Cannabinoid Receptor Modulation in Focal Ischemic Stroke: A Systematic Review and Meta-Analysis of Infarct Volume and Behavioral Deficits in Animal Models

Abstract

Objective

This systematic review and meta-analysis aimed to evaluate the impact of cannabinoid (CB) receptor modulation on infarct volume and behavioral deficits in animal models of focal ischemic stroke, with a primary focus on infarct outcomes.

Method

A comprehensive literature search was conducted following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, yielding 29 eligible studies for inclusion.

Results

The analysis revealed significant reductions in infarct volume with CB agonists, particularly CB1 and CB2 agonists, indicating their potential as neuroprotective agents. Subgroup analyses further highlighted specific agonists, such as ACEA and KN38-72717, as particularly effective in reducing infarct volume. Additionally, CB antagonists, particularly CB1 antagonists, such as SR141716, showed promising results in reducing infarct volume. Although improvements in neurological scores were observed with both agonists and antagonists, statistical significance was not reached, indicating the need for further investigation.

Conclusion

These results highlight the potential of CB receptor modulation as a neuroprotective strategy in ischemic strokes and underscore the need for further research to elucidate the underlying mechanisms and optimize therapeutic approaches.

Highlights of the Study

• Cannabinoid (CB) agonists, particularly CB1 and CB2 agonists, significantly reduced infarct volume, with standardized mean differences indicating robust neuroprotective effects.

• Among CB antagonists, CB1 antagonists like SR141716 significantly reduced infarct volume (SMD −2.93; p < 0.0001), while CB2 antagonists did not show similar effects, highlighting a differential role for CB1 versus CB2 receptors in mitigating ischemic injury.

• Both CB2 agonists and non-selective CB1/CB2 agonists demonstrated significant reductions in infarct volume, with SMDs of −3.69 and −3.60, respectively, suggesting a broader neuroprotective role for CB receptor activation across different subtypes.

Introduction

Stroke remains a predominant global health challenge, consistently escalating in prevalence across its subtypes, yet there exists no effective treatment to counteract its detrimental impacts [1–3]. Ischemic stroke, the most prevalent form, occurs when the blood flow to brain tissue becomes inadequate, causing cerebral infarction or damage. This often leads to lasting behavioral impairments. Animal models of ischemic stroke can be classified as global or focal [4]. Focal ischemic stroke can be induced in animal models using various methods, most commonly by obstructing the middle cerebral artery (middle cerebral artery occlusion [MCAO]). Global ischemic stroke can be achieved or simulated using several techniques, including 4-vessel occlusion (4-VO) or 2-vessel occlusion (2-VO), among other approaches [5, 6].

The endocannabinoid system (ECS) is a vital communication network between cells that regulates a range of physiological processes in the body, including synaptic transmission, memory processes, and inflammation [7]. This system primarily operates through cannabinoid receptor 1 (CB1R) and cannabinoid receptor 2 (CB2R), which are widely expressed in the central nervous and immune systems [7, 8]. Cannabinoids (CBs) are mainly classified into three categories: endocannabinoids, phytocannabinoids, and synthetic CBs [9]. In addition, research has shown that components of the ECS, such as CB receptors and endocannabinoids, are modified after stroke, suggesting their role in the body’s natural response to stroke damage [10, 11]. Preclinical studies have demonstrated the neuroprotective effects of various approaches to modulate the ECS, indicating a potential therapeutic benefit of targeting the ECS in stroke treatment [10, 11].

Although the meta-analysis by England et al. [9] provided foundational insights into CB therapies in experimental stroke, its inclusion of compounds with off-target mechanisms (i.e., effects independent of CB1/CB2 receptor activation) limits the ability to isolate neuroprotective pathways specific to canonical CB receptor signaling [9]. To address this gap, our systematic review employs a targeted methodological approach, restricting analysis to pharmacological agents that directly modulate CB1 or CB2 receptors. By excluding compounds with non-receptor-mediated mechanisms, this strategy enables a rigorous evaluation of the neuroprotective efficacy and therapeutic potential of selective CB1/CB2 receptor modulation in ischemic stroke, thereby clarifying their distinct roles in mitigating neuronal injury. The objective of this study was to systematically evaluate the effects of CB receptor modulation on infarct volume and behavioral deficits in animal models of focal ischemic stroke.

Methods

Search Strategy

Our search strategy adhered to established guidelines recommended by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A thorough examination of the existing literature was conducted, encompassing the electronic databases PubMed, Google Scholar, and ScienceDirect, from January 2000 to January 2024. The search was systematically executed utilizing Medical Subject Headings terms, specifically targeting relevant studies on stroke or ischemia in conjunction with CB-related terms, including “cannabinoid,” “endocannabinoids,” “CB1,” and “CB2.”

