Protective effect of sex steroid hormones on morphological and cellular outcomes after neonatal hypoxia-ischemia: A meta-analysis of preclinical studies

doi:10.21203/rs.3.rs-2203667/v1

Download PDF

Research Article

Protective effect of sex steroid hormones on morphological and cellular outcomes after neonatal hypoxia-ischemia: A meta-analysis of preclinical studies

https://doi.org/10.21203/rs.3.rs-2203667/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 02 Jan, 2023

Read the published version in Neuroscience & Biobehavioral Reviews →

Version 1

posted

You are reading this latest preprint version

Sex steroid hormones play an important role in fetal development, brain functioning and neuronal protection. Growing evidence highlights positive effects of these hormones against brain damages induced by neonatal hypoxia-ischemia (HI). This systematic review and meta-analysis aim to verify the efficacy of sex steroid hormones to prevent HI-induced brain damage in rodent models. The protocol was registered at PROSPERO and a total of 22 articles were included. Moderate to large effects were observed in HI animals treated with sex steroid hormones in reducing cerebral infarction size and cell death, increasing neuronal survival and mitigating neuroinflammatory responses and astrocyte reactivity. A small effect was evidenced for cognitive function, and there was no significant effect for motor function. In summary, published rodent data suggest that sex steroid hormones such as progesterone and 17β estradiol improve morphological and cellular outcomes following neonatal HI. Additional research is paramount to examine neurological function during neonatal HI recovery.

steroid hormone

hypoxia-ischemia

brain development

cell death

neuroinflammation

neuroprotection

The immature central nervous system (CNS) shows a higher vulnerability to insults, which can have negative repercussions throughout the lifespan, according to brain damage severity [1]. Although neonatal hypoxia-ischemia (HI) is recognized as the main cause of childhood disabilities as well as a worldwide public health problem [2], there is no effective clinical treatment for this condition [3, 4]. HI models induced in rodents at early postnatal periods are widely used to mimic morphological and cellular changes, as well as functional deficits observed in human infants [5, 6]. Factors such as stage of brain development and sex can influence the course of HI pathophysiology and are a challenge for designing new therapeutic strategies [7, 8]. Sex steroid hormones, including androgens (testosterone, dihydrotestosterone and androstenedione), estrogens (estradiol, estrone and estriol), and progesterone, act through nuclear and membrane-associated receptors, which facilitate the activation of mechanisms mediated by both slow and fast signaling pathways [9]. In addition, the brain contains all the enzymes required for converting steroid hormones to several metabolites and elevated tissue levels of steroid hormones are found in regions highly affected by neonatal HI, such as the hippocampus [10, 11].

Steroid hormones are synthesized by endocrine organs outside the nervous system, acting mainly on reproductive organs and modulating reproductive behavior [12]. Nevertheless, during the developmental period, sex steroid hormones play an important role in intrauterine homeostasis and maturation of fetal organ systems [9]. They are involved in trophic effects, cell proliferation, neuroinflammation, pro- and anti-apoptotic mechanisms, calcium signaling and synaptogenesis, among others [9, 11, 13]. Remarkably, progesterone and its metabolite allopregnanolone are able to activate gamma-aminobutyric acid type A (GABA_A) receptors. In fact, GABA exerts excitatory actions before becoming the main inhibitory neurotransmitter in the mammalian brain [12, 14]. In this sense, steroid hormones play an important role in balancing mechanisms crucial for development and normal brain function as well as neuronal protection [12, 15].

Growing evidence suggests that sex steroid hormones affect cellular events following brain insults, causing a neuroprotective effect on both immature and adult brain [16–18]. Steroid hormones such as estradiol and progesterone, for example, are able to reduce cell death and DNA fragmentation [19], act as potent antioxidant and anti-inflammatory agents [20, 21], prevent white matter injury [22], promote neurogenesis, increase neurotrophic support [23], and improve behavioral function [24–26]. Thereby, steroid hormones have shown to reduce infarct volume, mitochondrial and blood-brain barrier dysfunction, brain edema, neuroinflammation, and neurological and sensory-motor deficits [27, 28]. These effects become relevant given the limited number of interventions currently available for reducing the negative consequences of neonatal HI.

Studies in adult animals have provided strong evidence about the positive effects of steroid hormones and have also revealed that age-dependent changes contribute to increased brain vulnerability to stroke, which could be related to endogenous hormonal fluctuations both in males and females [27]. Nevertheless, there are significant differences between neonatal and adult brain in terms of maturation and functional organization, added to the role of steroid hormones in prenatal and early postnatal development. Moreover, divergence in the degree of injury according to the stage of neurodevelopment (more severe at later stages), as well as disparities in sexual maturation during early postnatal period, could modify the sex steroid response in the immature brain and should not be ignored.

Despite that, studies showing a general analysis of the data reported in the literature regarding the efficacy of sex steroid hormones when administrated at early stages of neurodevelopment and focused on neonatal HI are scarce. Therefore, the present study aimed to conduct a systematic review and meta-analysis to verify the efficacy of sex steroid hormones treatment to prevent HI-induced morphological, cellular, and behavioral impairments in rodent models.

2.1 Search

The literature review consisted in the search of experimental research studies in three electronic databases: EMBASE, EBSCO and PubMed, from January 1995 until 1st March 2022. The PICO framework questions (P: population, rodents exposed to neonatal HI models; I: intervention, sex steroid hormone; C: comparison, administration of vehicle substance; O: outcome, as primary outcomes brain tissue damage, cell death and neuronal survival. Secondary outcomes: astrocyte reactivity and neuroinflammation as well as motor and cognitive function) were used to answer the following question of analysis: What is the effectiveness of sex steroid hormone treatment on morphological, cellular, and functional parameters in rodents submitted to neonatal HI models?

MeSH terms, keywords and other free terms related to neonatal HI, sex steroid hormones and rodents were used with Boolean operators (OR, AND) to combine the searches: (“hypoxia” OR “ischemia” OR “hypoxia-ischemia” OR “anoxia” OR “excitotoxicity” OR “asphyxia” OR “hypoxia ischemia” OR “periventricular leukomalacia” OR “white matter injury”) AND (“brain” OR “cerebral” OR “central nervous system” OR “encephalon” OR “encephalic” OR “hippocampus” OR “cortex” OR “striatum”) AND (“sex steroid hormone” OR “progesterone” OR “testosterone” OR “estrone” OR “estradiol” OR “estriol” OR “dihydrotestosterone” OR “androstenedione”) AND (“rodent” OR “rat” OR “mouse” OR “mice” OR “guinea pigs”).

The protocol of the present systematic review and meta-analysis was registered a priori at the PROSPERO site (ID: CRD42022326604) and it was conducted according to the parameters of the Systematic Review Center for Laboratory Animal Experimentation (SYRCLE) statements [29] and the Preferred Reporting Items for Systematic Reviews and Meta-Analysis - PRISMA [30].

2.2 Inclusion Criteria

Experimental studies published in peer-reviewed journals with full-text available and reporting data of sex steroid hormones administration in rodent models of neonatal HI were considered. Moreover, studies had to be written in English and provide data on at least one of the following outcomes: (a) cerebral infarction size; (b) cell death; (c) neuronal survival; (d) astrocyte reactivity; (e) neuroinflammation and (f) motor and cognitive function. To determine the inter-rater reliability during the inclusion process, the kappa test was performed and interpreted as previously described [31].

2.3 Exclusion Criteria

Articles were rejected following a list of prioritized exclusion criteria in order to reduce discrepancies between reviewers: 1) narrative review; 2) research not using rodent models; 3) studies not using analogous neonatal HI models; 4) studies using only in vitro/cellular culture; 5) studies including neonatal HI without exposing animals to sex steroid hormones; 6) co-treatment without description of sex steroid hormones treatment alone or the control group comparison was not clearly described; and 7) animals exposed to corticosteroid treatment. Articles with at least two positive evaluations were included in the meta-analysis.

2.4 Risk of bias and study quality assessment

The possibility of publication bias was assessed through the funnel plot of standard error vs. Hedges g (SMD), followed by Egger’s test with the corrected standard error formula proposed by Pustejovsky and Rodgers [32, 33]. In case publication bias was assumed, the trim-and-fill-procedure was performed to impute “missing studies” and to estimate the effect of publication bias on the interpretation of the results [34].

