Abstract
Early identification of biomarkers that can predict the onset of sepsis-induced coagulopathy (SIC) in septic patients is clinically important. This study endeavors to examine the diagnostic and prognostic utility of serum C1q in the context of SIC. Clinical data from 279 patients diagnosed with sepsis at the Departments of Intensive Care, Respiratory Intensive Care, and Infectious Diseases at the Renmin Hospital of Wuhan University were gathered spanning from January 2022 to January 2024. These patients were categorized into two groups: the SIC group comprising 108 cases and the non-SIC group consisting of 171 cases, based on the presence of SIC. Within the SIC group, patients were further subdivided into a survival group (43 cases) and non-survival group (65 cases). The concentration of serum C1q in the SIC group was significantly lower than that in the non-SIC group. Furthermore, A significant correlation was observed between serum C1q levels and both SIC score and coagulation indices. C1q demonstrated superior diagnostic and prognostic performance for SIC patients, as indicated by a higher area under the curve (AUC). Notably, when combined with CRP, PCT, and SOFA score, C1q displayed the most robust diagnostic efficacy for SIC. Moreover, the combination of C1q with the SOFA score heightened predictive value concerning the 28-day mortality of SIC patients.
Keywords: sepsis, sepsis-induced coagulopathy, C1q, diagnosis, prognosis
Introduction
Sepsis-induced coagulopathy (SIC) is a life-threatening complication characterized by systemic activation of coagulation in sepsis, currently garnering widespread attention. 1 Literature findings indicate that between 50% and 70% of sepsis patients encounter coagulation dysfunction. End-stage SIC can potentially progress to disseminated intravascular coagulation (DIC), which significantly increases sepsis-related mortality.2–4 Extensive research on sepsis has clinically affirmed that the coagulation system undergoes further activation during sepsis onset, thereby fueling systemic inflammation. Concurrently, the activation of the secondary coagulation system is triggered by the release of numerous inflammatory factors, which creates a mutually reinforcing cycle that exacerbates the condition.5,6 Presently, there exists a dearth of biological markers for early prediction of coagulation disorders. Consequently, identifying biological indicators for early SIC prediction holds immense clinical significance.
Complement C1q serves as a crucial component of the complement system, exerting regulatory effects on various immune cells and playing a pivotal role in preserving autoimmune tolerance and modulating inflammatory responses. 7 Notably, a reciprocal cascade reaction exists between coagulation and complement systems. Complement activation enhances tissue factor activity, thereby initiating the exogenous clotting pathways and generating activated thrombin. Conversely, coagulation processes can stimulate complement factor activity. 8 The cascade reaction of coagulation commences with platelet activation. As a classical complement pathway pattern recognition molecule, C1q has been demonstrated to participate in platelet adhesion and aggregation. 9 Its involvement in coagulation occurs through inhibition of platelet adhesion and IgG adsorption, as well as by reducing the diffusion of adherent platelets. 10 Moreover, research has shown that C1q serves as a valuable marker for assessing the severity of sepsis and organ damage. 11
Presently, there is a lack of clinical research on the relationship between serum C1q levels and SIC. Thus, this study aims to assess the predictive and prognostic significance of C1q levels in SIC. The goal is to identify high-risk patients promptly and implement suitable intervention measures to enhance disease prognosis.
Patients and Methods
Study Population
Clinical data were collected from 279 sepsis patients treated at the Departments of Intensive Care Medicine, Respiratory Intensive Care, and Infectious Disease at Renmin Hospital of Wuhan University from January 2022 to January 2024. 108 patients were classified into the SIC group, while 171 patients were categorized into the non-SIC group. Exclusion criteria were applied, including (1) History of blood system disease, immune system disease, or malignant tumor. (2) Presence of other diseases causing thrombocytopenia, or use of medications resulting in platelet reduction. (3) Incomplete clinical records. (4) Pregnancy or lactation. (5) Death within 24 h of admission. This study received approval from the Clinical Research Ethics Committee of Renmin Hospital of Wuhan University (WDRY2020-K223) and obtained informed consent from patients or their families.
Definitions of serum C1q Level and SIC
According to the serum C1q reference range of 159.0-233.0 mg/L, patients were categorized into three groups: low C1q level (< 159.0 mg/L), normal C1q level (159.0-233.0 mg/L), and high C1q level (> 233.0 mg/L). 12 To diagnose SIC, the International Society on Thrombosis and Haemostasis (ISTH) provides a scoring system, 13 including platelet count (1 point:≥100 × 109/L to＜150 × 109/L; 2 points: < 100 × 109/L), PT-INR (1 point: > 1.2 to ≤ 1.4; 2 points:＞1.4), and SOFA score (1 point: 1; 2 points: ≥ 2). If the total score is ≥4 and the sum of PT-INR and platelet scores is greater than 2, the patient is diagnosed with SIC.
