modelHit&Run injury was initially characterized at three different severities.Arousal time for mild TBI animals did not differ significantly from shamcontrol animals, whereas moderate and severe TBI animals exhibitedsignificantly longer arousal latencies. Out of366total TBI treated mice,delayed death was observed in1mild TBI mouse,1moderate TBI mouse and14severe TBI mice.2Blood brain barrier disruption and infarction2.1Depressed skull fractures were evident in10.8%(16out of148) insevere injury group, but not observed in either the moderate or the mild injurygroup. Ipsilateral subdural and/or subarachnoid hemorrhage and contusionwere evident in moderate and severe animals. Hemorrhages were confined to the site of impact in the moderate group, while in severe TBI animals, theinjury was more extensive, spreading into the deep cortex and underlyingwhite matter.2.2Mice in the mild group did not exhibit any appreciable infarction,while those with moderate injury exhibited only small infarcts. Compared toboth mild and moderate injury, severe TBI resulted in dramatically largerinfarcts, encompassing nearly75%of the volume of the contralateralhemisphere.3Evolution of reactive gliosis after injuryThe disrupted cortex at three days was continuous with a wide surroundingfield of reactive astrogliosis and microgliosis. Between7and28days afterinjury, the glial scar did not change appreciably in dimensions with extensivesurrounding astrogliosis evident at7and14days post injury that began toresolve by28days. Severe TBI resembled moderate injury, but was moreextensive, encompassing much of the ipsilateral cortex, subcortical whitematter, striatum and portions of the anterior hippocampus. More pronouncedcontrecoup injury was observed in severe TBI animals.4Neuronal loss and axonal degeneration after injury4.1After moderate TBI, a delayed loss of cortical tissue was observedbeginning7days after injury. In severe TBI animals, a similar loss of corticalvolume was evident immediately at3days. Neuronal loss was plainly evidentin the gliotic regions at both3and7days in all injury grades, including mildTBI, and did not appear to systematically differ between grades. In the cortexof the moderate and severe groups, neuronal loss did not appear to progressfrom3to7days, while a significant decline was observed in the mild groupbetween these two time points.4.2Immunolabeling for phosphorylated neurofilament (SMI-34)14daysafter injury revealed evidence of axonal degeneration in the ipsilateral cortexand subcortical white matter. While axon degeneration was more widespreadafter moderate TBI. 4.3MBP labeling in regions of GFAP-immunoreactivity surroundingmoderate lesions was both less intense and less continuous compared tomirror-image regions of the contralateral cortex. In the cortex surroundingsevers TBI lesions; demyelization was much more extensive at14dayspost-injury.5Brain water content was significantly elevated at1and3days post-TBIin both the ipsilateral and contralateral hemispheres of severe TBI animals, inaddition to the ipsilateral hemisphere of moderate TBI animals. Thecontralateral moderate TBI brain did not exhibit significant cerebral edema. At7days post-TBI, cerebral edema resolved in the contralateral hemisphere ofthe severe TBI and the ipsilateral hemisphere of the moderate TBI animals.6Intracranial pressures (ICP) quickly increased and peaked3dayspost-injury. ICP pulse wave amplitude significantly elevated at1daypost-injury. After TBI, ICP values normalized between4-6days post-injury.7There was no statistically significant increase in seizure-like activity.The absence of detectable post-traumatic seizure-like activity in the moderateTBI grade is consistent with the lack of frank hippocampal disruption and thegrossly preserved neuronal numbers within the CA1and CA3regions.8Behavioral evaluations8.1Control animals exhibited no deficits in global neuroscore (score of0)at any time points. At3hours post-injury, all TBI groups exhibited significantdeficits in neuroscore values. These deficits were most pronounced in thesevere group, with moderate and mild groups exhibiting similar levels ofdeficit. Mild TBI animals recovered quickly, returning to control valuesbetween24hrs and3days post-injury. Moderate and severe TBI animalsrecovered gradually between3and7days post-injury.8.2Rotarod test: Control animals improved in performance of the rotarodtest. In contrast, moderate/severe TBI animals failed to improve in the rotorodtask, exhibiting significant motor deficit beginning1week post-injury, andpersisting for the full duration of the experiment. Mild TBI animals exhibiteda deficit in the Rotarod task that was indistinguishable from that observed in moderate/severe animals.8.3Open field test: Severe TBI showed a trending deficit and motorfunction and anxiety.8.4The novel object recognition test: Control animals exhibited a stablenovel object preference in all five weeks of study. Moderate/severe TBIanimals exhibited a significant and persistent decline in novel objectrecognition test performance. Mild TBI animals did not exhibit a deficit innovel object recognition test performance at any time point.8.5Barnes maze test: Control animals exhibited a declining latency tocomplete the maze task. As observed with novel object recognition,moderate/severe TBI mice exhibited a significant cognitive deficit, showingno change in latency to complete the task on subsequent trial days. Also inaccordance with novel object recognition test findings, mild TBI animals didnot show any cognitive deficit compared to the control cohort. Mild TBIanimals had not deficit in cognitive function but suffered a motor function ormotor learning deficit that was as great as observed in moderate and severeTBI.Conclusions:We have developed and characterized a new TBI model, termed ‘Hit&Run’ TBI, in which the mouse head and body are freely mobile during impact,while the brain is exposed to angular and linear acceleration forces similar tothose in real-life accidents. Additionally, the Hit&Run model has otheradvantages, including (1) no pre-injury surgery is required;(2) the mice areanesthetized for only2-3min prior to impact;(3) the skull remains closed andskull fractures are avoided;(4) the model results in widespread reactive gliosisand delayed axonal degeneration, as well as a consistent contrecoup injury;(5)the severity of injury can be modulated, including the production of mildinsult characterized by diffuse injury and white matter degeneration. In all,this model will be highly conducive to large experimental trials ofneuroprotection. Part2Complex patterns of AQP4disorder after closed-skull traumaticbrain injury (TBI)Objective: To investigate the dynamics of AQP4disorder after both mildand moderate closed-skull injury and complex changes in post-traumaticAQP4expression and localization in reactive astrocytes that may makeimportant contributions to edema formation and resolution after traumaticinjury.Methods: Male C57Bl/6mice, aged10-12weeks were used in ourexperiments. The device used for “Hit&Run” TBI was modified from thecommercially-available Controlled Cortical Impact Device (CCI), Histologyand image analysis after labeling of glial fibrillary acidic protein (GFAP),CD68, NeuN, phosphorylated neurofilament (SMI-34), aquaproin-4(AQP4)and myelin basic protein (MBP).Results:1Hit&Run device and injury modelHit&Run injury was characterized at two different severities.‘Mild’ and‘moderate’ injuries utilized cortical impactor velocities of4.8m/s and5.2m/s,respectively, while the impact depth and contact times were held constant.2Development of post-traumatic reactive gliosisIn mild TBI animals3days post-injury, diffuse GFAP (reactive astrocytes)and CD68(reactive microglia) labeling were evident local to the impact site(Fig.1B). No defined glial scar was apparent in mild TBI animals at any timepoint. Cortical disruption was apparent in moderate TBI animals3dayspost-injury (Fig.1F) that evolved over the first week to form a well-definedglial scar surrounding the lesion cavity and reached a peak within3-7dayspost-injury.3Demyelination and axonal degeneration after TBINo measurable changes in cortical or subcortical volume were evident atany time point (Fig.2A-B). After moderate TBI, a delayed loss of corticaltissue was observed beginning7days after injury (Fig.2A;*P<0.05vs. control,2-way ANOVA). Loss of striatal tissue was muted compared to the cortex (Fig.1B;*P<0.05vs. control,2-way ANOVA). Both mild and moderate TBI, atransient atrophy of the subcortical white matter including the corpus callosumand the external capsule was evident that peaked at7days post-injury (Fig.2C;*P<0.05vs. control,2-way ANOVA) and normalized within14days of injury.In regions of GFAP-immunoreactivity surrounding moderate lesions, MBPlabeling was both less extensive and less continuous compared tomirror-image regions of the contralateral cortex (Fig.2D-F;*P<0.05vs.control,1-way ANOVA). Immunolabeling for phosphorylated neurofilament(SMI-34)14days after injury revealed evidence of axonal degeneration in theipsilateral cortex and subcortical white matter (Fig.2G, I, arrows). While axondegeneration was more widespread after moderate TBI (Fig.2G), it was alsoobserved in mild TBI animals (Fig.2I).4AQP4expression and localization are undergoing complex changeswithin the cortex and striatum after TBI. A mild increase in global AQP4expression coincided with a marked decrease in perivascular AQP4expression.Reactive gliosis in the chronic phase may provide a therapeutic avenue tonormalize AQP4expression and reduce post-traumatic seizure susceptibilityand improve clearance of interstitial wastes after moderate or severe TBI.Conclusion:We find that post-traumatic reactive astrocytes exhibit markeddepolarization of AQP4expression, with localization shifting from theperivascular endfoot processes to the wider somal compartments. Our findingssuggest that AQP4disorder likely does not contribute to the development ofpost-traumatic cerebral edema and changes in ICP, but may represent acompensatory response to these events. The observation that AQP4disorderresolved with reactive astrogliosis after mild TBI suggests that targetingreactive gliosis in the chronic phase may provide a therapeutic avenue tonormalize AQP4expression and reduce post-traumatic seizure susceptibilityand improve clearance of interstitial wastes after moderate or severe TBI. Part3Anatomical study and clinical treatment of frontal sinuscerebrospinal fluid rhinorrhea after trauma brain injury (TBI)Objectives: In this study, we describe the frontal sinus drainage pathwayof cerebrospinal fluid (CSF) and examined the effectiveness of percutaneousinjected fibrin glue as a treatment for frontal sinus CSF rhinorrhea in a seriesof4cases.Methods: All four cases were identified by experienced neurosurgeonsas frontal sinus CSF rhinorrhea from425patients, percutaneous injected fibringlue to frontal sinus after standard topical anesthesia. Postoperative care andefficacy evaluation were carry out by follow-up visits including CTexamination and routine physical examination, at2weeks and3,6, and12months after the surgery.Results: Display of the FSDP is useful when evaluating the cause andpotential surgical therapy for obstruction of the frontal sinus. The frontalostium forms the upper border of the superior compartment of FSDP. Theinferior-compartment of the FSDP is formed by the ethmoid infundibulum.In our case series, percutaneous injection of fibrin glue was successful inthe treatment of frontal sinus CSF rhinorrhea in all four patients. The surgerytime is27.6±7.98min; average hospitalization is11.25±4.99days. Norecurrence was found in an average of11.5months following injection offibrin glue. Frontal defects were found in3of the4cases (patients1,2and4);the glue was dispelled two days after the repair in the first case, butsubsequent re-repair was successful. Case2recovered well and received rightfrontal cranioplasty6months after repair. One patient (case3) eventually diedof tumor relapse in12month despite of successful repair of CSF rhinorrheawithout recurrence through10-month follow-up.Conclusion: In summary, display of the FSDP is useful when evaluatingthe cause and potential surgical therapy for obstruction of the frontal sinus. Wehave developed a novel method for stopping CSF rhinorrhea from theposterior wall of the frontal sinus that only requires examination of theanterior-posterior wall and the frontal ostium under CT image. This techniquecauses minimal injury, and thus could be used more aggressively in a widearray of clinical scenarios. An additional advantage is the fact that patients do not have to wait until the outcome of conventional conservative treatmentbecomes clear. The success of this novel yet simple method for stopping CSFrhinorrhea in our case series has encouraged further use. |
PDF Full Text Request