The risk of bias and quality of studies were characterized following a previously published method, based on a 9-point scoring scale based on the Collaborative Approach to Meta-Analysis and Review of Animal Data from Experimental Studies (CAMARADES) score [35]. In addition, Animal Research Reporting In Vivo Experiments (ARRIVE) essential guidelines 10: compliance questionnaire was used in order to verify the quality and reliability of published research [36, 37]. The evaluation process was carried out by Da Silva JL and Colucci ACM independently. Disagreements were resolved by consensus among reviewers when assessing the full-text article.

2.5 Data extraction

All abstracts identified by the search were imported to Rayyan software. Duplicates were removed and two authors (Durán-Carabali LE and Da Silva JL) independently assessed title and abstract against inclusion/exclusion criteria. After this process, data were extracted using a predefined table, including first author’s name, year of publication, strain, sex, age, HI model, type of sex steroid hormone, doses, administration time, duration and treatment onset regarding days after brain insult as well as HI timing (pre- and/or post injury induction - Table 1). Eventual conflicts were resolved by consensus. A minimum of three studies for each outcome were used for meta-analysis.

The primary outcome measures for the meta-analysis were differences in cerebral infarction size, cell death and neuronal survival between animals suffering HI and exposed to sex steroid hormones in comparison to control group. These parameters were chosen since they are widely employed outcomes used to confirm injury in models of HI. Secondary outcome measures were the differences in astrocyte reactivity and neuroinflammation as well as motor and cognitive function between experimental groups.

Data extracted from each parameter were: number of animals per group (HI-vehicle and HI-sex steroid hormone), mean, standard deviation (SD) or standard error of the mean. Outcomes presented only graphically were measured using WebPlotDigitizer software [1] For articles with separated serial measures, the mean value was estimated in order to determine general injury progression and hormone effects. Whenever key information was missing, the corresponding author was contacted by email and asked to provide it.

2.6 Moderator variables

To identify whether some variables could be significantly associated with the pooled effect size between experimental groups, as well as the heterogeneity across the studies, a random meta-regression analysis under the test of moderator was conducted. In addition, to assess the significance of predictors, the Knapp-Hartung adjustment was used. Thereby, variables considered for this analysis were: sex, type of hormone, doses, HI timing (pre and/or post-injury), duration of the treatment (1 day, 2 to 5 days, 6 to 9 days and ≥ 10 days) and quality of the study.

2.7 Data analysis

Data were analyzed using: “esc”, “meta”, “metafor” and “dmetar” packages in the Rstudio (RStudio Team − 2020, version 4.1.2. RStudio: Integrated Development for R. RStudio, PBC, Boston, MA) as previously described [37]. Effect size was depicted as small: 0.2; medium: 0.5; and large: 0.8 [38] and was calculated as standardized mean difference (SMD) with Hedges g in order to control potential bias in small studies [39]. To pool effect sizes, a random-effect model was performed, while for subgroup analysis a mixed-effect model was used. Heterogeneity was calculated using I² statistics, Cochran’s Q and prediction intervals. I² values greater than 50% were considered as an indicator of substantial heterogeneity between studies. Outlier and influencer cases were verified by means of graphic display of heterogeneity analysis (GOSH) to determine if any studies pushed the effect of analysis into one direction. In addition, pooled effect estimates were reevaluated after such outliers had been removed from the analysis. Publication bias and moderator analysis were described in sections 2.4 and 2.6, respectively. Graphs were performed using Neo4J Cypher Graph database (version 1.4.15).

3.1 Characteristics of included studies

Figure 1 depicts the flowchart of the screening process of eligible articles. Preliminary search of three databases identified 1459 articles. After removing 798 duplicates, 619 articles were withdrawn based on title and abstract information. During full-text screening 20 were removed considering the following exclusion criteria: 3 papers do not have the full-text available, 2 papers were not written in English, 6 studies did not use comparable neonatal HI models, 3 studies did not expose animals to sex steroid hormone treatment, 3 studies did not report relevant data considering primary and secondary outcomes and 1 study did not report data. For 2 eligible articles some information was requested from the corresponding authors, however, no response was obtained [22, 24]. Thereby, 22 articles were included in the meta-analysis and their main characteristics are described in Table 1. A substantial strength of agreement between reviewers was estimated by means of Kappa index (k = 0.647, CI= [0.4121–0.8819]; p = 0.001; %-agree = 82.9). Additional information regarding morphological, cellular and function effects was briefly summarized and can be found in the Supplementary Table 1.

Twenty studies used rats and two employed mice. One of the studies did not mention the strain of the animals used. As shown in Table 1, most studies (86%, n = 19) induced neonatal HI by using the Rice-Vannucci model [25, 26, 40–57]. One study used permanent unilateral common carotid ligation without hypoxic environment exposure [58] and two studies used perinatal asphyxia to induce brain damage [59, 60]. Fifty percent of the studies (n = 11) used mixed sexes for analysis. In turn, 22.7% of studies (n = 5) used female and male animals considering sex dimorphism and the other 27.3% used only males.

The sex steroid hormones used were progesterone (n = 11), 17β estradiol (n = 7), estetrol (n = 2), testosterone propionate (n = 2) and allopregnanolone (n = 1). Eleven studies evaluated the effectiveness of hormone treatment before HI induction, 12 studies started hormone treatment after brain damage and 5 studies used the combination of pre- and post-injury treatments. Regarding the doses, there was not a clear consensus for each hormone; however, doses of 8 (n = 6) and 10 mg/kg body weight (n = 7) were mostly used (Fig. 2b and Table 1). Eleven studies used a unique administration of the hormone, while 11 studies administered the sex steroid hormone from 2 to 5 days, followed by studies administering from 6 to 9 days (n = 3). Only one study used hormone administration for more than 10 days. A graph with the main studies characteristics and their relationship is shown in Fig. 2a.

3.2 Quality assessment

No study achieved a score ≤ 5, which is classified as poor-quality. Thirteen studies were classified as moderate (score 6 to 10), eight studies displayed high quality (score 11 to 15) and only one study was scored as very high quality (score ≥ 16). Although all studies reported that randomization processes were performed, no study clearly described the method by which subjects were allocated to the control and treatment group (Supplementary Table 2). In addition, only one study provided information regarding a sample size calculation method and indicated if a blinded procedure was used during injury induction [58]. The percentage of studies reporting the items based on the CAMARADES score is depicted in Fig. 3.

3.3 Primary outcome assessment

3.3.1 The effect of sex steroid hormones on cerebral infarction size

The effect of sex steroid hormones on cerebral infarction size was evaluated in twelve of the studies included in the present review (Supplementary Table 3.1): 67% of the studies evaluated morphological changes in the ipsilateral brain hemisphere to the insult (n = 8), as well as in the hippocampus (n = 3) and cerebral cortex (n = 1). Meta-analysis showed that sex steroid hormones significantly reduced tissue brain damage in rodents exposed to neonatal HI compared to the control group (overall effect (SMD) = 1.6579, CI= [0.8595–2.4563], p = 0.0002, Q = 155.76 (p < 0.0001), I² = 75.6%, prediction interval= [-3.3189–6.6348] - Supplementary Fig. 1.1). Subgroup analysis considering hormone treatment indicated a large effect induced by progesterone (SMD = 1.9035, CI= [0.5896–3.2174], I² = 78.1%) and 17β estradiol (SMD = 1.7967, CI= [0.4390–3.1544], I² = 75.4%), as well as a moderate effect of allopregnanolone (SMD = 0.7669, CI= [0.6500–0.8837], I² = 0.0%), in reducing brain tissue loss.

During the identification of potential outliers and influencer cases, the graphic display of heterogeneity (GOSH) diagnostic detected six studies (Supplementary Fig. 2.1) with a possible large impact on the pooled effect and heterogeneity observed. The removal of these studies shrinks considerably the I² heterogeneity from 75.6–65.3%. In addition, the general pooled effect of SMD = 0.8650 was smaller than the initial estimated SMD = 1.6579. Considering hormone subgroup, the exclusion of these data also reduced the heterogeneity on progesterone (I² from 78.1% to 62.2; SMD from 1.9035 to 0.8006) and 17β estradiol (I² from 75.4% to 70.8; SMD from 1.9035 to 1.7967 – Fig. 4) effects.