Data Collection
This study employed a retrospective approach to compile demographic and laboratory data based on patients’ initial examinations upon admission. Platelet (PLT), white blood cell (WBC) levels were determined using the Sysmex XN-9000 automatic blood cell analyzer. CRP and SAA were detected by H780-3 automatic specific protein analyzer produced by Shenzhen Xilaiheng Company. NT-proBNP levels were assessed utilizing the Siemens Dimension Exl200 analyzer, with a compatible kit. Procalcitonin (PCT) levels were measured employing the Cobas 801 automatic chemiluminescence immunoassay analyzer, with a corresponding kit from Roche. Coagulation parameters, including prothrombin time (PT), prothrombin time activated (PT-act), prothrombin time international normalized ratio (PT-INR), activated partial thromboplatin time (APTT), thrombin time (TT), fibrinogen (FIB), D-dimer, and antithrombin III (AT-III), were analyzed using the Sysmex CA7000 automated coagulation analyzer. Total bilirubin (TBIL), creatinine and C1q were detected by Siemens ADVIA 2400 biochemical analyzer and related reagents. Notably, complement C1q levels were determined via immunoturbidimetry, with no instances of missing values recorded in serum C1q measurements.
Statistical Analysis
The statistical analysis was conducted using SPSS 26.0 software, MedCalc 20.1, and GraphPad 9.5.0. Initially, the single-sample Kolmogorov-Smirnov test was utilized to evaluate data distribution. For normally distributed data, the independent sample t-test was applied, and results were presented as mean ± standard deviation. Non-normally distributed data were analyzed using nonparametric tests and expressed as median and quartile [M (Q1, Q3)]. Categorical variables were compared using the chi-square test and reported as percentages (%). Multicollinearity analysis was performed through univariate and multivariate logistic regression analysis to identify factors influencing SIC occurrence. A correlation coefficient > 0.7 indicated high correlation between variables, while a variance inflation factor (VIF) > 10 was indicative of multicollinearity. Receiver operating characteristic (ROC) curves were constructed to assess the diagnostic and 28-day prognostic value of C1q and other indicators in SIC patients. Statistical significance was defined as P < .05.
Results
Characteristics of Study Population
The characteristics of all patients are summarized in Table 1. There were no noticeable differences in age, sex, or underlying diseases between the SIC and non-SIC groups. However, the SIC group exhibited significantly elevated SOFA score and APACHE II score compared to the non-SIC group, with a concomitant decrease in GCS score (all P＜.05). Moreover, CRP, PCT, pro-BNP, TBIL, and creatinine levels were notably higher in the SIC group compared to the non-SIC group, while RBC and C1q levels were lower in the SIC group (all P＜.05). Additionally, within the SIC cohort, the survival group displayed higher SOFA score, APACHE II score, CRP, and PCT compared to the SIC death group, alongside lower WBC and C1q (all P＜.05). Further investigation into the primary infection site among the groups revealed no statistically significant differences.
Table 1.
Characteristics of the Study Population.
| Variables | Non-SIC （n = 171） |
SIC （n = 108） |
P | SIC组 | ||
|---|---|---|---|---|---|---|
| Survival (n = 43) | Non-survival (n = 65) | P | ||||
| Age, years, median (IQR) | 67.19 ± 13.15 | 66.94 ± 13.66 | 0.883 | 64.02 ± 13.48 | 68.88 ± 13.54 | .070 |
| Sex，male，n (%) | 108(63.16) | 65(60.19) | .618 | 23(53.49) | 42(64.62) | .248 |
| Medical history, n (%) | ||||||
| Diabetes | 57(33.33) | 25(23.15) | .069 | 10(23.26) | 15(23.08) | .983 |