Study Selection

Two authors independently devised the search approach and resolved discrepancies with a third investigator. Eligible studies included original empirical research with quantitative assessment of behavioral deficits and/or infarct volume, utilizing CB1 or CB2 receptor-targeting agents for in vivo administration, focusing exclusively on brain injury outcomes, employing accepted focal ischemia induction methods, and intervention with CB receptor agonists or antagonists. The exclusion criteria were in vitro experiments, global ischemia models, neonatal stroke models, adjunctive treatments lacking CB-related mechanisms, CBs primarily acting on non-CB receptors, samples with genetic modulation of CB1 or CB2 receptors, and sample groups administered two drugs affecting the same receptor. Exclusion of global ischemia models and neonatal stroke ensured homogeneity in pathophysiology, while restricting to CB1/CB2-targeted agents isolated receptor-specific effects.

Outcome Measures and Data Extraction

The data extraction process was conducted by a team of four members, following rigorously established protocols. To ensure data accuracy, verification was performed by all four members, and any inconsistencies were resolved through consultation with two other team members. Pertinent study parameters, including study name, publication year, geographical location, sample size, drug classification, animal species, methodological approach to inducing stroke, injury duration, therapeutic regimen, frequency of administration, assessment time points, and Stroke Therapy Academic Industry Roundtable (STAIR) score, were meticulously collected. The primary outcomes under scrutiny were brain infarct volume and behavioral deficit measurements between the experimental and control cohorts.

Assessment of Risk of Bias

Studies that met our predefined criteria underwent a meticulous assessment to ensure the reliability of our findings. This evaluation utilized an eight-criterion scale adapted from the STAIR criteria and England et al. [9, 12, 13]. The key methodological aspects evaluated included the implementation of randomization to minimize selection bias, continuous temperature monitoring to control for confounding variables, blinding of outcome assessment to mitigate performance bias, evaluation at both acute and subacute/chronic phases to capture comprehensive outcomes, assessment beyond infarct size to consider broader impact measures, testing multiple doses to elucidate dose-response relationships, and repeated outcome assessments at various time points to validate findings longitudinally. By systematically addressing these criteria, we comprehensively evaluated the risk of bias within the included studies, thereby enhancing the reliability and validity of the meta-analysis. Detailed quality assessments of all included studies are shown in Table 1.

Table 1.

Quality assessment and risk of bias evaluation for each included study based on the Stroke Therapy Academic Industry Roundtable (STAIR) criteria

Quality Assessment by STAIR criteria
study name	A	B	C	D	E	F	G	H	total score
Yang et al. [14] (2020)	No	No	No	Yes	Yes	Yes	Yes	Yes	5
Bai et al. [15] (2017)	Yes	No	Yes	No	Yes	Yes	No	Yes	5
Caltana et al. [16] (2015)	No	Yes	Yes	Yes	Yes	No	Yes	No	5
Sun et al. [17] (2013)	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	7
Yokubaitis et al. [18] (2021)	No	Yes	No	No	Yes	Yes	No	Yes	4
Yang et al. [19] (2017)	Yes	Yes	Yes	No	Yes	Yes	Yes	No	6
Pottier et al. [20] (2017)	No	No	No	Yes	Yes	No	No	Yes	3
Ronca et al. [21] (2015)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	8
Yu et al. [22] (2015)	No	Yes	Yes	Yes	Yes	Yes	No	Yes	6
Choi et al. [23] (2013)	No	Yes	No	No	Yes	Yes	No	Yes	4
Zarruk et al. [24] (2012)	Yes	Yes	Yes	No	Yes	Yes	No	Yes	6
Shirazi et al. [25] (2021)	No	No	Yes	No	Yes	Yes	Yes	Yes	5
Murakami et al. [26] (2013)	No	Yes	No	No	Yes	Yes	No	Yes	4
Suzuki et al. [27] (2012)	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	7
Schmidt et al. [28] (2012)	No	Yes	Yes	Yes	Yes	Yes	No	Yes	6
Reichenbach et al. [29] (2016)	No	Yes	No	No	Yes	No	Yes	No	3
Jalin et al. [30] (2015)	Yes	No	Yes	No	Yes	Yes	No	Yes	5
Bravo-Ferrer et al. [31] (2016)	No	Yes	No	Yes	Yes	Yes	Yes	Yes	6
Zhang et al. [32] (2012)	No	Yes	No	No	Yes	Yes	Yes	Yes	5
Zhang et al. [33] (2009)	No	Yes	No	No	Yes	No	Yes	Yes	4
Zhang et al. [34] (2007)	No	Yes	No	Yes	Yes	Yes	Yes	Yes	6
Zhang et al. [35] (2008)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	8
Hu et al. [36] (2010)	Yes	Yes	Yes	No	Yes	No	No	No	4
Berger et al. [37] (2004)	Yes	Yes	No	No	Yes	No	No	No	3
Muthian et al. [38] (2004)	Yes	Yes	No	No	Yes	Yes	Yes	No	5
Hayakawa et al. [39] (2007)	No	Yes	No	No	Yes	Yes	Yes	Yes	5
Leker et al. [40] (2003)	No	Yes	Yes	No	Yes	Yes	No	Yes	5

Stroke Therapy Academic Industry Roundtable (STAIR) criteria: A, implementation of randomization to minimize selection bias; B, continuous temperature monitoring to control for confounding variables; C, blinding of outcome assessment to mitigate performance bias, D, evaluation at both acute and subacute/chronic phases to capture comprehensive outcomes; E, assessment beyond infarct size to consider broader impact measures; F, testing multiple doses to elucidate dose-response relationships; G, repeated outcome assessments at various time points to validate findings longitudinally; H, therapeutic time window relationship of a particular agonist conducted.

