Biomarkers and clinical factors associated with sepsis-induced coagulopathy: A systematic review and meta-analysis

Abstract

Objective

Sepsis-induced coagulopathy (SIC) is a life-threatening complication that significantly impacts patient prognosis. This systematic review and meta-analysis aimed to evaluate clinical and biological indicators associated with the development of SIC to facilitate early risk identification.

Methods

We searched nine databases for studies reporting predictive factors for SIC. Data were pooled using random-effects or fixed-effects models as appropriate. Methodological quality was assessed using the Newcastle-Ottawa Scale.

Results

A total of 34 studies involving 37,459 patients were included. Meta-analysis revealed that several markers showed a significant association with SIC. Glycocalyx degradation emerged as a strong predictor (OR = 11.85, 95% CI: 8.59–16.34). Other associated markers included reduced Fibrinogen (OR = 4.35, 95% CI: 2.47–7.69), Albumin (OR = 2.5, 95% CI: 1.82–3.45), Neutrophil Extracellular Traps (OR = 2.4, 95% CI: 1.81–3.19), increased circulating Microparticles (OR = 1.61, 95% CI: 1.33–1.95), elevated Lactate(OR = 1.18, 95% CI: 1.13–1.23) and Thrombomodulin (OR = 1.10, 95% CI: 0.97–1.26), Procalcitonin, APACHE II score and C-reactive protein were also demonstrated predictive value for SIC occurrence.

Conclusion

Our findings identify multiple biological and clinical predictors associated with SIC. However, these results should be interpreted as statistical associations rather than direct causal links. These markers may assist clinicians in early recognition of high-risk patients, though further prospective trials are needed to confirm their clinical utility.

Registration

This study was registered on the PROSPERO (CRD420250653820).

Keywords

Sepsis sepsis-induced coagulopathy associated factors systematic review and meta-analysis

Introduction

Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection.¹ A hallmark of its complex pathogenesis is the profound disruption of the coagulation system, primarily driven by vascular endothelial injury and excessive systemic inflammation.^2,3 Sepsis remains a major global health challenge, affecting over 30 million people and causing 5 million deaths annually.^4,5 Sepsis-induced coagulopathy (SIC) represents a critical stage of this dysfunction, where endothelial damage triggers a cascade of hemostatic imbalances that can progress to multiple organ failure.⁶

Pathophysiologically, SIC evolves from beneficial immunothrombosis, which is a localized host defense mechanism, to dysregulated thromboinflammation.⁷Persistent systemic inflammation triggers extensive microvascular thrombosis, aggravating endothelial dysfunction, capillary leakage, and consumptive coagulopathy, which eventually culminates in disseminated intravascular coagulation (DIC) and multiple organ dysfunction syndrome (MODS).⁸Currently, SIC affects 24% to 68% of sepsis patients globally and is associated with a twofold increase in mortality.^9–13

Given its multifactorial nature, identifying the exact predictors of SIC and understanding their interactions are essential for early detection and clinical management.¹⁴ Despite extensive research,¹⁵ significant gaps remain in characterizing these markers within the evolving landscape of sepsis care.

With the evolution of sepsis management, an updated and comprehensive analysis of factors associated with SIC is necessary. By integrating heterogeneous data, this study aims to provide robust evidence on the most critical predictors of SIC to guide clinical decision-making and optimize patient outcomes.

Methods

Search strategy and selection criteria

This systematic review and meta-analysis was conducted following the PRISMA 2020 guidelines,¹⁶ and registered on the PROSPERO (CRD420250653820). A systematic search was conducted across international (PubMed, Embase, Web of Science, Cochrane Library, Ovid-Medline, Scopus) and Chinese databases (CNKI, Wanfang, CBM) for relevant studies published up to December 31, 2025. We employed a comprehensive search strategy combining controlled vocabulary (e.g. MeSH terms in PubMed and Emtree in Embase) and free-text keywords related to “sepsis,” “coagulopathy,” and “associated factors.” No language restrictions were applied to minimize publication bias. The complete search strategy is detailed in the text and Supplementary Table 1.

Inclusion criteria: (1) Peer-reviewed observational studies (prospective or retrospective cohort studies, case-control studies) published in English or Chinese. (2) Adult patients ( $\geq$ 18 years) diagnosed with sepsis (according to the sepsis 3.0¹) and SIC (according to the International Society on Thrombosis and Haemostasis diagnostic criteria 2017).¹⁷ (3) Comparison of clinical or biological factors between patients with SIC and those without SIC. (4) Reported adjusted odds ratios (OR) from multivariable analyses or provided sufficient raw data to calculate ORs with 95% confidence intervals (CI).

Exclusion criteria: (1) Publication types were reviews, conference abstracts, case reports, editorials, and animal experiments. (2) Studies with overlapping populations, inaccessible full texts, or insufficient data for effect size extraction. (3) Patients with significant co-morbidities (e.g. hematological malignancies) that could confound coagulation markers.

Literature screening process

Literature records were managed using EndNote X9, with duplicates systematically removed. Two reviewers (YW and FX) independently conducted a two-stage screening process: an initial title and abstract review followed by a comprehensive full-text evaluation based on the pre-defined eligibility criteria. Any discrepancies at either stage were resolved through consensus or, if necessary, consultation with a third reviewer (XP or ZY).

