Abstract
Background
This research focused on exploring the association between coagulopathy scores and the survival outcomes, both short-term and long-term, in individuals diagnosed with liver cirrhosis complicated by sepsis.
Methods
This study retrospectively analyzed data from individuals with liver cirrhosis and sepsis who were admitted to the intensive care unit (ICU) at Beth Israel Deaconess Medical Center between 2008 and 2022. The main outcome of interest was all-cause mortality within 28 days post-admission, while the secondary outcome assessed mortality within 90 days. We used the Kaplan-Meier analysis to compare the mortality risk among the different groups. To evaluate the relationship between coagulopathy score and mortality risk in patients with liver cirrhosis and sepsis, a multivariate Cox proportional hazards regression analysis was performed. The predictive performance of the coagulopathy score for short- and long-term all-cause mortality was assessed using receiver operating characteristic (ROC) curve analysis, which included evaluation of its sensitivity, specificity, and area under the curve. Subgroup analyses were performed to evaluate the relationship between coagulopathy score and survival across different groups.
Results
The study included a total of 2,278 patients. Kaplan-Meier survival analysis demonstrated that individuals with elevated coagulopathy scores exhibited markedly higher rates of ICU mortality, in-hospital mortality, as well as 28-day and 90-day mortality, with all log-rank tests yielding P-values of less than 0.001. The results of the multivariate Cox regression analysis showed that an elevated coagulopathy score was independently linked to higher 28-day and 90-day all-cause mortality, both before and after controlling for potential confounders. ROC curve analysis showed that although the coagulopathy score was slightly less predictive of prognosis than the Model for End-stage Liver Disease score, it significantly outperformed the Sequential Organ Failure Assessment score and the Sepsis-induced Coagulopathy score. Subgroup analysis revealed no significant interaction between the coagulopathy score and survival across the different subgroups.
Conclusions
Higher coagulopathy scores in critically ill patients with liver cirrhosis and sepsis were independently associated with poor prognosis. Due to its simplicity and potential predictive value, the coagulopathy score can serve as an effective complement to existing clinical tools for managing critically ill patients with liver cirrhosis and sepsis.
Keywords: Liver cirrhosis, Sepsis, Coagulopathy score, Intensive care unit, Prognosis
Background
Sepsis is a clinical syndrome caused by dysregulated host immune and inflammatory responses to infection, leading to life-threatening organ dysfunction [1–3]. Sepsis is characterized by local inflammation triggered by pathogens and involves systemic inflammatory responses, immune imbalances, and metabolic disturbances. It is a severe manifestation of infectious diseases and remains one of the most common and fatal conditions in intensive care units (ICUs) [1, 4]. Liver cirrhosis is prevalent. Due to impaired liver synthetic function leading to immune protein deficiency, splenomegaly-induced leukopenia, and intestinal barrier dysfunction caused by portal hypertension, patients with liver cirrhosis often exhibit acquired immune deficiencies. Consequently, they are highly susceptible to infections and prone to developing sepsis [5–7]. Previous studies have shown that hospitalized patients with liver cirrhosis are four to five times more likely to develop infections than the general population [8], and those who progress to sepsis have poorer outcomes and higher mortality rates. Therefore, identifying reliable prognostic indicators for patients with liver cirrhosis and sepsis is crucial for early and accurate detection of disease progression, enabling timely intervention.
Patients with liver cirrhosis often experience coagulation dysfunction due to impaired liver function, which affects the synthesis of coagulation factors (such as factors II, V, VII, IX, X, and XI) [9–11]. In addition, patients with liver cirrhosis often experience thrombocytopenia and coagulation factor depletion, making them more susceptible to coagulation abnormalities [12, 13]. Sepsis is commonly accompanied by widespread coagulation system activation due to a systemic inflammatory response, leading to disseminated intravascular coagulation (DIC) [14, 15]. During sepsis, excessive activation of endogenous coagulation factors results in microthrombus formation and simultaneous depletion of coagulation factors, ultimately leading to bleeding [14, 16]. Therefore, in patients with liver cirrhosis, the occurrence of sepsis complicates the coagulation state because of preexisting coagulation factor deficiencies. Studies have confirmed that coagulation dysfunction in patients with liver cirrhosis and sepsis is related to disease severity and prognosis [17].
The coagulopathy score is a scoring system derived from a series of coagulation parameters such as the international normalized ratio (INR), activated partial thromboplastin time (APTT), and platelet count (PLT), which are used to assess coagulation dysfunction in patients [18]. Previous studies have revealed that the coagulopathy score is related to an increased mortality risk in patients with various critical conditions, such as acute intracerebral hemorrhage [19], severe acute respiratory distress syndrome [20], and acute respiratory failure [21]. However, the relationship between the coagulopathy score and prognosis in critically ill patients with liver cirrhosis and sepsis has not been thoroughly explored. Considering the pathophysiological interactions between liver cirrhosis and sepsis, we proposed that the coagulopathy score could function as a valuable biomarker for predicting prognosis in this patient population.
Therefore, this study retrospectively analyzed the data of critically ill patients with liver cirrhosis and sepsis to investigate the relationship between coagulopathy score and prognosis, to aid early clinical identification and improving outcomes.
Methods
Data source and study population
This study performed a retrospective analysis utilizing data extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database, version 3.0, targeting patients diagnosed with liver cirrhosis and sepsis. MIMIC-IV (v3.0), a publicly accessible dataset curated by the Laboratory for Computational Physiology at the Massachusetts Institute of Technology, includes medical records of patients admitted to Beth Israel Deaconess Medical Center between 2008 and 2022 [22]. The research team successfully completed the Collaborative Institutional Training Initiative (CITI) program, including examinations on “Conflict of Interest” and “Research Use of Data or Specimens,” thereby gaining authorization to access the MIMIC-IV (version 3.0) database.
Using the International Classification of Diseases, 9th Revision and 10th Revision codes, we extracted the inpatient data for all patients with liver cirrhosis and sepsis. Notably, based on the Sepsis 3.0 criteria, individuals with a suspected infection and a Sequential Organ Failure Assessment (SOFA) score of 2 or higher were classified as having sepsis [23]. In this study, we implemented strict exclusion criteria to ensure robust results, including [1] patients under 18 years of age at the time of admission [2], patients with an ICU stay of less than 24 h, and [3] patients without recorded INR, APTT, or PLT values within 24 h of admission. Furthermore, for patients with multiple ICU admissions, only data from their initial hospitalization were included in the analysis. Ultimately, this study encompassed a total of 2,278 patients (Fig. 1).
Fig. 1.
Flowchart for participants
Coagulopathy score calculation, patient grouping, and clinical outcomes
The coagulopathy score is determined by summing the scores from three components: APTT score (0 point: APTT < 29 s; 1 point: 29 s to 34 s; 2 points: APTT > 34 s), INR score (0 point: INR < 1.4; 1 point: 1.4 to 2.6; 2 points: INR > 2.6), and platelet count score (0 point: PLT > 150 × 10⁹/L; 1 point: 100 to 150 × 10⁹/L; 2 points: PLT < 100 × 10⁹/L).
In this study, patients with a coagulopathy score of 0–2 were classified into Group 1 (low coagulopathy score group), those with a score of 3–4 into Group 2 (middle coagulopathy score group), and those with a score of 5–6 into Group 3 (high coagulopathy score group).
The main endpoint of this study was all-cause mortality within 28 days of admission, while the secondary endpoint focused on mortality within 90 days.
