Sepsis is a life-threatening clinical syndrome characterised by a dysregulated host response to infection, resulting in extensive tissue damage, multi-organ dysfunction and high global mortality (Suzuki et al., 2025; Jinzhong Wang and Jian Fu, 2025). Among the various complications caused by sepsis, septic cardiomyopathy (SCM) is a particularly severe manifestation and a significant factor in morbidity and mortality in critically ill patients (Mohammad et al., 2022; Wang et al., 2024). The condition is defined as acute myocardial dysfunction associated with sepsis, typically manifested by reduced left ventricular ejection fraction (LVEF), impaired myocardial contractility and elevated cardiac biomarkers in the absence of coronary artery obstruction or direct myocardial injury (Yuan et al., 2019).

Abdominal trauma and subsequent abdominal sepsis are common clinical conditions leading to SCM, with unique pathophysiological mechanisms that exacerbate myocardial injury and dysfunction (Pan et al., 2024; Shen et al., 2024; Su et al., 2024). Mechanistically, abdominal sepsis induces systemic inflammation, characterised by significant elevations in proinflammatory cytokines (e.g., tumour necrosis factor-α [TNF-α], interleukin-6 [IL-6]), excessive oxidative stress, mitochondrial dysfunction and subsequent cardiac cell apoptosis and dysfunction (Latifi and Smiley, 2023; Adenuga and Adeyeye, 2023). Despite advances in sepsis management, including early antibiotic treatment, fluid resuscitation and vasoactive drug support, targeted therapies specifically targeting SCM remain limited, highlighting the urgent need for new effective preventive or therapeutic strategies.

In recent years, increasing preclinical evidence has shown that active small molecules, including polyphenols, flavonoids, saponins, antioxidants, glycosides and other bioactive compounds, may have significant cardioprotective potential against septic myocardial injury through multiple mechanisms, including anti-inflammatory effects, anti-oxidative stress response, mitochondrial protection and reduced cardiomyocyte apoptosis (Wang et al., 2021; Bayly-Jones et al., 2022; Lee et al., 2022). For example, resveratrol, a phenotypic polysaccharide, has been shown to have mitochondrial protective effects and the ability to significantly reduce cardiac injury biomarkers in rodent models of endotoxemia. Flavonoids such as quercetin and luteolin have also been reported to reduce cardiac dysfunction and inflammation in animal sepsis models, although their efficacy varies, depending on the experimental setting (Zhao et al., 2023). Despite preliminary evidence of the promising efficacy of these small molecules, a comprehensive quantitative review of their efficacy and mechanistic characteristics in animal models of SCM remains limited.

Abdominal trauma and subsequent intra-abdominal sepsis are major causes of critical illness in surgical and trauma populations and are associated with a high incidence of multiple organ dysfunction, including SCM (Sartelli et al., 2024; Mureşan et al., 2018). Abdominal-origin sepsis has pathophysiological characteristics such as ongoing peritoneal contamination, bacterial translocation and the potential for intra-abdominal hypertension/abdominal compartment physiology, which may produce distinct inflammatory and haemodynamic profiles compared with other sepsis sources (Mureşan et al., 2018; MacFie, 2004).

Accordingly, this systematic review and meta-analysis aimed to comprehensively evaluate the preventive effects of various active small molecules on SCM induced by abdominal trauma or abdominal sepsis in animal models. Specifically, we aimed to systematically evaluate the overall efficacy of these small molecules, identify potential sources of heterogeneity through subgroup analyses (e.g., molecular class, sepsis model, outcome measures), explore potential protective mechanisms and rigorously evaluate the methodological quality of different studies. We hypothesised that active small molecules could significantly prevent SCM through anti-inflammatory, antioxidant and mitochondrial protective mechanisms, although the extent of their effects varied, based on the specific molecular class and experimental model. Our findings aim to provide strong preclinical evidence to guide future translational research and clinical trials targeting SCM.

In this systematic review and meta-analysis, we comprehensively evaluated the protective effects of various active small molecules on SCM induced by abdominal trauma or sepsis in animal models. We included 17 studies covering multiple molecular classes and animal models, and performed quantitative analysis, which showed that SCM had an overall positive effect. Specifically, the pooled SMD for cardiac function and injury outcomes was 1.47 (95% CI: −0.52–3.47). Although the pooled effect suggested a potential cardioprotective effect, the wide confidence interval and high heterogeneity (I² = 92.9%) highlighted significant differences in the strength and consistency of these effects among the included studies.