| Hypertensive | 94(54.97) | 52(48.15) | .266 | 21(48.84) | 31(47.69) | .907 |
| Coronary heart disease | 15(8.77) | 14(12.96) | .264 | 6(13.95) | 8(12.31) | .803 |
| Cerebral ischemic stroke | 26(15.20) | 8(7.41) | .052 | 4(9.30) | 4(6.15) | .541 |
| Hyperlipoidemia | 10(5.85) | 5(4.63) | .660 | 1(2.33) | 4(6.15) | .354 |
| GCS score | 12.00(9.00,13.0) | 7.00(10.00,13.00) | ＜.001 | 10.00(4.50,12.00) | 7.00(3.00,10.00) | .056 |
| SOFA score | 7.00(5.00,9.00) | 10.00(7.00,13.00) | ＜.001 | 8.00(6.00,11.00) | 11.00(9.00,14.00) | .001 |
| APACHEII score | 14.00(12.00,19.00) | 18.00(14.00,24.25) | ＜.001 | 17.00(14.00,20.50) | 19.00(14.00,26.00) | .119 |
| Infection site, n (%) | ||||||
| Lung | 115(67.25) | 72(66.67) | .919 | 24(55.81) | 48(73.85) | .052 |
| Urinary tract | 31(18.13) | 11(10.19) | .071 | 6(13.95) | 5(7.69) | .292 |
| Intra-abdominal | 11(6.43) | 12(11.11) | .166 | 7(16.28) | 5(7.69) | .165 |
| Other | 14(8.19) | 13(12.04) | .289 | 6(13.95) | 7(10.77) | .619 |
| Smoking (%) | 26(15.20) | 8(7.41) | .052 | 2(4.65) | 6(9.23) | .374 |
| Alcohol drinking (%) | 20(11.70) | 8(7.41) | .246 | 3(6.98) | 5(7.69) | .889 |
| Laboratory results | ||||||
| WBC (109/L) | 11.67(8.55,16.03) | 9.87(6.57,19.38) | .285 | 13.65(9.27,20.19) | 8.69(5.86,16.29) | .018 |
| RBC (109/L) | 3.61 ± 0.85 | 3.24 ± 0.88 | ＜.001 | 3.24 ± 0.74 | 3.23 ± 0.97 | .962 |
| CRP (mg/L) | 88.34(43.07,136.49) | 132.75(79.86,187.11) | ＜0.001 | 118.27(61.89,172.32) | 145.70(105.76,214.60) | .018 |
| SAA (mg/L) | 300.00(230.01,300.00) | 300.00(149.42,300.00) | .187 | 300.00(111.47,300.00) | 300.00(199.79,300.00) | .476 |
| PCT (ng/mL) | 1.59(0.36,7.22) | 10.85(4.12,32.48) | ＜.001 | 7.43(2.69,13.48) | 16.33(5.32,54.70) | .011 |
| pro-BNP (pg/mL) | 1331.00(421.50,4502.00) | 5765.50(2133.75,14869.25) | ＜.001 | 7372.00(2357.00,19228.00) | 5010.00(2060.00,12992.00) | .244 |
| TBIL (µmol/L) | 17.00(9.69,40.44) | 22.27(13.98,43.38) | .027 | 20.70(14.17,36.50) | 26.70(13.20,48.80) | .478 |
| Creatinine (µmol/L) | 80.50(57.00,155.00) | 147.00(99.00,237.50) | ＜.001 | 146.00(92.50,242.50) | 148.50(104.50,231.50) | .952 |
| C1q (mg/L) | 178.30(150.40,213.15) | 131.80(108.40,154.18) | ＜.001 | 149.50(128.55,175.30) | 116.70(97.80,144.20) | ＜.001 |
Correlation Analysis of Serum C1q with Coagulation Indicators
According to the C1q concentration at ICU admission, the 279 patients were categorized into three groups: low C1q, normal C1q, and high C1q levels, comprising 140, 107, and 32 patients, respectively. Patients with low C1q demonstrated a more active clotting status compared to those with normal or high C1q levels. Specifically, they exhibited lower PLT, PT-act, and AT-III, along with higher PT, PT-INR, and APTT (all P＜.05) (Table 2 and Figure 1).
Table 2.
Coagulation Indicators for Each Serum C1q Level.
| Characteristics | Low C1q (N = 140) | Normal C1q (N = 107) | High C1q (N = 32) | P |
|---|---|---|---|---|
| PLT (109/L) | 94.00(68.25,160.25) | 159.00(103.50,238.00) | 186.00(120.75,257.25) | ＜.001 |
| PT (sec) | 15.15(13.60,17.98) | 12.90(11.83,15.28) | 12.35(10.78,14.35) | ＜.001 |
| PT-act (%) | 51.90(40.33,61.90) | 67.70(51.65,83.00) | 75.05(55.13,87.68) | ＜.001 |
| PT-INR | 1.31(1.81,1.57) | 1.12(1.02,1.32) | 1.07(0.93,1.25) | ＜.001 |
| APTT (sec) | 36.85(30.58,43.18) | 31.65(27.78,31.65) | 29.80(26.85,34.00) | ＜.001 |
| TT (sec) | 15.50(14.50,17.63) | 16.10(15.20,17.00) | 15.50(15.10,17.50) | .298 |
| FIB(g/L) | 4.27(2.63,6.32) | 4.89(3.44,6.61) | 5.03(3.05,6.77) | .072 |
| D-dimmer(mg/L) | 5.65(2.06,12.96) | 4.47(1.72,9.46) | 2.90(0.90,7.00) | .112 |
| AT-III (%) | 54.66 ± 21.65 | 67.85 ± 20.60 | 78.58 ± 20.20 | ＜.001 |
Figure 1.