Data extraction
Data extraction was performed using the PostgreSQL software (version 13.7.2) and Navicat Premium software (version 16). Structured Query Language was used for data extraction. Table 1 presents all covariates extracted for this study, which primarily included demographic characteristics, vital signs, comorbidities, etiology of liver cirrhosis, laboratory test results, clinical treatments, survival information, microbiological culture results, and severity scores. Notably, blood infection was defined based on the diagnostic criteria for bacteremia outlined in the International Classification of Diseases codes.
Table 1.
Covariates extracted in detail from MIMIC-IV (v 3.0)
| Items | Composition |
|---|---|
| Demographic variables | Age, Gender, Ethnicity |
| Comorbidities | Hypertension, Diabetes mellitus, Chronic obstructive pulmonary disease, Myocardial infarction, Ascites, Hepatorenal syndrome, Malignancy, Heart failure, Atrial fibrillation, Hepatic encephalopathy, Esophagealvarices with hemorrhage, Hepatic failure, Septic shock |
| Vital Signs | Heart rate, Systolic blood pressure, Diastolic blood pressure, Mean arterial pressure, Respiratory rate, SPO2, Body temperature |
| Laboratory parameters | Neutrophil cells, Lymphocyte cells, Red blood cells, White blood cells, Erythrocyte distribution width, Platelet, Hemoglobin, Lymphocyte percentage, Hematocrit, Creatinine, Blood urea nitrogen, Albumin, Total bilirubin, Direct bilirubin, Aspartate aminotransferase, Alanine aminotransferase, Glucose, Triglyceride, Total cholesterol, High density lipoprotein cholesterol, Low density lipoprotein cholesterol, Prothrombin time, International normalized ratio, Potassium, Sodium, Calcium, Anion gap, Lactate, PH, FiO2, PCO2, PaO2 |
| Clinical Treatments | Urinary catheter, Vasopressin, Ventilation, Continuous Renal Replacement Therapy, Norepinephrine |
| Clinical Outcomes | Length of ICU stay, Length of hospitalization, ICU mortality, Hospital mortality, 28-day all-cause mortality, 90-day all-cause mortality |
| Scores | Sequential Organ Failure Assessment, Model for End-Stage Liver Disease score |
| Others | Causes of liver cirrhosis, Infections positions, Microculture |
MIMIC-IV (v 3.0): Medical Information Mart for Intensive Care-IV database (version 3.0); ICU, intensive care unit
Management of abnormal and missing values
We handled abnormal values using the Winsorization method with the winsor2 command in STATA, applying 1% and 99% cutoff values. We used multiple imputation methods to address the missing values, excluding variables with missing data exceeding 15%.
Ethical statement
This study used publicly available data from the MIMIC-IV database (version 3.0). Because the dataset was de-identified, no additional Institutional Review Board approval was required for this study. All procedures were conducted following the relevant guidelines and regulations to protect patient privacy and confidentiality.
Statistical analysis
For normally distributed data, continuous variables were expressed as mean ± standard deviation, while skewed data were presented as medians (interquartile range). We compared continuous variables using t-tests or one-way analysis of variance. Categorical variables were expressed as frequencies (percentages) and analyzed using either the chi-square test or Fisher’s exact test for comparisons. Kaplan-Meier survival analysis was used to evaluate and compare the distribution and differences in 28-day and 90-day all-cause mortality between the different groups of patients with liver cirrhosis and sepsis. Cox proportional hazards models were utilized to quantify the association between the coagulopathy score and patient prognosis, yielding hazard ratios (HR) along with 95% confidence intervals (CI). Three models were used to adjust for potential confounders: Model 1 (baseline model, no adjustment for covariates), Model 2 (adjusted for age, sex, and ethnicity), and Model 3 (further adjusted for serum creatinine, white blood cell count, hypertension, heart failure, respiratory failure (RF), diabetes, vasopressor use, continuous renal replacement therapy (CRRT), and SOFA. Group 1 served as the reference category, and adjusted hazard ratios (HRs) for 28-day and 90-day all-cause mortality were calculated for the remaining groups, with comparisons made relative to the reference. The predictive performance of the coagulopathy score for 28-day and 90-day all-cause mortality was evaluated using receiver operating characteristic (ROC) analysis, with the area under the curve (AUC) computed to quantify its accuracy. Subgroup analyses were conducted to investigate the consistency of the prognostic value of the coagulopathy score across different subgroups stratified by age, sex, hypertension, diabetes, and RF. All statistical analyses were conducted using a two-tailed approach, with significance defined as P < 0.05. Data were processed and analyzed with R software (version 4.2.2), STATA (version 16.0), and IBM SPSS (version 22.0).
Results
Comparison of baseline characteristics in patients with liver cirrhosis and Sepsis
A total of 2,278 patients with liver cirrhosis and sepsis were included in this study (Fig. 1). The primary causes of liver cirrhosis were alcohol consumption and viral hepatitis. The most common sites of infection were the lungs, abdomen, and urinary system. Patients with higher coagulopathy scores had a higher prevalence of hypertension upon admission. However, during hospitalization, due to the effects of disease, patients with higher coagulopathy scores were more likely to experience lower systolic and diastolic blood pressures, mean arterial pressure, and body temperature. Significant variations were identified among the three groups in several parameters, including neutrophil count, red blood cell count, white blood cell count, red cell distribution width, PLT, hemoglobin, hematocrit, creatinine, blood urea nitrogen, total bilirubin, direct bilirubin, alanine aminotransferase, aspartate aminotransferase, glucose, triglycerides, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, prothrombin time, INR, sodium, calcium, anion gap, lactate, and partial pressure of carbon dioxide, with all P-values less than 0.05. Furthermore, patients with higher coagulopathy scores had significantly longer ICU stays and total hospitalizations than those with lower coagulopathy scores. Additional details are provided in Table 2.
Table 2.
