Introduction
Sepsis is often a complication that occurs in the clinical course of medical and surgical patients treated for other diseases [
1,
2], and remains one of the most significant causes of mortality in intensive care units. The most recent international sepsis guidelines entitled “Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock: 2012” recommend early diagnosis and treatment of sepsis to avoid multiple organ failure and other adverse outcomes [
3]. Sepsis is diagnosed based on evidence of infection along with the presence of systemic inflammatory response syndrome (SIRS) defined by the American College of Chest Physicians/Society of Critical Care Medicine (ACCP/SCCM) guidelines [
4]. And sepsis was diagnosed when patients met the criteria for SIRS and an infectious source was documented or strongly suspected based on clinical presentation.
The majority of ill patients with SIRS present with coagulation abnormalities. Inflammation and coagulation play pivotal roles in the pathogenesis of sepsis. Evidence of extensive cross-talk between these two systems has been increasing [
5]; inflammation leads to the activation of coagulation, which in turn considerably affects inflammatory activity. Since infection-induced disseminated intravascular coagulation (DIC) is closely associated with SIRS in a considerable percentage of patients, patients who exhibit two or more SIRS criteria for more than three consecutive days are frequently associated with DIC [
6]. Sepsis is the most common disease associated with DIC. Approximately 20 to 40% of all sepsis patients are complicated with DIC [
7‐
10]. There are some standard care procedures for sepsis, including the use of antibiotics, oxygen, fluid resuscitation and corticosteroids [
11]. However, the mortality rate is still within the range of 30 to 50% in patients with septic shock [
12,
13]. Gando
et al. [
6] reported that DIC is frequently associated with SIRS (83%) and that such patients have a high mortality rate (63%). Thus, the mortality rate of sepsis patients complicated with DIC is clearly higher than that of patients without DIC. Therefore, the early diagnosis and treatment of sepsis-induced DIC are critical for improving the prognosis.
However, there are different diagnostic systems and criteria for SIRS/sepsis and DIC. Furthermore, there are still no diagnostic criteria for sepsis-induced DIC, which may delay diagnosis and treatment initiation, and consequently be detrimental for the patient. Therefore, this study aimed to establish the diagnostic criteria for sepsis-induced DIC.
Materials and methods
This prospective single-center observational study was conducted at the Department of Emergency and Critical Care Medicine, Fukuoka University Hospital, Fukuoka, Japan – a 915-bed referral, tertiary hospital - from June 2010 to June 2011. This study was approved by the institutional ethics committee, and all participants provided informed consent prior to participation. Patients aged ≥18 years who met one or more SIRS criteria were enrolled in this study. The background of these patients included liver cirrhosis, warfarin treatment, continuing antibiotics and/or steroid use, traumatic injury and others. We excluded patients who lacked a concentration of biomarkers or apparent clinical manifestations. Patients were evaluated for the presence of SIRS and sepsis according to the ACCP/SCCM guidelines [
4]. The scoring system of the Japanese Association for Acute Medicine (JAAM) for DIC was used for the diagnosis of DIC in this study. This DIC diagnostic algorithm for scoring DIC includes the following variables: platelet count, prothrombin time, fibrin/fibrinogen degradation product level and SIRS criteria. The details of the algorithm have been published elsewhere [
14]. DIC was defined by a score of ≥4. Illness severity was evaluated according to the Acute Physiology and Chronic Health Evaluation (APACHE) II score [
15]. The APACHE II score assesses the illness severity of critical patients admitted to intensive care units on the basis of routine physiologic measurements, age and previous health status. It is used to predict the outcome of critical illnesses. Organ failure was assessed according to the Sequential Organ Failure Assessment (SOFA) score [
16]. The SOFA score estimates organ dysfunction related to various disease statuses, especially sepsis, and is calculated using readily available measurements to quantify the dysfunction of the six major organs. Furthermore, it is useful for evaluating the morbidity and mortality of critical illnesses. All patients were followed-up for 28 days after enrollment in the study, and 28-day all-cause mortality was assessed.
Study procedures
Blood samples for measuring the markers were collected on admission. Presepsin, procalcitonin (PCT), interleukin-6 (IL-6), C-reactive protein (CRP) and white blood cell (WBC) count were measured as inflammatory molecular markers in plasma. Antithrombin (AT), protein C (PC) activities, platelet count, prothrombin time (PT), D-dimer and thrombomodulin (TM) levels were measured as coagulation and fibrinolysis molecular markers. Platelet and WBC counts were measured in whole blood using an XT-1800i (Sysmex Co., Kobe, Japan). PT, D-dimer level, and PC and AT activities were measured in plasma using a Coapresta 2000 (Sekisui Medical, Tokyo, Japan). TM was measured using a STACIA (Mitsubishi Chemical Medience Corp., Tokyo, Japan). International normalized ratio (INR) was calculated using the following formula: INR = (patient PT/normal PT) × ISI, where normal PT represents the average of mean normal PT range of the laboratory result and ISI is the International Sensitivity Index, which is the correction coefficient of thromboplastin in commercial kits calculated according to international reference samples.