Statistical Analysis

For the statistical analysis, Review Manager (RevMan) version 5.4 and RStudio were utilized. By employing a random-effects model, calculation of the standard mean difference was carried out with 95% confidence interval (CI). Our analysis included constructing forest plots to illustrate primary outcomes, such as overall infarct size and behavioral deficit across CB agonists and antagonists. Further exploration included the creation of scatter diagrams to investigate the correlation between specific drug responses and infarct size. To assess the impact of time of administration, grouped bar plots with error bars and CIs were developed. A visual examination of the funnel plot was performed to evaluate publication bias. Statistical significance was set at a threshold of p < 0.05. Heterogeneity was evaluated using Higgins I², with values exceeding 50% considered statistically significant. A leave-one-out sensitivity analysis, conducted post hoc and guided by observed data patterns, assessed individual studies’ impact on overall heterogeneity. Subgroup analyses were prioritized over meta-regression due to heterogeneity in reported variables and sample size limitations.

Study Registration and Ethical Approval

This study was registered at the King Abdullah International Medical Research Center under protocol number NRR25/063/3.

Results

Literature Search

An extensive search across the databases yielded a total of 3,405 articles. After eliminating duplicates (n = 574), 2,831 unique articles remained for screening. Each of these 2,831 records underwent a thorough evaluation based on their titles and abstracts, resulting in the identification of 44 articles for further scrutiny. A subsequent detailed full-text examination led to the exclusion of 15 articles [41–55], citing reasons such as failure to meet the inclusion criteria and lack of relevant data. These reasons are documented in online supplementary Table 1 (for all online suppl. material, see https://doi.org/10.1159/000547277). Ultimately, 29 articles [14–40, 56, 57] were deemed appropriate for inclusion in the systematic review, representing a qualitative synthesis. Among these, 26 studies were eligible for incorporation into the meta-analysis, representing a quantitative synthesis. The selection process for the included studies is shown in online supplementary Figure 1.

Study Characteristics and Quality Assessment

The baseline characteristics of the included studies, detailed in Tables 2, revealed a diverse landscape spanning from 2003 to 2021 across several countries, including China, the USA, Argentina, Spain, Japan, South Korea, France, and Germany (Fig. 1–5). Predominantly employing rodent models, particularly Sprague-Dawley rats, C57Bl/6 mice, and Wistar rats, the studies investigated various drugs targeting CB receptors, such as ACEA, WIN55, 212-2, BCP, JWH-133, AM1241, TC, O-1966, O-3853, HU-210, SR141716A, KN38-72717, and SR144528. Stroke induction methods varied, including MCAO with or without reperfusion, photo thrombosis, and thromboembolic MCAO, with treatments administered either pre-stroke or post-stroke. The total number of animals included in the experimental group was 385, while the number of animals in the control or vehicle-treated group was 258. Outcome assessments were conducted at various post-stroke intervals, encompassing parameters such as infarct volume, neurological deficit scores, grip strength, ladder-rung walking, and foot-fault tests. The quality of evidence presented in each study was evaluated using the STAIR score, providing an assessment of study reliability and validity.

Table 2.