Data extraction and study quality assessment

Data extraction was performed independently by two researchers (YW and FX), capturing study characteristics, participant demographics, and adjusted effect sizes (OR with 95% CI). In this study, OR were extracted from the included literature as effect measures. Only multivariable-adjusted effect estimates were extracted when available. Importantly, multivariable-adjusted estimates were included regardless of their statistical significance, provided that the corresponding predictors were prespecified and reported in the original studies. Factors associated with SIC were excluded only when multivariable-adjusted effect estimates were unavailable or could not be reliably extracted.

Methodological quality was assessed by YW and LC using the Newcastle-Ottawa Scale (NOS). The NOS tool assesses studies through three major modules with a total of eight items, which specifically include: (1) selection of study population, (2) comparability between groups, (3) assessment of exposure/outcome. The NOS employs a semi-quantitative system with a maximum score of 9 stars, where scores of 7–9 indicate high-quality research. Two reviewers independently assessed study quality using the NOS. Inter-rater reliability was evaluated using Cohen's kappa statistic, which showed moderate agreement between reviewers (k = 0.8, P < 0.001), indicating acceptable consistency in quality assessment.

Data synthesis and analysis

Data synthesis was performed using Review Manager (version 5.3) and Stata (version 14.0). Adjusted OR and their 95% CI were extracted as the primary effect measures. Between-study heterogeneity was assessed using Cochran's Q statistic and the $I^{2}$ index. A fixed-effects model was applied if $I^{2} < 50 %$ or $P > 0.05$ for the Q test; otherwise, a random-effects model was employed. Sensitivity analyses were conducted by sequentially excluding individual studies to evaluate the stability of the pooled estimates. For indicators supported by at least 10 studies, potential publication bias was assessed visually using funnel plots and formal statistical testing was performed using Egger's regression test ( $P < 0.1$ was considered to indicate significant publication bias). For factors with fewer than 10 studies, where statistical tests have limited power, the results were interpreted primarily as hypothesis-generating.

Results

Study characteristics

A total of 34 studies met the inclusion criteria and were incorporated into the final analysis after title/abstract screening and full-text evaluation (Figure 1).

Figure 1.

Flowchart for study selection process.

The characteristics of the eligible studies are summarized in Table 1. These comprised 14 prospective and 20 retrospective investigations, including 26 cohort studies and 8 case–control studies. Most included studies were conducted in Asian populations, predominantly from China and Japan, the majority originated from Asia (n = 26; primarily China and Japan), followed by North America (n = 3) and Europe (n = 5), while only a limited number originated from other regions.

Table 1.

Characteristics of the included studies.

Author (year)	Country	Study design	Setting	Total sample	Age	SIC sample (%)	Associated factors
Annalisa B¹⁸ (2019)	Italy	Case-control	ICU	80	64 (48.5–71)	12 (15.0)	①
Xavier D¹⁹ (2017)	France	Prospective	ICU	20	65 ± 17	10 (50.0)	①②
Emmanuelle M²⁰ (2016)	France	Prospective	ICU	259	66.8 ± 13.3	93 (35.9)	①
Shishuai M²¹ (2021)	China	Case-control	ICU	60	56 (19–86)	30 (50.0)	①
Xiaoya W²² (2021)	China	Prospective	EICU	62	67.91 ± 17.11	32 (51.6)	①⑨
Cao R²³ (2023)	China	Prospective	ICU	78	5.63 (3.08–10.09)	36 (46.1)	②
Fujita N²⁴ (2024)	Japan	Prospective	ICU	145	79	26 (17.9)	③
Xiao H²⁵ (2020)	China	Retrospective	ICU	45	56 (48–65)	10 (22.2)	③
Mitsunori I²⁶ (2017)	Japan	Prospective	ICU	54	73 (65–81)	13 (24.1)	③
Shaokang Z²⁷ (2023)	China	Prospective	ICU	80	78.00 ± 10.77	24 (30.0)	③
Yaojie F²⁸ (2024)	China	Retrospective	ICU	336	58.00 ± 16.96	32 (9.5)	④
Gao Y²⁹ (2024)	China	Retrospective	PICU	422	9 (4.06–13.46)	140 (33.1)	④⑧
Francesco G³⁰ (2022)	Italy	Prospective	ED	763	80 (72–87)	38 (5.0)	④
Xiaofang G³¹ (2022)	China	Case-control	ICU	145	69.48 ± 11.91	62 (42.8)	④⑦
Qingbo Z³² (2021)	China	Case-control	ICU	172	66 (53 ± 78)	68 (39.5)	④⑤
Ken T³³ (2022)	Japan	Retrospective	ICU	753	70 (63–78)	181 (24.0)	④⑨
Weiqing H³⁴ (2023)	China	Retrospective	ICU	166	52.70 ± 17.11	66 (39.8)	④
Yingying Y³⁵ (2023)	USA	Case-control	ICU	2745	71 (62.8–79.7)	1152 (41.9)	④
Jing T³⁶ (2021)	China	Retrospective	ICU	90	77.2 ± 3.6	45 (50.0)	⑤
Xiaowei B³⁷ (2021)	China	Prospective	ICU	80	68.3 ± 18.75	28 (35.0)	⑤
Ikhwan R³⁸ (2022)	Indonesia	Retrospective	ICU	248	49.76 ± 13.97	50 (20.2)	⑥
Yukun C³⁹ (2022)	China	Case-control	ICU	200	58.94 ± 10.35	138 (69.0)	⑥⑨⑩
Ye Z⁴⁰ (2024)	China	Retrospective	ICU RICU	279	66.94 ± 13.6	108 (38.7)	⑦⑧
Suhui Q⁴¹ (2021)	China	Retrospective	ICU	57	61.05 ± 8.32	21 (36.8)	⑦⑩
Takamitsu M⁴² (2020)	Japan	Prospective	ICU	107	74.6 (62.0–91.0)	32 (29.9)	⑧
Liying Z⁴³ (2024)	China	Prospective	ICU	548	60.5 ± 16.9	302 (55.1)	④⑥
Yishan W⁴⁴ (2021)	China	Prospective	EICU RICU SICU	116	59.56 ± 17.13	49 (42.2)	⑩
Keisuke M⁴⁵ (2022)	Japan	Prospective	ICU	60	75 (69–80)	21 (35.0)	⑩
Jian-Yue Y⁴⁶ (2025)	USA	Case-control	ICU	12,455	63 (52–73)	5145 (41.3)	④
Jia S⁴⁷ (2025)	USA	Case-control	ICU	15,479	71 (59–81)	6036 (38.9)	⑥
Ruimin T⁴⁸ (2025)	China	Retrospective	ICU	847	70.01 ± 14.73	480 (56.7)	④⑥⑦
Wen- hao M⁴⁹ (2025)	China	Retrospective	ICU	236	66 (53–73)	87 (36.8)	④
Maijier M⁵⁰ (2025)	China	Retrospective	ICU	125	4 (2, 8)	64 (51.2)	④⑥⑦
Zhuowen D⁵¹ (2025)	China	Prospective	ICU	147	70 (58,78. 5)	62 (42.0)	②