Presepsin assay
Presepsin concentrations were measured using a compact automated immunoanalyzer, PATHFAST, based on a chemiluminescent enzyme immunoassay (CLEIA) (Mitsubishi Chemical Medience Corp., Japan) [
17,
18]. Whole blood was collected using a conventional blood collection tube (TERUMO, Tokyo, Japan) with EDTA-2 K as an anticoagulant and used as a sample within 4 h after collection.
PCT assay
PCT concentrations were measured by the Elecsys BRAHMS PCT assay (Roche Diagnostics, Tokyo, Japan) using EDTA plasma as a sample.
Interleukin-6 assay
IL-6 concentrations were measured using the Immulyze 2000 assay system (Siemens Healthcare Diagnostics, Tokyo, Japan) using EDTA plasma as a sample.
CRP assay
CRP concentrations were measured by CRP-LATEX (II) X2 “SEIKEN” (Denka Seiken Co., Ltd., Tokyo, Japan) using EDTA plasma as a sample.
Statistical analysis
Unless otherwise indicated, all data are expressed as mean ± standard deviation (SD). SPSS 15.0 J (SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. Comparisons between the two groups were made using unpaired Students
t-test and either the χ
2 test or Fisher’s exact test if necessary. Multiple datasets were analyzed by one-way ANOVA. The relationships between the measured variables and prognosis were analyzed by stepwise multiple logistic regression analysis with sepsis and/or DIC as the dependent variable. The results are reported as odds ratios (ORs) and 95% confidence intervals (95% CIs). Receiver-operating curve (ROC) analysis including the area under the ROC (AUC) was used to compare prognostic methods as predictors of sepsis and DIC. The standard error of the ROC was calculated using the formula based on Hanley and McNeil [
19]. The level of significance was set at
P <.05.
Discussion
Sepsis is the most common cause of death in hospitalized patients and affects >18 million people worldwide; its incidence is expected to increase by 1% annually [
20]. The mortality of patients who meet the severe sepsis criteria in the first 24 h after admission to the intensive care unit is 30 to 40% before intensive care unit discharge. Moreover, 45 to 50% of these patients die during their hospital stay [
10,
12,
21]. In sepsis patients, bacterial products and cytokines also activate coagulation by increasing tissue factor (TF) synthesis and preventing fibrinolysis by increasing the level of Plasminogen Activator Inhibitor Type-1 (PAI-1). Sepsis is frequently complicated with DIC, which arises from fibrin accumulation in small vessels and occlusion of capillaries with microthrombi [
22]. Activation of coagulation that arises from sepsis is accompanied by impaired function of major anticoagulant mechanisms including antithrombin, the PC tissue factor pathway inhibitor system and fibrinolysis [
23,
24]. Pro-inflammatory cytokines and other mediators are capable of activating the coagulation system and down-regulating important physiological anticoagulant pathways [
25]. These processes collectively result in increased levels of intravascular fibrin and the formation of microvascular thrombosis and, thus, ischemic multiple organ dysfunctions leading to necrosis [
26].
DIC is associated with high mortality in patients with severe sepsis. The results of this study indicate the 28-day all-cause mortality of sepsis-induced DIC patients meeting the sepsis and JAAM DIC criteria was significantly higher than that of patients who did not meet these criteria (Table
6). Although the effectiveness of anticoagulant therapy in septic patients remains controversial, some studies suggest that rapid diagnosis and early treatment of DIC improve outcomes for these patients. In particular, therapeutic intervention directly against coagulation and inflammation in DIC associated with severe sepsis is effective [
8,
27]. Furthermore, it is generally accepted that early aggressive treatment of the underlying disease is important.
In this study, we proposed new diagnostic criteria for sepsis-induced DIC. We assembled a cohort of patients with ≥1 SIRS criterion and investigated a series of biomarkers with the overall goal of creating a panel capable of diagnosing sepsis-induced DIC. Using an innovative approach to establish clinical utility, we created a panel of biomarkers that could provide clinicians with a tangible estimate of the increased risks of sepsis and DIC. The present results indicate that the optimal biomarkers for identifying sepsis-induced DIC are presepsin and PC, which represent inflammatory, and coagulation and fibrinolysis molecular markers, respectively. Presepsin and PC were obviously superior biomarkers for evaluating sepsis and DIC, respectively. This panel is biologically plausible as it incorporates biomarkers involved in key components of the pathophysiology of sepsis and DIC, including infection (presepsin) and activation of coagulation (PC). Moreover, we created new sepsis-induced DIC diagnostic criteria (Figure
1A), which included presepsin >900 pg/mL and PC <45%. This multi-marker approach offers a distinct mechanistic advantage over single-marker approaches.