Correlation between C1q levels and Coagulation indicators. (A) PLT; (B) PT; (C) PT-act; (D) PT-INR; (E) APTT; (F) TT; (G) FIB; (H) D-dimmer; (I)AT-III. *P < .05, compared with Low C1q; #P < .05, compared with Normal C1q.
Correlation Between C1q and SIC Score
We divided the patients into two groups based on their SIC score. The C1q levels for SIC score ≤ 4 and SIC score > 4 were 178.20 mg/L (150.3 mg/L, 213.15 mg/L) and 132.2 mg/L (108.4 mg/L, 155.4 mg/L), respectively. As the SIC score increased, a decreasing trend in C1q levels was observed, and the statistical analysis revealed a significant difference between the two groups (P < .05, Figure 2A). Furthermore, Pearson correlation analysis revealed a negative correlation between C1q levels and SIC score, with a correlation coefficient of −0.354 (P < .001, Figure 2B).
Figure 2.
Correlation between C1q level and SIC. (A) C1q levels across different SIC rating groups are presented using the IQR to describe the data. (B) A scatter plot illustrating the Pearson correlation between C1q and SIC score. *P＜.05.
Diagnosis of SIC Using C1q and Other Indicators
In the univariate logistic regression analysis, GCS, SOFA, APACHE II, CRP, PCT, pro-BNP, and C1q were identified as being correlated with the occurrence of SIC. Subsequently, multicollinearity analysis was conducted on these variables, revealing correlations below 0.5, tolerance values above 0.1, and VIF values below 2, indicating no collinearity among the variables (Table 3, Figure 3). Further multivariate logistic regression analysis identified CRP, PCT, SOFA, and C1q as independent risk factors for diagnosing SIC (Table 4). To further explore the diagnostic value of C1q for SIC, two diagnostic models were established: Model 1 (comprising SOFA, CRP, and PCT) and Model 2 (including Model 1 variables plus C1q). The results revealed that the AUC values for CRP, PCT, SOFA, C1q, Model 1, and Model 2 were 0.792, 0.641, 0.714, 0.788, 0.814, and 0.864, respectively (Figure 4, Table 5). According to the ROC comparison results, Model 1 showed superior performance compared to single diagnoses using other indicators, while Model 2 exhibited even greater diagnostic efficacy than Model 1 and other indicators. (Table 6) (all P＜.05).
Table 3.
The Tolerance and VIF for the Logistic Regression Analysis.
| Characteristics | Tolerance | VIF |
|---|---|---|
| GCS | 0.652 | 1.534 |
| SOFA | 0.734 | 1.362 |
| APACHEII | 0.694 | 1.440 |
| CRP | 0.950 | 1.052 |
| PCT | 0.826 | 1.211 |
| pro-BNP | 0.864 | 1.158 |
| C1q | 0.857 | 1.166 |
Figure 3.
Correlation of variables in logistic regression analysis.
Table 4.
Logistic Regression Analysis of SIC.
| Variables | Univariate Logistic regression | Multivariate Logistic regression | ||||||
|---|---|---|---|---|---|---|---|---|
| B | OR | 95%CI | P | B | OR | 95%CI | P | |
| GCS | −0.194 | 0.823 | 0.770∼0.881 | ＜.001 | −0.092 | 0.912 | 0.829∼1.003 | .059 |
| SOFA | 0.238 | 1.269 | 1.167∼1.379 | ＜.001 | 0.118 | 1.125 | 1.012∼1.251 | .029 |
| APACHEII | 0.090 | 1.094 | 1.051∼1.139 | ＜.001 | 0.054 | 1.055 | 0.991∼1.123 | .092 |
| WBC | 0.005 | 1.005 | 0.976∼1.035 | .746 | / | / | / | / |
| CRP | 0.007 | 1.007 | 1.004∼1.011 | ＜.001 | 0.009 | 1.009 | 1.004∼1.014 | ＜.001 |
| PCT | 0.056 | 1.058 | 1.035∼1.081 | ＜.001 | 0.054 | 1.056 | 1.030∼1.083 | ＜.001 |
| pro-BNP | 0.000 | 1.000 | 1.000∼1.000 | .003 | 0.000 | 1.000 | 1.000∼1.000 | .226 |
| TBIL | 0.003 | 1.003 | 0.995∼1.010 | .481 | / | / | / | / |
| Creatinine | 0.001 | 1.001 | 1.000∼1.003 | .084 | / | / | / | / |
| C1q | −0.023 | 0.978 | 0.971∼0.985 | ＜.001 | −0.021 | 0.979 | 0.972∼0.987 | ＜.001 |
Figure 4.