Baseline characteristics of patients with cirrhosis and sepsis
| Variable | Overall (n = 2278) | Group 1 (n = 547) | Group 2 (n = 967) | Group 3 (n = 764) | P value |
|---|---|---|---|---|---|
| Demographics | |||||
| Age, years | 58 (50–65) | 61 (55–69) | 58 (50–66) | 55 (49–62) | < 0.001 |
| Men, n (%) | 1471 (64.57) | 351 (64.17) | 638 (65.98) | 482 (63.09) | 0.45 |
| Ethnicity, n (%) | |||||
| Asian populations | 57 (2.50) | 18 (3.29) | 21 (2.17) | 18 (2.36) | 0.03 |
| White populations | 1467 (64.40) | 367 (67.09) | 643 (66.49) | 457 (59.82) | |
| Black populations | 156 (6.85) | 36 (6.58) | 61 (6.31) | 59 (7.72) | |
| Others | 598 (26.25) | 126 (23.03) | 242 (25.03) | 230 (30.10) | |
| Causes of liver cirrhosis, n (%) | |||||
| Alcohol, n (%) | 1002 (43.99) | 197 (36.01) | 442 (45.71) | 363 (47.51) | < 0.001 |
| Viral, n (%) | 636 (27.92) | 134 (24.50) | 256 (26.47) | 246 (32.20) | |
| Biliary, n (%) | 46 (2.02) | 21 (3.84) | 15 (1.55) | 10 (1.31) | |
| Others, n (%) | 594 (26.08) | 195 (35.65) | 254 (26.27) | 145 (18.98) | |
| Comorbidities | |||||
| Hypertension, n (%) | 1508 (66.20) | 340 (62.16) | 630 (65.15) | 538 (70.42) | 0.005 |
| Diabetes mellitus, n (%) | 707 (31.04) | 213 (38.94) | 278 (28.75) | 216 (28.27) | < 0.001 |
| Chronic obstructive pulmonary disease, n (%) | 145 (6.37) | 42 (7.68) | 65 (6.72) | 38 (4.97) | 0.12 |
| Atrial fibrillation, n (%) | 441 (19.36) | 115 (21.02) | 206 (21.30) | 120 (15.71) | 0.007 |
| Malignancy, n (%) | 433 (19.01) | 132 (24.13) | 181 (18.72) | 120 (15.71) | < 0.001 |
| Respiratory failure, n (%) | 1019 (44.73) | 220 (40.22) | 437 (45.19) | 362 (47.38) | 0.03 |
| Decompensated events at admission | |||||
| Ascites, n (%) | 1179 (51.76) | 194 (35.47) | 495 (51.19) | 490 (64.14) | < 0.001 |
| Hepatorenal syndrome, n (%) | 363 (15.94) | 23 (4.20) | 148 (15.31) | 192 (25.13) | < 0.001 |
| Hepatic encephalopathy, n (%) | 244 (10.71) | 45 (8.23) | 98 (10.13) | 101 (13.22) | 0.01 |
| Hepatic failure, n (%) | 681 (29.89) | 84 (15.36) | 285 (29.47) | 312 (40.84) | < 0.001 |
| Septic shock, n (%) | 664 (29.15) | 124 (22.67) | 269 (27.82) | 271 (35.47) | < 0.001 |
| Bleeding events | |||||
| Esophagealvarices with hemorrhage, n (%) | 296 (12.99) | 71 (12.98) | 129 (13.34) | 96 (12.57) | 0.89 |
| Acute gastric ulcer with hemorrhage, n (%) | 5 (0.22) | 1 (0.18) | 2 (0.21) | 2 (0.26) | 0.95 |
| Acute gastric mucosal injury with hemorrhage, n (%) | 154 (6.76) | 33 (6.03) | 62 (6.41) | 59 (7.72) | 0.41 |
| Infections position | |||||
| Blood, n (%) | 89 (3.91) | 15 (2.74) | 40 (4.14) | 34 (4.45) | 0.26 |
| Lung, n (%) | 668 (29.32) | 153 (27.97) | 290 (29.99) | 225 (29.45) | 0.71 |
| Abdomen, n (%) | 410 (18.00) | 58 (10.60) | 180 (18.61) | 172 (22.51) | < 0.001 |
| Urinary, n (%) | 383 (16.81) | 78 (14.26) | 163 (16.86) | 142 (18.59) | 0.12 |
| Skin, n (%) | 99 (4.35) | 28 (5.12) | 34 (3.52) | 37 (4.84) | 0.24 |
| Microculture | |||||
| Fungal, n (%) | 388 (17.03) | 75 (13.71) | 166 (17.17) | 147 (19.24) | 0.03 |
| Bacterial, n (%) | 1166 (51.19) | 254 (46.44) | 504 (52.12) | 408 (53.40) | 0.03 |
| Bacterial and fungal, n (%) | 246 (10.80) | 52 (9.51) | 108 (11.17) | 86 (11.26) | 0.53 |
| Vital sign | |||||
| Heart rate, beats/min | 92.5 (79–106) | 90 (77–105) | 94 (80–107) | 93 (80–107) | 0.01 |
| Systolic blood pressure, mmHg | 114 (100–131) | 118 (102–134) | 114.5 (101–132) | 111 (98–128) | < 0.001 |
| Diastolic blood pressure, mmHg | 63 (54–75) | 66 (56–77) | 63 (54–75) | 62 (52–72) | < 0.001 |
| Mean arterial pressure, mmHg | 80.67 (70.67–92.67) | 83.33 (73-95.33) | 81.67 (70.67–93.33) | 78.33 (69-89.67) | < 0.001 |
| Respiratory rate, times/min | 19 (16–23) | 19 (16–23) | 19 (16–23) | 19 (16–23) | 0.45 |
| SPO2, % | 98 (95–100) | 98 (95–100) | 98 (95–100) | 98 (95–100) | 0.23 |
| Body temperature, ℃ | 36.72 (36.44–37.06) | 36.78 (36.5-37.11) | 36.78 (36.44–37.11) | 36.7 (36.44-37) | < 0.001 |
| Laboratory parameters | |||||
| Neutrophil cells, 109/L | 5.5 (3.36–9.66) | 5.79 (3.61–10.45) | 5.77 (3.47–9.79) | 5.04 (2.87–9.03) | 0.002 |
| Lymphocyte cells, 109/L | 1.1 (0.69–1.69) | 1.13 (0.7–1.75) | 1.12 (0.69–1.72) | 1.06 (0.69–1.59) | 0.40 |
| Red blood cells, 109/L | 10.2 (6.5–15.7) | 11.3 (7.2–16.5) | 10.7 (6.8–16.8) | 9.1 (5.8–13.6) | < 0.001 |
| White blood cells, 109/L | 2.97 (2.53–3.48) | 3.25 (2.79–3.78) | 3.01 (2.6–3.48) | 2.71 (2.34–3.2) | < 0.001 |
| Erythrocyte distribution width, % | 16.9 (15.2–19) | 15.9 (14.5–18) | 16.8 (15.4–18.9) | 17.7 (15.85–19.9) | < 0.001 |
| Platelets, 109/L | 101 (65–154) | 162 (122–227) | 116 (82–156) | 64.5 (46–82) | < 0.001 |
| Hemoglobin, g/L | 9.4 (8-10.8) | 9.9 (8.4–11.3) | 9.5 (8.1–10.8) | 8.8 (7.7–10.3) | < 0.001 |
| Lymphocyte percentage, % | 9.8 (5.7-16.55) | 10 (5.7–17.5) | 9.7 (5.5–16) | 9.8 (6-16.8) | 0.72 |
| Hematocrit, % | 28 (24.3–32.6) | 29.6 (26.1–34.3) | 28.3 (24.7–32.6) | 26.5 (23.1–30.7) | < 0.001 |
| Creatinine, mg/dL | 1.3 (0.8–2.2) | 1.1 (0.8–1.8) | 1.2 (0.8–2.3) | 1.5 (0.9–2.7) | < 0.001 |
| Blood urea nitrogen, mg/dL | 27 (16–48) | 24 (16–42) | 26 (16–48) | 31 (17–51) | 0.001 |
