Prognostic value of monocyte chemoattractant protein-1 in sepsis: a systematic review and meta-analysis Authors: Zhonghui Huang, Nannan Li, Wei Wang, Qing Wang, Wen Ye Published: 2025 ABSTRACT AbstractBackgroundMonocyte chemoattractant protein-1 (MCP-1) has been utilized as a prognostic indicator for sepsis patients; however, the findings have been inconsistent. This meta-analysis seeks to elucidate the prognostic significance of MCP-1 in individuals diagnosed with sepsis.MethodsWe systematically queried four databases (Web of Science, PubMed, EMBASE, Cochrane Library) to identify studies published from the establishment of the databases until January 2025. Hazard ratios (HRs) with 95% CIs were employed to evaluate the prognostic significance of MCP-1 in this patient population. In our study, the Newcastle-Ottawa Quality Assessment Scale was evaluated for quality assessment, and Funnel plot was used for publication offset.ResultsA total of 592 patients from eight studies were included in the evaluation. The findings revealed that MCP-1 levels were significantly elevated in non-survivors compared to those who survived [SMD = 0.54, 95% CI: 0.25–0.82, P = 0.0002]. Our study demonstrated that elevated levels of MCP-1 were significantly associated with an unfavorable prognosis in patients with sepsis, with a hazard ratio (HR) of 1.36 (95% confidence interval: 1.25–1.48; P < 0.00001) using a random-effects model. The funnel plot indicated no publication bias.ConclusionThis meta-analysis suggests that elevated levels of MCP-1 could serve as a valuable prognostic indicator in sepsis patients, however, this conclusion still requires validation through higher-quality studies with longer-term follow-up.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12865-025-00783-8.Keywords: Sepsis, Monocyte chemoattractant protein-1, MCP-1, Prognosis, Meta-Analysis Introduction The term “sepsis” originates from Greek word for putrefaction. Its initial medical documentation dates back over 250 years to Homer’s writings where it was characterized as a systemic putrefaction resulting from infection. Over time, sepsis has evolved through three significant stages from sepsis 1.0 to sepsis 3.0. Currently, sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection [1]. Sepsis and septic shock affect millions of people worldwide each year [2–5], previous study shows that the global burden of sepsis is even greater than previously estimated, with a larger proportion of cases and deaths occurring in low- and middle-income countries, where the disease burden is particularly high [5]. The presentation and outcomes of sepsis are influenced by various factors, including patient characteristics, the causative microorganism, and the site of infection [6]. The improper regulation of the host’s inflammatory response to pathogens is a critical determinant of mortality in sepsis patients. Prognostic information plays a crucial role in reducing mortality rates. Despite extensive research into various inflammatory cytokines, their application in clinical practice remains limited due to the complex and heterogeneous nature of sepsis. Thus, identifying reliable biomarkers is crucial for facilitating timely and effective interventions in the management of sepsis. Monocyte chemoattractant protein-1 (MCP-1), a member of the CC chemokine family, is ubiquitously present in various human organs and tissues. Its production is attributed to epithelia cells, monocytes/macrophages, endothelial cells, and other cell types [7]. MCP-1 plays a critical role in numerous pathological conditions, including cardiovascular disease, encephalopathy, bone and joint disease, respiratory tract infections, and cancer [8]. Its involvement in sepsis has garnered significant attention in research. Pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs) induce overproduction of MCP-1 in sepsis, resulting in a dysregulated systemic inflammatory response and immunosuppression, ultimately worsening tissue injury