ROC curves for different parameters and Models in the diagnosis of SIC.
Table 5.
Diagnostic Value of Different Parameters and Models in Sepsis.
| Variable | AUC | 95%CI | Sensitivity | Specificity | Cut-off value | P | Youden index |
|---|---|---|---|---|---|---|---|
| PCT | 0.792 | 0.740∼0.838 | 0.611 | 0.667 | 114.28 (mg/L) | ＜.001 | 0.278 |
| CRP | 0.641 | 0.582∼0.697 | 0.972 | 0.480 | 1.208 (ng/mL) | ＜.001 | 0.452 |
| SOFA | 0.714 | 0.657∼0.766 | 0.648 | 0.678 | 8.5 points | ＜.001 | 0.326 |
| C1q | 0.788 | 0.735∼0.834 | 0.796 | 0.673 | 160.25 mg/L | ＜.001 | 0.469 |
| Model 1 | 0.814 | 0.763∼0.858 | 0.731 | 0.789 | - | ＜.001 | 0.520 |
| Model 2 | 0.864 | 0.819∼0.902 | 0.769 | 0.836 | - | ＜.001 | 0.605 |
Table 6.
Pairwise Comparisons of Different ROC Curves Between Different Parameters and Models.
| Pairwise comparison | DBA | SE | 95%CI | Z statistic | P |
|---|---|---|---|---|---|
| CRP∼PCT | 0.151 | 0.042 | 0.068∼0.234 | 3.750 | ＜.001 |
| CRP∼SOFA | 0.073 | 0.048 | −0.021∼0.166 | 1.522 | .128 |
| CRP∼C1q | 0.147 | 0.047 | 0.055∼0.238 | 3.150 | .002 |
| PCT∼SOFA | 0.079 | 0.036 | 0.008∼0.149 | 2.194 | .028 |
| PCT∼C1q | 0.005 | 0.041 | −0.075∼0.085 | 0.116 | .908 |
| SOFA∼C1q | 0.074 | 0.042 | −0.008∼0.156 | 1.779 | .075 |
| Model 1∼CRP | 0.173 | 0.0328 | 0.109∼0.237 | 5.277 | ＜.001 |
| Model 1∼PCT | 0.022 | 0.024 | −0.026∼0.069 | 0.884 | .377 |
| Model 1∼SOFA | 0.100 | 0.025 | 0.051∼0.149 | 3.993 | ＜.001 |
| Model1∼C1q | 0.026 | 0.040 | −0.052∼0.104 | 0.658 | .516 |
| Model 2∼CRP | 0.224 | 0.036 | 0.152∼0.295 | 6.147 | ＜.001 |
| Model 2∼PCT | 0.072 | 0.027 | 0.019∼0.125 | 2.658 | .008 |
| Model 2∼SOFA | 0.151 | 0.030 | 0.092∼0.210 | 5.015 | ＜.001 |
| Model 2∼C1q | 0.077 | 0.024 | 0.031∼0.123 | 3.265 | .001 |
| Model 2∼Model1 | 0.051 | 0.021 | 0.010∼0.092 | 2.433 | .015 |
Abbreviations: Model 1: SOFA + CRP + PCT; Model 2: Model1 +C1q.
The Predictive Value of C1q on the 28-Day Outcome of Patients with SIC
Multicollinearity analysis was performed on the variables SOFA, CRP, PCT, and C1q, all of which exhibited significant results in the univariate logistic regression analysis. The pairwise correlations for these variables were all less than 0.5, with tolerance values greater than 0.1 and VIF values less than 2. These findings indicate the absence of collinearity between the variables (Table 7 and Figure 5). Further multivariate regression analysis revealed that C1q and SOFA were independent risk factors for SIC death (Table 8). Regarding the ROC results, the AUC values for C1q, SOFA, and their combined AUC were 0.692, 0.705, and 0.736, respectively, with corresponding cutoff values of 10.5 points and 126.90 mg/L(all P＜.05) (Figure 6 and Table 9).
Table 7.
The Tolerance and VIF for the Logistic Regression Analysis.
| Characteristics | Tolerance | VIF |
|---|---|---|
| SOFA | 0.903 | 1.108 |
| CRP | 0.960 | 1.042 |
| PCT | 0.921 | 1.086 |
| C1q | 0.991 | 1.009 |
Figure 5.
