Traumatic brain injury (TBI) is one of the leading causes of death and disability related to trauma in the United States. In 2014, the CDC estimated that TBI accounted for 2.87 million Emergency Department visits, 288,000 hospitalizations, and 56,800 deaths (1). The Global Burden of Disease Study found that the age-adjusted incidence rate of TBI in 2016 was 369 per 100,000. The global prevalence of individuals living with TBI-related disability was 759 per 100,000; TBI alone accounted for 8.1 million years lived with disability in 2016 (2). TBI is characterized by two phases of injury: primary injury and secondary injury. Primary injury is the initial insult via physical impact and damage to brain tissue. Secondary injury from TBI is characterized by an increase in neuronal excitotoxicity and progression of neuroinflammation (3).

Current treatment modalities for severe TBI focus on control of intracranial hypertension, maintenance of adequate cerebral perfusion pressure, and supportive care. Treatment options include surgical decompression, hyperosmolar therapy, temporizing ventilator strategies, optimization of nutrition, infection prophylaxis, and anti-epileptic medication for seizure prophylaxis (4). Despite the benefits of these options, few truly address slowing or halting the progression of TBI. Thus, research in TBI has emphasized understanding its pathophysiology in order to develop targeted therapies that limit injury progression and promote resolution (3).

Sexual dimorphism may influence the progression of TBI and may need to be considered in pre-clinical and clinical studies. Some clinical studies have suggested that there is a difference in outcomes with the female sex deriving some benefit in mortality and complication risk (5); other clinical studies have indicated that there is no significant difference (6). TBI severity may influence these observed differences (7). Pediatric studies have suggested that females may have some survival advantage compared to males after sustaining a TBI. Improved survival in pre- and post-pubescent stage females has been observed in retrospective studies of severe TBI (8, 9). Some retrospective studies allude to a variance in post-injury functional and psychological outcomes. For example, males and females who have sustained a TBI report different clinical symptoms, such as sleep disturbances and noise sensitivity (10, 11). Scott et al. found that psychological issues after TBI may differ in males and females as well; males were more likely to report substance abuse, and females were more likely to report experiencing anxiety or depression (11). In contrast to studies reporting differences, a recently published retrospective study of data from the Chinese Head Trauma Data Bank indicated no significant difference between male and female adult patients in mortality or long-term unfavorable outcomes after acute TBI (12).

Pre-clinical studies have demonstrated potential variability in the secondary injury of TBI. Doran et al. found that the increase in both peripheral myeloid influx and resident microglia proliferation in the injured cerebral hemisphere after controlled-cortical impact (CCI) injury was higher in male rats compared to female rats. Bromberg et al. found that Iba-1 staining of microglia, was higher in male rats in the paraventricular nucleus 7 days after sustaining fluid-percussion injury (FPI), indicating a higher number of microglia (13, 14). Armstead et al. have indicated that cerebral autoregulation may have sex-based differences (15–17). Different mechanisms to induce TBI are used in animal studies, which can influence the results. In a review of animal studies, 55% of studies using CCI as an injury mechanism reported females having better outcomes in post-injury evaluation such as behavioral and motor function studies. When using a closed-head injury (CHI) mechanism, 31% of studies showed females with improved outcomes in post-injury evaluations like cytoskeletal degradation, reactive astrogliosis, and cortical blood flow. Studies using fluid-percussion injury (FPI) reported females having worse survival and regulation of cell apoptosis in 60% of studies. However, these apparent differences may be in part depend on the outcomes measured (7).

After review of the information above and other studies ( Tables 1 , 2 ), we used our CCI model with our specific methodological outcome measures to uncover any sex-based differences in neuroinflammation. Experimental groups of male and female Sprague-Dawley rats sustaining either CCI or sham injury were compared. The progression of neuroinflammation at various time intervals after injury was evaluated based on the following features: microglial and splenocyte cell analysis with flow cytometry, blood-brain barrier (BBB) permeability, and tissue cytokine analysis ( Figure 1 ).

Our data found little difference in neuroinflammatory markers of injury when comparing males and females in a CCI model in the acute time period. These data are important as many have called for evaluating sex as a biological variable, by testing both males and females in experimental protocols. In this study, we sought to compare multiple properties of neuroinflammation between male and female sex, in a TBI rodent model. The properties compared were blood-brain barrier permeability, splenocyte activation, microglial activation, and inflammatory cytokine production. Each study was conducted at certain time intervals after injury ( Figure 1 ) to account for the progression of neuroinflammation after a TBI. Time points for each study were selected based on our prior experience with each test.

Splenocyte activation markers did not differ significantly between sexes at 7 days post-injury. Microglial activation markers 7-day post injury had some differences noted on t-SNE, as there may have been more typical non-activated microglia in the female CCI group. However, their density plots did show that this group also had a nearly equivalent number of activated microglia; morphometry of the microglia in male and female CCI groups was largely similar. Our multiple cytokine assay and traditional ELISA showed some minor differences between the male and female groups at the 6-hour time point post injury, but none that reached statistical significance. Of all the studies, only a subset of the BBB permeability analysis showed a significant difference between sexes.