and multi-organ failure [9, 10]. MCP-1 has important immunomodulatory functions in sepsis to maintain a balance between proinflammatory and anti-inflammatory cytokines. In experimental models of acute septic shock, endogenous MCP-1 expression is markedly increased following lipopolysaccharide (LPS) administration, a response that can be suppressed by the anti-inflammatory drug dexamethasone [11]. Zhai et al. demonstrated a significant increase in MCP-1 levels in myocardial tissue of aged wild-type mice with sepsis, correlating with heightened cardiac dysfunction and mortality [12]. In this context, MCP-1, as an inflammatory mediator, has emerged as a potential target for sepsis prognosis research. The clinical utility of MCP-1 as a prognostic biomarker for sepsis is a topic of debate. Zhu et al.’s cohort studies have indicated a potential link between elevated MCP-1 levels and increased 28-day mortality in sepsis patients [13], while other investigations have failed to establish a significant relationship. Discrepancies in study outcomes may stem from variations in sample sizes, population heterogeneity, or inconsistencies in assay standardization. A comprehensive synthesis of available data is imperative to elucidate the prognostic significance of MCP-1. Meta-analysis offers a robust approach to enhance statistical power and mitigate biases by amalgamating findings from diverse studies, thereby furnishing superior-quality evidence for clinical decision-making. This investigation aimed to assess the association between MCP-1 concentrations and sepsis patient outcomes via a systematic review and meta-analysis, to underpin risk stratification, personalized therapeutic interventions, and the identification of novel biomarkers. Materials and methods This study was conducted according to the Cochrane handbook [14] and PRISMA checklist [15] and was registered on PROSPERO with the identifier CRD42024613233. This study was not a registered clinical trial. Therefore, clinical trial number: not applicable. Search strategy The investigators conducted a comprehensive search across four major electronic databases—PubMed, Embase, Cochrane Central Register of Controlled Trials, and Web of Science—from their inception dates up to January 2025. The specific search strategy employed, illustrated in Table 1 using PubMed as an example, was meticulously designed to capture relevant studies. Given that this meta-analysis involves synthesizing data from already published studies, ethical approval was deemed unnecessary. Table 1Search strategy on pubmed#1Sepsis [MeSH Terms]#2((((((((((((((((sepsis) OR (Bloodstream Infection)) OR (Bloodstream Infections)) OR (Infection, Bloodstream)) OR (Pyemia)) OR (Pyemias)) OR (Pyohemia)) OR (Pyohemias)) OR (Pyaemia)) OR (Septicemia)) OR (Septicemias)) OR (Poisoning, Blood)) OR (Blood Poisoning)) OR (Blood Poisonings)) OR (Poisonings, Blood)) OR (Severe Sepsis)) OR (Sepsis, Severe)#3# 1OR #2#4Chemokine CCL2[MeSH Terms]#5(((((((((((((Chemokine CCL2[Title/Abstract]) OR (CCL2, Chemokine[Title/Abstract])) OR (“Monocyte Chemotactic and Activating Factor“[Title/Abstract])) OR (Monocyte Chemoattractant Protein-1[Title/Abstract])) OR (Chemoattractant Protein-1, Monocyte[Title/Abstract])) OR (Monocyte Chemoattractant Protein 1[Title/Abstract])) OR (Chemokines CCL2[Title/Abstract])) OR (CCL2, Chemokines[Title/Abstract])) OR (CCL2 Chemokine[Title/Abstract])) OR (Chemokine, CCL2[Title/Abstract])) OR (Chemokine (C-C Motif) Ligand 2[Title/Abstract])) OR (Monocyte Chemotactic Protein-1[Title/Abstract])) OR (Chemotactic Protein-1, Monocyte[Title/Abstract])) OR (Monocyte Chemotactic Protein 1[Title/Abstract])#6#4 OR #5#7#3 AND #6 Search strategy on pubmed Inclusion criteria (1) Observational studies (cohort or case-control) involving sepsis patients diagnosed by Sepsis criteria; (2) Studies reporting serum/plasma MCP-1 levels and their association with at least one prognostic outcome (e.g., 28-day mortality, ICU length of stay); (3) Availability of extractable effect sizes (HR, OR) or raw data. We specifically included observational studies as they are the optimal design for assessing the prognostic value of a biomarker, allowing for the evaluation of the association between baseline MCP-1 levels and subsequent clinical outcomes without the confounding influence of an experimental intervention. Exclusion criteria (1) Studies lacking complete or reported data; (2) Literature reviews, meta-analyses, case reports, case series, conference abstracts, and letters; (3) Animal studies, cell-based experiments, and dissertations. For articles published multiple times, the most recent versions or those with the largest sample sizes were selected. Study selection The initial screening of the literature was conducted using EndNote, a reference management software. Two researchers independently reviewed the titles to exclude duplicates, conference papers, review articles, protocols, and correspondence. Then, the abstracts of the literature were reviewed to identify articles for inclusion and exclude irrelevant ones. Finally, both researchers reviewed the full texts of the remaining articles to determine their inclusion. During this process, both researchers independently screened the literature, compared their results, and included articles if there was agreement. Discrepancies were resolved through discussion with the involvement of a third researcher. Data extraction The information obtained from the study encompasses: (1) general information: country, author, year of publication, mean age, sex ratio, SOFA score, and APACHE II score; (2) primary outcomes: number of survivors and non-survivors, MCP-1 levels, HR values, and 95% CIs. Two reviewers independently extracted the data. Ayn discrepancies were resolved by discussion, and if necessary, by consultation with a third reviewer. In cases where MCP-1 values were not directly provided, we used images to indirectly extract the MCP-1 values using Engauge Digitizer software. Assessment of risk of bias The quality of the included studies was evaluated using the Newcastle-Ottawa Quality Assessment Scale (NOS). The NOS is composed of three main components: the selection of the study population, the comparability between groups, and the assessment of outcomes. The specific criteria and their corresponding scores are as follows: representativeness of the exposed cohort (1 point); selection of the non-exposed cohort (1 point); ascertainment of exposure (1 point); demonstration that the outcome of interest was not present at the start of the study (1 point); comparability of cohorts based on the study design or analysis (up to 2 points); assessment of the outcome (1 point); sufficient follow-up duration to allow outcomes to occur (1 point); and adequacy of follow-up of cohorts (1 point). Each criterion is scored based on a “yes” response. Studies are categorized into three quality levels based on their total scores: high quality (7–9 points), moderate quality (4–6 points), and low quality (0–3 points). Statistical analysis The meta-analysis was performed using Review Manager 5.3 software. Standardized mean differences (SMDs) and their corresponding 95% confidence intervals (CIs) were utilized to evaluate the differences in MCP-1 levels among sepsis patients. The SMD was used to combine the results of MCP-1 levels from different studies, as the included studies reported MCP-1 in various units or scales. SMD allows for the comparison of effect sizes across studies by standardizing the differences to a common scale. Hazard ratios (HRs) with 95% CIs were employed to quantify the prognostic value of MCP-1 in sepsis. Heterogeneity across studies was evaluated using the I² statistic. We considered an I² value greater than 50% to represent substantial heterogeneity [16]. When substantial heterogeneity was present (I² >50%), a random-effects model was applied to calculate the SMDs and HRs. In the absence of substantial heterogeneity (I² <50%), a fixed-effects model was employed. To assess whether publication bias influences the validity of the estimates, funnel plots were constructed and evaluated both visually and statistically using Egger’s linear regression test [16]. Subgroup analyses were conducted based on sample size, sepsis diagnostic criteria, study design, MCP-1 assay method, incorporation of other parameters, and geographic location. For the purpose of this investigation, studies with sample sizes exceeding 100 were categorized as large sample cohorts, while those with sample sizes below 100 were designated as small sample cohorts. Statistical significance was determined using a two-sided P-value, with a threshold of less than 0.05. Results Study identification and selection A total of 2,900 articles were retrieved by searching four databases based on predefined selection criteria. Of these, 816 duplicate articles were excluded. Next, 2,032 articles were excluded based on the screening of titles and abstracts. Subsequently, after a thorough review of the full texts, an additional 44 studies were excluded. Among these, 11 studies were excluded due to incomplete data, 13 were excluded because they were conference proceedings, and 20 were excluded as review articles. Finally, eight studies were included for meta-analysis. Figure 1 presents a flowchart summarizing the selection process. PRISMA flowchart showing the literature search and studies selection Study characteristics and quality assessment Among the eight included studies, a total of 592 patients were enrolled, all of whom met the international diagnostic criteria for sepsis. Two studies used sepsis 1.0, three studies used sepsis 2.0, and three studies used sepsis 3.0. The fatality rate of sepsis patients ranged from 18% to 74.3%, with 282 deaths and 310 survivors. The severity of patients’ conditions was evaluated using the Acute Physiology and Chronic Health Evaluation II (APACHE II) or Sequential Organ Failure Assessment (SOFA) scales. The basic characteristics of the included studies are listed in Table 2. The Newcastle-Ottawa Scale was used to assess the quality of the included studies in our study. Specific details will be presented in Table 3. Table 2Characteristic of the studies included in the meta-analysisAuthorCountryYearAge)Total/male/femaleSOFAAPACHE IIMCP-1 level and unitStephen P. J. MacdonaldAustralia201467 (59, 80)43 (25, 75)NA7(5,9)0(0,1)NR517(268,1646)pg/mL426(90.0,1239)pg/mLMax BarreGermany201867 (26–88)64 (26–88)93/68/2543/27/1710.9 ± 0.57.0 ± 0.625.3 ± 0.915.3 ± 1.62765 ± 496pg/mL1313 ± 435pg/mLTingting ZhuChina201760.96 ± 15.0856.63 ± 11.5756/38/1887/50/379.1 ± 4.011.1 ± 4.519.9 ± 7.424.4 ± 7.6187.05[100.71–487.96.71.96]pg/mL67.86[40.20–119.94.20.94]pg/mLFernando A BozzaBrazil200764.0(58.5–3.5)55(48.5–1.50)29/17/1231/19/1211(8.5–12)7 (5–9)22.0(20.0–28.0)17.0(14.5–20.0)757.78(324.61–1966.88.61.88)pg/mL268.36(0.00–759.30.00.30)pg/mLDavid Andaluz-OjedaSpain201270.162.212/7/517/10/78.77.72619.5278.3[677.8]pg/mL80.6[166.7]pg/mLTsai-Hsia HongChina201472.0 ± 13.6NA5/4/119/9/10NRNR3590.87[6192.76]pg/mL251.54[65.86]pg/mLYunjiao DuanChina202379 (70, 84)73 (69, 83)101/51/5035/21/149 (6, 11)5 (3, 5)22.0 (17.0, 27.0)16.0 (13.0, 19.5)205.19(144.42,374.92)pg/mL140.49(116.72,182.64)pg/mLTakeshi EbiharaJapan2021NRNRNRNR447.92[1557.94]pg/mL206.06[848.32]pg/mLSOFA: sequential organ failure assessment scale, APACHE II: Acute Physiology, Age, Chronic Health Evaluation II, NA not assessed, NR not recorded. For each study, data are presented separately for the non-survivor group (upper row) and the survivor group (lower row) Characteristic of the studies included in the meta-analysis 67 (59, 80) 43 (25, 75) 7(5,9) 0(0,1) 517(268,1646)pg/mL 426(90.0,1239)pg/mL 67 (26–88) 64 (26–88) 93/68/25 43/27/17 10.9 ± 0.5 7.0 ± 0.6 25.3 ± 0.9 15.3 ± 1.6 2765 ± 496pg/mL 1313 ± 435pg/mL 60.96 ± 15.08 56.63 ± 11.57 56/38/18 87/50/37 9.1 ± 4.0 11.1 ± 4.5 19.9 ± 7.4 24.4 ± 7.6 187.05[100.71–487.96.71.96]pg/mL 67.86[40.20–119.94.20.94]pg/mL 64.0(58.5–3.5) 55(48.5–1.50) 29/17/12 31/19/12 11(8.5–12) 7 (5–9) 22.0(20.0–28.0) 17.0(14.5–20.0) 757.78(324.61–1966.88.61.88)pg/mL 268.36(0.00–759.30.00.30)pg/mL 70.1 62.2 12/7/5 17/10/7 8.7 7.7 26 19.5 278.3[677.8]pg/mL 80.6[166.7]pg/mL 72.0 ± 13.6 NA 5/4/1 19/9/10 3590.87[6192.76]pg/mL 251.54[65.86]pg/mL 79 (70, 84) 73 (69, 83) 101/51/50 35/21/14 9 (6, 11) 5 (3, 5) 22.0 (17.0, 27.0) 16.0 (13.0, 19.5) 205.19(144.42,374.92)pg/mL 140.49(116.72,182.64)pg/mL 447.92[1557.94]pg/mL 206.06[848.32]pg/mL SOFA: sequential organ failure assessment scale, APACHE II: Acute Physiology, Age, Chronic Health Evaluation II, NA not assessed, NR not recorded. For each study, data are presented separately for the non-survivor group (upper row) and the survivor group (lower row) Table 3Quality assessment of included studies, according to the eight domains of the Newcastle-Ottawa-ScaleStudy, first author, yearIs the case definition adequate?Representativeness of the casesSelection of controlsDefinition of controlsComparabilityControl forimportant factorAscertainment of exposureSame method of ascertainment for cases and controlsNon-response rateScoreStephen P. J. Macdonald 2014★★★★★★★7Max Barre 2018★★★★★★★★8Tingting Zhu 2017★★★★★★★7Fernando A Bozza 2007★★★★★7David Andaluz-Ojeda 2012★★★★★★6Tsai-Hsia Hong 2014★★★★★★★7Yunjiao Duan 2023★★★★★★★7Takeshi Ebihara 2021★★★★★★6 Quality assessment of included studies, according to the eight domains of the Newcastle-Ottawa-Scale Comparison of MCP-1 levels between non-survivors and survivors Eight studies compared the MCP-1 levels between the non-survivor and survivor groups, involving a total of 592 patients. The heterogeneity test indicated no significant heterogeneity among the results (P = 0.03, I² = 54%), and the random-effects model was applied for meta-analysis. The findings revealed a significant statistical discrepancy between the two groups [SMD = 0.54, 95% CI: 0.25–0.82, P = 0.0002] (Fig. 2). Comparison of MCP-1 in non-survivors and survivors among patients with sepsis Association of MCP-1 with prognosis in patients with sepsis The studies assessed the prognostic value of MCP-1 in sepsis patients. The combined analysis revealed that elevated levels of MCP-1 were linked to an unfavorable prognosis in patients with sepsis (random-effects model: HR = 1.36, 95% CI: 1.25–1.48, P < 0.00001). However, substantial heterogeneity was noted across the studies (I² = 99%, P < 0.00001) (Fig. 3). The hazard ratios of MCP-1 associated with prognosis in sepsis patients Publication bias analysis Publication bias was assessed for the studies included. Standard error (SE) of the logarithm of HR (log HR) was plotted on the y-axis, with HR values on the x-axis, and a funnel plot was generated. The analysis indicated the absence of significant publication bias across the studies as assessed by Egger’s test (p = 0.950 for Fig. 4 and p = 0.176 for Fig. 5). Funnel plot analysis of MCP-1 levels in non-survivors versus survivors in sepsis patients Funnel plot analysis assessing the association between elevated MCP-1 levels and mortality risk in sepsis patients Subgroup analysis In this study, we explored the heterogeneous effects of different factors on the association between MCP-1 and sepsis prognosis by subgroup analysis. The results were as follows. Sample Size Subgroup Analysis. Sample sizes exceeding 100 are considered large, while those below 100 are deemed small. The pooled hazard ratio (HR = 1.01, 95% CI [1.00,1.02], P = 0.07) in the large sample subgroup (n = 3, weight 66.6%) indicated a nearly neutral effect with minimal heterogeneity (I²=72%). Conversely, the small sample subgroups (n = 5, weight 33.4%) exhibited a significantly elevated risk (HR = 2.05, 95%CI [1.02, 4.12], P = 0.05) alongside substantial heterogeneity (I²=99%). The intergroup distinctions approached significance (P = 0.05) (Fig. 6). Subgroup analysis based on sample size Sepsis Diagnostic Criteria Grouping. Studies based on sepsis 1.0 or 2.0 diagnostic criteria (HR = 1.35, 95% CI [0.89, 2.06], P = 0.16) and studies defined by sepsis 3.0 (HR = 1.18, 95% CI [0.85, 1.62], P = 0.32) did not show significant overall effects and no significant differences between subgroups (P = 0.60) (Fig. 7). Subgroup analysis based on sepis diagnostic criteria Study Design Grouping. Prospective study subgroups (n = 6, weight 77.2%) had a weak effect (HR = 1.01, 95% CI [1.00, 1.02], P = 0.07) and moderate heterogeneity (I²=51%). The retrospective study subgroup (n = 2, weight 22.8%) showed a very significant high risk (HR = 3.31, 95%CI [2.99, 3.67], P < 0.00001) and very low heterogeneity (I²=0%). There was significant difference between subgroups (P < 0.00001) (Fig. 8). Subgroup analysis based on study design MCP-1 Test Method Grouping. Subgroups tested using ELISA (HR = 1.01, 95%CI [1.00, 1.02], P = 0.05) had near-neutral effects and moderate heterogeneity (I²=58%). The subgroup not tested by ELISA (HR = 2.55, 95% CI [1.28, 5.06], P = 0.008) had a significantly higher risk and very high heterogeneity (I²=88%). There were significant differences between subgroups (P = 0.008) (Fig. 9). Subgroup analysis based on MCP-1 detection method MCP-1 Detection Mode Grouping. The effect was not significant in the subgroup that examined MCP-1 alone (HR = 1.01, 95% CI [1.00, 1.02], P = 0.07). The subgroup combined with other markers (HR = 2.05, 95% CI [1.01, 4.13], P = 0.05) showed a borderline significant risk increase. The differences between subgroups were borderline significant (P = 0.05) (Fig. 10). Subgroup analysis based on MCP-1 combination Geographical Grouping Analysis. The European study demonstrated a significant risk (HR = 2.07, 95% CI [1.21, 3.55], P = 0.008) with low heterogeneity (I²=42%). In contrast, Asian studies revealed a weak but significant effect (HR = 1.33, 95% CI [1.22, 1.45], P < 0.00001) with very high heterogeneity (I²=99%). No significant difference between subgroups was observed (P = 0.11) (Fig. 11). Subgroup analysis based on geographic region Discussion The present meta-analysis of eight studies [13, 17–23], including 592 patients with sepsis, showed that MCP-1 levels were significantly higher in non-survivors compared to survivors, suggesting a significant association between elevated MCP-1 levels and poor sepsis outcomes. Multidimensional subgroup analysis showed that sample size, study design, detection method, and co-indicator application significantly affected the prognostic power of MCP-1. However, sepsis diagnostic criteria and geographic region may not affect the prognostic value of MCP-1 in sepsis. Notably, the subgroup of MCP-1 administered in combination with other biomarkers showed stronger effect values, which was consistent with previous studies [24–26]. The results support the use of a multi-marker combination strategy to improve the predictive accuracy of sepsis outcomes. Significant heterogeneity between studies may be due to methodological differences: first, measurement bias may be introduced by different assay techniques, and multiple immunoassays may have confounding effects due to simultaneous detection of other inflammatory factors; second, small sample studies may have exaggerated effect values due to random error or selective reporting; and third, differences in sepsis treatment regimens across centers may alter the dynamic expression trajectory of MCP-1. These findings suggest the need to establish standardized test systems before clinical translation and to verify their independent predictive value in large cohorts. MCP-1 is involved in sepsis through CCR2 receptor-mediated immune regulatory network: (1) chemotactic monocytes/macrophages to inflammatory sites [27, 28], forming a positive feedback loop with IL-6 and G-CSF [29], amplifying early inflammatory storms; (2) gene polymorphism (e.g., rs1024611) may increase susceptibility to sepsis in certain populations (e.g., diabetic patients) [30]; and (3) late induction of monocyte depletion, resulting in an immune paralytic