Note: ED: emergency department; EICU: emergency intensive care units; RICU: respiratory intensive care units; SICU: surgery intensive care units.

①Microparticles; ②Neutrophil extracellular traps; ③Glycocalyx; ④Lactate; ⑤Thrombomodulin; ⑥Albumin; ⑦Procalcitonin; ⑧C-reactive protein; ⑨APACHE II; ⑩Fibrinogen.

Quality assessment

The methodological quality of the included studies was assessed using the NOS. Nine studies achieved the maximum score of 9, fourteen scored 8, eight scored 7, and three scored 6, reflecting an overall high quality and supporting the robustness and reliability of the pooled findings. Supplementary Table 1 provides a detailed breakdown of each study across domains of participant selection, comparability between groups, and outcome assessment, together with the overall score. In addition, two representative studies are presented in the Supplementary Materials to illustrate the step-by-step NOS scoring process at the item level.

Results of synthesis

These studies investigated a range of contributing factors, including Glycocalyx,^24–27 Fibrinogen (FIB),^39,41,44,45 Albumin (ALB),^{38,39,47,48,50} Neutrophil Extracellular Traps (NETs),^19,23,51 Microparticles (MPs),^18–22 Lactate,^28–35^{,43,46,48–50} Thrombomodulin (TM),^32,36,37 Procalcitonin (PCT),^{31,40,41,48,50} Acute Physiology and Chronic Health Evaluation II score (APACHE II)^22,33,39,51and C-reactive Protein (CRP).^{23,29,40,42,51}

Overall, ten factors were synthesized using random-effects models, with effect estimates expressed as OR comparing SIC versus non-SIC patients (Table 2). These factors were categorized into three groups based on their pathophysiological roles. Endothelial Injury Markers (Glycocalyx, TM, MPs),Clinical Severity and Metabolic Indicators(FIB, ALB, APACHE II, Lactate) and Inflammation and Immunothrombosis Markers(NETs, PCT, CRP).

Table 2.

Summary of associated factors for SIC.

Associated factors	Number of studies	OR	95% CI	Heterogeneity			Overall effect test
Associated factors	Number of studies	OR	95% CI	τ²	P	I² (%)	Z-value	P-value
Glycocalyx	4	11.85	8.59∼16.34	0.00	0.41	0	15.09	P < 0.05
FIB	4	4.35	2.47∼7.69	0.00	0.84	0	5.07	P < 0.05
ALB	5	2.50	1.82∼3.45	0.00	0.62	0	5.61	P < 0.05
NETs	3	2.40	1.81∼3.19	0.00	0.62	0	6.04	P < 0.05
MPs	5	1.61	1.33∼1.95	0.01	0.33	13	4.84	P < 0.05
Lactate	13	1.18	1.13∼1.23	0.00	0.16	28	7.53	P < 0.05
TM	3	1.10	0.97∼1.26	0.01	0.24	29	1.52	P < 0.05
PCT	5	1.06	1.03∼1.09	0.00	0.27	23	3.99	P < 0.05
APACHE II	4	1.06	1.02∼1.10	0.00	0.31	16	3.18	P < 0.05
CRP	5	1.01	1.00∼1.01	0.00	0.33	13	4.84	P < 0.05

Endothelial injury markers

Markers reflecting endothelial injury and Glycocalyx degradation showed the most robust associations with SIC development (Figure 2).

Figure 2.