Baseline characteristics of included studies

Author	Year	Country	Drug	Mechanism of action	Animal	Stroke method	Therapy type	Timing of administrations	Outcome	Assessment time points	Significant improvements
Yang et al. [14]	2020	China	ACEA	CB1 agonist	Sprague-Dawley rats	MCAO/r	Pre-treatment	30 min before injury	Substantial reduction compared to control	Day 1	Longa’s neurological deficit score
Bai et al. [15]	2017	China	ACEA	CB1 agonist	Sprague-Dawley rats	MCAO/r	Post-treatment	1 h after MCAO	Significant decrease compared to control	Day 3	Garcia JH 18-point neurological deficit score
Caltana et al. [16]	2015	Argentina	ACEA	CB1 agonist	C57Bl/6 mice	MCAO	Post-treatment	3, 24, and 48 h after injury	Significant decrease compared to control	Day 28	Neurological score
Sun et al. [17]	2013	China	WIN55,212-2	CB1 agonist	Sprague-Dawley rats	MCAO	Post-treatment	2 h after injury	Significant decrease compared to vehicle treated	Day 1	Neurological score
Yokubaitis et al. [18]	2021	USA	BCP	CB2 agonist	C57Bl/6 mice	Photo thrombosis	Pre-/post-treatment	1 h prior to injury and 24 h post-injury	Significant decrease compared to vehicle treated	Day 3	Grip strength test
Yang et al. [19]	2017	China	BCP	CB2 agonist	C57BL/6 mice	MCAO/r	Pre-/post-treatment	3 days before injury and 2 h post-injury	Significant decrease compared to control	Day 2	Longa’s neurological deficit score
Pottier et al. [20]	2017	France	JWH-133	CB2 agonist	Sprague-Dawley rats	MCAO/r	Post-treatment	Daily from 1 h after MCAO to day 7	Significant improvement compared to non-treated samples	Day 7	N/A
Ronca et al. [21]	2015	USA	O-1966	CB2 agonist	C57B/6 mice	Photo thrombosis	Pre-/post-treatment	1 h prior to injury and 2.5 days following or 1 h prior to injury and 2 days following	Significant decrease compared to control	Day 1, 3, and 7	Novel object recognition task, food-motivated operant task
Yu et al. [22]	2015	USA	AM1241	CB2 agonist	Sprague-Dawley rats	MCAO/r	Pre-/post-treatment	5 min before injury or from day 2–5 following	Significant decrease compared to vehicle-treated (pre-treatment), non-significant improvement (post-treatment)	Day 2 and 6	Bederson’s score
Choi et al. [23]	2013	South Korea	TC	CB2 agonist	Sprague-Dawley rats	MCAO/r	Post-treatment	3 h after injury	Significant decrease compared to control	Day 1	N/A
Zarruk et al. [24]	2012	Spain	JWH-133	CB2 agonist	Swiss mice	MCAO	Post-treatment	10 min or 3 h post-injury	Significant decrease compared to vehicle treated in both groups of treatment	Day 2	mNSS
Shirazi et al. [25]	2021	New Zealand	Novel N-acylethanolamine derivatives	CB1/CB2 agonist	C57BL/6J mice	Photo thrombosis	Post-treatment	1 h post-injury	Significant decrease compared to vehicle treated (NAE-18:4n-6, NAE-18:5n-3); non-significant improvement compared to vehicle treated (NAE-22:5n-6, NAE-22:6n-3, NAE-20:5n-3, NAE-20:4n-6)	Day 7	NAE-18:4n-6, NAE-18:5n-3
Murakami et al. [26]	2013	Japan	TAK-937	CB1/CB2 agonist	Sprague-Dawley rats	Photo thrombosis	Post-treatment	Continuous administration from 3 h, 5 h, or 8 h–24 h	Significant decrease compared to vehicle treated (3–24 h and 5–24 h); no improvement (8–24 h)	Day 2	N/A
Suzuki et al. [27]	2012	Japan	TAK-937	CB1/CB2 agonist	Sprague-Dawley rats or cynomolgus monkeys	MCAO/r or Thromboembolic MCAO	Post-treatment	Continuous infusion for 22 h following perfusion (rats), continuous infusion for 23.5 h following clot injection (monkeys)	Significant decrease compared to vehicle-treated (rats), non-significant improvement compared to vehicle treated (monkeys)	Day 1	Foot-fault test, Bederson’s score
Schmidt et al. [28]	2012	Germany	KN38-72717	CB1/CB2 agonists	Sprague-Dawley rats	mMCAO	Pre-/post-treatment	2 h before injury, during eMCAO, 30 min, 4 h, 3 days, 6 days after injury	Significant decrease compared to vehicle treated in multiple timepoints	Day 7 and 21	Ladder-rung walking test
Reichenbach et al. [29]	2016	USA	SR141716A	CB1 antagonist	C57Bl/6 mice	Photo thrombosis	Pre-/post-treatment	1 h before or after injury	Significant decrease compared to vehicle treated (pre-treatment); non-significant improvement (post-treatment)	Day 1	Hata et al. five point neurological score
Jalin et al. [30]	2015	Spain	Hinokiresinols	CB1/CB2 antagonist	Sprague-Dawley rats	MCAO/r	Post-treatment	2 h and 7 h after injury	Significant decrease compared to vehicle treated	Day 1	N/A
Bravo-Ferrer et al. [31]	2016	Spain	JWH-133 or SR144528	CB2 agonist or CB2R antagonist (respectively)	C57BL/6 mice	MCAO	Post-treatment	2 days following injury till day 6	No improvement compared to vehicle treated (both drugs)	Day 14 and 28	NR
Zhang et al. [32]	2012	USA	O-1966, SR141716A, both, or SR144528	CB2 agonist, CB1 antagonist, CB1 antagonist/CB2 agonist, or CB2 antagonist (respectively)	C57BL/6 mice	MCAO/r	Pre-treatment	1 h before injury	O-1966: significant decrease compared to control; SR141716A: significant decrease compared to control; O-1966 and SR141716A: significant decrease compared to control; SR144528: no improvement	Day 1	N/A
Zhang et al. [33]	2009	USA	O-1966	CB2 agonist	C57BL/6 mice	MCAO/r	Pre-/post-treatment	1 h before injury	Significant decrease with 5 mg dose only compared to vehicle treated	Day 1	Significant improvement compared with vehicle group
Zhang et al. [34]	2007	USA	O-3853, O-1966	CB2 agonist	C57BL/6 mice	MCAO/r	Pre-/post-treatment	1 h before and after the injury	Significant decrease in size with both pre- and post-treatment	Day 1	Significant improvement with both pre- and post-treatment group
Zhang et al. [33]	2009	USA	O-1966, SR141716A, SR144528	CB1 and CB2 antagonist	C57BL/6 mice	MCAO/r	Pre-treatment	1 h before injury	CB2 receptor activation and CB1 receptor inhibition were neuroprotective, with the greatest protection achieved by combining a CB2 agonist and CB1 antagonist	Day 1	CB2 receptor activation and CB1 receptor inhibition were neuroprotective, with the greatest protection achieved by combining a CB2 agonist and CB1 antagonist
Hu et al. [36]	2010	China	WIN 55,212-2, U0126	CB agonist, CB antagonist	Sprague-Dawley rats	MCAO/r	Pre-treatment	1 day before the injury	CB agonist has neuroprotective function while the antagonist proves the opposite effect	Day 1, 3, 6	Pretreatment with CB agonist has protective function
Berger et al. [37]	2004	Germany	SR141716A	CB1 antagonist	Wistar rats	MCAO/r	Post-treatment	30 min after injury	CB1 antagonist reduces infarct size significantly	Day 1	NR
Muthian et al. [38]	2004	USA	WIN 55212-2, SR141716, LY320135	CB1 antagonist	Wistar rats	MCAO/r	Pre-treatment	5 min pre-onset	CB1 antagonist reduces infarct size significantly, while no effect with CB1 agonist	Day 1	Significantly improved neurological function with CB1 receptor antagonist
Hayakawa et al. [39]	2007	Japan	Delta(9)-THC, SR141716	CB agonist, CB antagonist	ddY mice	MCAO/r	Pre-treatment	Pre-onset	Delta 9 THC significantly decreased the infarct volume	Day 1	NR
Leker et al. [40]	2003	Jerusalem	HU-210	CB antagonist	Sprague-Dawley rats	pMCAO	Post-treatment	1, 2, 4, or 6 h	Motor disability and infarct volumes were significantly reduced in animals treated with HU-210	Day 3	Motor disability and infarct volumes were significantly reduced in animals treated with HU-210