In the analysis of BBB permeability at 72 hours post-injury, the female CCI group recorded a higher mean integrated density compared to the male CCI group, but not a higher mean intensity. This result does suggest that there may be some sex-based difference in how a traumatic injury affects BBB permeability. We considered that the significantly smaller body weights of the female rats may have factored into this, thus the reason for our correlation analysis between weights and integrated density values. However, there was not a statistically significant correlation of body weight with a lower integrated density in all CCI rats. We also looked for correlations between brain weights and BBB permeability after injury, but there were no trends or significant correlations observed. These lack of correlations between brain and body weight and BBB permeability make the potential for weight as a confounding factor unlikely. The result does suggest that BBB permeability may be greater in female rats after CCI, but this was only shown in one specific subset of the BBB permeability analysis, and only after a certain range of fluorescence intensities was set prior to the analysis. One could argue that if this was a true difference on the basis of sex, BBB permeability would be a consistent finding across all analyses using common threshold parameters. In contrast to our results, Julienne et al. used a similar CCI injury model, and did not find a difference in BBB alteration (27). Interestingly, they did find that there may be a difference in vascular repair, as male rats had an increased activation and expression of β-catenin, a marker of angiogenesis and vascular repair, as well as an increase in cortical vascular density (27). A recent study by Bharadwaj et al. found results regarding BBB permeability similar to ours. In a CCI-model of mice, they found that females had a greater extravasation of a macromolecular HRP tracer at 3 hours and 24 hours post injury, compared to males; they did not observe a difference in neuroglial response (28). An earlier study by Duvdevani et al. used a CCI rodent model as well, using Evans Blue dye to evaluate BBB permeability; they found no significant sex difference in BBB permeability, and no effect of hormonal status at 1 day post injury (29). O’Connor et al. found that BBB permeability may be elevated in male rats after a diffuse injury. In a frontal impact diffuse TBI injury model, O’Connor et al. found an increase in Evans Blue dye extravasation in male rats compared to female rats. Interestingly, administration of exogenous estrogen or progesterone reduced BBB permeability in males, and also attenuated the increase in cerebral edema post-TBI in males and ovariectomized females (21).

Our study does have limitations. We utilized a sham-injury model that does not include a craniectomy, as studies have found that craniectomy procedures create a risk of confounding injury to the brain and cerebral blood vessels and our study largely focuses on the difference between male and female CCI-injured groups (37, 38). In each individual experiment, we studied changes in neuroinflammation and TBI pathology at specified time points. These time points were selected based on prior studies using these tests in our lab and upon our interpretation of discussions in the literature of the importance of certain time points in the progression of neuroinflammation in TBI. It is possible however, that differences may be seen at different time points in each individual test. While the male and female rats were of equal age, we did not coordinate the injury to occur at a specific stage of estrous cycle of female rats. There was some variability in the female rat cytokine analysis, and it is possible that this variability may be due to individual rats being at various stages of the estrous cycle. However, it is worth noting that this variability was not limited to the female rats, as it was also noted in certain other neuroinflammatory markers for the male rat groups. Some markers of flow cytometry, such as CD45 and CD11b appear to have larger variability in the male CCI groups compared to the female groups. For this study, we selected time points used in previous studies of neuroinflammation following CCI in rats, however, there is a reasonable possibility that our assays and time points were not sufficient to capture dynamic or temporal differences between male and females in TBI.

Our study sought to elucidate whether there were clear and significant differences in outcomes of neuroinflammation in our CCI-model between male and female rats. As stated above, the only significant difference detected was a potential increase in BBB permeability. In other pre-clinical studies, it seems that potential sex-based differences are varied. These differences can extend to multiple different features of neuroinflammation, and may vary with different models of TBI ( Table 2 ). Similar to the human studies, pre-clinical animal studies have yielded varying results. Neurodegeneration in rats after TBI may evolve at different rates in each sex, but it does not appear that one sex incurs a higher level of neurodegeneration (30, 31). In a series of experiments with piglets, Armstead et al. found that cerebral autoregulation males may be impaired after FPI, which is thought to be due to an upregulation of endothelin-1 (15–17). In contrast to our findings regarding microglia, Doran et al. found some functional differences between sexes in rats sustaining CCI injury. After CCI injury, microglia from female rats had decreased phagocytic activity and production of reactive oxygen species; however, they found no sex-based difference in the expression of pro-inflammatory cytokine expression by microglia (14). Villapol et al. found results that were similar; in their rodent CCI-model, there was a faster influx and activation of microglia in males compared to females, but the differences became indistinguishable after 1-week post-injury (32). Some studies have also indicated that it may be females with a more rapid pro-inflammatory response. In a meta-analysis of their data from a closed head injury of mice, Späni et al. found that females had a greater increase in pro-inflammatory markers like IL-1β, IL-6, and TNF-α in the 6-hour window post-injury; they also found that the anti-inflammatory cytokine IL-10 was elevated in males, but not females (51). In a CCI rat model, Taylor et al. found that ovariectomized females had an increased expression of IL-1β and IL-6 in the ipsilateral hippocampus, but males had a higher expression of TNF-α (22).