Forest plot for endothelial injury markers. (2A: Glycocalyx, 2B: TM, 2C: MPs).

Glycocalyx^24–27: Demonstrated the strongest association with SIC (4 studies; pooled OR = 11.85, 95% CI: 8.59–16.34), with no detectable heterogeneity ( $I^{2} = 0 %$ ).

TM^32,36,37: Was significantly associated with an increased likelihood of SIC (3 studies; OR = 1.10, 95% CI: 0.97–1.26), with mild heterogeneity $(I^{2} = 29 %$ ).

MPs^18–22: Showed a consistent association with SIC (5 studies; OR = 1.61, 95% CI: 1.33–1.95), with mild heterogeneity ( $I^{2} = 13 %$ ).

Clinical severity and metabolic indicators

Factors reflecting systemic metabolic distress and overall disease severity exhibited substantial predictive value (Figure 3).

Figure 3.

Forest plot for clinical severity and metabolic indicators. (3A: FIB, 3B: ALB, 3C: APACHE II, 3D: Lactate).

FIB^39,41,44,45: Reduced levels were markedly associated with SIC and likely represent a consequence of consumptive coagulopathy rather than a primary upstream trigger. (4 studies; OR = 4.35, 95% CI: 2.47–7.69), with no detectable heterogeneity ( $I^{2} = 0 %$ ).

ALB^{38,39,47,48,50}: Hypoalbuminemia was strongly associated with SIC (5 studies; OR = 2.50, 95% CI: 1.82–3.45), with no detectable heterogeneity ( $I^{2} = 0 %$ ).

APACHE II^22,33,39,51: Higher scores were significantly associated with SIC occurrence (4 studies; OR = 1.06, 95% CI: 1.02–1.10), with mild heterogeneity ( $I^{2} = 16 %$ ).

Lactate^28–35^{,43,46,48–50}: Elevated Lactate levels were associated with SIC (13 studies; OR = 1.18, 95% CI: 1.13–1.23), with mild heterogeneity ( $I^{2} = 28 %$ ).

Inflammation and immunothrombosis markers

Markers related to the host's inflammatory response and immunothrombotic pathways showed consistent positive associations (Figure 4).

Figure 4.

Forest plot for inflammation and immunothrombosis markers. (4A: NETs, 4B: PCT, 4C: CRP).

NETs^19,23,51: Were significantly associated with SIC (3 studies; OR = 2.40, 95% CI: 1.81–3.19), with mild heterogeneity ( $I^{2} = 0 %$ ).

PCT^{31,40,41,48,50}: Showed a positive association with SIC (5 studies; OR = 1.06, 95% CI: 1.03–1.09), with mild heterogeneity ( $I^{2} = 23 %$ ).

CRP^{23,29,40,42,51}: Demonstrated a small but statistically significant association(5 studies; OR = 1.01, 95% CI: 1.00–1.01), with mild heterogeneity( $I^{2} = 13 %$ ).

Conceptual framework

Based on the synthesized results, a conceptual diagram (Figure 5) was developed to illustrate the interplay between these multi-dimensional factors and the development of SIC.

Figure 5.

Conceptual framework of the interplay between these multi-dimensional factors and the development of SIC.

Sensitivity analysis

Leave-one-out sensitivity analyses were performed for the ten factors associated with SIC. By sequentially excluding each study and recalculating the pooled effect sizes using a random-effects model, we observed that the pooled estimates remained consistent across all iterations. These results suggest that the overall findings were robust and not materially driven by any individual study.

Meta-regression analysis for lactate

Following the methodological guidance from the Cochrane Handbook (version 6.4), meta-regression is generally recommended only when at least ten studies are available per covariate to ensure adequate statistical power. Among the ten factors analyzed, only Lactate (n = 13) met this criterion. For all other biomarkers, the number of contributing studies was $\leq 5$ , rendering meta-regression statistically underpowered and potentially misleading. To avoid overfitting and spurious precision, we adhered to the “one-covariate-per-ten-studies” rule and performed univariate meta-regression for Lactate rather than a multivariable model. The detailed results are presented in Table 3. In summary, none of the examined covariates (including SIC incidence rate, geographic region, or clinical setting) significantly explained the between-study heterogeneity for Lactate, as all P-values exceeded 0.05 and adjusted $R^{2} \leq 0 %$ .

Table 3.

Univariate meta-regression estimates for potential sources of heterogeneity of lactate.

Covariates	t	P	I² (%)	Adj. R²	τ²	β (95% CI)
SIC_rate	0.67	0.52	35.8	−61%	0.001	0.10 (−0.22 to 0.41)
Continent (vs North America)
Asia	0.23	0.82	42.2	−87%	0.001	0.02 (−0.14 to 0.17)
Europe	0.00	0.999				0.00 (−0.25 to 0.25)
Setting
ICU vs Non-ICU	0.14	0.894	37.2	−32%	0.001	0.013 (−0.19 to 0.22)

Publication bias

Publication bias was assessed with caution due to the limited number of studies available for most associated factors. Funnel plots were generated to visually explore potential publication bias; however, formal interpretation was restricted to factors supported by an adequate number of studies. Among the ten identified factors, only Lactate was reported in a sufficient number of studies (n = 13) to permit formal statistical assessment of small-study effects. For Lactate, Egger's regression test did not indicate significant publication bias (bias coefficient = 1.48, P = 0.105), and visual inspection of the funnel plot showed approximate symmetry, suggesting a low risk of publication bias for this factor (Figure 6).For the remaining factors, the number of contributing studies was limited (generally fewer than 10), which substantially reduces the reliability of both funnel plot symmetry assessment and formal tests such as Egger's regression. Therefore, no definitive conclusions regarding publication bias were drawn for these factors, and their pooled results should be interpreted cautiously and considered hypothesis-generating rather than confirmatory.