Pre-treatment, treatment administered before inducing stroke; post-treatment, treatment administered after inducing stroke; CB, cannabinoid; CB1, cannabinoid receptor type 1; CB2, cannabinoid receptor type 2; MCAO/r, middle cerebral artery occlusion with reperfusion; MCAO, middle cerebral artery occlusion; NR, not reported; N/A, not applicable; STAIR, Stroke Therapy Academic Industry Roundtable; h, hours; min, minutes.

Fig. 1.

PRISMA flowchart illustrating the systematic process of identification, screening, eligibility, and inclusion of studies in a systematic review and meta-analysis [58].

Fig. 5.

Forest plot based on specific CB antagonist drug used vs. control group, showing the standard mean difference in infarct volume.

Fig. 2.

Forest plot illustrating the standard mean difference in brain infarct volume between the experimental and control groups, exclusively utilizing CB agonist drugs.

Fig. 3.

Subgroup analysis based on specific CB agonist drug used vs. control group, showing the standard mean difference in infarct volume.

Fig. 4.

Subgroup analysis based on drugs targeting specific CB ligands, illustrating the standard mean difference in infarct volume.

Infarct Volume with CB Agonist

The analysis of infarct volume in response to CB agonist drugs showed a significant reduction in infarct volume within the experimental group compared to controls. The combined data from various agonist drugs revealed a substantial decrease, with a standardized mean difference (SMD) of −3.61 (95% CI: −4.41 to −2.81; p < 0.00001; I² 83%), as illustrated in online supplementary Figure 2. This suggests a significant effect of CB agonists in reducing the infarct volume. Due to heterogeneity, a leave-one-out analysis was conducted, and a forest plot is included in online supplementary Figure 2.

Further investigation through subgroup analysis categorized by specific drug types revealed varying levels of efficacy among different agonists. Notably, both ACEA and KN38-72717 showed a significant reduction in infarct volume, with a SMD of −5.40 (95% CI: −7.07, −3.74; p < 0.00001; I² 15%) for ACEA and −13.12 (95% CI: −17.86, −8.38; p < 0.00001; I² 83%) for KN38-72717, as depicted in online supplementary Figure 3. Additionally, TAK-937 (−8.50) and O-1966 (−5.35) also demonstrated significant reductions in infarct volume, highlighting their potential as therapeutic agents for ischemic events. Other compounds such as WIN 55212-2 (−2.22), NAE-18:4n-6 (−4.69), BCP (−3.46), NAE-18:5n-3 (−1.33), NAE-22:5n-6 (−0.24), JWH-133 (−1.86), and NAE-20:5n-3 (−1.35) have also exhibited a reduction in infarct volume but were not as significant as ACEA and KN38-72717.

Subgroup analysis based on ligand type also provided insight into the efficacy of CB agonists in reducing infarct volume. CB1 agonists emerged as particularly effective, producing a substantial reduction with an SMD of −4.55 (95% CI: −5.54, −3.57; p < 0.00001; I² 10%), as illustrated in online supplementary Figure 4. This highlights the specific targeting of CB1 receptors as a promising strategy for mitigating ischemic damage. Furthermore, CB2 agonists and non-selective agonists (CB1/CB2) also displayed significant effects, with SMDs of −3.69 (95% CI: −4.71, −2.67; p < 0.00001) and −3.60 (95% CI: −5.83, −1.36; p = 0.002), respectively.