| Albumin, g/dL | 2.9 (2.5–3.3) | 2.9 (2.5–3.3) | 2.9 (2.5–3.3) | 2.9 (2.4–3.4) | 0.52 |
| Total bilirubin, mg/dL | 3.2 (1.4–7.9) | 1.25 (0.7–2.4) | 3 (1.5–6.8) | 6.4 (3.1–14.3) | < 0.001 |
| Direct bilirubin, mg/dL | 1.7 (0.7–4.6) | 0.8 (0.4–1.7) | 1.8 (0.7–4.7) | 2.6 (1-6.1) | < 0.001 |
| Aspartate aminotransferase, U/L | 75 (41–172) | 50 (30–104) | 79 (43–178) | 91.5 (49–238) | < 0.001 |
| Alanine aminotransferase, U/L | 35 (21–79) | 29 (18–60) | 35 (21–85) | 39 (23–94) | < 0.001 |
| Glucose, mg/dL | 126 (102–166) | 130 (109–174) | 126 (103–163) | 122 (96–164) | < 0.001 |
| Triglyceride, mg/dL | 112 (78–172) | 116 (81–201) | 121 (81-176.5) | 101 (74–150) | 0.006 |
| Total cholesterol, mg/dL | 136 (99–173) | 148 (114–185) | 140 (104–177) | 124 (84–163) | < 0.001 |
| High density lipoprotein cholesterol, mg/dL | 40 (26–55) | 44 (31–57) | 41 (25–57) | 38 (22–53) | < 0.001 |
| Low density lipoprotein cholesterol, mg/dL | 74 (50–101) | 79.5 (52-105.5) | 75 (52–103) | 69 (45–97) | 0.005 |
| Prothrombin time, s | 18.7 (15.5–23.8) | 14.6 (13.2–15.8) | 18.5 (16.1–22.1) | 23.8 (19.95–30.9) | < 0.001 |
| International normalized ratio | 1.7 (1.4–2.2) | 1.3 (1.2–1.4) | 1.7 (1.5-2) | 2.2 (1.8–2.9) | < 0.001 |
| Potassium, mmol/L | 4.2 (3.7–4.8) | 4.2 (3.8–4.8) | 4.2 (3.7–4.8) | 4.1 (3.7–4.7) | 0.14 |
| Sodium, mmol/L | 137 (133–140) | 138 (134–141) | 137 (133–140) | 136 (131–140) | < 0.001 |
| Calcium, mg/dL | 8.2 (7.6–8.8) | 8.2 (7.7–8.7) | 8.2 (7.6–8.8) | 8.3 (7.6–8.9) | 0.02 |
| Anion gap, mmol/L | 15 (12–19) | 14 (12–18) | 15 (12–19) | 16 (12–20) | < 0.001 |
| Lactate, mmol/L | 2.4 (1.6-4) | 1.9 (1.3–2.9) | 2.4 (1.6–3.7) | 2.9 (2-4.6) | < 0.001 |
| PH | 7.36 (7.3–7.42) | 7.36 (7.3–7.42) | 7.37 (7.29–7.42) | 7.36 (7.3–7.42) | 0.86 |
| FiO2 | 50 (49–100) | 50 (40–80) | 50 (50–100) | 50 (48.5–100) | 0.76 |
| PCO2 | 38 (33–44) | 40 (34–45) | 39 (33–45) | 37 (32–43) | 0.001 |
| PaO2 | 83 (49–155) | 85 (46–147) | 87 (51–166) | 78 (49–150) | 0.11 |
| Scores | |||||
| SOFA | 3 (1–6) | 1 (0–4) | 3 (1–6) | 5 (1–8) | < 0.001 |
| MELD | 23 (16–31) | 15 (11–20) | 23 (16–29) | 30 (24–38) | < 0.001 |
| Treatment | |||||
| Urinary catheter, n (%) | 600 (26.34) | 119 (21.76) | 264 (27.30) | 217 (28.40) | 0.02 |
| Vasopressin, n (%) | 445 (19.53) | 63 (11.52) | 186 (19.23) | 196 (25.65) | < 0.001 |
| Ventilation, n (%) | 1990 (87.36) | 474 (86.65) | 846 (87.49) | 670 (87.70) | 0.84 |
| Continuous Renal Replacement Therapy, n (%) | 404 (17.73) | 54 (9.87) | 158 (16.34) | 192 (25.13) | < 0.001 |
| Norepinephrine, n (%) | 912 (40.04) | 185 (33.82) | 371 (38.37) | 356 (46.60) | < 0.001 |
| Clinical Outcomes | |||||
| LOS ICU, day | 3.68 (2.02–7.49) | 3.33 (1.95–6.8) | 3.69 (2.06–7.64) | 3.89 (2.03–7.94) | 0.01 |
| LOS Hospital, day | 11.73 (6.13–21.42) | 10.21 (6-18.33) | 11.88 (6.17–21.33) | 12.15 (6.35–23.94) | 0.03 |
| ICU mortality, n (%) | 433 (19.01) | 55 (10.05) | 164 (16.96) | 214 (28.01) | < 0.001 |
| Hospital mortality, n (%) | 654 (28.71) | 83 (15.17) | 267 (27.61) | 304 (39.79) | < 0.001 |
| 28-day mortality, n (%) | 703 (30.86) | 97 (17.73) | 281 (29.06) | 325 (42.54) | < 0.001 |
| 90-day mortality, n (%) | 921 (40.43) | 137 (25.05) | 379 (39.19) | 405 (53.01) | < 0.001 |
SOFA, Sequential Organ Failure Assessment; MELD, Model for End-Stage Liver Disease score; ICU, intensive care unit
Kaplan-Meier survival curves for the three groups of patients with liver cirrhosis and Sepsis
In this study, 703 patients succumbed within 28 days, while 921 deaths occurred within 90 days. Kaplan-Meier survival curves demonstrated that patients with higher coagulopathy scores had significantly higher 28-day and 90-day all-cause mortality rates than those with lower coagulopathy scores (Fig. 2, all log-rank P < 0.0001).
Fig. 2.
Kaplan–Meier survival curves showing cumulative incidence of 28-day (A) and 90-day (B) all-cause mortality, stratified by coagulopathy score categories
Relationship between different levels of coagulopathy score and clinical outcomes
We constructed multivariate Cox regression models to evaluate the association between coagulopathy score and 28-day and 90-day all-cause mortality in patients with cirrhosis and sepsis (Table 3). The findings confirmed that an increase in coagulopathy score was closely associated with a higher risk of 28-day all-cause mortality in patients with cirrhosis and sepsis. For instance, in the unadjusted baseline model (Model 1), compared to Group 1, the HRs (95% CI) for Group 2 and Group 3 were 1.78 (95% CI: 1.42–2.25) and 2.88 (95% CI: 2.30–3.62), respectively. In Model 2, after adjusting for age, gender, and ethnicity, the adjusted HRs (95% CI) for Group 2 and Group 3 were 1.96 (95% CI: 1.55–2.47) and 3.33 (95% CI: 2.64–4.20), respectively, with Group 1 as the reference group. Further adjustment for creatinine, white blood cell count, hypertension, heart failure, RF, diabetes, vasopressors, CRRT, and SOFA scores in Model 3 confirmed that this significant association persisted. In Model 3, the adjusted HRs (95% CI) for Group 2 and Group 3 were 1.71 (95% CI: 1.36–2.17) and 3.04 (95% CI: 2.39–3.86), respectively, with Group 1 as the reference group.
Table 3.