Figure 6.

Funnel plot for lactate.

Discussion

This systematic review and meta-analysis identified a multi-dimensional set of clinical and biological predictors associated with SIC. By synthesizing data from both routine clinical parameters and novel biomarkers, our findings highlight three core pathological axes: endothelial injury, immune-inflammatory dysregulation, and systemic metabolic distress.

The endothelial injury axis: the initiating event of SIC

Endothelial dysfunction is increasingly recognized as the central mechanism in the pathogenesis of SIC.¹⁴ In our analysis, markers of endothelial Glycocalyx degradation demonstrated the most potent predictive strength (OR = 11.85). The Glycocalyx is a delicate, carbohydrate-rich layer coating the vascular endothelium that regulates permeability and prevents the adhesion of platelets and leukocytes.⁵² Under the systemic “insult” of sepsis, inflammatory mediators trigger the shedding of Glycocalyx components (such as Syndecan-1) into the circulation. This structural breakdown exposes the underlying endothelium to pro-thrombotic factors, initiating microvascular thrombosis.^53,54 The finding in our study underscores that monitoring Glycocalyx integrity is paramount for the early identification of SIC.However, this pooled estimate was derived from only four studies, and extreme effect sizes in meta-analyses may arise from small-study effects or residual publication bias. Therefore, although the association appears strong, it should be interpreted cautiously and considered hypothesis-generating pending validation in larger, multicenter cohorts. TM (OR = 1.1) and MPs (OR = 1.61) serve as significant indicators within this axis. Under physiological conditions, TM acts as a key anticoagulant on the endothelial surface; however, widespread endothelial damage causes TM to be cleaved and released into the plasma.⁵⁵ This shedding not only signifies the loss of local anticoagulant capacity but also reflects a pathological “phenotype shift” of the endothelium from a thromboresistant to a procoagulant state.⁵⁶ Similarly, MPs act as pro-coagulant vesicles that transport tissue factor, amplifying the coagulation cascade intravascularly.⁵⁷ Integrating these markers allows for a sensitive biological assessment of the vascular endo-environment.

The immune-inflammatory cascade: driving thromboinflammation

Sepsis is characterized by a dysregulated host immune response, where inflammation and coagulation are inextricably linked.² Our findings demonstrate that NETs (OR = 2.4) are significantly associated with SIC development. As a cornerstone of “immunothrombosis,” NETs provide a physical scaffold for fibrin deposition and platelet entrapment, bridging the gap between innate immunity and pathological thrombosis.⁵⁸ Among traditional inflammatory markers, PCT (OR = 1.06) and CRP (OR = 1.01) showed statistically significant, albeit more modest, associations with SIC. These findings suggest that a persistent systemic inflammatory state acts as a continuous driver of hemostatic deterioration.^59,60 Although their individual predictive power is lower than that of endothelial markers, their high clinical accessibility makes CRP and PCT valuable tools for monitoring infection progression and the secondary risk of coagulopathy at the bedside.

Clinical severity and metabolic indicators: reflections of systemic crisis

Clinical severity scores and metabolic markers are indispensable for refining the predictive landscape of SIC.⁶¹ In our study, the APACHE II score (OR = 1.06) and Lactate levels (OR = 1.18) emerged as robust indicators of the cumulative physiological burden and tissue hypoxia. High APACHE II scores often reflect multiorgan dysfunction, where circulatory and respiratory failures synergistically exacerbate hemostatic imbalances.^62,63 The stability of this association across different cohorts suggests that the overall severity of sepsis, as captured by this scoring system, is a universal prerequisite for the transition into a coagulopathic state. Complementing this, Lactate acts as a critical biomarker of metabolic distress and cellular hypoxia. In sepsis, elevated Lactate signifies a transition toward decompensated shock and mitochondrial dysfunction.⁶⁴ Our univariate meta-regression on the 13 studies involving Lactate further reinforced its reliability, suggesting its association with SIC remains stable across diverse healthcare environments and populations. In contrast, markers such as FIB (OR = 4.35) and ALB (OR = 2.50) provide complementary information regarding the consumptive and exudative phases of the disease,^65,66Notably, decreased fibrinogen levels in sepsis most likely reflect ongoing coagulation activation and consumption, indicating an established coagulopathic state rather than serving as an early upstream predictor. Low albumin levels, in particular, are often a harbinger of capillary leak syndrome, which parallels the destruction of the endothelial Glycocalyx.⁶⁷

Methodological heterogeneity and clinical implications

A significant challenge in interpreting our findings is the lack of standardization across studies regarding detection methods and diagnostic thresholds. For key indicators such as Glycocalyx, NETs, and MPs, various assays were employed, which likely contributed to the statistical heterogeneity observed. Furthermore, the “optimal cut-off” for many biomarkers varied across cohorts, complicating the establishment of a universal threshold for clinical use. To date, no human trials have demonstrated that targeting these markers reduces SIC incidence or improves clinical outcomes. Consequently, the predictive values identified in this meta-analysis should be viewed with caution. Future research must prioritize the standardization of laboratory techniques and the validation of consensus-based thresholds before these markers can be formally integrated into SIC diagnostic protocols.