Infarct Volume with CB Antagonist

Analysis of infarct volume in response to CB antagonists revealed a significant overall effect between the experimental and control groups, with a SMD of −2.12 (95% CI: −3.20, −1.04; p = 0.0004). A leave-one-out analysis was conducted, and a forest plot is included in the supplementary file. Subgroup analysis provided further insights into the efficacy of specific CB antagonists. SR141716, a CB1 antagonist, demonstrated a significant reduction in infarct volume with an SMD of −2.93 (95% CI: −4.37, −1.49; p < 0.0001), while LY320135, another CB1 antagonist, also showed a notable effect with an SMD of −2.55 (95% CI: −4.42, −0.67; p = 0.008). However, SR144528, a CB2 antagonist, did not exhibit a significant effect on infarct volume, with an SMD of 0.96 (95% CI: −0.32, 2.24; p = 0.14). In contrast, U0126, a CB antagonist, demonstrated a significant reduction in infarct volume, with an SMD of −1.23 (95% CI: −2.21, −0.26; p = 0.01). These findings highlight the varying efficacies of CB antagonists in reducing infarct volume, with CB1 antagonists showing more prominent effects than CB2 antagonists (online suppl. Fig. 5).

Neurological Scores

Neurological scores for both CB agonists and antagonists were evaluated across the experimental and control groups. The SMD for neurological scores with CB agonists was −0.63 (95% CI: −1.57, 0.31; p = 0.19), suggesting a potential improvement trend, although statistical significance was not reached. Similarly, CB antagonists showed an SMD of −0.84 (95% CI: −1.81, 0.13; p = 0.09), indicating a possible enhancement in neurological function, but not statistically significant (online suppl. Fig. 5). These findings suggest a promising yet inconclusive pattern towards enhanced neurological scores with both CB agonists and antagonists. Further investigations with larger cohorts may be necessary to unveil the precise impact of CB modulation on neurological outcomes. Such insights could pave the way for potential therapeutic interventions targeting conditions that affect the central nervous system.

Summary of Effects of CB Agonist and Antagonist Drugs on Infarct Volume

This study examined the impact of CB agonist and antagonist drugs on infarct volume using SMD analysis, along with 95% CI. Among CB agonists, drugs such as ACEA and WIN 55212-2 demonstrated negative SMD values, suggesting a potential decrease in infarct volume. Conversely, some agonists, such as KN38-72717 and O-1966, displayed substantial negative SMD values, indicating a more pronounced effect on reducing infarct volume (online suppl. Fig. 6a) For the antagonist drugs, SR141716 and LY320135 exhibited negative SMD values, indicating a potential reduction in infarct volume, whereas SR144528 showed a positive SMD value, suggesting a potential increase. U0126 also displayed a negative SMD value, implying a potential reduction in the infarct volume (online suppl. Fig. 6b) These findings underscore the complex and varied effects of CB agonists and antagonists on infarct volume, highlighting the importance of considering both the central tendency and uncertainty in interpreting the results.

Effect of CB Agonist and Antagonist Drugs on Experimental Infarct Volume over Time

The grouped bar plot illustrates the effect of the time of administration on the experimental infarct volume, categorized by CB agonist and antagonist drug classes. Overall, it can be observed that the SMD in infarct volume varies across different time points for both agonist and antagonist drug classes. For the CB agonist group (online suppl. Fig. 7a), the SMD tends to decrease over time, with the most substantial reduction observed between 2–3 h and 3–4 h post-administration. Conversely, in the CB antagonist group (online suppl. Fig. 7b), SMD fluctuated across time points, with no clear trend observed. Notably, the error bars representing the 95% CIs indicate the uncertainty associated with the SMD estimates, and values crossing zero are deemed statistically insignificant.

Publication Bias

Publication bias was assessed through the examination of funnel plots. Notably, the funnel plot representing the overall CB agonist data exhibited slight asymmetry. Subsequently, an Egger test was conducted, yielding an intercept value of −7.05, a standard error of 1.22, a t value of 5.79, and a p value of 0.000001. To further validate our findings, a Duval and Tweedie trim-and-fill test was performed, revealing no trimmed studies and minimal impact on the results, thus enhancing the accuracy of our findings. Similarly, a funnel plot was generated for the CB antagonist data, with an Egger intercept value of −5.04, a standard error of 1.51, a t value of 3.33, and a p value of 0.003. Once again, a Duval and Tweedie trim-and-fill analysis was undertaken, resulting in no trimmed studies and negligible alteration to the findings. All funnel plots have been included (online suppl. Fig. 2, 4).

Discussion

CBs have demonstrated therapeutic and neuroprotective effects in various models of stroke [59]. In particular, the impact of CBs on the outcomes of focal ischemic stroke has been a subject of debate for over 2 decades [60]. A systematic review and meta-analysis conducted by England et al. [9] not only revealed the significant influence of CBs on focal ischemic strokes but also the outcomes associated with drug dose and therapeutic time window [9]. It has been demonstrated that multiple CB drugs, including oral cannabidiol (CBD), oleoylethanolamide (OEA), and palmitoylethanolamide (PEA), exert their effects through receptors other than CB receptors [9, 54, 61–63]. Given the diverse actions of different CBs, further investigation of the role of CB receptors in focal ischemic strokes is essential to establish a comprehensive understanding of their involvement in the therapeutic cascade. This study aimed to investigate the function of CB receptor-related medications in the management of focal stroke.