Cox proportional hazard ratios (HR) for all-cause mortality
| Model 1 | Model 2 | Model 3 | ||||||
|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | |||
| 28-day mortality | ||||||||
| Coagulopathy score categories | ||||||||
| 0–2 | Reference | Reference | Reference | |||||
| 3–4 | 1.78 (1.42–2.25) | < 0.001 | 1.96 (1.55–2.47) | < 0.001 | 1.71 (1.36–2.17) | < 0.001 | ||
| 5–6 | 2.88 (2.30–3.62) | < 0.001 | 3.33 (2.64–4.20) | < 0.001 | 3.04 (2.39–3.86) | < 0.001 | ||
| P for trend | < 0.001 | < 0.001 | < 0.001 | |||||
| Coagulopathy score | ||||||||
| 0 | Reference | Reference | Reference | |||||
| 1 | 1.90 (0.91–3.96) | 0.09 | 1.98 (0.95–4.13) | 0.07 | 2.18 (1.04–4.54) | 0.04 | ||
| 2 | 1.62 (0.80–3.28) | 0.18 | 1.68 (0.83–3.40) | 0.15 | 1.80 (0.89–3.65) | 0.10 | ||
| 3 | 2.93 (1.49–5.74) | 0.002 | 3.26 (1.66–6.41) | < 0.001 | 2.93 (1.49–5.76) | 0.002 | ||
| 4 | 2.82 (1.44–5.54) | 0.002 | 3.28 (1.67–6.43) | < 0.001 | 3.26 (1.65–6.41) | < 0.001 | ||
| 5 | 3.85 (1.98–7.51) | < 0.001 | 4.59 (2.35–8.96) | < 0.001 | 4.72 (2.41–9.27) | < 0.001 | ||
| 6 | 7.64 (3.87–15.06) | < 0.001 | 9.50 (4.80–18.80) | < 0.001 | 8.15 (4.10-16.24) | < 0.001 | ||
| 90-day mortality | ||||||||
| Coagulopathy score categories | ||||||||
| 0–2 | Reference | Reference | Reference | |||||
| 3–4 | 1.75 (1.44–2.12) | < 0.001 | 1.93 (1.58–2.35) | < 0.001 | 1.74 (1.43–2.13) | < 0.001 | ||
| 5–6 | 2.70 (2.22–3.28) | < 0.001 | 3.13 (2.57–3.82) | < 0.001 | 2.95 (2.40–3.62) | < 0.001 | ||
| P for trend | < 0.001 | < 0.001 | < 0.001 | |||||
| Coagulopathy score | ||||||||
| 0 | Reference | Reference | Reference | |||||
| 1 | 2.08 (1.10–3.91) | 0.02 | 2.18 (1.16–4.10) | 0.02 | 2.41 (1.28–4.54) | 0.007 | ||
| 2 | 1.77 (0.97–3.26) | 0.06 | 1.85 (1.01–3.41) | 0.047 | 2.00 (1.09–3.68) | 0.02 | ||
| 3 | 3.08 (1.72–5.51) | < 0.001 | 3.48 (1.94–6.24) | < 0.001 | 3.22 (1.79–5.78) | < 0.001 | ||
| 4 | 3.03 (1.69–5.43) | < 0.001 | 3.55 (1.98–6.36) | < 0.001 | 3.70 (2.05–6.65) | < 0.001 | ||
| 5 | 3.91 (2.19–6.98) | < 0.001 | 4.71 (2.64–8.42) | < 0.001 | 5.05 (2.81–9.06) | < 0.001 | ||
| 6 | 7.94 (4.40-14.33) | < 0.001 | 9.56 (5.50-18.02) | < 0.001 | 8.77 (4.81–15.97) | < 0.001 | ||
Model 1: Unadjusted;
Model 2: Adjusted age, gender, and ethnicity;
Model 3: Adjusted age, gender, ethnicity, white blood cells, creatinine, platelet, hypertension, heart failure, respiratory failure, diabetes mellitus, vasopressin, continuous renal replacement therapy, and Sequential Organ Failure Assessment
Using Cox regression models to evaluate the association between coagulopathy score and 90-day all-cause mortality in cirrhotic patients with sepsis yielded consistent findings. Comprehensive results are provided in Table 3.
Predictive value of coagulopathy score for All-Cause mortality in patients with cirrhosis and Sepsis
ROC curves were generated for coagulopathy score, SOFA, the Sepsis-induced Coagulopathy (SIC) score, and Model for End-stage Liver Disease (MELD) scores to assess their ability to predict 28-day and 90-day all-cause mortality among patients with cirrhosis and sepsis, as illustrated in Fig. 3. The results revealed that although the coagulopathy score demonstrated a slightly lower predictive value for prognosis than the MELD score, it was significantly superior to SIC score and the SOFA score. Specifically, when predicting 28-day all-cause mortality, the AUC for Coagulopathy score was 62.89% (95% CI: 60.47–65.31), for SOFA was 58.19% (95% CI: 55.63–60.75), for SIC score was 58.44% (95% CI: 56.02–60.86), and for MELD was 71.54% (95% CI: 59.25–73.83). Similarly, when predicting 90-day all-cause mortality, the AUC for Coagulopathy score was 62.87% (95% CI: 60.60-65.14), for SOFA was 57.75% (95% CI: 55.36–60.14), for SIC score 57.93% (95% CI: 55.63–60.22), and for MELD was 71.58% (95% CI: 69.43–73.73). More detailed findings are presented in Fig. 3; Table 4.
Fig. 3.
Receiver operating characteristic (ROC) curves evaluating the predictive ability of the coagulopathy score for 28-day (A), and 90-day (B) all-cause mortality, with corresponding area under the curve (AUC) values
Table 4.
Information of ROC curves in Fig. 3
| Variables | AUC (%) | 95% CI (%) | AUC difference | Threshold | Sensitivity | Septicity |
|---|---|---|---|---|---|---|
| 28-day mortality | ||||||
| Coagulopathy score | 62.89 | 60.47–65.31 | Reference | 4.50 | 0.72 | 0.46 |
| SOFA | 58.19 | 55.63–60.75 | -0.05 (-0.08 to -0.02), P = 0.002 | 5.50 | 0.74 | 0.39 |
| MELD | 71.54 | 59.25–73.83 | 0.09 (0.06 to 0.11), P < 0.001 | 27.50 | 0.75 | 0.60 |
| SIC | 58.44 | 56.02–60.86 | -0.04 (-0.06 to -0.03), P < 0.001 | 3.50 | 0.32 | 0.82 |
| 90-day mortality | ||||||
| Coagulopathy score | 62.87 | 60.60-65.14 | Reference | 4.50 | 0.74 | 0.44 |
| SOFA | 57.75 | 55.36–60.14 | -0.05 (-0.08 to -0.02), P < 0.001 | 3.50 | 0.58 | 0.54 |
| MELD | 71.58 | 69.43–73.73 | 0.09 (0.06 to 0.11), P < 0.001 | 25.50 | 0.72 | 0.63 |
| SIC | 57.93 | 55.63–60.22 | -0.05 (-0.07 to -0.03), P < 0.001 | 3.50 | 0.33 | 0.81 |
SOFA, Sequential Organ Failure Assessment; MELD, Model for End-Stage Liver Disease score; SIC, Sepsis-induced Coagulopathy score
Subgroup analysis
We further investigated the relationship between coagulopathy score and the risk of all-cause mortality at 28 and 90 days post-admission across different patient subgroups. Stratified analyses based on age, sex, hypertension, diabetes, and RF were conducted. Forest plots indicated no significant interaction between the coagulopathy score and any of the subgroups (All P for interaction > 0.05; Fig. 4). These results confirm the robustness of the findings.
Fig. 4.
Forest plots displaying the results of stratified analyses of the association between coagulopathy score and 28-day (A), and 90-day (B) all-cause mortality
Discussion
To our knowledge, this study is the first to evaluate the prognostic value of the coagulopathy score in critically ill patients with cirrhosis and sepsis, both in the short- and long-term. Our findings demonstrated that higher coagulopathy scores were strongly associated with an increased risk of all-cause mortality at 28 and 90 days. The subgroup analysis further confirmed the robustness of these findings.
Patients with cirrhosis are particularly susceptible to infections closely related to its pathophysiological mechanisms, including hepatic microcirculatory dysfunction, gut microbiota imbalance, localized and systemic inflammatory responses, and immune system disturbances [24]. Studies have shown that Infections are a major trigger for systemic inflammation and organ dysfunction in decompensated cirrhosis, leading to a four-fold increase in the risk of mortality [25]. These factors contribute to a significantly impaired immune function, making cirrhotic patients more prone to infections in various sites, including the abdomen, respiratory tract, and urinary tract [5, 17]. Common infections in this population include spontaneous bacterial peritonitis caused by ascites, pneumonia, and urinary tract infections [5, 17]. However, these infections are often not diagnosed through blood cultures, which explains the relatively low proportion of blood infections observed in our study. Infections are a major trigger for systemic inflammation and organ dysfunction in decompensated cirrhosis, increasing the risk of mortality by four-fold. Once cirrhotic patients progress to sepsis, both short- and long-term mortality risks rise significantly [26]. Therefore, identifying a clinically meaningful and easily accessible predictor for early prognosis in patients with cirrhosis and sepsis is of considerable clinical significance.
In addition, several factors may contribute to the high incidence of fungal infections observed in our study. Firstly, patients in the ICU often require invasive procedures, such as vascular access and mechanical ventilation, which increase the risk of fungal infections by providing a direct pathway for pathogens to enter the body. Additionally, the frequent use of broad-spectrum antibiotics and corticosteroids in critically ill patients disrupts the balance of the normal microbiota, creating an environment conducive to fungal overgrowth. These treatments not only eliminate or suppress bacterial flora but also impair the host’s immune function, further increasing susceptibility to opportunistic infections. Renal replacement therapies, commonly employed in ICU settings, can also contribute to immune dysregulation and provide a further challenge to the host’s defense mechanisms. Moreover, sepsis itself has a profound impact on the immune system, often leading to immune suppression and dysfunction. This state of immune dysregulation, characterized by impaired phagocytosis, reduced cytokine response, and altered T-cell function, predisposes patients to fungal infections by compromising their ability to mount an effective immune response. Together, these factors create a complex interplay of conditions that heighten the risk of fungal infections in critically ill patients, emphasizing the need for vigilant monitoring and targeted antifungal therapy in this high-risk population.
The coagulopathy score was initially developed to quantify coagulopathy and was first used for diagnosing specific conditions such as DIC [18]. As research progressed, the prognostic value of the coagulopathy score in various other diseases has been explored. Tang et al. [27] found that patients with congestive heart failure and higher coagulopathy scores had lower survival rates and shorter life expectancies. Li et al. [20] reported that hospital mortality rates progressively increased in patients with acute respiratory distress syndrome when the coagulopathy score exceeded 2. Therefore, regular monitoring of coagulation-related biomarkers is beneficial for daily management of these patients. Furthermore, Wu et al. [21] demonstrated that the coagulopathy score outperformed single indicators (such as PLT count, INR, and APTT) in predicting in-hospital mortality in patients with acute respiratory failure. This study is the first to reveal the relationship between the coagulopathy score and both short- and long-term survival of patients with cirrhosis complicated by sepsis. Higher Coagulopathy scores in these patients were linked to a poorer prognosis and increased mortality risk, and the coagulopathy score proved to be a useful predictor of both 28-day and 90-day all-cause mortality in this cohort.
Morevover, our results show that, compared to SIC score, the coagulopathy score demonstrates better predictive ability in patients with liver cirrhosis complicated by sepsis, including higher predictive value for both short-term and long-term outcomes. The SIC score is specifically designed for assessing coagulopathy associated with sepsis and includes indicators such as the SOFA score and platelet count. Its primary value lies in its multidimensional assessment (organ function + coagulation markers) to early identify the hypocoagulable state in septic patients, which provides a basis for anticoagulation therapy decisions [11]. However, liver cirrhosis patients inherently exhibit thrombocytopenia (due to splenomegaly or impaired platelet production), prolonged PT (due to coagulation factor synthesis dysfunction), and falsely elevated INR (due to vitamin K deficiency), leading to potential false positives in the SIC score [5, 7, 24]. Additionally, liver cirrhosis patients often exhibit a dual nature of coagulopathy, with both hypocoagulation (e.g., decreased anticoagulant proteins such as ATIII, PC/PS) and hypercoagulation (e.g., increased factor VIII, vWF, and fibrinolysis inhibition) coexisting [6, 9]. As a result, the SIC score fails to identify hypercoagulable states, which may delay thrombosis prevention. Moreover, the chronic organ dysfunction in cirrhosis patients (such as elevated bilirubin and abnormal creatinine) can be difficult to distinguish from acute liver injury in sepsis, potentially affecting the accuracy of the SOFA score used in the SIC assessment.
We hypothesized that the coagulopathy score could be used to assess the prognosis of critically ill patients with cirrhosis and sepsis primarily because coagulopathy plays a crucial role in their pathophysiology. Cirrhosis is linked to significant coagulation dysfunction, and sepsis further exacerbates this disorder, leading to the coexistence of both thrombosis and bleeding, which have a profound impact on clinical outcomes [28].
Liver cirrhosis is a disease caused by long-term chronic liver damage, and its most prominent change compared with a normal liver is a decline in liver function, particularly the loss of synthetic and detoxifying functions. First, reduced liver function in patients with cirrhosis leads to decreased synthesis of coagulation factors, especially coagulation factors I (fibrinogen), II (prothrombin), V, VII, IX, and X. Deficiency in various coagulation factors significantly increases the risk of bleeding in cirrhotic patients [29]. However, liver dysfunction in cirrhosis also leads to activation of the fibrinolytic system, which enhances fibrin degradation, further raising the risk of bleeding [30]. At the same time, cirrhosis is associated with a reduced or dysregulated synthesis of thrombopoietin and other platelet production factors by the liver, resulting in a decreased platelet count or platelet dysfunction, which contributes to a worsened bleeding tendency [9]. Importantly, the deficiency of anticoagulant proteins such as protein C and protein S, which are produced by the liver, disrupts the balance of coagulation and anticoagulation, further complicating the clinical picture by increasing the risk of thrombosis [31]. Moreover, cirrhosis-induced portal hypertension can cause microvascular damage, exacerbating both coagulation disorders and the formation of microthrombi [32]. This results in a precarious balance between bleeding and thrombosis, with patients at risk for both outcomes. Despite these alterations in coagulation parameters, it is important to note that the deterioration of these parameters in cirrhosis does not always correlate with clinically significant bleeding events, a feature that underscores the complexity of coagulopathy in cirrhotic patients. Therefore, patients with liver cirrhosis often experience varying degrees of coagulation dysfunction, even in the absence of other comorbidities [29]. This complex interplay between bleeding and thrombosis highlights the need for careful management and monitoring in cirrhotic patients to prevent both bleeding complications and thrombotic events.
Sepsis-induced inflammatory responses affect coagulation. First, the systemic inflammatory response triggered by sepsis activates the coagulation system through the release of large amounts of inflammatory mediators, such as tumor necrosis factor-alpha and interleukins, leading to the rapid consumption of coagulation factors and increased platelet aggregation, ultimately the development of DIC [14]. Changes induced by DIC, such as thrombocytopenia, fibrinogen depletion, and elevated D-dimer levels, further exacerbate coagulation dysfunction [33]. Therefore, sepsis aggravates coagulation disturbances in patients with cirrhosis, significantly increasing the risks of bleeding, thrombosis, and organ failure [34].
The coagulopathy score quantifies coagulation-related laboratory parameters to assess the severity of coagulation dysfunction, helping determine whether the coagulation disorder is mild or severe. The severity of coagulation abnormalities directly affects the risk of multiorgan failure [35]. This is particularly important in critically ill patients with cirrhosis and sepsis, as they often present with DIC, a condition that severely affects coagulation and also exacerbates the occurrence of organ failure [9, 36]. The Coagulopathy score allows for the identification of high-risk patients, which aligns with our finding that patients with higher coagulopathy scores generally have poorer prognoses. For these patients, clinicians may need to implement timely interventions (such as hemostatic therapy, anticoagulation, or plasmapheresis) to prevent or mitigate multiorgan failure and ultimately improve patient outcomes.
The Coagulopathy score demonstrates significant advantages in assessing both the short- and long-term prognoses of patients with cirrhosis and sepsis. First, it offers greater specificity by focusing on coagulopathy, a critical pathophysiological feature in this patient population. Compared with the SOFA score, the coagulopathy score is more sensitive to changes in coagulation function, making it a more targeted tool for assessing prognosis in these patients. Moreover, this scoring system is simple and easy to use, relying on routine coagulation parameters (such as PT and INR) without the need for complex calculations or the integration of multiple indicators. This allows for faster outcomes, particularly in emergency settings or resource-limited environments, where quick patient assessment is crucial. Additionally, although the MELD score demonstrated a higher AUC than the coagulopathy score in this study, the latter may offer a distinct advantage in detecting early coagulation abnormalities, providing an early indication of mortality risk, and aiding clinicians in timely intervention and treatment optimization.
It is worth mentioning that the coagulopathy score can effectively complement to both MELD and SOFA scores. Combining these scoring systems provides a more comprehensive reference for clinical decision-making, further enhancing predictive accuracy and clinical value. In conclusion, the coagulopathy score offers a valuable method for prognostic assessment in cirrhosis patients with sepsis due to its strong specificity, ease of use, ability to detect early changes, and potential as a multidimensional supplement. To further confirm its reliability and applicability, as well as to enhance its utility across different clinical scenarios, future research involving large-scale, multicenter studies is essential.
The primary strength of this study is its innovative investigation into the association between the coagulopathy score and both short- and long-term survival outcomes in patients with cirrhosis complicated by sepsis. Utilizing the extensive and diverse dataset from the MIMIC-IV (v 3.0) database allowed for comprehensive statistical adjustments, thereby reducing the influence of potential confounders. In addition, our study is the first to apply this scoring system to this particular patient cohort. Future multi-center prospective trials are needed to confirm the effectiveness of this scoring system and further evaluate its potential clinical utility.
Despite providing new insights into the prognostic value of the coagulopathy score in patients with cirrhosis and sepsis, this study had several limitations. First, since this study was a retrospective analysis conducted at a single center, the generalizability of the results may be restricted. Prospective, multicenter studies are needed in the future to confirm the prognostic utility of the coagulopathy score for predicting all-cause mortality in this patient group. Second, although the MIMIC-IV (v 3.0) database includes a large cohort of critically ill patients with rich clinical information, it was primarily sourced from the Beth Israel Deaconess Medical Center and its affiliated hospitals in the Boston area, introducing potential regional bias. This may have affected the external validity of our findings, particularly their applicability to other regions or populations. Therefore, future studies should validate our findings in a broader geographical context and across diverse patient groups. Finally, The different stages of liver cirrhosis significantly influence infection rates and patient prognosis. Decompensated cirrhosis, in particular, is associated with more pronounced immune dysfunction, metabolic abnormalities, and organ failure, all of which heighten susceptibility to infections and lead to worse outcomes. However, due to the limitations of the MIMIC-IV database, we were unable to obtain detailed data on the proportion of patients with decompensated cirrhosis in our study. Future studies should further investigate the correlation between cirrhosis staging, infection incidence, and prognosis to provide a more comprehensive understanding of these dynamics.
Conclusion
Patients with cirrhosis complicated by sepsis, particularly those with high coagulopathy scores, generally have poor prognoses. These patients often require early and more intensive medical interventions.
Acknowledgements
We are appreciative of the MIMIC-IV (v 3.0) participants and staff. We thank all reviewers who participated in the review.
Abbreviations
- ICUs
Intensive Care Units
- DIC
Disseminated Intravascular Coagulation
- INR
International Normalized Ratio
- APTT
Activated Partial Thromboplastin Time
- PLT
Platelet Count
- MIMIC
IV-Medical Information Mart for Intensive Care IV
- SOFA
Sequential Organ Failure Assessment
- HR
Hazard Ratio
- CI
Confidence Interval
- RF
Respiratory Failure
- CRRT
Continuous Renal Replacement Therapy
- ROC
Receiver Operating Characteristic
- AUC
Area Under the Curve
- MELD
Model for End-Stage Liver Disease
- PT
Prothrombin Time
- SIC
Sepsis-induced Coagulopathy
Author contributions
(I)Conception and design: Tao Wang, Decai Wang, Ruizi Shi, Xintao Zeng; (II) Administrative support: Xintao Zeng, Pei Yang, Xi Chen, Chidan Wan, Jianjun Wang; (III) Provision of study material: All authors; (IV) Collection and assembly of data: All authors; (V) Data analysis and interpretation: Tao Wang, Decai Wang, Xintao Zeng, Sirui Chen, Chuan Qin; (VI) Manuscript writing: All authors; (VII) Final approval of manuscript: All authors. Tao Wang, Decai Wang, and Ruizi Shi contributed equally to this work and share the first authorship. Chidan Wan and Jianjun Wang contributed equally to this work and share the corresponding authorship.
Funding
This study was supported by National Natural Science Foundation of China (NSFC) (Grant no.82400758), NHC Key Laboratory of Nuclear Technology Medical Transformation (Mianyang Central Hospital) (Grant no.2023HYX032), Health Commission of Sichuan Province Medical Science and Technology Program (Grant no.24QNMP028), Clinical Special Project of Mianyang Central Hospital (Grant No.2024LC007), and Medical Research Youth Innovation Project of Sichuan Province, China (Grant no.Q23046).
Data availability
Raw data supporting the conclusions of this paper will be made available by the authors without reservation and may be requested from the corresponding author.
Declarations
Ethics approval and consent to participate
This study used publicly available data from the MIMIC-IV database (version 3.0). Because the dataset was de-identified, no additional Institutional Review Board approval was required for this study. All procedures were conducted following the relevant guidelines and regulations to protect patient privacy and confidentiality.
Consent for publication
Not applicable.
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.
Tao Wang, Decai Wang and Ruizi Shi contributed equally to this work.
Contributor Information
Chidan Wan, Email: chidanwanjsr@163.com.
Jianjun Wang, Email: wangjianjunmch@163.com.