Correlation of variables in logistic regression analysis.
Table 8.
Logistic Regression Analysis of SIC 28-day Mortality.
| Variables | Univariate logistic regression | Multivariate logistic regression | ||||||
|---|---|---|---|---|---|---|---|---|
| B | OR | 95%CI | P | B | OR | 95%CI | P | |
| SOFA | −0.228 | 0.796 | 0.695∼0.912 | 0.001 | −0.197 | 0.821 | 0.707∼0.954 | .010 |
| WBC | 0.024 | 1.024 | 0.985∼1.065 | 0.227 | / | / | / | / |
| CRP | 0.007 | 1.007 | 1.001∼1.013 | 0.022 | 0.006 | 1.006 | 0.999∼1.012 | .085 |
| PCT | −0.017 | 0.983 | 0.969∼0.997 | 0.016 | −0.013 | 0.987 | 0.972∼1.002 | .90 |
| C1q | 0.012 | 1.012 | 1.003∼1.021 | 0.010 | 0.013 | 1.013 | 1.003∼1.023 | .009 |
Figure 6.
ROC curves for C1q, SOFA and combination in the prediction of 28-day mortality.
Table 9.
Diagnostic Value of C1q, SOFA and Combination in the SIC 28-Day Mortality.
| Variable | AUC | 95%CI | Sensitivity | Specificity | Cut-off value | P | Youden index |
|---|---|---|---|---|---|---|---|
| SOFA | 0.692 | 0.593∼0.791 | 0.585 | 0.721 | 10.50 points | .001 | 0.306 |
| C1q | 0.705 | 0.606∼0.805 | 0.585 | 0.791 | 126.90 mg/L | ＜.001 | 0.376 |
| Combination | 0.736 | 0.643∼0.829 | 0.662 | 0.744 | - | ＜.001 | 0.406 |
Discussions
SIC affects multiple physiological processes, including coagulation, inflammation, host immunity, and endothelial cells.14,15 Predicting clinical outcomes in sepsis patients upon admission is challenging due to the complex pathophysiology of SIC patients and the variability in disease progression, which significantly increases the risk of mortality.16,17 This study analyzed 279 patients to assess the clinical significance of complement C1q in SIC. Our findings indicate that C1q levels were notably lower in the SIC group compared to the non-SIC group, with even lower levels observed in the deceased SIC subgroup. Furthermore, C1q exhibited associations with coagulation function and SIC score. The combined assessment of CRP, PCT, SOFA, and C1q demonstrated superior diagnostic efficacy for SIC, while combining C1q with SOFA score showed the highest prognostic value.
C1q is a multifunctional molecule that plays a pivotal role in enhancing phagocytosis, regulating cytokine production, and influencing T lymphocyte maturation. 18 In this study, serum C1q levels were notably diminished in patients with SIC. Firstly, downstream activation of the C1q complement and subsequent production of inflammatory mediators (such as C3a and C5a) can stimulate circulating neutrophils and endothelial cells, leading to up-regulated tissue factor (TF) expression. The upregulation of TF expression facilitates the activation of the hemostatic contact system, consequently increasing thrombosis. This mechanism may contribute to the development of SIC, highlighting the close association between C1q levels and the occurrence of SIC.19,20 Secondly, C1q can bind to lipopolysaccharide components of bacteria, thereby activating the complement system. In cases of immune dysfunction, extensive consumption of C1q by the complement system occurs in an effort to counteract inflammatory damage, ultimately leading to reduced C1q levels. 21 Furthermore, we investigated the correlation between C1q and coagulation parameters. The study findings revealed that low C1q levels were associated with coagulation dysfunction, characterized by decreased PLT count, reduced PT-act and AT-III, prolonged PT, elevated PT-INR, and APTT. Studies have indicated that the C1q and von Willebrand factor (vWF) complex can recruit human platelets, with C1q significantly inhibiting platelet adhesion and reducing the diffusion of adsorbed platelets.22,23 C1q contributes to coagulation by inhibiting platelet adhesion, IgG adsorption, and reducing the diffusion of adherent platelets. 24 Moreover, research indicates that C1q can directly influence blood coagulation. Bleeding experiments involving C1q-deficient mice and wild-type mice revealed that mice lacking C1q exhibited prolonged bleeding times and increased blood loss, underscoring the direct involvement of C1q in thrombosis during the coagulation process. 9 As a diagnostic criterion for SIC, the SIC score reflects the severity of SIC to some extent. Findings from this study demonstrate that a higher SIC score correlates with lower C1q levels, suggesting that C1q may partly assess the severity of SIC in patients.