This comprehensive meta-analysis has demonstrated that CB receptor agonists, such as CB1 and CB2 agonists, significantly decrease infarct volume in both transient and permanent ischemic models and enhance both early and late functional outcomes. Our meta-analysis demonstrated significant outcomes in terms of diminishing infarct volume with either CB1 or CB2 agonist, as illustrated in online supplementary Figure 2. Moreover, by conducting a subgroup analysis categorized by specific drug types, we discovered varying levels of efficacy among the different agonists. Importantly, ACEA, KN38-72717, and TAK-937 were found to be the most effective compounds, showing a considerable decrease in infarct volume with a SMD [14, 15, 64]. CB1 agonists alone reduce infarct volume more than CB2 agonists and non-selective CB receptor agonists [16, 65]. The role of the CB receptor is not well understood as significant results were also obtained for reducing infarct volume with a CB1 antagonist [29, 31]. The role of CB receptors is very well established, but it is not completely understood and is far from being clear. Therefore, future studies are of utmost importance to describe these discrepancies and their exact mechanisms and roles. Neurological scores were evaluated across experimental studies and for all the CB agonists and antagonists they showed a trend towards improvement but still not significant [5, 29, 57, 64, 66, 67].

The ECS has proved substantial immunomodulatory and protective effects on glial cells, and the pharmacological modulation of this signaling pathway has demonstrated promising neuroprotective effects in various neurological diseases [68–70]. Research suggests that the ECS plays a crucial role in regulating glial cell function, not only by influencing cell reactivity or phenotype polarization but also by promoting cell survival and preventing functional impairment [11]. Research on the ECS and CB2 receptors, particularly, has been extensive owing to their significant role in neuroinflammation and their presence in microglial cells [71]. This area of study has garnered much attention from researchers owing to its potential implications in treating various neurological disorders [71]. Multiple investigations have demonstrated encouraging results related to the activation of the CB2 receptor in reducing protein aggregation-related pathologies as well as in mitigating inflammation and several dementia-related symptoms [71, 72].

The neuroprotective effects of CBs may also be attributed to other receptor targets, such as the serotonin 1A receptor (5HT1A) [61]. For instance, CBD’s effects, which are not influenced by CB1, CB2, the transient receptor potential cation channel subfamily V member 1 (TRPV1) antagonism, are partially mediated through 5HT1A, leading to decreased infarct volume and improved cerebral blood flow [54, 62, 63]. Additionally, anti-inflammatory effects have yet to be linked to specific target sites, and further investigation is needed for known CB target sites such as TRPs and the peroxisome proliferator-activated receptors PPARs [73]. Endocannabinoids PEA and OEA have been associated with reduced cell death, edema, and inflammation [74]. OEA’s infarct-reducing effects are believed to be mediated through PPARα, as demonstrated by their absence in PPARα−/− mice and inhibition by a PPARα antagonist [67, 75, 76].

The neuroprotective effects of CBs have been also linked to their role in maintaining cerebral blood flow and preserving blood-brain barrier (BBB) integrity [51, 77]. For instance, a recent study revealed a notable elevation in IgG levels within the ipsilateral cortex of mice subjected to MCAO compared to sham-control mice [44]. Interestingly, the administration of VCE-004.8 effectively mitigated this increase, indicating a potential protective effect on the BBB. Specifically, the levels of IgG were significantly lower in the VCE-004.8 treated group compared to the vehicle group. This suggests that VCE-004.8 treatment may prevent the extravasation of immunoglobulins, thus safeguarding the integrity of the BBB. Furthermore, the study observed that VCE-004.8 treatment led to a reduction in active-matrix metalloproteinase-9 (MMP-9) levels in the brain [44]. MMP-9 is recognized for its involvement in ischemic stroke models, where it plays a crucial role in the degradation of tight junction proteins and extracellular matrix proteins, contributing to BBB breakdown and edema formation [78]. This finding underscores the potential of VCE-004.8 in modulating MMP-9 activity, thereby potentially mitigating BBB disruption associated with ischemic stroke.

While England et al. [9] identified neuroprotective effects of CBs like CBD and PEA (which act via non-CB receptors), our analysis isolated CB1/CB2 agonists (e.g., ACEA, KN38-72717) as uniquely effective in reducing infarct volume. Notably, CB1 antagonists (e.g., SR141716) also showed efficacy, a finding not emphasized in prior work, underscoring the complexity of CB receptor signaling [9].

Clinical research investigating CB1 and CB2 modulation for ischemic stroke remains limited and predominantly preclinical in scope. While preclinical studies highlight neuroprotective mechanisms such as reduced neuroinflammation and oxidative stress, few compounds have advanced to human trials. Dexanabinol, a synthetic intravenous CB, progressed to Phase III trials for traumatic brain injury but failed to demonstrate efficacy in improving neurological outcomes [79]. In contrast, EPIDIOLEX (CBD) received FDA approval in 2018 for two severe forms of seizers (Lennox-Gastaut and Dravet syndromes), validating CB modulation in neurological disorders [80]. Based on England et al. [9] and the current meta-analysis, taking a step in clinical ischemic stroke research is warranted. Trials must address critical barriers, including optimal therapeutic time windows, receptor subtype specificity and compatibility with reperfusion therapies like thrombectomy.