References
- 1.Martin-Loeches I, Singer M, Leone M. Sepsis: key insights, future directions, and immediate goals. A review and expert opinion. Intensive Care Med. 2024;50(12):2043–9. 10.1007/s00134-024-07694-z. [DOI] [PubMed] [Google Scholar]
- 2.Kullberg RFJ, Haak BW, Chanderraj R, et al. Empirical antibiotic therapy for sepsis: save the anaerobic microbiota. Lancet Respir Med. 2025;13(1):92–100. 10.1016/S2213-2600(24)00257-1. [DOI] [PubMed] [Google Scholar]
- 3.Willmann K, Moita LF. Physiologic disruption and metabolic reprogramming in infection and sepsis. Cell Metab. 2024;36(5):927–46. 10.1016/j.cmet.2024.02.013. [DOI] [PubMed] [Google Scholar]
- 4.Giamarellos-Bourboulis EJ, Aschenbrenner AC, Bauer M, et al. The pathophysiology of sepsis and precision-medicine-based immunotherapy. Nat Immunol. 2024;25(1):19–28. 10.1038/s41590-023-01660-5. [DOI] [PubMed] [Google Scholar]
- 5.Piano S, Bunchorntavakul C, Marciano S, et al. Infections in cirrhosis. Lancet Gastroenterol Hepatol. 2024;9(8):745–57. 10.1016/S2468-1253(24)00078-5. [DOI] [PubMed] [Google Scholar]
- 6.Durand F, Kellum JA, Nadim MK. Fluid resuscitation in patients with cirrhosis and sepsis: A multidisciplinary perspective. J Hepatol. 2023;79(1):240–6. 10.1016/j.jhep.2023.02.024. [DOI] [PubMed] [Google Scholar]
- 7.Bajaj JS, Kamath PS, Reddy KR. The evolving challenge of infections in cirrhosis. N Engl J Med. 2021;384(24):2317–30. 10.1056/NEJMra2021808. [DOI] [PubMed] [Google Scholar]
- 8.Piano S, Tonon M, Angeli P. Changes in the epidemiology and management of bacterial infections in cirrhosis. Clin Mol Hepatol. 2021;27(3):437–45. 10.3350/cmh.2020.0329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zanetto A, Campello E, Senzolo M, et al. The evolving knowledge on primary hemostasis in patients with cirrhosis: A comprehensive review. Hepatology. 2024;79(2):460–81. 10.1097/HEP.0000000000000349. [DOI] [PubMed] [Google Scholar]
- 10.Crăciun R, Grapă C, Mocan T, et al. The bleeding edge: managing coagulation and bleeding risk in patients with cirrhosis undergoing interventional procedures. Diagnostics (Basel). 2024;14(22):2602. 10.3390/diagnostics14222602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Williams B, Zou L, Pittet JF, et al. Sepsis-Induced coagulopathy: A comprehensive narrative review of pathophysiology, clinical presentation, diagnosis, and management strategies. Anesth Analg. 2024;138(4):696–711. 10.1213/ANE.0000000000006888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zermatten MG, Fraga M, Moradpour D, et al. Hemostatic alterations in patients with cirrhosis: from primary hemostasis to fibrinolysis. Hepatology. 2020;71(6):2135–48. 10.1002/hep.31201. [DOI] [PubMed] [Google Scholar]
- 13.Wei X, Tu Y, Bu S, et al. Unraveling the intricate web: complement activation shapes the pathogenesis of Sepsis-Induced coagulopathy. J Innate Immun. 2024;16(1):337–53. 10.1159/000539502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Girardis M, David S, Ferrer R, et al. Understanding, assessing and treating immune, endothelial and haemostasis dysfunctions in bacterial sepsis. Intensive Care Med. 2024;50(10):1580–92. 10.1007/s00134-024-07586-2. [DOI] [PubMed] [Google Scholar]
- 15.Iba T, Helms J, Maier CL, et al. The role of thromboinflammation in acute kidney injury among patients with septic coagulopathy. J Thromb Haemost. 2024;22(6):1530–40. 10.1016/j.jtha.2024.02.006. [DOI] [PubMed] [Google Scholar]
- 16.Kappelmayer J, Debreceni IB, Fejes Z, et al. Inflammation, sepsis, and the coagulation system. Hamostaseologie. 2024;44(4):268–76. 10.1055/a-2202-8544. [DOI] [PubMed] [Google Scholar]
- 17.Wong F, Bernardi M, Balk R, et al. Sepsis in cirrhosis: report on the 7th meeting of the international Ascites club. Gut. 2005;54(5):718–25. 10.1136/gut.2004.038679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Long Y, Tong Y, Miao R, et al. Early coagulation disorder is associated with an increased risk of atrial fibrillation in septic patients. Front Cardiovasc Med. 2021;8:724942. 10.3389/fcvm.2021.724942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Xie Z, Zhu S, Wang J, et al. Relationship between coagulopathy score and ICU mortality: analysis of the MIMIC-IV database. Heliyon. 2024;10(14):e34644. 10.1016/j.heliyon.2024.e34644. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Li J, Pan G. Association of coagulation function with the risk of in-hospital mortality in patients with severe acute respiratory distress syndrome. Am J Med Sci. 2024;368(2):143–52. 10.1016/j.amjms.2024.04.012. [DOI] [PubMed] [Google Scholar]
- 21.Wu Y, Zhu G. Association between coagulation disorder scores and in-hospital mortality in ARF patients: a retrospective analysis from the MIMIC-IV database. Front Med (Lausanne). 2023;10:1184166. 10.3389/fmed.2023.1184166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Johnson AEW, Bulgarelli L, Shen L, et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci Data. 2023;10(1):1. 10.1038/s41597-022-01899-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Singer M, Deutschman CS, Seymour CW, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA. 2016;315(8):801–10. 10.1001/jama.2016.0287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Incicco S, Angeli P, Piano S. Infectious complications of portal hypertension. Clin Liver Dis. 2024;28(3):525–39. 10.1016/j.cld.2024.03.007. [DOI] [PubMed] [Google Scholar]
- 25.Fernández J, Acevedo J, Castro M, et al. Prevalence and risk factors of infections by multiresistant bacteria in cirrhosis: a prospective study. Hepatology. 2012;55(5):1551–61. 10.1002/hep.25532. [DOI] [PubMed] [Google Scholar]
- 26.Arvaniti V, D’Amico G, Fede G, et al. Infections in patients with cirrhosis increase mortality four-fold and should be used in determining prognosis. Gastroenterology. 2010;139(4):1246–56. 10.1053/j.gastro.2010.06.019. [DOI] [PubMed] [Google Scholar]
- 27.Tang Y, Chen Q, Liang B, et al. A retrospective cohort study on the association between early coagulation disorder and short-term all-cause mortality of critically ill patients with congestive heart failure. Front Cardiovasc Med. 2022;9:999391. 10.3389/fcvm.2022.999391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Basili S, Ferro D, Violi F. Endotoxaemia, hyperfibrinolysis, and bleeding in cirrhosis. Lancet. 1999;353(9158):1102. 10.1016/S0140-6736(05)76463-1. [DOI] [PubMed] [Google Scholar]
- 29.O’Leary JG, Greenberg CS, Patton HM, et al. AGA clinical practice update: coagulation in cirrhosis. Gastroenterology. 2019;157(1):34–e431. 10.1053/j.gastro.2019.03.070. [DOI] [PubMed] [Google Scholar]
- 30.Jairath V, Burroughs AK. Anticoagulation in patients with liver cirrhosis: complication or therapeutic opportunity? Gut. 2013;62(4):479–82. 10.1136/gutjnl-2012-303088. [DOI] [PubMed] [Google Scholar]
- 31.Flores B, Trivedi HD, Robson SC, et al. Hemostasis, bleeding and thrombosis in liver disease. J Transl Sci. 2017;3(3). 10.15761/JTS.1000182. [DOI] [PMC free article] [PubMed]
- 32.Airola C, Pallozzi M, Cerrito L, et al. Microvascular thrombosis and liver fibrosis progression: mechanisms and clinical applications. Cells. 2023;12(13):1712. 10.3390/cells12131712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Danckwardt S, Trégouët DA, Castoldi E. Post-transcriptional control of haemostatic genes: mechanisms and emerging therapeutic concepts in thrombo-inflammatory disorders. Cardiovasc Res. 2023;119(8):1624–40. 10.1093/cvr/cvad046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Choudhury A, Kedarisetty CK, Vashishtha C, et al. A randomized trial comparing terlipressin and noradrenaline in patients with cirrhosis and septic shock. Liver Int. 2017;37(4):552–61. 10.1111/liv.13252. [DOI] [PubMed] [Google Scholar]
- 35.Carcillo JA, Shakoory B. Cytokine storm and Sepsis-Induced multiple organ dysfunction syndrome. Adv Exp Med Biol. 2024;1448:441–57. 10.1007/978-3-031-59815-9_30. [DOI] [PubMed] [Google Scholar]
- 36.Zanetto A, Campello E, Pelizzaro F, et al. Haemostatic alterations in patients with cirrhosis and hepatocellular carcinoma: laboratory evidence and clinical implications. Liver Int. 2022;42(6):1229–40. 10.1111/liv.15183. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Raw data supporting the conclusions of this paper will be made available by the authors without reservation and may be requested from the corresponding author.