Presepsin is a 13-kDa protein that is a truncated N-terminal fragment of CD14, the receptor for lipopolysaccharide (LPS)/LPS-binding protein (LBP) complexes [
28,
29]. Its levels specifically increase in the blood of septic patients. The measurement of presepsin concentrations is reported to be useful for the diagnosis of sepsis, evaluating the severity of sepsis, and monitoring clinical responses to therapeutic interventions [
30‐
34]. Most recently, multicenter clinical studies reported that presepsin is the most valuable predictive marker of sepsis between PCT and IL-6, and is superior to blood culture [
35]. Our results corroborate the notion that presepsin is currently the most valuable predictive marker of sepsis. In addition, our results indicate that presepsin is the best predictive marker of DIC compared to other inflammatory molecular markers, including PCT, IL-6 and CRP.
Coagulation activation with subsequent diffuse intravascular fibrin deposition is implicated as an etiological factor in multiple organ dysfunction syndromes in patients with sepsis as well as in transplant and trauma patients [
36]. Septic shock progression is associated with even greater mortality rates, ranging from 50% to 70% [
37,
38]. The PC system plays a crucial role in the control of microvascular coagulation and inflammation; it is one of the basic regulatory systems of homeostasis, as it has potent anticoagulant, profibrinolytic and anti-inflammatory properties [
39,
40]. PC is converted into activated PC (APC) under the formation of thrombin-thrombomodulin complexes with endothelial PC receptors (EPCRs) in the presence of protein S. Normal levels of circulating PC range from 2,800 to 5,600 ng/mL (80 to 140%), and the protein has a 10-h half-life. In contrast, APC has normal circulating levels from 1 to 3 ng/mL and a 20-minute half-life [
39‐
42]. Activated PC inactivates coagulation factors Va and VIIIa and neutralizes the effects of PAI-l [
43‐
45]. In addition, APC is capable of direct anti-inflammatory activity that reduces cytokine production (TNF, migration inhibitory factor (MIF)), thereby inhibiting the adhesion of leukocytes to the blood vessel endothelium [
39,
40,
45]. As a result of all of the above mentioned mechanisms, APC significantly reduces the processes of microvascular thrombosis and endothelial dysfunction [
46]. Numerous studies demonstrate depressed PC concentrations in both pediatric and adult septic patients are associated with increased morbidity and mortality [
47‐
49]. The present study indicates that PC is more useful for evaluating DIC compared to other coagulation and fibrinolysis molecular markers.
These proposed criteria are useful for the diagnosis of sepsis-induced DIC with extremely high precision (AUC: approximately 0.9). Since the cutoff point for these criteria yielded the optimal sensitivity and specificity (80.7% and 87.5%, respectively), they were able to identify patients who likely developed sepsis-induced DIC. The severity of patients’ condition (DIC, APACHE II and SOFA scores) differed significantly not only between the mild and severe groups, but also between the mild and moderate groups. The severity of these criteria allows the differentiation of patients with respect to DIC severity, physiological illness (APACHE II), multiple organ failure (SOFA) and mortality. Besides the severe group, treatment should be ready to be initiated anytime in the moderate group.
In this study, we diagnosed DIC using the JAAM DIC diagnostic criteria. The JAAM DIC study group retrospectively analyzed patients with sepsis complicated with DIC (DIC diagnosed according to the JAAM DIC scoring system) using databases from two previous multicenter studies [
14,
50]. They tested the hypothesis that the JAAM DIC scoring system constitutes a continuum dependent on the International Society on Thrombosis and Haemostasis (ISTH) overt DIC (DIC diagnosed according to the ISTH overt DIC scoring system) and that the JAAM DIC scoring system can predict full-blown DIC in a group of patients associated with systemic inflammation caused by infection. In other words, JAAM DIC with a stressed but compensated hemostatic system continuously progresses to ISTH overt DIC with a stressed but decompensated hemostatic system [
51]. When JAAM DIC patients meet the ISTH overt DIC criteria, the risks of multiple organ dysfunction syndrome (MODS) and death are increased approximately 1.5 times. Therefore, the JAAM DIC scoring system is useful for determining patients with sepsis at a stage of stressed but compensated DIC [
6].
This study has some limitations that should be noted. This study was a small, prospective, single-center, observational study. The present results suggest that there is no significant association between the severity of sepsis-induced DIC classification (severe, moderate and mild) and mortality. However, we have already initiated a validation study and planned a prospective multicenter study.
Competing interests
All authors declare that they have no competing interests.
Authors’ contributions
HI contributed to the study design, analysis, interpretation of the results, drafting of the manuscript and critical revisions of the manuscript for intellectual content. TN and AM were involved in data acquisition and carried out the immunoassays. YN, YI and JT acquired data and YN performed the statistical analysis. TU participated in the study design, development and methodology, and helped to draft the manuscript. All authors read and approved the final manuscript.