The potential protective mechanisms of active small molecules against SCM involve multiple biological pathways, particularly reducing inflammation, alleviating oxidative stress, protecting mitochondrial function and, subsequently, improving myocardial injury markers and cardiac function parameters. In subgroup analyses, polyphenols showed a significant but imprecise protective effect (SMD = 4.81), while the results for flavonoids were more ambiguous (SMD = −0.45). These differential findings suggest the existence of different molecular mechanisms, bioavailability, pharmacokinetic or dose-related factors that may explain the differences between the categories.

Previous preclinical and clinical studies have highlighted the pleiotropic effects of polyphenolic compounds such as resveratrol and rosmarinic acid, especially their potent antioxidant and anti-inflammatory activities. Zhao et al. reported that resveratrol significantly increased myocardial ATP and mitochondrial CCO activity in a rat LPS-induced sepsis model, highlighting mitochondrial protection as a key mechanism of cardioprotection (Zhao et al., 2015). These results are consistent with existing evidence from other cardiovascular studies that attribute the efficacy of polyphenols to antioxidant pathways targeting mitochondria, and the subsequent attenuation of oxidative stress-induced myocardial dysfunction (Su et al., 2024; Xu et al., 2020; Yang et al., 2024).

In contrast, the results for flavonoids in this meta-analysis were inconsistent. Although some individual flavonoids, such as luteolin and apigenin, showed moderate efficacy in specific cardiac outcomes, the overall results for flavonoids were relatively neutral. This finding is in stark contrast to previous studies that have highlighted the potent anti-inflammatory properties of flavonoids in various inflammatory disease models (Hu et al., 2015; Wu et al., 2023a; Wang et al., 2023). These variances may reflect differences in the severity of SCM animal models, differences in dosing regimens or inherent bioavailability limitations of flavonoids.

Subgroup analyses based on the method of sepsis induction further demonstrated that the efficacy of active small molecules varied across models. The LPS-induced endotoxemia model showed a greater protective effect (SMD = 2.86) than the CLP model (SMD = 0.16). This difference may stem from the pathophysiological differences between endotoxemia (acute inflammatory stimulation, systemic cytokine surge) and CLP-induced polymicrobial sepsis, which is closer to the clinical sepsis state (Wang et al., 2021; Rosado-Franco et al., 2021; Yuan et al., 2022). Therefore, molecules with acute anti-inflammatory effects, such as resveratrol and rosmarinic acid, may show higher efficacy in the LPS model, reflecting their rapid modulation of cytokine profiles and acute inflammation, but their sustained effects may be weaker in the more complex and prolonged CLP-induced sepsis environment.

Further analysis based on outcome measures found significant differences between biochemical markers (CK, CK-MB) and functional cardiac parameters (LVEF, MAP). Biochemical indicators such as CK-MB showed sustained protective effects, while functional indicators such as LVEF showed mixed and unclear efficacy. This observation is consistent with existing literature suggesting that biochemical markers are sensitive indicators of early myocardial injury and recovery, whereas echocardiographic parameters such as LVEF may reflect more complex, multifactorial changes in myocardial function and haemodynamics, which evolve over time (Wang et al., 2022; Wu et al., 2023b; Sun et al., 2024).

Comparison with previous literature: Previous systematic reviews have highlighted the potent antioxidant and mitochondrial protective properties of polyphenols, especially resveratrol, which are well aligned with our current findings (Zeng et al., 2023; Taha et al., 2023). Similarly, flavonoids, despite their variable overall efficacy in our analysis, have been widely reported to exhibit cardioprotective properties through anti-inflammatory and antioxidant mechanisms, although results have been inconsistent across experimental contexts (Huajie, 2015).

Septic cardiomyopathy remains an important unmet clinical need. Current clinical management is primarily supportive (including careful fluid resuscitation, vasopressors and inotropes when indicated), and there are no widely accepted, evidence-based pharmacologic agents that specifically prevent or reverse sepsis-induced myocardial dysfunction (L'Heureux et al., 2020). Recent preclinical data indicate that certain small molecules (notably, polyphenolic compounds such as resveratrol and related agents) can modulate key pathophysiologic pathways implicated in SCM, including pro-inflammatory signalling, oxidative stress and mitochondrial dysfunction, and thereby protect cardiac tissue in animal models. However, translating these findings to human therapy faces several practical barriers: many polyphenols and small molecules exhibit poor oral bioavailability and limited tissue penetration; robust safety/toxicity and PharmacoKinetics and PharmacoDynamics (PK/PD) characterisation in relevant models is often lacking (Di Lorenzo et al., 2021); dosing regimens and timing relative to sepsis induction require careful definition; and functional cardiac endpoints (rather than biochemical markers only) must be demonstrated in reproducible, cross-species studies. To facilitate translation, we propose a staged roadmap: (1) standardised, high-quality preclinical studies that adhere to ARRIVE recommendations and include PK/PD and formal toxicity arms; (2) the demonstration of reproducibility across species and in more clinically relevant models (e.g., CLP and larger animal models, where feasible); and (3) early-phase clinical studies that prioritise safety and objective cardiac functional endpoints (e.g., echocardiographic strain measures, haemodynamic indices and clinically relevant outcomes) (Percie du Sert et al., 2020; Nicol et al., 2021). These steps aim to bridge promising preclinical signals and rigorous human evaluation.