CRP is an acute-phase reactant synthesized by liver cells in response to inflammation. 25 It activates the mononuclear phagocyte system, potentially binding to phospholipid components, as well as necrotic and apoptotic cells originating from various pathogens like bacteria, parasites, and fungi, thereby facilitating pathogen clearance. 26 This study identified independent risk factors for the occurrence of CRP and SIC. It has been observed that the combination of C1q and CRP enhances platelet adhesion to IgG compared to C1q alone. This phenomenon may result from CRP's regulatory effect on complement activation via its globular region interacting with C1q. 27 Thus, CRP may promote C1q involvement in regulating platelet adhesion and aggregation, thereby contributing to thrombosis formation.
PCT expression remains low in the absence of inflammation but undergoes significant elevation during severe systemic inflammation or bacterial infection, serving as a widely used biomarker for diagnosing, treating, and prognosticating sepsis. 28 Research indicates that inflammation and coagulation interact during infection, with PCT levels notably increased in patients with sepsis complicated by SIC. 29 Consistent with these findings, our study reveals significantly higher PCT levels in the SIC group compared to the non-SIC group. The SOFA, employed as a diagnostic criterion for defining sepsis and SIC, reflects the severity of sepsis. 30 Our study findings identify SOFA as an independent risk factor for SIC. Furthermore, serum C1q demonstrates potential clinical value in predicting the prognosis of SIC patients within 28 days of admission. Combining C1q with SOFA score further enhances prediction accuracy.
Limitations
The study was subjected to the common limitations of a retrospective study. Firstly, it is a small-sample study conducted at a single center, potentially introducing selection bias or unmeasured confounding factors due to missing data. Therefore, further validation of our findings is warranted with larger sample sizes. Additionally, future studies are needed to elucidate the specific mechanisms underlying the relationship between C1q levels and SIC.
Conclusions
To the best of our knowledge, this study represents the first investigation into the relationship between serum C1q levels and SIC. The findings underscore the diagnostic and prognostic significance of C1q in SIC, suggesting its potential integration with conventional laboratory and clinical parameters to develop a more precise diagnostic and prognostic model for sepsis.
Acknowledgments
We would like to express our gratitude to all participants in charge with data collection, laboratory measurement and statistical analysis All authors have agreed with the content of this study.
Footnotes
Authors Contributions: Ye zhang wrote the full text, prepared all graphs, collected data, statistics. Li Wang helped collect data, collect literature, and provide ideas. Xiandong Kuang helped with statistics. Dongling Tang: Resources, Project administration, Funding acquisition. Pingan Zhang revised the literature and provided funding.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Health Commission of Hubei Province, National Natural Science Foundation of China, (grant number No. WJ2023M073, No. 81773444).
ORCID iD: Ye Zhang https://orcid.org/0009-0009-7490-1905
References
- 1.Yu S, Ma X, Li X. Phenotype-oriented anticoagulant therapy for sepsis: Still a work in progress. Int J Hematol. 2022;116(1):48-54. [DOI] [PubMed] [Google Scholar]
- 2.Iba T, Umemura Y, Wada H, et al. Roles of coagulation abnormalities and microthrombosis in sepsis: Pathophysiology, diagnosis, and treatment. Arch Med Res. 2021;52(8):788-797. [DOI] [PubMed] [Google Scholar]
- 3.Martin JV, Liberati DM, Diebel LN. Excess sodium is deleterious on endothelial and glycocalyx barrier function: A microfluidic study. J Trauma Acute Care Surg. 2018;85(1):128-134. [DOI] [PubMed] [Google Scholar]
- 4.Iba T, Ito T, Maruyama I, et al. Potential diagnostic markers for disseminated intravascular coagulation of sepsis. Blood Rev. 2016;30(2):149-155. [DOI] [PubMed] [Google Scholar]
- 5.Li S, Yu S, Peng M, et al. Clinical features and development of sepsis in Klebsiella pneumoniae infected liver abscess patients: A retrospective analysis of 135 cases. BMC Infect Dis. 2021;21(1):597-599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kudo D, Goto T, Uchimido R, et al. Coagulation phenotypes in sepsis and effects of recombinant human thrombomodulin: An analysis of three multicentre observational studies. Crit Care. 2021;25(1):114-117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Coss SL, Zhou D, Chua GT, et al. The complement system and human autoimmune diseases. J Autoimmun. 2023;137:102979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yanase Y, Matsubara D, Takahagi S, et al. Basophil characteristics as a marker of the pathogenesis of chronic spontaneous Urticaria in relation to the coagulation and complement systems. Int J Mol Sci. 2023;24(12):10320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Donat C, Kölm R, Csorba K, et al. Complement C1q enhances primary hemostasis. Front Immunol. 2020;11:1522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ekdahl KN, Teramura Y, Hamad OA, et al. Dangerous liaisons: Complement, coagulation, and kallikrein/kinin cross-talk act as a linchpin in the events leading to thromboinflammation. Immunol Rev. 2016;274(1):245-269. [DOI] [PubMed] [Google Scholar]