The analysis of publication bias through funnel plots and Egger’s regression test (Intercept = 2.654, 95% CI= [1.58–3.73], p < 0.0002) indicated an asymmetrical distribution in the size effect. The trim and fill procedure after outliers’ removal added a total of four studies. The estimated corrected effect was lower than the initially calculated one (SMD = 0.6502, CI= [ 0.0538–1.2466], p = 0.0335, Q = 117.57 (p < 0.0001), I² = 69.4%, prediction interval= [-3.0302–4.3307] – Supplementary Fig. 3.1). Nevertheless, an overall moderate effect was still observed in HI animals receiving administration of sex steroid hormones when the degree of brain tissue damage was considered. Moderator analysis did not show significant differences when moderators such as sex, hormone, duration, study quality and treatment onset (days) and HI-timing were considered (all p > 0.05).

3.3.2 Impact of sex steroid hormones on cerebral cell death

Meta-analysis showed that sex steroid hormones intervention significantly reduced the levels of cell death biomarkers (i.e Caspases, Bax, TUNEL staining) after neonatal HI insult (SMD = 2.2523, 95% CI= [ 0.5561–3.9485], p = 0.0121, Q = 134.45 (p < 0.0001), I² = 86.6%, prediction interval= [-5.7669–9.9551] - Supplementary Table 3.2 and Supplementary Fig. 1.2). Considering hormone type, a large effect was observed for progesterone (SMD = 2.9534, CI= [0.5806–5.3261], I² = 88.4%); in turn, a moderate pooled effect was evidenced for 17β estradiol (SMD = 0.6248, CI= [-1.1638–2.4134], I² = 54.8%). The effect size was not estimated for testosterone due to the reduced number of studies using this hormone. Outlier and influencer check procedures detected three studies corresponding to progesterone hormone. The exclusion of these studies reduced the outcome of the overall analysis, however, the estimative kept being as a large effect in favor of sex steroid hormones (SMD = 1.0830, 95% CI= [ 0.3448–1.8211], p = 0.0069, Q = 40.79 (p = 0.0003), I² = 63.2%, prediction interval= [-1.6351–3.8010] - Fig. 5).

Funnel plot and Eggers' test generated for this outcome indicated an asymmetry spread of the reported SMD, suggesting a risk of publication bias (Intercept = 3.526, 95% CI= [0.87–6.18], p = 0.0185). After trimming and filling analysis, 4 studies were added. This finding evidenced that the initial effect was overestimated due to publication bias (SMD = 0.6062, CI= [-0.2467–1.4592], p = 0.1533, Q = 61.23 (p < 0.0001), I² = 69.0%, prediction interval= [-3.0017–4.2141] – Supplementary Fig. 2.2 e 3.2). Moderator analysis revealed no significant differences in the overall effect estimates of sex steroid hormones on cell death considering sex, hormone, duration, treatment onset (all p > 0.05).

3.3.3 Sex steroid hormones effect on neuronal survival

For the neuronal survival outcome, exposure to sex steroid hormones increased the number of intact cells and the levels of phosphorylated AKT, thus evidencing a large effect size of the treatment in animal suffering neonatal HI (SMD= -1.4758, 95% CI= [-1.9385 - -1.0131], p < 0.0001, Q = 74.93 (p = 0.0001), I² = 53.3%, prediction interval= [-4.1988–1.2473] - Supplementary Table 3.3 and Fig. 1.3). Considering hormones as a subgroup, a larger effect was observed to progesterone (SMD= -2.2377, CI= [-4.5322–0.0568], I² = 74.6%), 17β estradiol (SMD= -2.6191, CI= [-4.3670 - -0.8712], I² = 0.0%) and estetrol (SMD= -1.1980, CI= [-1.6107 - -0.7852], I² = 42.7%). Three studies were identified as possible influencer of the heterogeneity (Supplementary Fig. 2.3). Although there was a decrease in the pooled effect size of neuronal survival, that pattern did not induce significant changes on the large effect observed in favor of HI animals treated with sex steroid hormones (SMD= -1.2348, 95% CI= [-1.5077 - -0.9618], p < 0.0001, Q = 36.64 (p = 0.2621), I² = 12.7%, prediction interval= [-2.4898–0.0203] – Fig. 6).

Assessment for publication bias evidenced an asymmetrical pattern in the data regarding neuronal survival (Eggers' test: Intercept= -3.54, 95% CI= [-4.58 - -2.5], p < 0.0001). After trim and fill analysis, eleven studies were imputed. This adjustment also indicated a large overall effect (SMD= -0.9050, CI= [-1.2333 - -0.5768], p < 0.0001, Q = 69.29 (p = 0.0067), I² = 37.9%, prediction interval= [-2.8082–0.9981] - Supplementary Fig. 3.3).

Meta-regression analysis evidenced significant regressions of SMD on timing of hormone administration (pre- and post injury induction) (R² = 20.76%, p = 0.0041, estimate= -0.7761, t= -3.0853), treatment onset after HI in days (R² = 31.39%, p = 0.0004, estimate= -0.0108, t= -3.9546) and duration of sex steroid hormone administration (R² = 19.47%, p = 0.0042, estimate = 0.3060, t = 3.0740). When a meta-regression was performed using them as predictors, an increase in the value of R² was observed (76.79%, p = 0.0098), indicating that 77% of the difference in true effect sizes can be explained for these predictors. No significant differences were detected for quality study (p = 0.2587), hormone (p = 0.7099) and sex (p = 0.1929). Therefore, when neuronal survival was considered, data suggested that pre-injury intervention, early treatment onset and chronic treatment were significantly associated with effect size in favor of sex steroid hormones.

Secondary outcomes assessment

3.3.4 The effect of sex steroid hormones on neuroinflammatory response

Inflammatory response was evaluated in 5 of 11 studies using progesterone hormone (Supplementary Table 3.4). HI caused an increase in proinflammatory biomarkers such as cytokines IL-6, IL1β, TNFα, among others. Nevertheless, meta-analysis estimated a large overall effect, in which HI animals exposed to hormone treatment were benefited (SMD = 2.9771, 95% CI= [1.6707–4.2834], p = 0.0003, Q = 83.34 (p < 0.0001), I² = 85.6%, prediction interval= [-1.8082–7.7623] – Supplementary Fig. 1.4). A mild reduction in the overall effect was observed after removal of two studies identified during outlier and influences analysis (SMD = 2.3288, 95% CI= [1.5148–3.1429], p < 0.0001, Q = 51.85 (p < 0.0001), I² = 80.7%, prediction interval= [-0.3390–4.9966] – Fig. 7a and Supplementary Fig. 2.4). Moreover, a funnel plot asymmetry was detected (Eggers' test: Intercept= -2.555, 95% CI= -0.67–5.78], p = 0.1485). The exclusion of outlier studies leaded to funnel plot symmetry and no study was added during trim and fill analysis (Supplementary Fig. 3.4). Meta-regression analysis showed significant differences when HI timing (R² = 27.46%, p = 0.0465, estimate= -2.5740, t= -2.2419) and duration (R² = 33.95%, p = 0.0273, estimate= -0.4362, t= -2.5440) were used as predictors. Using both variables into the model, the results indicated that they explain R² = 66.09% of the heterogeneity (p = 0.0498). No significant differences were detected for treatment onset after HI in days (p = 0.1369), quality study (p = 0.2745) and sex (p = 0.2348).