Lipopolysaccharide-induced endotoxemia and CLP-induced polymicrobial sepsis differ markedly in terms of their kinetics of cytokine responses, bacterial dissemination and clinical course; consequently, therapeutic efficacy observed in LPS models may not translate to CLP models or to human sepsis cases. Previous comparative studies have documented these kinetic and immunologic differences (Remick et al., 2020; Seemann et al., 2017) accordingly, we suggest prioritising CLP models for translational studies, while recognising the value of LPS models for testing acute anti-inflammatory effects. Importantly, our review focused on abdominal-origin models (including abdominal trauma and abdominal-source sepsis); therefore, the findings may not be generalisable to sepsis arising from other sources (e.g., pulmonary or urinary tract infections), and it is possible that the inclusion of studies focused on other sepsis origins could yield different pooled estimates due to source-specific pathophysiology.

Notably, although previous studies have frequently documented significant protective effects of antioxidants such as vitamin C, our meta-analysis included only limited data (Cui et al., 2024), highlighting the need for further exploration and the validation of such interventions in animal SCM models. Similarly, the favourable effects observed for saponins (astragaloside IV, ginsenoside Rc) are consistent with previous findings documenting anti-inflammatory, anti-apoptotic and mitochondrial protective properties in other experimental models of myocardial injury.

Our meta-analysis has several strengths. First, to our knowledge, this is the first systematic, comprehensive study of active small molecules in animal models of abdominal trauma/sepsis-induced cardiomyopathy. In addition, we used a rigorous and comprehensive approach, including extensive subgroup analyses, sensitivity testing and systematic risk of bias assessment, using the validated SYRCLE tool.

Several important limitations should, however, temper the interpretation of our findings. First, substantial heterogeneity persisted across studies (I² = 92.9%), which likely reflects multiple sources of clinical and methodological variability, that is, inconsistent reporting of dosing regimens and timing relative to sepsis induction, heterogeneity in sepsis models and model severity, differences in animal species/strains, as well as variable definitions and timing of the outcome measurements. Second, many of the included studies incompletely reported key experimental design features, increasing the risk of selection and detection bias. Moreover, these methodological shortcomings not only threaten internal validity by inflating effect sizes but also reduce the reproducibility and external generalisability of the findings. Third, the modest number of studies overall, combined with sparse and inconsistent reporting of quantitative dosing, route and timing, limits the feasibility and reliability of meta-regression or stratified analyses. Fourth, the available literature placed greater emphasis on biochemical markers than on standardised functional cardiac outcomes; biomarker improvements may not equate to meaningful or durable functional recovery. Fifth, adverse effects and formal toxicity assessments were generally not reported, restricting any assessment of benefit/risk for the compounds studied. Sixth, small-study effects and possible publication bias cannot be excluded, and this may have led to the overestimation of efficacy. Seventh, many studies lacked pre-registration or publicly available raw data, limiting transparency and reproducibility. Finally, several included studies originated from a limited number of research groups or geographic regions, which may have reduced methodological diversity and increased the risk of confirmation bias. Future work should follow ARRIVE recommendations, include PK/PD and toxicity arms, pre-register experimental protocols, use harmonised endpoints, and make raw datasets available to enable reliable synthesis.

Our findings highlight the potential of active small molecules, especially polyphenols, as preventive treatments for SCM and highlight their mechanisms of inflammation reduction, oxidative stress reduction and mitochondrial protection. However, the large differences observed suggest that further standardised, rigorously designed preclinical trials are urgently needed. Future studies should prioritise methodological rigour, including detailed reporting of randomisation, allocation concealment and blinding, as well as systematic investigation of dose-response relationships, timing and long-term cardiac outcomes.

In summary, this meta-analysis demonstrates that active small molecules have generally positive but heterogeneous protective effects against abdominal trauma or sepsis-induced cardiomyopathy in animal models. Polyphenols showed particularly promising cardioprotective effects through antioxidant and mitochondrial protective mechanisms, although large variability remained. Future studies should prioritise methodological rigour, improved standardisation and translation-oriented studies to validate these promising preclinical findings and guide clinical application.