- 11.Li H, Chen J, Hu Y, et al. Serum C1q levels have prognostic value for sepsis and are related to the severity of sepsis and organ damage. J Inflamm Res. 2021;14:4589-4600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Liu MC, Li JL, Wang YF, et al. Association between serum complements and kidney function in patients with diabetic kidney disease. Front Endocrinol (Lausanne). 2023;14:1195966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Iba T, Nisio MD, Levy JH, 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(9):e017046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ortega-Martorell S, Olier I, Johnston BW, et al. Sepsis-induced coagulopathy is associated with new episodes of atrial fibrillation in patients admitted to critical care in sinus rhythm. Front Med (Lausanne). 2023;10:1230854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Iba T, Levy JH, Warkentin TE, et al. Diagnosis and management of sepsis-induced coagulopathy and disseminated intravascular coagulation. J Thromb Haemost. 2019;17(11):1989-1994. [DOI] [PubMed] [Google Scholar]
- 16.Walborn A, Hoppensteadt D, Syed D, et al. Biomarker profile of sepsis-associated coagulopathy using biochip assay for inflammatory cytokines. Clin Appl Thromb Hemost. 2018;24(4):625-632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.de Bont CM, Boelens WC, Pruijn GJM. NETosis, complement, and coagulation: A triangular relationship. Cell Mol Immunol. 2019;16(1):19-27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Thielens NM, Tedesco F, Bohlson SS, et al. C1q: A fresh look upon an old molecule. Mol Immunol. 2017;89:73-83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Yanase Y, Takahagi S, Ozawa K, et al. The role of coagulation and complement factors for mast cell activation in the pathogenesis of chronic spontaneous Urticaria. Cells. 2021;10(7):1759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Miao D, Li DY, Chen M, et al. Platelets are activated in ANCA-associated vasculitis via thrombin-PARs pathway and can activate the alternative complement pathway. Arthritis Res Ther. 2017;201(1):252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pont S, Fraikin N, Caspar Y, et al. Bacterial behavior in human blood reveals complement evaders with some persister-like features. PLoS Pathog. 2020;16(12):e1008893. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kölm R, Schaller M, Lubka T, et al. Von willebrand factor interacts with surface-bound C1q and induces platelet rolling. J Immunol. 2016;197(9):3669-3679. [DOI] [PubMed] [Google Scholar]
- 23.Schulz K, Donat C, Punjabi M, et al. Complement C1q and von willebrand factor interaction in atherosclerosis of human carotid artery. Front Immunol. 2023;14:1265387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Freda CT, Yin W, Ghebrehiwet B, et al. Complement component C1q initiates extrinsic coagulation via the receptor for the globular head of C1q in adventitial fibroblasts and vascular smooth muscle cell. Immun Inflamm Dis. 2023;11(1):e769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Cantey JB, Bultmann CR. C-Reactive protein testing in late-onset neonatal sepsis: Hazardous waste. JAMA Pediatr. 2020;174(3):235-236. [DOI] [PubMed] [Google Scholar]
- 26.Ticinesi A, Lauretani F, Nouvenne A, et al. C-reactive protein (CRP) measurement in geriatric patients hospitalized for acute infection. Eur J Intern Med. 2017;37:7-12. [DOI] [PubMed] [Google Scholar]
- 27.Roumenina LT, Ruseva MM, Zlatarova A, et al. Interaction of C1q with IgG1, C-reactive protein and pentraxin 3: Mutational studies using recombinant globular head modules of human C1q A, B, and C chains. Biochemistry. 2006;45(13):4093-4104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Fu G, Zhan HC, Li HL, et al. Association between procalcitonin and acute kidney injury in patients with bacterial septic shock. Blood Purif. 2021;50(6):790-799. [DOI] [PubMed] [Google Scholar]
- 29.Long X, Zhang Z, Zou W, et al. Coagulopathy of patients with COVID-19 is associated with infectious and inflammatory markers. Risk Manag Healthc Policy. 2020;13:1965-1975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Li J, Liu H, Wang N, et al. Persistent high sepsis-induced coagulopathy and sequential organ failure assessment scores can predict the 28-day mortality of patients with sepsis: A prospective study. BMC Infect Dis. 2024;24(1):282. [DOI] [PMC free article] [PubMed] [Google Scholar]