This systematic review and meta-analysis have significantly advanced our understanding of the role of CB in managing stroke patients. Our study’s specific inclusion criteria, rigorous analysis, and clear presentation of results represent a significant milestone in this area of research. Unlike previous systematic reviews, such as those by England et al. [9], our study focused exclusively on CB receptor-related drugs, enhancing the precision and relevance of our findings. Our extensive evaluation revealed several inadequacies in the existing literature, highlighting the need for further investigation, particularly in CB2 antagonists against mixed CB1/CB2 agonists, which could offer significant therapeutic benefits. Moreover, it is worth mentioning that the expression of CB2 receptors experiences a decline in the first 3 h post-MCAO, followed by a gradual rise within the 24-h timeframe [35].

Our findings face several limitations that may impact the way our findings are interpreted. Initially, there was substantial heterogeneity owing to the variety in the design of the individual studies. To address this, we used a random-effects model for analysis. Moreover, we conducted a subgroup analysis to further examine the studies and pinpoint those causing the heterogeneity. Additionally, we implemented a leave-one-out analysis to adjust for the heterogeneity, enhancing the robustness of our findings. Several studies exhibit relatively small sample sizes, emphasizing the significance of future research initiatives with larger participant cohorts. Undertaking studies with substantial sample sizes is essential to mitigate potential biases, increase statistical power, and ensure the overall quality of the research outcomes. Furthermore, the limited size of the study groups resulted in imprecise estimates of variance and, subsequently, the SMD. It is worth mentioning that we opted for the use of SMD rather than weighted mean differences to combine diverse scales to assess the same parameter. Although interpreting SMD may be less intuitive, this choice has enabled the inclusion of a considerably greater number of studies in this analysis. A limitation of our study is the absence of human-level investigations. We solely relied on animal-level studies due to the extensive body of research conducted at this level, whereas human-level studies were not included in our analysis.

Looking ahead, the implications of our findings hold promise for the development of novel therapeutic interventions targeting CB receptors in the management of ischemic stroke. The observed reduction in infarct volume with CB agonists, particularly CB agonists such as ACEA, KN38-72717, and TAK-937, suggests a potential avenue for the development of neuroprotective agents. Further investigation of the specific mechanisms underlying the neuroprotective effects of CBs could lead to the development of targeted therapies aimed at preserving neural tissue integrity and improving functional outcomes following stroke. Additionally, the varying efficacy of CB antagonists in reducing infarct volume highlights the complexity of CB receptor signaling and warrants further exploration of their therapeutic potential. Future research should focus on elucidating the precise mechanisms of CB receptor modulation and optimizing therapeutic strategies to enhance their clinical involvement in practice. Further studies assessing behavioral outcomes are warranted to have a broader understanding to the effects of CB receptors modulation in ischemic brain injury. Ultimately, the translation of these preclinical findings into clinical practice has the potential to revolutionize stroke management and improve patient outcomes.

Conclusion

This systematic review and meta-analysis underscore the significant potential of CB receptor modulation in mitigating infarct volume in animal models of focal ischemic stroke. Our findings revealed a consistent reduction in infarct volume with CB agonists, particularly CB agonists such as ACEA, KN38-72717, and TAK-937, suggesting their promising therapeutic efficacy. The subgroup analyses provided further insight into the varying levels of efficacy among different agonists and emphasized the need for continued exploration of their specific mechanisms of action. Moreover, CB antagonists, particularly CB1 antagonists such as SR141716, showed significant reductions in infarct volume but with varying degrees of effectiveness. Although promising trends were observed, the impact on neurological scores remained inconclusive, indicating the need for more extensive investigation with larger cohorts. Our study not only contributes to advancing the understanding of the role of CB receptors in stroke management but also underscores the need for future research to elucidate the underlying mechanisms and optimize therapeutic strategies.

Statement of Ethics

No interventions involving animals or humans were conducted, and no confidential data were utilized.

Conflict of Interest Statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.

Funding Sources

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author Contributions

Hussain Al Dera contributed to conception and design; supervised the study; approved the final version of the manuscript on behalf of all authors; reviewed submitted version of manuscript; critically revised the article; drafted the article, and performed acquisition, analysis, and interpretation of data. Abdulrahman M. Khojah contributed to conception and design; supervised the study; reviewed submitted version of manuscript; critically revised the article; drafted the article; and performed acquisition, analysis, and interpretation of data. Abdulrazak Sakhakhni reviewed submitted version of manuscript, drafted the article, and contributed to acquisition of data. Syed Aga and Faisal Alamri reviewed submitted version of manuscript and critically revised the article. Abdulaziz Alqahtani drafted the article and contributed to acquisition of data. Fahad Alsharef and Shadell Alghamdi contributed to administrative/technical/material support and acquisition of data.

Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data Availability Statement

The data used during the current study are available from the corresponding author upon reasonable request.