Based on these findings, we propose a two-tier risk stratification strategy for the early identification and management of SIC. First-tier screening relies on routine, readily available parameters, including APACHE II, Lactate, FIB, ALB, and CRP, for rapid bedside assessment. Patients flagged as high-risk through these routine tools should then undergo second-tier precision assessment using specialized biomarkers such as Glycocalyx and TM. This hierarchical approach allows for an “early-warning” phase, where endothelial damage can be detected before the manifestation of irreversible clinical abnormalities, such as severe platelet consumption or multi-organ failure. This strategy not only optimizes resource allocation but also provides a roadmap for future predictive model construction.

Limitations

This study provides valuable insights into the associated factors associated with SIC; however, several limitations should be acknowledged. First, Given the observational nature of the included studies, the identified factors should be interpreted as strong statistical associations rather than definitive causal links. Second, the number of eligible studies was relatively limited, which constrained our ability to fully elucidate the specific roles and predictive value of certain factors, thereby restricting their immediate clinical applicability. Third, the literature search was confined to studies published in English and Chinese, raising the possibility that relevant evidence reported in other languages may have been overlooked. Fourth, biological collinearity remains a challenge in sepsis research, and multivariable models in original studies may not have fully eliminated these inter-relationships. Last, most of the included studies were conducted in Asian populations, which may introduce regional bias and limit the generalizability of our findings to more diverse patient cohorts.

Conclusion

In conclusion, the development of SIC is driven by a triad of endothelial injury, inflammatory surges, and metabolic crises. Among the 10 predictors identified, endothelial glycocalyx markers and lactate demonstrate high predictive value and clinical robustness. Future research should focus on developing integrated, dynamic scoring systems that incorporate these multi-dimensional indicators, and prospective trials are needed to determine if early interventions targeting these pathways can improve clinical outcomes in sepsis.

Supplemental Material

sj-docx-1-sci-10.1177_00368504261435291 - Supplemental material for Biomarkers and clinical factors associated with sepsis-induced coagulopathy: A systematic review and meta-analysis

Supplemental material, sj-docx-1-sci-10.1177_00368504261435291 for Biomarkers and clinical factors associated with sepsis-induced coagulopathy: A systematic review and meta-analysis by Wei You, Xiaoyu Fan, Ping Xu, Cheng Lei and Yanli Zeng in Science Progress

Footnotes

Acknowledgements

The authors would like to thank all investigators and participants of the original studies included in this systematic review and meta-analysis.

ORCID iDs

Wei You

Xiaoyu Fan

Ping Xu

Cheng Lei

Yanli Zeng

Author contributions

Wei You: conceptualization, methodology, software, visualization, and writing—original draft. Xiaoyu Fan: methodology, data curation, and visualization. Ping Xu and Cheng Lei: methodology, data curation, and visualization. Yanli Zeng: conceptualization, methodology, supervision, and writing—review and editing.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

All data analyzed in this study are derived from previously published articles. The extracted data supporting the findings of this study are available from the corresponding author upon reasonable request.

Supplemental material

Supplemental material for this article is available online.

References

Singer

, et al. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA 2016; 315: 801–810.

Font

Thyagarajan

Khanna

. Sepsis and septic shock – basics of diagnosis, pathophysiology and clinical decision making. Med Clin North Am 2020; 104: 573–585.

Cheng

, et al. Unveiling the research advances of sepsis: pathogenesis, precise intervention and clinical perspective. Int J Surg 2025; 111: 6260–6289.

Adhikari

, et al. Critical care and the global burden of critical illness in adults. Lancet 2010; 376: 1339–1346.

Fleischmann

, et al. Assessment of global incidence and mortality of hospital-treated sepsis. Current estimates and limitations. Am J Respir Crit Care Med 2016; 193: 259–272.

Zhu

, et al. The mechanisms of sepsis induced coagulation dysfunction and its treatment. J Inflamm Res 2025; 18: 1479–1495.

Zaid

Merhi

. Implication of platelets in immuno-thrombosis and thrombo-inflammation. Front Cardiovasc Med 2022; 9: 863846.

Iba

, et al. Disseminated intravascular coagulation: the past, present, and future considerations. Semin Thromb Hemost 2022; 48: 978–987.

Ding

, et al. Comparison of a new criteria for sepsis-induced coagulopathy and International Society on Thrombosis and Haemostasis disseminated intravascular coagulation score in critically ill patients with sepsis 3.0: a retrospective study. Blood Coagul Fibrinolysis 2018; 29: 551–558.

10.

Schmoch

, et al. The prevalence of sepsis-induced coagulopathy in patients with sepsis – a secondary analysis of two German multicenter randomized controlled trials. Ann Intensive Care 2023; 13: 3.

11.

Singh

, et al. Trends in the incidence and outcomes of disseminated intravascular coagulation in critically ill patients (2004-2010): a population-based study. Chest 2013; 143: 1235–1242.