As mentioned previously, one limitation of our study is that we did not control for the stage of estrous cycle in female rats. However, the literature on the relationship between hormones and pathology presents various findings. Certain levels of circulating endogenous hormones have been linked to improved outcomes in TBI. Roof et al. found that pro-estrous female rats recorded a lower brain edema content at injury site compared to males. This difference was even greater in female rats with induced pseudopregnancy. They found that the decrease in cerebral edema after injury was more likely due to progesterone and not estrogen, as this decrease was not seen in ovariectomized rats until progesterone supplementation was administered (23). In a diffuse TBI model, Roof et al. found that cortical blood flow (CBF) decreased in male rats at a greater rate than females after injury; CBF was lower in ovariectomized females compared to intact females. However, the decrease of CBF in ovariectomized rats was not corrected with the administration of estrogen replacement (24). As mentioned above, Duvdevani et al. did not find an effect by hormonal status on BBB permeability (29), whereas O’Connor et al. found a treatment effect of estrogen and progesterone on reducing BBB permeability in male rats (21).

In another TBI model using ovariectomized rats, Bramlett et al. showed that intact female rats had smaller contusion volumes compared to male rats and ovariectomized female rats. Neuroprotection in this specific study was not limited to a specific stage of the estrous cycle, as the average contusion volume was similar between pro-estrous and non-pro-estrous female rats (25). Wagner et al. found that female rats achieved better functional outcomes, but the difference did not depend on estrous cycle stage. Using the CCI-injury model, Wagner et al. found that regardless of estrous phase, female rats performed better than male rats in motor performance tests. However, that difference in behavioral performance was not influenced by serum estrogen and progesterone levels at the time of injury (26).

Interestingly, there is some literature that supports the notion that it is actually the injury itself that induces disruption of hormonal signaling. In an observational human study, Wagner et al. found that progesterone levels increased in male patients after injury, and in female patients the progesterone level was equivalent to control patients in the luteal or follicular phase of the menstrual cycle. They found that differences in hormone levels between the male and female groups were small and did not reach significance. In their combined study sample of male and female patients, high progesterone levels post-injury were associated with an increase in acute care mortality (18).. This does raise the question of whether the endogenous hormones alone or the disruptions in the hormone signaling due to cerebral injury create these potential differences in potential outcomes. Follow-up studies with TBI patients have found that injury is associated with chronic hypopituitarism (19, 20). Amongst all of this, testosterone may also need to be considered, as it can be an estradiol metabolite. In the previously mentioned study by Wagner et al., they reported that testosterone levels in male patients were lower after TBI compared to their controls, but in females, testosterone was elevated after injury. As with progesterone, they found a similar association with increased mortality in the patients that recorded elevated testosterone levels after injury (18). What our study does not evaluate is the potential response to any given treatment or intervention which may be influenced by sex of the animal.

Retrospective studies of sex-based differences in TBI have produced various findings of mortality rates after TBI ( Table 1 ). A large retrospective clinical study of patients in the Chinese Head Trauma Data Bank of TBI was conducted by the Chinese Head Trauma Study Collaborators. In their study of 7145 acute TBI patients, the mortality rate in males and females was similar. When the study group stratified for severe TBI, they again found similar mortality rates (12). In another large retrospective study of 72,294 patients with moderate to severe TBI, males were compared to females. There was a lower risk in mortality in female patients overall, but when the female patients were stratified, it was found that this difference was present in perimenopausal or post-menopausal females. There was no difference in mortality when comparing pre-menopausal female patients to males, downplaying the influence of endogenous hormones on TBI outcomes (5).

Our results fit with the current literature discussing the potential influence of sex on TBI ( Tables 1 , 2 ). Some sex-based differences have been noted in numerous studies utilizing various injury models. However, a distinct and common difference that is consistent across pre-clinical models has not been established in the most current literature. As our lab and others have discussed in prior reviews, the sex differences that may occur in TBI in the pre-clinical and clinical settings are wide, varied, and appear to have some dependence on the model or mechanism of TBI (51–53). TBI outcomes based on sex can also be dependent on the specific outcomes studied. In relation to our own model of TBI, our results indicate that the difference in outcomes between male and female rats may be minor, and difficult to control. In relating this to our review of the literature, further controlling our model on the basis of the hormonal balance or sex of rats may only yield small differences. These small differences may not reflect a significant or true variation in the progression of neuroinflammation based upon our preferred markers.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

The animal study was reviewed and approved by the Animal Welfare Committee University of Texas Health Science Center at Houston, TX, USA.

MS and KP wrote the main manuscript text and prepared figures. MS and SO prepared Figures 1 – 5 . AW prepared Tables 1 and  2 . The manuscripts was reviewed by all authors. All authors contributed to the article and approved the submitted version.

MS was supported by the National Institute of General Medical Sciences of the National Institutes of Health under award number 2T32GM008792. SO was supported by the National Institute of Neurological Disorders and Stroke under award number R21NS116302. CC and SO have received research support from Athersys, CBR Systems, Hope Bio, and Biostage. The funders were not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication.

CC is on the Scientific Advisory Board of Cellvation and CBR.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.