3.3.5 The effect of sex steroid hormones on astrocyte reactivity

Astrocyte reactivity was evaluated mainly in studies using progesterone and estetrol as treatment. Meta-analysis shows that sex steroid hormones reduced the expression of astrocyte biomarkers (GFAP and S100B) in animals submitted to neonatal HI when compared to control groups (SMD = 2.1079, 95% CI= [1.6597–2.5560], p < 0.0001, Q = 49.14 (p = 0.0018), I² = 51.2%, prediction interval= [0.1935–4.0222] – Supplementary Table 3.5 and Supplementary Fig. 1.5). Three outliers were detected (Supplementary Fig. 2.5); after their exclusion, lower heterogeneity, as well as a subtle increase in the overall effect were estimated (SMD = 2.1796, 95% CI= [1.8328–2.5264], p < 0.0001, Q = 22.91 (p = 0.3490), I² = 8.3%, prediction interval= [0.9205–3.4387]– Fig. 7b). Both funnel plot and Eggers' test (Intercept = 4.954, 95% CI= [2.54–7.37], p = 0.0005) generated by this outcome indicated asymmetrical spread of the reported SMD. Moreover, an overestimation of the size effect due to publication bias was showed following trim and fill analysis, in which 6 studies were added (Supplementary Fig. 3.5). Thereby, the most appropriate effect when controlling for selective publication should be SMD = 1.8704 (CI= [1.4771–2.2637], p < 0.0001, Q = 42.33 (p = 0.0306), I² = 36.2%, prediction interval= [0.0982–3.6426]) rather than SMD = 2.1796. Moderator analysis indicated HI timing (pre- and post-injury - R² = 16.47%, p = 0.0228, estimate = 0.8582, t = 2.4398) and treatment onset after HI in days (R² = 21.00%, p = 0.0142, estimate = 0.0126, t = 2.6531) as significant predictors of the effect size of sex steroid hormones treatment when astrocyte reactivity was considered. The analysis using both variables as predictors showed that the amount of heterogeneity accounted was R² = 52.92% (p = 0.0267).

3.3.6 Impact of sex steroid hormones on functional parameters

As for the memory function, administration of sex steroid hormones showed a small effect reducing HI-induced cognitive deficits (SMD = 0.4564, 95% CI= [-0.0027–0.9155], p = 0.050, Q = 30.12 (p = 0.0027), I² = 60.2%, prediction interval= [-1.0593–1.9721] – Supplementary Table 3.6 and Supplementary Fig. 1.6). Considering spatial and non-spatial memory function, when evaluated by means of Morris water maze (MWM - SMD = 0.1898, 95% CI= [-0.2201–0.5997], I² = 8.1%) and novel object recognition test (NOR - SMD = 0.8343, 95% CI= [-0.1792–1.8477], I² = 76.6%), an overall small and large effect were observed, respectively.

After exclusion of one outlier, no significant overall effect was observed for the spatial MWM task (SMD = 0.3284, 95% CI= [-0.0630–0.7198], p = 0.0918, Q = 20.44 (p = 0.0396), I² = 46.2%, prediction interval= [-0.8578–1.5146] – Supplementary Fig. 2.6). Otherwise, a moderate effect was estimated for non-spatial memory function (SMD = 0.5784, 95% CI= [ -0.4389–1.5957], I2 = 70.0% - Fig. 8a). Publication bias by funnel plot and Egger’s test was detected (Intercept = 5.662, 95% CI= [1.99–9.33], p = 0.01155) and one study was imputed after trim-and-fill adjustment (Supplementary Fig. 3.6). For motor function, non-overall effect was observed (SMD=-0.0073, 95% CI= [-0.5928–0.5781], p = 0.9772, Q = 11.93 (p = 0.1030), I² = 41.3%, prediction interval= [-1.4833–1.4686] – Fig. 8b and Supplementary Table 2.7). Meta-regression evidenced no differences in the effect size on behavioral outcomes when moderator predictors were used (all p > 0.05).

The present meta-analysis investigated the neuroprotective effects of sex steroid hormones treatment in rodents suffering neonatal HI. Results suggested moderate to large size effects of hormone administration against morphological and cellular consequences of HI, while a small impact was observed on functional deficits. The analysis also indicated a high degree of heterogeneity among studies, suggesting that guidelines that address methodological aspects are required to reduce the risk of spurious findings.

During the developmental period, sex hormones are responsible to sculping sexually dimorphic neuronal structures and circuits, that determine behavioral responses to hormones and sensory stimuli at adulthood [2]. The functional role of sex hormones subserve unique effects on neurodevelopmental processes and hormone receptor expression dynamically changes over development. Structures such as the cerebral cortex display a high concentration of estrogen receptors at early postnatal period; however, there is a decline in the number of these receptors after 14 days [3]. Interestingly, some studies have indicated that exposure to sex hormones during critical periods of development causes irreversible changes on the animal’s brains and its subsequent sexual behavioral (e.g animals without typical mating stance). Unfortunately, data are controversial when comparisons with humans are performed [3] .

The influence of sex steroid involves mechanism such as epigenetic changes, neural pruning, apoptosis, myelination and dendritic spine remodeling, processes that present a organizational effect and lead to the maturation and remodeling of cerebral networks [2]. As the brain undergoes changes in structural and functional parameters, which are related to the action of circulating hormones [10], there is significant concern with regard to the impact of exposure to exogenous sex hormone administration during sensitive periods in neuronal development and their potential impact on brain and behavioral development. In addition, the responses of newborn animals to sex hormone treatment after HI insult is likely to be unique from other forms of HI treatment.

For example, progesterone and estradiol play important roles in the fetal-placental unit, CNS development, signaling activity and immune system [9, 61, 62]. and have shown an important effect as protective agents after cerebral ischemia in adults [61, 63]. These characteristics have led to a growing number of studies focusing on the analysis of their potential effects on mitigating HI-induced damage at the early stages of neurodevelopment, as shown in the Table 1. In contrast, a reduced number of studies evaluated the effectiveness of estetrol or testosterone as hormone therapy against neonatal HI. Estetrol is a natural estrogenic steroid synthesized exclusively during pregnancy by the fetal liver [64, 65]. Moreover, fetal levels of estetrol are higher in comparison with maternal levels, suggesting it as a safe and efficacious candidate for the treatment of early brain insults. In turn, testosterone is tied to the masculinization of neuronal morphology and behavioral responses [66] and testosterone administration before HI insult causes a detrimental effect in females, inducing an intrinsic vulnerability to brain injury [43]. Future research using testosterone would be useful to better understand the mechanism underlying dysmorphic outcomes at the early stages of neurodevelopment.

Although dose-response curves were not run for sex steroid hormones, progesterone was mainly used in a dose of 8mg/kg, agreeing with previous studies that indicated it as an optimal dose to induce beneficial effects after stroke [61, 67]. Nevertheless, a wider range of doses was used regarding other steroid hormones, which may be associated with the heterogeneity observed and making it difficult to compare individual protocols. Despite of not statistical significance, the observation of dose-response outcomes in studies evaluating different hormone concentrations suggest that high doses of sex hormone (e.g estrogen) are neeed to induce neuroprotective effects in newborn rats [4],being related with a dose-dependent pattern [5].

Although several models of HI have been developed in order to mimic the pathophysiological mechanisms observed in term and preterm infants after brain insults [5], the Rice-Vannucci model performed at postnatal day 7 was widely used between the articles included in the present analysis. Interestingly, no paper using this model induced HI insult at postnatal days 1 to 5 in rats in order to simulate a brain lesion in preterm babies, even though prematurity s a potential cause of neonatal HI in humans and the its current treatment – therapeutic hypothermia - shows adverse effects for neonates born preterm [68].. Some authors proposed the supplementation of estradiol and progesterone in preterm infants to maintain conditions similar to those found in the intrauterine milieu and reduce the risk of neurological complications [9]. However, physiological characteristics such as the presence of fetal zone steroids (only present in humans and higher primates, such as baboons) in the adrenal cortex is a limitation for the use of rodent models to investigate the role of sex steroids hormones in conditions as prematurity.

Furthermore, there is clear evidence that sex is an important factor to determine histological, molecular and functional responses to neonatal HI and experimental treatments [7]. In addition, newborns differ by sex in regard to placental growth and structure, physical characteristics and brain vulnerability to perinatal complications [69]. It is plausible that sex hormones treatment could have collateral off-target effects on brain development given differences in prenatal exposure to sex hormones associated with sexual differentiation of the fetus as well as their potential effects on SNC maturation and organization. Despite that, 50% of the papers enrolled in the present study included both sexes into the same analysis, which did not allow to determine differences between males and females, neither to evidence if hormone treatment could be an alternative intervention with a sex dysmorphic profile. Only five studies showed a comparison between sexes, fact that limit the statistical power of sex as a predictor of steroid hormones effect. Thus, sex differences related to steroid hormone treatment and outcomes following neonatal HI remain poorly understood. It is mandatory that future research with translational potential include sufficient numbers of both males and females in preclinical work to be able to analyse sex differences.

Experimental studies often use lesion volume as the main outcome evaluated after cerebral ischemia [70]. Results presented here showed that estradiol, progesterone, or its metabolite, allopregnanolone exert a moderate effect to reduce cerebral infarction size as compared to the control group. Interestingly, it has been reported that progesterone increases the incidence of stroke-related death [71]. This fact was not verified due to the low number of articles reporting the mortality rate after hormone treatment. Nevertheless, our data in favor of sex steroid hormones were also confirmed after influencer analysis and publication bias adjustment, and because of similarities with previous meta-analysis and systematic reviews using stroke models in adult rodents [70–72]. They are suggestive that morphological parameters, such as degree of brain damage, can benefit from steroid hormonal intervention after a neonatal hypoxic-ischemic insult.

Another interesting outcome to be evaluated is activation of cell death pathways. Apoptosis is one of the main mechanisms of cell death following neonatal HI and cysteine-aspartic proteases (caspases), Bax and DNA fragmentation staining techniques are important biomarkers of damage [75]. Several studies suggest that steroid hormones can modulate apoptotic pathways, inflammatory responses, mitochondrial dysfunction, blood-brain barrier disruption and neurogenesis [11, 12, 27, 62, 73, 74]. Findings presented here show a positive effect of sex steroid hormones for this parameter, which is in agreement with data reported in conditions such as TBI and stroke [70]. In addition, an enhancement of the expression of anti-apoptotic factors such as phosphorylated AKT, Bcl-2 and the number of intact cells was also observed. There is strong evidence showing an improvement in neuronal survival and neurogenesis after brain damage, especially when progesterone and estradiol were used [27, 76].

Neonatal hypoxic-ischemic insults also trigger an inflammatory response which results in microglia activation and increase of inflammatory mediators including interleukin-1β, tumor necrosis factor alpha (TNFα) and nuclear factor kappa B (NF-κB) [77]. Progesterone is a well-known suppressor of the injury-induced inflammatory response [70], being the only hormone used to assess this parameter among the studies evaluated here. As expected, progesterone treatment showed a large size effect. On the other hand, some authors have reported estrogen also plays an important role in ischemic neuroprotection by down-regulating TNFα expression [78], which could be an interesting issue for further investigation. Altogether, it seems the neuroprotective role of steroid hormones in reducing infarct size is especially associated with the effect of progesterone and estradiol in modulating cell death and inflammatory pathways following neonatal HI. In addition, these findings highlight the need to further evaluate the effectiveness of steroid hormones in a temporal context, i.e., following injury progression for longer periods as well as evaluating different stages of brain development.

Animal stroke models have demonstrated that sex hormones play a pivotal role in the prognostic after brain insult [6]. A critical point in attempting to identify new therapeutic targets is to identify whether the treatment is rescuing the effects of HI or engaging separate neurodevelopmental processes that appear to be rescuing HI insult. To restate this, is the sex hormone treatment saving cells or merely enhancing cell growth and suppressing natural death of cells associated with neurodevelopmental pruning? Compiling the data presented here as well as evidence available in adult animals, it is clear that neurosteroids can modulated different points of the ischemic cascade reducing reactive gliosis, decreasing pro-inflammatory cytokines migration, promoting the activation of neuronal survival pathways, among others [7–9]. Therefore, it may be reasonable to suggest synergic neuroprotective effects, enhancing neurodevelopmental pathways and reducing HI-dependent damage.

In addition to these morphological and biochemical data, it is also worth mentioning brain damage caused by neonatal HI leads to long-term impairments in both motor and cognitive function [79]. Therefore, functional assessment should be considered in order to determine the efficacy of several treatments after brain insult [80, 81]. In addition, some divergences between histopathological results and behavioral outcomes have been reported, indicating that rodents submitted to HI models could present functional improvement without significant tissue preservation or vice-versa [72, 82]. Despite the relevance of functional responses, there are a dearth of studies including more exhaustive functional measures with regard to either short term or long-term outcomes. The estimation of neuroprotective effect of sex steroid hormones in the functional parameters was small or null, for cognitive and motor function, respectively. One possible explanation to that could be the methodological differences among studies, considering the test used or the variable reported, as well as the time point at which the measurement was performed. In this sense, this review highlights the relevance of further research including behavioral tests in order to confirm whether this result is a consequence of high heterogeneity among data or sex steroid hormones are protective under certain stages of the injury but are limited regarding functional prognosis. Therefore, the functional significance of sex steroid hormones treatment after neonatal HI remains to be clarified.

Treatment onset regarding the HI insult, as well as duration of hormone administration, were identified as relevant moderators of the effect of sex steroid hormones on neuronal survival and neuroinflammatory responses. Administration of steroid hormones immediately after HI insult (post-injury period) and during some days displayed more beneficial effects than hormone treatment prior to the injury and as a unique dose. These outcomes showed similarity with previous reports on stroke models [70, 72], suggesting that steroid hormones could be modulating cell death and inflammatory pathways at earlier stages of the insult. Thereby, further attention should be given to steroid hormones, particularly progesterone and estrogen since the encouraging therapeutic effects against neonatal brain insult would be of interest for translating it into pediatric clinical practice.

Concerning the quality of the studies, the papers analyzed here were classified with moderate and high scores. However, methodological limitations were identified in terms of description of the sample size calculation, randomization process and blinded assessment, agreeing with previous meta-analysis using animal models [37, 72]. The lack of relevant methodological information introduces bias and also influences the outcomes. From this perspective, the detailing of methodological procedures and protocols used in the studies should be encouraged.

The positive effects of sex steroid hormones shown here, estimated as a moderate effect size, point to the use neurosteroids as a promissory strategy against HI-induced brain damage. However, some limiting aspects should be taken into consideration. The analysis of funnel plot and Egger’s test evidenced the presence of publication bias in the studies included. Consequently, in some cases the benefits of sex steroid hormone on morphological and cellular outcomes following neonatal HI were overestimated. This may be related to the fact that only studies with available data could be included and consequently there was a small number of studies analyzed. Moreover, the inclusion of studies using different sex steroid hormones and the variability of protocols regarding hormone administration could impact the effect of hormonal treatment and the results should be interpretated with caution.

Overall, the present meta-analysis provides statistical evidence for the pleiotropic beneficial effects of sex steroid hormones on the immature brain suffering from neonatal HI. The promising neuroprotective effects of these hormones involve morphological and cellular changes associated to neuronal and glial function that counteract HI-induced damage. Future studies involving steroid hormones should prioritize standardization of protocols according to the available literature, especially for progesterone and estradiol – the two most used steroid hormones - as well as the detailing of methodological aspects. This review has identified fundamental areas where experimental evidence is still required, such as sex-specific analysis and long-term behavioral responses, among others, which could be relevant for potential clinical trials involving newborns going through neonatal HI.

Funding: This study was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES).

Conflict of Interest: The authors declare no conflict of interest.

Ethical approval: This article is a review; therefore, it does not contain any experiments with human participants or animals performed by any of the authors.

Consent for publication: all authors have read and approved the final version of the manuscript.

Availability of data and materials: The datasets generated and analyzed during the current study are detailed in the supplementary material. Additional information is available from the corresponding author on reasonable request.

Authors’ contributions: Conception and design of research: D-C, LE; dF, LS, literature search: D-C, LE; dS, JL; C, ACM; studies evaluation: dS, JL; C, ACM; data extraction: D-C, LE; dS, JL; C, ACM; analyzed data: D-C, LE; drafted manuscript: D-C, LE; dF, LS; edited and revised manuscript: D, C-LE; dF, LS; N, CA.

Acknowledgement: none

Semple BD, Blomgren K, Gimlin K, Ferriero DM, Noble-Haeusslein LJ. Brain development in rodents and humans: Identifying benchmarks of maturation and vulnerability to injury across species. Prog Neurobiol [Internet]. Elsevier Ltd; 2013;106–107:1–16. Available from: http://dx.doi.org/10.1016/j.pneurobio.2013.04.001
Scafidi J, Fagel DM, Ment LR, Vaccarino FM. Modeling premature brain injury and recovery. Int J Dev Neurosci. 2009;27:863–71.
Frajewicki A, Laštůvka Z, Borbélyová V, Khan S, Jandová K, Janišová K, et al. Perinatal Hypoxic-Ischemic Damage: Review of the Current Treatment Possibilities. Physiol. Res. 2020.
Davidson JO, Gonzalez F, Gressens P, Gunn AJ. Update on mechanisms of the pathophysiology of neonatal encephalopathy. Semin. Fetal Neonatal Med. 2021.
Hamdy N, Eide S, Sun HS, Feng ZP. Animal models for neonatal brain injury induced by hypoxic ischemic conditions in rodents. Exp Neurol [Internet]. Elsevier; 2020;334:113457. Available from: https://doi.org/10.1016/j.expneurol.2020.113457
Mallard C, Vexler Z. Modeling ischemia in the immature brain: how translational are animal models? Stroke. 2015;46:1922–2013.
Netto CA, Sanches E, Odorcyk FK, Duran-Carabali LE, Weis SN. Sex-dependent consequences of neonatal brain hypoxia-ischemia in the rat. J Neurosci Res. 2017;95:409–21.
Andersen SL. Trajectories of brain development: Point of vulnerability or window of opportunity? Neurosci Biobehav Rev. 2003;27:3–18.
Hübner S, Reich B, Heckmann M. Role of sex steroids and their receptors in human preterm infants: Impacts on future treatment strategies for cerebral development. Biochem Pharmacol. 2015;98:556–63.
McEwen BS, Davis PG, Parsons B, Donald W. Pfaf. The brain as a target for steroid hormone action. Ann Rev NeuroscL. 1979;65–112.
Fester L, Rune GM. Sex neurosteroids: Hormones made by the brain for the brain. Neurosci Lett. 2021;753.
Nuñez J. Sex and steroid hormones in early brain injury. Rev Endocr Metab Disord. 2012;13:173–86.
Singh M, Su C. Progesterone-induced neuroprotection: Factors that may predict therapeutic efficacy. Brain Res. 2013.
Ben-Ari Y. Excitatory actions of gaba during development: the nature of the nurture. Nat Rev Neurosci. 2002;3:728–39.
Jevtovic-Todorovic V. Testosterone: much more for the brain than a sex hormone. Br. J. Anaesth. 2022.
Hurn PD, Macrae IM. Estrogen as a neuroprotectant in stroke. J. Cereb. Blood Flow Metab. 2000.
Wise PM. Estrogens: Protective or risk factors in brain function? Prog. Neurobiol. 2003.
Merchenthaler I, Dellovade TL, Shughrue PJ. Neuroprotection by Estrogen in Animal Models of Global and Focal Ischemia. Ann N Y Acad Sci. 2003.
Jover T, Tanaka H, Calderone A, Oguro K, Bennett MVL, Etgen AM, et al. Estrogen protects against global ischemia-induced neuronal death and prevents activation of apoptotic signaling cascades in the hippocampal CA1. J Neurosci. 2002;22.
He J, Evans CO, Hoffman SW, Oyesiku NM, Stein DG. Progesterone and allopregnanolone reduce inflammatory cytokines after traumatic brain injury. Exp Neurol. 2004;189.
Gélinas S, Martinoli MG. Neuroprotective effect of estradiol and phytoestrogens on MPP+-induced cytotoxicity in neuronal PC12 cells. J Neurosci Res. 2002;70.
Liu G, Yan Y, Shi B, Huang J, Mu H, Li C, et al. Benefits of progesterone on brain immaturity and white matter injury induced by chronic hypoxia in neonatal rats. J Thorac Cardiovasc Surg [Internet]. Elsevier Inc.; 2020;160:e55–66. Available from: https://doi.org/10.1016/j.jtcvs.2020.03.057
Brown CM, Suzuki S, Jelks KAB, Wise PM. Estradiol is a potent protective, restorative, and trophic factor after brain injury. Semin. Reprod. Med. 2009.
Kawarai Y, Tanaka H, Kobayashi T, Shozu M. Progesterone as a Postnatal Prophylactic Agent for Encephalopathy Caused by Prenatal Hypoxic Ischemic Insult. Endocrinology. 2018;159:2264–74.
Waddell J, Hanscoma M, Edwards NS, McKenna MC, McCarthy MM. Sex differences in cell genesis, hippocampal volume and behavioral outcomes in a rat model of neonatal HI. Exp Neurol [Internet]. Elsevier B.V.; 2016;275:285–95. Available from: http://www.embase.com/search/results?subaction=viewrecord&from=export&id=L607445889
Peterson BL, Won S, Geddes RI, Sayeed I, Stein DG. Sex-related differences in effects of progesterone following neonatal hypoxic brain injury. Behav Brain Res [Internet]. Elsevier B.V.; 2015;286:152–65. Available from: http://dx.doi.org/10.1016/j.bbr.2015.03.005
Guennoun R, Zhu X, Fréchou M, Gaignard P, Slama A, Liere P, et al. Steroids in Stroke with Special Reference to Progesterone. Cell. Mol. Neurobiol. 2019.
Suzuki S, Brown CM, Wise PM. Neuroprotective effects of estrogens following ischemic stroke. Front. Neuroendocrinol. 2009.
de Vries RBM, Hooijmans CR, Langendam MW, van Luijk J, Leenaars M, Ritskes-Hoitinga M, et al. A protocol format for the preparation, registration and publication of systematic reviews of animal intervention studies. Evidence-based Preclin Med. 2015;2:e00007.
Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;339.
McHugh ML. Interrater reliability: The kappa statistic. Biochem Medica. 2012;22.
Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. Br Med J. 1997;315:629–34.
Pustejovsky JE, Rodgers MA. Testing for funnel plot asymmetry of standardized mean differences. Res Synth Methods. 2019;10.
Duval S, Tweedie R. Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics. 2000;56:455–63.
Macleod MR, O’Collins T, Howells DW, Donnan GA. Pooling of animal experimental data reveals influence of study design and publication bias. Stroke. 2004. p. 1203–8.
Hu K, Deng W, Yang J, Wei Y, Wen C, Li X, et al. Intermittent hypoxia reduces infarct size in rats with acute myocardial infarction: A systematic review and meta-analysis. BMC Cardiovasc. Disord. 2020.
Durán-Carabali LE, Odorcyk FK, Sanches EF, de Mattos MM, Anschau F, Netto CA. Effect of environmental enrichment on behavioral and morphological outcomes following neonatal hypoxia-ischemia in rodent models: A systematic review and meta-analysis. Mol Neurobiol [Internet]. Springer US; 2022;59:1970–91. Available from: https://doi.org/10.1007/s12035-022-02730-9
Cohen J. Statistical Power Analysis for the Behavioural Science (2nd Edition) [Internet]. Stat. Power Anaylsis Behav. Sci. 1988. Available from: http://repositorio.unan.edu.ni/2986/1/5624.pdf
Harrer M, Cuijpers P, Furukawa T., Ebert DD. Doing Meta-Analysis in R: A Hands-on Guide. Prot Lab. 2019;
Li Z, JX F, CY Z, H K, L S. Protection of androgen against hypoxic-ischemic brain damage in neonatal rats and possible mechanisms. chin J comtemp pediatr. 2006;8 (6).
Gerstner B, Lee J, DeSilva TM, Jensen FE, Volpe JJ, Rosenberg PA. 17β-Estradiol Protects Against Hypoxic/Ischemic White Matter Damage in the Neonatal Rat Brain. J Neurosci Res. 2009;87:2078–86.
Feng Y, Fratkins JD, LeBlanc MH. Estrogen attenuates hypoxic-ischemic brain injury in neonatal rats. Eur J Pharmacol. 2005;507:77–86.
Hill C, Threlkeld SW, Fitch RH. Early testosterone modulated sex differences in behavioral outcome following neonatal hypoxia ischemia in rats. Int J Dev Neurosci [Internet]. International Society for Developmental Neuroscience; 2011;29:621–8. Available from: http://dx.doi.org/10.1016/j.ijdevneu.2011.03.005
Khan AA, Ramli NB, Ismail ZM. The effects of Progesterone on Hypoxic Ischemic Injuries in the Cornu Ammonis (CA) region of the Hippocampus of Neonatal Rats. Int J Morphol. 2015;33:962–70.
Li X, Zhang J, Chai S, Wang X. Progesterone alleviates hypoxic-ischemic brain injury via the Akt/GSK-3β signaling pathway. Exp Ther Med. 2014;8:1241–6.
Li X, Zhang J, Zhu X, Hou R, Li X, Dong X, et al. Effects of progesterone on hippocampal ultrastructure and expression of inflammatory mediators in neonatal rats with hypoxic-ischemic brain injury. Exp Ther Med. 2014;7:1311–6.
Dong S, Zhang Q, Kong D, Zhou C, Zhou J, Han J, et al. Gender difference in the effect of progesterone on neonatal hypoxic/ischemic brain injury in mouse. Exp Neurol. 2018;306:190–8.
Wang X, Li X, Li D, Li X, Zhu X, Guo X. Neuroprotective effect of progesterone in newborn rats with hypoxic-ischemic encephalopathy. Int J Phys Sci. 2011;6:2894–900.
Wang X, Zhang J, Yang Y, Dong W, Wang F, Wang L, et al. Progesterone attenuates cerebral edema in neonatal rats with hypoxic-ischemic brain damage by inhibiting the expression of matrix metalloproteinase-9 and aquaporin-4. Exp Ther Med. 2013;6:263–7.
Fabres RB, da Rosa LA, de Souza SK, Cecconello AL, Azambuja AS, Sanches EF, et al. Effects of progesterone on the neonatal brain following hypoxia-ischemia. Metab Brain Dis. Metabolic Brain Disease; 2018;33:813–21.
Fabres RB, Montes NL, Camboim Y de M, de Souza SK, Nicola F, Tassinari ID, et al. Long-Lasting Actions of Progesterone Protect the Neonatal Brain Following Hypoxia-Ischemia. Cell Mol Neurobiol [Internet]. Springer US; 2020;40:1417–28. Available from: https://doi.org/10.1007/s10571-020-00827-0
Müller MM, Middelanis J, Meier C, Surbek D, Berger R. 17Β-Estradiol Protects 7-Day Old Rats From Acute Brain Injury and Reduces the Number of Apoptotic Cells. Reprod Sci. 2013;20:253–61.
Nuñez J, Yang Z, Jiang Y, Grandys T, Mark I, Levison SW. 17β-Estradiol protects the neonatal brain from hypoxia-ischemia. Exp Neurol. 2007;208:269–76.
Tskitishvili E, Nisolle M, Munaut C, Pequeux C, Gerard C, Noel A, et al. Estetrol attenuates neonatal hypoxic-ischemic brain injury. Exp Neurol [Internet]. Elsevier Inc.; 2014;261:298–307. Available from: http://dx.doi.org/10.1016/j.expneurol.2014.07.015
Tskitishvili E, Pequeux C, Munaut C, Viellevoye R, Nisolle M, Noël A, et al. Use of estetrol with other steroids for attenuation of neonatal hypoxic-ischemic brain injury: To combine or not to combine? Oncotarget. 2016;7:33722–43.
Tsuji M, Taguchi A, Ohshima M, Kasahara Y, Ikeda T. Progesterone and allopregnanolone exacerbate hypoxic-ischemic brain injury in immature rats. Exp Neurol [Internet]. Elsevier Inc.; 2012;233:214–20. Available from: http://dx.doi.org/10.1016/j.expneurol.2011.10.004
Fang C, Yang Y, Wang Q, Yao Y, Zhang X, He X. Intraventricular injection of human dental pulp stem cells improves hypoxic-ischemic brain damage in neonatal rats. PLoS One [Internet]. 2013;8:e66748. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3682969&tool=pmcentrez&rendertype=abstract
Atif F, Yousuf S, Stein DG. Progesterone in the treatment of neonatal arterial ischemic stroke and acute seizures: Role of BDNF/TrkB signaling. Neuropharmacology [Internet]. Elsevier Ltd; 2016;107:317–28. Available from: http://dx.doi.org/10.1016/j.neuropharm.2016.03.052
Saraceno GE, Bertolino MLA, Galeano P, Romero JI, Garcia-Segura LM, Capani F. Estradiol therapy in adulthood reverses glial and neuronal alterations caused by perinatal asphyxia. Exp Neurol [Internet]. Elsevier B.V.; 2010;223:615–22. Available from: http://dx.doi.org/10.1016/j.expneurol.2010.02.010
Saraceno GE, Bellini MJ, Garcia-Segura LM, Capani F. Estradiol activates PI3K/Akt/GSK3 pathway under chronic neurodegenerative conditions triggered by perinatal asphyxia. Front Pharmacol. 2018;9:1–10.
Guennoun R. Progesterone in the brain: Hormone, neurosteroid and neuroprotectant. Int J Mol Sci. 2020;21:1–21.
McCarthy MM. Estradiol and the developing brain. Physiol. Rev. 2008.
Ström JO, Ingberg E. Impact of methodology on estrogens’ effects on cerebral ischemia in rats: An updated meta-analysis. BMC Neurosci. 2014;15.
Holinka CF, Diczfalusy E, Coelingh Bennink HJT. Estetrol: A unique steroid in human pregnancy. J Steroid Biochem Mol Biol. 2008;110.
Schwers J, Eriksson G, Diczfalusy E. Metabolism of oestrone and oestradiol in the human foeto-placental unit at midpregnancy. Acta Endocrinol (Copenh). 1965;49.
Knickmeyer RC, Baron-Cohen S. Fetal testosterone and sex differences. Early Hum. Dev. 2006. p. 755–60.
Wali B, Ishrat T, Won S, Stein DG, Sayeed I. Progesterone in experimental permanent stroke: A dose-response and therapeutic time-window study. Brain. 2014;137.
Singer D. Pediatric Hypothermia : An Ambiguous Issue. Int J Environ Res Public Heal. 2021;18(21):11484.
Worsham W, Dalton S, Bilder DA. The Prenatal Hormone Milieu in Autism Spectrum Disorder. Front Psychiatry. 2021;12:1–18.
Gibson CL, Coomber B, Rathbone J. Is progesterone a candidate neuroprotective factor for treatment following ischemic stroke? Neuroscientist. 2009;15:324–32.
Wong R, Renton C, Gibson CL, Murphy SJ, Kendall DA, Bath PMW. Progesterone treatment for experimental stroke: An individual animal meta-analysis. J Cereb Blood Flow Metab [Internet]. Nature Publishing Group; 2013;33:1362–72. Available from: http://dx.doi.org/10.1038/jcbfm.2013.120
Gibson CL, Gray LJ, Bath PMW, Murphy SP. Progesterone for the treatment of experimental brain injury; a systematic review. Brain. 2008;131:318–28.
McCarthy MM. The two faces of estradiol: Effects on the developing brain. Neuroscientist. 2009.
Li X, Bai R, Zhang J, Wang X. Effect of progesterone intervention on the dynamic changes of AQP-4 in hypoxic-ischaemic brain damage. Int J Clin Exp Med. 2015;8:18831–6.
Kari S, Subramanian K, Altomonte IA, Murugesan A, Yli-Harja O, Kandhavelu M. Programmed cell death detection methods: a systematic review and a categorical comparison. Apoptosis [Internet]. Springer US; 2022; Available from: https://doi.org/10.1007/s10495-022-01735-y
Zhang Z, Yang R, Cai W, Bai Y, Sokabe M, Chen L. Treatment with progesterone after focal cerebral ischemia suppresses proliferation of progenitor cells but enhances survival of newborn neurons in adult male mice. Neuropharmacology. 2010;58.
Liu F, McCullough LD. Inflammatory responses in hypoxic ischemic encephalopathy. Acta Pharmacol Sin [Internet]. Nature Publishing Group; 2013;34:1121–30. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3764334&tool=pmcentrez&rendertype=abstract
Liao SL, Chen WY, Chen CJ. Estrogen attenuates tumor necrosis factor-α expression to provide ischemic neuroprotection in female rats. Neurosci Lett. 2002;330.
Millar LJ, Shi L, Hoerder-Suabedissen A, Molnár Z. Neonatal hypoxia ischaemia: Mechanisms, models, and therapeutic challenges. Front. Cell. Neurosci. 2017.
Dixon G, Badawi N, Kurinczuk JJ, Keogh JM, Silburn SR, Zubrick SR, et al. Early developmental outcomes after newborn encephalopathy. Pediatrics. 2002;109:26–33.
Hobbs CE, Oorschot DE. Neonatal rat hypoxia-ischemia: long-term rescue of striatal neurons and motor skills by combined antioxidant-hypothermia treatment. Brain Pathol [Internet]. 2008;18:443–54. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18371175
Reglodi D, Tamás A, Lengvári I. Examination of sensorimotor performance following middle cerebral artery occlusion in rats. Brain Res Bull. 2003;59.

Table 1: Summary of the main characteristics of the studies included in the meta-analysis (n=22), focused on the investigation of sex steroid hormones effects in neonatal hypoxia-ischemia (HI) in rodent models. Abbreviation: PND: Postnatal day, pUCCL: Permanent unilateral common carotid ligation; Mixed: studies using both sexes without sexual dimorphism analysis.

Author (Year)	Strain	Experimental model	Injury age (PND)	Sex	Sex steroid hormone	Treatment period	Doses	Age of tx onset (PND)	Treatment onset regarding to injury (time)	Doses administration time	Duration (days)
(Atif et al., 2016)	CD1 mice	pUCCL	12	Female/Male	Progesterone	Post-injury	8 mg/kg	12	1h	1h, 3h and every 24h for 6 days	7
(Dong et al., 2018)	C57BL/6J mice	Neonatal HI (75 min)	7	Female/Male	Progesterone	Post-injury	16 mg/kg	23	1h	1h, 6h and every 24h for 7 days	8
(Fabres et al., 2018)	Wistar rats	Neonatal HI (90 min)	7	Male	Progesterone	Pre, post, and pre/post- injury	10 mg/kg	7	Pre tx: -2h Post tx: 6h Both: -2h	Pre tx: -2h. Post tx: 6h and 24h. Both: -2h, 6h and 24h	1
(Fabres et al., 2020)	Wistar rats	Neonatal HI (60 min)	7	Male	Progesterone	Pre, post, and pre/post- injury	10 mg/kg	7	Pre tx: -2h Post tx: 6h Both: -2h	Pre tX: -2h. Post tx: 6h and 24h. Both: -2h, 6h and 24h	1
(Feng et al., 2005)	Sprague-Dawley rats	Neonatal HI (150 min)	7	Mixed	17β Estradiol	Pre /post-injury	Group 1: 25 μg/kg Group 2: 50 μg/kg Group 3: 100 μg/kg Group 4: 2.4 μg/day	4	All groups: -72h	Every 24h	Groups 1-3: 5 Group 4: 21
(Gerstner et al., 2009)	Long-Evans rats	Neonatal HI (60 min)	6	Male	17β Estradiol	Pre-injury	Group 1: 300 µg/kg Group 2: 600 µg/kg	6	-12h	-12h.	1
(Hill et al., 2011)	Wistar rats	Neonatal HI (120 min)	7	Female/Male	Testosterone propionate	Pre-injury	0,1 mg	1	-144h	Every 24h	5
(Khan et al., 2015)	Sprague-Dawley rats	Neonatal HI (60 min)	7	Male	Progesterone	Post-injury	4 mg/kg	7	10 min	10 min, 6h, 24h, 48h	3
(Li et al., 2006)	Sprague-Dawley rats	Neonatal HI (150 min)	7	Mixed	Testosterone propionate	Post-injury	25 mg/kg	7	0h	0h	1
(Li et al., 2014b)	Wistar rats	Neonatal HI (150 min)	7	Mixed	Progesterone	Pre-injury	8 mg/kg	7	-30 min	-30 min	1
(Li et al., 2014a)	Wistar rats	Neonatal HI (150 min)	7	Mixed	Progesterone	Pre-injury	8 mg/kg	7	-30 min	-30 min	1
(Müller et al., 2013)	Wistar rats	Neonatal HI (80 min)	7	Mixed	17β Estradiol	Pre /post-injury	0,05 μg/g	4	-64h	-64h, -40h, -16h, 3h	3
(Nuñez et al., 2007)	Wistar rats	Neonatal HI (90 min)	6	Female/Male	17β Estradiol	Pre-injury	50 μg	Group 1:1 Group 2: 5	Group 1: -120h Group 2: -24h	Group 1: -120h, -72h, -24h Group 2: -24h.	Group 1: 3 Group 2: 1
(Peterson et al., 2015)	Sprague-Dawley rats	Neonatal HI (210 min)	7	Mixed	Progesterone	Post-injury	8 mg/kg	7	0h	0h, 2h and every 24h for 7 days	8
(Saraceno et al., 2010)	Sprague-Dawley rats	Perinatal asphyxia (19 min)	0	Male	17β Estradiol	Post-injury	250 µg/kg	117	117 days	Every 24h	3
(Saraceno et al., 2018)	Sprague-Dawley rats	Perinatal Asphixia (19 min)	0	Male	17β Estradiol	Post-injury	250 μg/kg	117	117 days	Every 24h	3
(Tskitishvili et al., 2014)	Sprague-Dawley rats	Neonatal HI (55 min)	7	Mixed	Estetrol	Pre- and post-injury	Group 1: 1 mg/kg Group 2: 5 mg/kg Group 3: 10 mg/kg Group 4: 50 mg/kg	Pre tx: 4 Post tx: 7	Pre tx: -72h Post tx: 0	Pre tx: -72h, -48h, -24h and -30 min Post tx: 0h	Pre tx: 4 Post tx: 1
(Tskitishvili et al., 2016)	Sprague-Dawley rats	Neonatal HI (55 min)	7	Mixed	Estetrol	Pre- and post-injury	Group 1: 5 mg/kg Group 2: 10 mg/kg	Pre tx: 4 Post tx: 7	Pre tx: -72h Post tx: 0	Pre tx: -72h, -48h, -24h and -30 min Post tx: 0h	Pre tx: 4 Post tx: 1
(Tsuji et al., 2012)	Wistar rats	Neonatal HI (120, 80 or 50 min)	7, 14 or 21	Mixed	Progesterone Allopregnanolone	Pre/post-injury	10 mg/kg	7, 14 or 21	-120,-80 and -50.	Immediately before, 6h and 24h.	2
(Waddell et al., 2016)	Rats (strain not mentioned)	Neonatal HI (60 min)	10	Female/Male	17β Estradiol	Post-injury	10 µg	10	0	0h and 24h.	2
(Wang et al., 2011)	Wistar rats	Neonatal HI (150 min)	7	Mixed	Progesterone	Pre-injury	8 mg/kg	7	-30 min	-30 min	1
(Wang et al., 2013)	Wistar rats	Neonatal HI (150 min)	7	Mixed	Progesterone	Pre-injury	8 mg/kg	7	-30 min	-30 min	1

No competing interests reported.

Supplementary.pdf

Download PDF

Journal Publication

published 02 Jan, 2023

Read the published version in Neuroscience & Biobehavioral Reviews →

Version 1

posted

You are reading this latest preprint version

Protective effect of sex steroid hormones on morphological and cellular outcomes after neonatal hypoxia-ischemia: A meta-analysis of preclinical studies

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Methods

2.1 Search

2.2 Inclusion Criteria

2.3 Exclusion Criteria

2.4 Risk of bias and study quality assessment

2.5 Data extraction

2.6 Moderator variables

2.7 Data analysis

3. Results

3.1 Characteristics of included studies

3.2 Quality assessment

3.3 Primary outcome assessment

3.3.1 The effect of sex steroid hormones on cerebral infarction size

3.3.2 Impact of sex steroid hormones on cerebral cell death

3.3.3 Sex steroid hormones effect on neuronal survival

Secondary outcomes assessment

3.3.4 The effect of sex steroid hormones on neuroinflammatory response

3.3.5 The effect of sex steroid hormones on astrocyte reactivity

3.3.6 Impact of sex steroid hormones on functional parameters

4. Discussion

Conclusion

Declarations

References

Table

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1