12.

Tanaka

, et al. Validation of sepsis-induced coagulopathy score in critically ill patients with septic shock: post hoc analysis of a nationwide multicenter observational study in Japan. Int J Hematol 2021; 114: 164–171.

13.

Yamakawa

, et al. External validation of the two newly proposed criteria for assessing coagulopathy in sepsis. Thromb Haemost 2019; 119: 203–212.

14.

Wang

, et al. Sepsis-Induced endothelial barrier dysfunction: mechanisms, pathology, and therapeutic advances. Research (Wash DC) 2025; 8: 0997.

15.

Wang

, et al. Global research trends in sepsis-associated coagulopathy: a web of science-based bibliometric analysis (2000-2024). J Multidiscip Healthc 2025; 18: 7235–7257.

16.

Page

, et al. PRISMA 2020 Explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. Br Med J 2021; 372: n160.

17.

Iba

, et al. New criteria for sepsis-induced coagulopathy (SIC) following the revised sepsis definition: a retrospective analysis of a nationwide survey. BMJ Open 2017; 7: e017046.

18.

Boscolo

, et al. Levels of circulating microparticles in septic shock and sepsis-related complications: a case-control study. Minerva Anestesiol 2019; 85: 625–634.

19.

Delabranche

, et al. Evidence of netosis in septic shock-induced disseminated intravascular coagulation. Shock 2017; 47: 313–317.

20.

Mercier

, et al. Early detection of disseminated intravascular coagulation during septic shock: a multicenter prospective study. Crit Care Med 2016; 44: e930–e939.

21.

Meng

, et al. Preliminary study of microparticle coagulation properties in septic patients with disseminated intravascular coagulation. J Int Med Res 2021; 49: 30–31.

22.

. Study on the changes and significance of tissue factor and high mobility group box 1 protein in patients with coagulation dysfunction in sepsis. Dalian, Liaoning, China: Dalian Medical University, 2021.

23.

Cao

. The predictive value of peripheral blood C3aR1 and NETs expression levels for sepsis-induced coagulopathy. Chinese Journal of Contemporary Pediatrics 2023; 25: 1259–1264.

24.

Fujita

, et al. Endothelial glycocalyx injury is a risk factor for disseminated intravascular coagulation development in urological sepsis. Eur Urol, 2024; 85: 154–155.

25.

Huang

, et al. Plasma endothelial glycocalyx components as a potential biomarker for predicting the development of disseminated intravascular coagulation in patients with sepsis. J Intensive Care Med 2021; 36: 1286–1295.

26.

Ikeda

, et al. Circulating syndecan-1 predicts the development of disseminated intravascular coagulation in patients with sepsis. J Crit Care 2018; 43: 48–53.

27.

Zhang

. The application value of heparin-binding protein and glycocalyx in the clinical diagnosis of sepsis-induced coagulopathy. Dalian, Liaoning, China: Dalian Medical University, 2023.

28.

, et al. A predictive model for disseminated intravascular coagulopathy in sepsis: an observational study. Int J Gen Med 2024; 17: 4845–4855.

29.

Gao

, et al. Novel nomogram for the prediction of sepsis-induced coagulopathy in the PICU: a multicentre retrospective study. Thromb Res 2024; 243: 109152.

30.

Gavelli

, et al. Are baseline levels of Gas6 and soluble mer predictors of mortality and organ damage in patients with sepsis? the need-Speed trial database. Biomedicines 2022; 10: 198.

31.

Guo

. Analysis of the incidence and risk factors of coagulation dysfunction in patients with sepsis related to abdominal infection. Journal of Clinical Emergency (China) 2022; 23: 260–264.

32.

Zeng

. The diagnostic value of thrombomodulin in sepsis-induced coagulopathy. Med J Chin PLA 2021; 46: 593–597.

33.

Tonai

, et al. Association between hypomagnesemia and coagulopathy in sepsis: a retrospective observational study. BMC Anesthesiol 2022; 22: 59.

34.

Hong

. Analysis of influencing factors of coagulation dysfunction associated with sepsis. China Modern Medicine 2023; 30: 51–56.

35.

. Establishment and validation of a prediction model for sepsis-related coagulation dysfunction in diabetic patients based on the MIMIC-IV database. Hangzhou, Zhejiang, China: Zhejiang Chinese Medical University, 2023.

36.

Tao

, et al. Increased level of thrombomodulin is associated with endothelial injury in patients with sepsis-Induced disseminated intravascular coagulation. Clin Lab 2021; 67. doi:https://doi.org/10.7754/Clin.Lab.2020.201204

37.

Bai

, et al. The early diagnostic value of plasma TM, TAT, PIC and t-PAIC in sepsis-induced DIC. Journal of Tongji University (Medical Science) 2021; 42: 327–332. +342.

38.

Rinaldi

Sudaryo

Prihartono

. Disseminated intravascular coagulation in sepsis and associated factors. J Clin Med 2022; 11: 6480.

39.

Cao

. Study on risk factors inducing coagulation dysfunction in patients with sepsis. Xinxiang, Henan, China: Xinxiang Medical University, 2022.

40.

Zhang

, et al. Diagnostic and prognostic value of C1q in sepsis-induced coagulopathy. Clin Appl Thromb Hemost 2024; 30: 10760296241257517.