state, is associated with an increased risk of secondary infections [31, 32]. Animal experiments confirmed that blocking MCP-1/CCR2 pathway could reduce inflammation and improve prognosis. In addition, regulation of MCP-1-related pathways, such as PI3K/AKT, may be a potential therapeutic strategy [33]. Comparison with previous studies shows that most studies support the prognostic value of MCP-1 [20, 22], especially Tian et al. found that it is significantly associated with the prognosis of critically ill patients [34]. A meta-analysis examining the diagnostic value of MCP-1 in adult sepsis demonstrated its high potential for early sepsis diagnosis [35]. However, it should be noted that Zhi et al.‘s Mendelian randomization study suggested a negative correlation trend [36]. Macdonald et al. ‘s study also failed to find an association between MCP-1 and sepsis prognosis [17]. This contradiction may be due to differences in study design, interference with comorbidities and selection of detection timing. As far as we are aware, this study represents the first meta-analysis to investigate the prognostic significance of MCP-1 in patients with sepsis. Furthermore, our study combined data from multiple studies that accounted for potential confounders, making the results more reliable than those from individual studies. Given the inconsistency in previous studies, our results may be invaluable for clinicians in identifying high-risk sepsis patients and providing appropriate treatment, thereby improving survival rates. Nevertheless, several limitations should be acknowledged in our study. First, substantial heterogeneity among the studies undermines the strength of the conclusions, even when employing a random-effects model. Second, the retrospective nature of some studies may introduce biases such as recall bias and selection bias. Third, publication bias is an inherent challenge in meta-analyses, as studies with positive results are more likely to be published. Additionally, the heterogeneity in sepsis etiologies (e.g., bacterial vs. fungal infections) across studies might confound the interpretation of biomarker-outcome relationships. Thus, the results should be interpreted with caution. The relationship between MCP-1 and sepsis prognosis may be influenced by numerous factors. Since no single biomarker has shown 100% accuracy in predicting outcomes, future studies that combine biomarkers and severity scores may yield more accurate results. Conclusions Our meta-analysis suggests that MCP-1 could be a promising prognostic biomarker for sepsis, with elevated MCP-1 levels potentially indicating poorer prognoses in sepsis patients. More robust studies on the sensitivity and specificity of MCP-1 as a biomarker are required to establish the clinical significance of MCP-1 as a prognostic tool for sepsis. Supplementary Information Supplementary Material 1. Acknowledgements Not applicable. Authors’ contributions Zhonghui HUANG conceived and designed the study; Nanna LI and Wei WANG performed the search; Qing WANG, Zhonghui HUANG and Wen YE performed the statistical analysis; Zhonghui HUANG and Wen YE wrote the manuscript; all authors read and approved the final manuscript. Funding This research was funded by China Armed police force science and technology innovation team independent research projects grant number ZZKY20242101. Data availability All data generated or analyzed in this research are included in this manuscript. Specific data research results are available from the corresponding author. Declarations Ethics approval and consent to participate All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration ant its later amendments or comparable ethical standards. Patient consent: not applicable. Consent for publication Not required. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement All data generated or analyzed in this research are included in this manuscript. Specific data research results are available from the corresponding author.