41.

Qin

S-H

. Analysis of risk factors for onset and prognosis in patients with sepsis-associated coagulation dysfunction. Chinese J Gen Pract 2021; 19: 2020–2023.

42.

Masuda

Shoko

. Clinical investigation of the utility of a pair of coagulation-fibrinolysis markers for definite diagnosis of sepsis-induced disseminated intravascular coagulation: a single-center, diagnostic, prospective, observational study. Thromb Res 2020; 192: 116–121.

43.

. Development and validation of a nomogram prediction model for the risk of sepsis-induced coagulopathy. Changchun, Jilin, China: Jilin University, 2024.

44.

Wang

, et al. Plasma hsa-miR-92a-3p in correlation with lipocalin-2 is associated with sepsis-induced coagulopathy. BMC Infect Dis 2020; 20: 55.

45.

Mori

, et al. Decreasing plasma fibrinogen levels in the intensive care unit are associated with high mortality rates in patients with sepsis-induced coagulopathy. Clin Appl Thromb Hemost 2022; 28: 10760296221101386.

46.

Yang

. Association analysis of sepsis progression to sepsis-induced coagulopathy: a study based on the MIMIC-IV database. BMC Infect Dis 2025; 25: 73.

47.

Sun

, et al. A machine learning model for robust prediction of sepsis-induced coagulopathy in critically ill patients with sepsis. Front Cell Infect Microbiol 2025; 15: 1579558.

48.

Tan

, et al. Development and validation of a nomogram for early prediction of sepsis-induced coagulopathy: a multicenter study. Front Med (Lausanne) 2025; 12: 1653699.

49.

, et al. Development and validation of a nomogram prediction model for sepsis-induced coagulopathy: a multicenter retrospective study. Curr Med Sci 2025; 45: 867–876.

50.

Maihemuti

Simijiang

. Predictors of sepsis-induced coagulopathy in children and development of a prediction model. Zhongguo Dang Dai Er Ke Za Zhi 2025; 27: 1506–1513.

51.

Dai

Wang

, et al. The value of peripheral blood neutrophil extracellular traps as a marker for sepsis-induced coagulopathy. Chinese Journal of Blood Transfusion 2025; 38: 1340–1347.

52.

Lupu

Kinasewitz

Dormer

. The role of endothelial shear stress on haemodynamics, inflammation, coagulation and glycocalyx during sepsis. J Cell Mol Med 2020; 24: 12258–12271.

53.

Halbgebauer

, et al. Hemorrhagic shock drives glycocalyx, barrier and organ dysfunction early after polytrauma. J Crit Care 2018; 44: 229–237.

54.

Sieve

Münster-Kühnel

Hilfiker-Kleiner

. Regulation and function of endothelial glycocalyx layer in vascular diseases. Vascul Pharmacol 2018; 100: 26–33.

55.

Yamakawa

Murao

Aihara

. Recombinant human soluble thrombomodulin in sepsis-induced coagulopathy: an updated systematic review and meta-analysis. Thromb Haemost 2019; 119: 56–65.

56.

Kato

, et al. Efficacy and safety of recombinant human soluble thrombomodulin in patients with sepsis-induced disseminated intravascular coagulation – A meta-analysis. Thromb Res 2023; 226: 165–172.

57.

Williams

, et al. Sepsis-Induced coagulopathy: a comprehensive narrative review of pathophysiology, clinical presentation, diagnosis, and management strategies. Anesth Analg 2024; 138: 696–711.

58.

Mao

, et al. Effects of neutrophil extracellular traps in patients with septic coagulopathy and their interaction with autophagy. Front Immunol 2021; 12: 757041.

59.

Coloretti

, et al. Protein C in adult patients with sepsis: from pathophysiology to monitoring and supplementation. J Anesth Analg Crit Care 2025; 5: 21.

60.

Zincircioğlu

, et al. Diagnostic value of procalcitonin and C reactive protein for infection and sepsis in elderly patients. Turk J Med Sci 2021; 51: 2649–2656.

61.

, et al. Development and validation of a nomogram for predicting sepsis-induced coagulopathy in septic patients: mixed retrospective and prospective cohort study. Thromb Haemost 2025; 125: 108–119.

62.

Wang

Zhang

. Comparison of the prognostic value of four different critical illness scores in patients with sepsis-induced coagulopathy. Open Life Sci 2023; 18: 20220659.

63.

Zubair

, et al. Precision medicine and patient outcomes in intensive care unit: culture sensitivity, antibiotics, and APACHE score IV. J Educ Health Promot 2024; 13: 23.

64.

Zhu

, et al. Hierarchical capability in distinguishing severities of sepsis via serum lactate: a network meta-Analysis. Biomedicines 2024; 12: 447.

65.

Zhou

, et al. Fibrinogen-like protein 2 controls sepsis catabasis by interacting with resolvin Dp5. Sci Adv 2019; 5: eaax0629.

66.

Zhan

, et al. Fibrinogen-to-albumin ratio is associated with the prognosis of patients with septic acute kidney injury. Clin Kidney J 2025; 18: sfaf095.

67.

Wiedermann

, et al. Temporal decline in intravascular albumin mass and its association with fluid balance and mortality in sepsis: a prospective observational study. J Clin Med 2025; 14: 5255.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB