Eur J Anaesthesiol 2015; 32:199–206
ORIGINAL ARTICLE
Presepsin in the prognosis of infectious diseases and diagnosis of infectious disseminated intravascular coagulation A prospective, multicentre, observational study Gaku Takahashi, Shigehiro Shibata, Hiroyasu Ishikura, Masanao Miura, Yasuo Fukui, Yoshihiro Inoue and Shigeatsu Endo BACKGROUND Few prospective studies have described the prognostic accuracy of presepsin for 28-day mortality during days 0 to 7, or its role in the diagnosis of disseminated intravascular coagulation (DIC) in patients with infection. OBJECTIVE We aimed to evaluate the clinical usefulness of presepsin levels by comparing infection markers such as procalcitonin, interleukin-6 and C-reactive protein, as well as markers of DIC such as fibrin degradation products (FDPs) and D-dimer, from days 0 to 7. DESIGN A prospective, multicentre, observational study. SETTING Four medical institutions between June 2010 and June 2011. PATIENTS A total of 191 patients who fulfilled at least one of the systemic inflammatory response syndrome (SIRS) criteria were enrolled in the study. MAIN OUTCOME MEASURES The presepsin levels were evaluated for their diagnostic accuracy in discriminating between SIRS and sepsis, the prognostic accuracy for 28-day mortality from days 0 to 7 and the diagnostic
accuracy for DIC in patients with infection by comparison with other infection markers. RESULTS The diagnostic accuracy for discriminating between SIRS and sepsis from combining the presepsin and procalcitonin measurements [area under the curve (AUC), 0.91; likelihood ratio, 4.96] was higher than that of presepsin (AUC, 0.89; likelihood ratio, 4.75) or procalcitonin (AUC, 0.85; likelihood ratio, 3.18) alone. Not only the correlation coefficient between the presepsin level and the sequential organ failure assessment (SOFA) score but also the prognostic accuracy of presepsin for 28-day mortality increased with the elapsed time, and both were highest at day 7. The diagnostic accuracy for DIC generated by combining presepsin and FDP (AUC, 0.84; likelihood ratio, 3.57) was higher than that of FDP (AUC, 0.82; likelihood ratio, 2.64) or presepsin (AUC, 0.80; likelihood ratio, 2.94) alone. CONCLUSION The prognosis and severity of infection may be assessed more accurately by measuring the presepsin levels until day 7. Presepsin is a useful diagnostic tool for DIC with infection. Published online 9 November 2014
Introduction Cluster of differentiation 14 (CD14) is a glycoprotein expressed on the membrane surface of macrophages and monocytes and serves as a receptor complex for lipopolysaccharides and signal transduction via Toll-like receptor 4.1 CD14 is divided into two types of soluble isoforms (49 and 55 kDa), as reported by Bazil and Strominger.2
We have previously reported that the levels of soluble CD14 (55 kDa) were elevated in patients with multiple organ failure.3 Presepsin (soluble CD14-ST) has recently been identified as a 13-kDa truncated N-terminal fragment of CD14 produced by stimuli such as phagocytosis in response to bacterial infection.4 Several studies have
From the Department of Critical Care Medicine, Iwate Medical University, Morioka (GT, SS, YI, SE), Department of Emergency and Critical Care Medicine, Faculty of Medicine, Fukuoka University, Fukuoka (HI), Anesthesiology, Emergency and Critical Care Center, Kariya Toyota General Hospital, Kariya (MM) and Department of Gastroenterological Surgery, Kochi Health Sciences Center, Kochi, Japan (YF) Correspondence to Gaku Takahashi, Department of Critical Care Medicine, Iwate Medical University, 19-1 Uchimaru, Morioka 020-8505, Japan E-mail: [email protected] 0265-0215 ß 2015 Copyright European Society of Anaesthesiology
DOI:10.1097/EJA.0000000000000178
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
200 Takahashi et al.
confirmed the usefulness of presepsin as a marker for the diagnosis of sepsis,5–7 but the levels of presepsin have varied. For example, the mean presepsin levels in patients with systemic inflammatory response syndrome (SIRS), sepsis and severe sepsis/septic shock were reported to be 2516.4, 2866.7 and 3167.1 pg ml1, respectively,6 while the median presepsin levels in patients with SIRS, sepsis, severe sepsis and septic shock were described as 212, 325, 787 and 1084 pg ml1, respectively.7 Few prospective studies have measured presepsin levels at not only day 0 but also until day 7 when, in the early stages of illness, the patient may be unstable. In a previous study, we found that the presepsin levels were elevated in patients with local infection (infection without SIRS) as well as sepsis.8 Also, disseminated intravascular coagulation (DIC) is often associated with infection.9 Therefore, we prospectively evaluated the prognostic accuracy of presepsin measured from days 0 to 7 for 28-day mortality and its diagnostic accuracy for DIC. We compared presepsin in this role with infection markers such as procalcitonin (PCT), interleukin (IL)-6 and C-reactive protein (CRP), as well as markers of DIC such as fibrin degradation products (FDPs) and D-dimer, from days 0 to 7 at four medical institutions.
Materials and methods Ethics
Approval for this study was provided by the Ethical Committee of Iwate Medical University, Morioka, Japan on 6 May 2010 (Ethical Committee No. H24-42). Informed consent was obtained from all patients in accordance with the guidelines of each institution. Patients
Blood samples were collected from patients admitted to the emergency rooms at Iwate Medical University Hospital, Fukuoka University Hospital, Kochi Health Sciences Centre or the ICU at Kariya Toyota General Hospital between June 2010 and June 2011.10 We measured the presepsin, PCT, IL-6, CRP, FDP and D-dimer levels as well as the platelet count in all blood specimens collected upon admission (on arrival; day 0) and on days 1, 3, 5 and 7, at 7 : 00 a.m. The blood specimens were categorised into six groups according to the patient conditions (SIRS, non-SIRS, infection without SIRS, sepsis, severe sepsis and septic shock).11 The diagnoses of SIRS, sepsis, severe sepsis and septic shock were made according to the seven criteria described by the American College of Chest Physicians/Society of Critical Care Medicine (ACCP/ SCCM).10 Non-SIRS was defined as patients without infection and a SIRS score less than 2, while infection without SIRS was defined as patients with infection but a SIRS score less than 2. The Acute Physiology and Chronic Health Evaluation II (APACHE II) score12 and sequential organ failure assessment (SOFA) score13
were used as indices of severity. The diagnosis of DIC was made according to the Japanese Association for Acute Medicine (JAAM) DIC diagnostic criteria.14 The definitive diagnoses were deliberated and confirmed by two infection control doctors certified by the Japanese College of Infection Control Doctors, and the clinical research coordinator. The presepsin concentrations were measured using a compact automated immunoanalyser, PATHFAST, based on a chemi-luminescent enzyme immunoassay (Mitsubishi Chemical Medience Co., Tokyo, Japan). Briefly, whole blood was collected using a conventional blood collection tube (Terumo Co., Tokyo, Japan) containing EDTA-2K as an anticoagulant, and the sample was assayed within 4 h after collection using the PATHFAST presepsin assay. The PCT concentrations were measured using an Elecsys BRAHMS PCT assay (Roche Diagnostics, Tokyo, Japan) with EDTA-plasma as the sample. The IL-6 concentrations were measured using an Immulyze 2000 assay system (Siemens Healthcare Diagnostics K.K., Tokyo, Japan) with EDTA-plasma as the sample. The CRP concentrations were measured using CRP-LATEX(II)X2’SEIKEN’ (Denka Seiken Co., Tokyo, Japan) with EDTA-treated plasma as the sample. The FDP concentrations were measured using an LPIAS500 FDP-P assay (Mitsubishi Chemical Medience Co.) with EDTA-plasma as the sample. Statistical analysis
Results are expressed as mean SD or median (Q1 to Q3). Two-group comparisons were performed using the Mann–Whitney U-test with the Bonferroni correction, and multiple comparisons were performed using the Kruskal–Wallis H test or one-way repeated-measures analysis of variance (ANOVA) with Dunnett’s multiple comparison tests. We used the cut-off value obtained by a receiver-operating characteristic (ROC) analysis. The patterns of the changes were investigated using interactions and two-way ANOVA by log-transformed values.15,16 Values of P less than 0.05 were considered to indicate statistical significance. Group comparisons and ROC analyses were performed using JMP software (SAS Institute, Cary, North Carolina, USA). Analyses of Spearman correlation coefficients and interactions were performed using Dr SPSS II software (SPSS, Chicago, Illinois, USA).
Results Patient backgrounds
A total of 191 patients (111 men and 80 women) were enrolled in this study. The mean ages of the men and women were 87 19 and 71 16 years, respectively. The levels of the biomarkers did not correlate with age, and there was no difference in the biomarker levels according to sex (P > 0.05). The diagnoses were as follows: gastroenterological disease (n ¼ 45); respiratory disease
Eur J Anaesthesiol 2015; 32:199–206 Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
Presepsin for infection prognosis and disseminated intravascular coagulation diagnosis 201
(n ¼ 30); trauma (n ¼ 26); circulatory disease (n ¼ 13); renal disease (n ¼ 9); burns (n ¼ 8); cerebral haemorrhage (n ¼ 8); malignant disease (n ¼ 7); hepatobiliary disease (n ¼ 3); cellulitis/phlegmon (n ¼ 3); drug poisoning (n ¼ 3); and others (n ¼ 36). Nine patients had died by day 7 (one at day 1, three at day 3, one at day 5 and four at day 7), and 11 died between day 8 and day 28. Overall, 20 patients died between day 0 and day 28. A total of 848 blood specimens were collected at different time points and categorised into six groups according to the patients’ illness (non-SIRS, SIRS, infection without SIRS, sepsis, severe sepsis and septic shock). The characteristics and diseases of the patients are summarised in Table 1. Of the 848 blood specimens, 489 and 359 samples were diagnosed as nonsepsis (non-SIRS, SIRS and infection without SIRS) and sepsis (sepsis, severe sepsis and septic shock), respectively. In addition, 292 and 556 samples were diagnosed as noninfection (non-SIRS and SIRS) and infection (infection without SIRS, sepsis, severe sepsis and septic shock), respectively. Median presepsin, procalcitonin, interleukin-6 and C-reactive protein levels in systemic inflammatory response syndrome, sepsis, severe sepsis and septic shock
The presepsin levels were measured in 456 samples, and the values of the measurements indicated the presence of SIRS, sepsis, severe sepsis and septic shock (Fig. 1). There was a significant correlation between presepsin levels and severity. The presepsin levels in patients with SIRS were significantly lower than those with sepsis (P ¼ 0.0004), severe sepsis (P < 0.0001) and septic shock (P < 0.0001). Conversely, the PCT levels in patients with SIRS were significantly lower than those in patients with severe sepsis (P ¼ 0.0002) or septic shock (P < 0.0001), although no difference in the PCT levels was observed between patients with SIRS and those with sepsis (P ¼ 0.3654). The IL-6 levels in patients with SIRS were Table 1
significantly lower than those in patients with septic shock (P < 0.0001), although no differences in the IL-6 levels were observed between patients with SIRS and those with sepsis or between patients with sepsis and those with severe sepsis. Although the CRP levels in patients with SIRS were significantly lower than those in patients with sepsis (P < 0.0001), severe sepsis (P < 0.0001) or septic shock (P < 0.0001), the levels did not correlate with severity. To discriminate between patients with SIRS and sepsis (including sepsis, severe sepsis and septic shock), the diagnostic accuracies of presepsin, PCT, IL-6 and CRP were analysed using ROC curves (Table 2). The area under the curve (AUC) and likelihood ratio for presepsin (0.89 and 4.75, respectively) were higher than those for PCT, IL-6 and CRP. The AUC and likelihood ratio generated by combining presepsin and PCT (0.91 and 4.96, respectively) were higher than those for presepsin or PCT alone. Relationships of time-course changes between presepsin, procalcitonin, interleukin-6 and C-reactive protein levels and severity in patients with infection
Figure 2a shows the relationships of the time-course changes between the presepsin, PCT, IL-6 and CRP levels and the SOFA scores indicating severity in 100 survivors with infection. The presepsin levels and SOFA scores seemed to decrease slightly from days 0 to 7. There was no significantly different pattern for the changes between the presepsin level and the SOFA score (P ¼ 0.841 for interaction). The PCT, IL-6 and CRP levels at day 7 tended to be much lower than those at day 0. There were significantly different patterns for the changes between the PCT, IL-6 and CRP levels and the SOFA score (P < 0.01 for interaction). To confirm the relationships of the time-course changes between the presepsin, PCT, IL-6 and CRP levels and the SOFA score, we investigated the Spearman correlation coefficients between these biomarkers and the SOFA score.
Patient characteristics
APACHE II SOFA Diagnosis GD Respiratory disease Trauma CD Renal disease Burns CH MD HD CP Drug poisoning Others Total of points (%)
Non-SIRS
SIRS
Infection without SIRS
Sepsis
Severe sepsis
Septic shock
P
11 3
15 5
12 4
15 5
19 10
21 10
ORIGINAL ARTICLE
Presepsin in the prognosis of infectious diseases and diagnosis of infectious disseminated intravascular coagulation A prospective, multicentre, observational study Gaku Takahashi, Shigehiro Shibata, Hiroyasu Ishikura, Masanao Miura, Yasuo Fukui, Yoshihiro Inoue and Shigeatsu Endo BACKGROUND Few prospective studies have described the prognostic accuracy of presepsin for 28-day mortality during days 0 to 7, or its role in the diagnosis of disseminated intravascular coagulation (DIC) in patients with infection. OBJECTIVE We aimed to evaluate the clinical usefulness of presepsin levels by comparing infection markers such as procalcitonin, interleukin-6 and C-reactive protein, as well as markers of DIC such as fibrin degradation products (FDPs) and D-dimer, from days 0 to 7. DESIGN A prospective, multicentre, observational study. SETTING Four medical institutions between June 2010 and June 2011. PATIENTS A total of 191 patients who fulfilled at least one of the systemic inflammatory response syndrome (SIRS) criteria were enrolled in the study. MAIN OUTCOME MEASURES The presepsin levels were evaluated for their diagnostic accuracy in discriminating between SIRS and sepsis, the prognostic accuracy for 28-day mortality from days 0 to 7 and the diagnostic
accuracy for DIC in patients with infection by comparison with other infection markers. RESULTS The diagnostic accuracy for discriminating between SIRS and sepsis from combining the presepsin and procalcitonin measurements [area under the curve (AUC), 0.91; likelihood ratio, 4.96] was higher than that of presepsin (AUC, 0.89; likelihood ratio, 4.75) or procalcitonin (AUC, 0.85; likelihood ratio, 3.18) alone. Not only the correlation coefficient between the presepsin level and the sequential organ failure assessment (SOFA) score but also the prognostic accuracy of presepsin for 28-day mortality increased with the elapsed time, and both were highest at day 7. The diagnostic accuracy for DIC generated by combining presepsin and FDP (AUC, 0.84; likelihood ratio, 3.57) was higher than that of FDP (AUC, 0.82; likelihood ratio, 2.64) or presepsin (AUC, 0.80; likelihood ratio, 2.94) alone. CONCLUSION The prognosis and severity of infection may be assessed more accurately by measuring the presepsin levels until day 7. Presepsin is a useful diagnostic tool for DIC with infection. Published online 9 November 2014
Introduction Cluster of differentiation 14 (CD14) is a glycoprotein expressed on the membrane surface of macrophages and monocytes and serves as a receptor complex for lipopolysaccharides and signal transduction via Toll-like receptor 4.1 CD14 is divided into two types of soluble isoforms (49 and 55 kDa), as reported by Bazil and Strominger.2
We have previously reported that the levels of soluble CD14 (55 kDa) were elevated in patients with multiple organ failure.3 Presepsin (soluble CD14-ST) has recently been identified as a 13-kDa truncated N-terminal fragment of CD14 produced by stimuli such as phagocytosis in response to bacterial infection.4 Several studies have
From the Department of Critical Care Medicine, Iwate Medical University, Morioka (GT, SS, YI, SE), Department of Emergency and Critical Care Medicine, Faculty of Medicine, Fukuoka University, Fukuoka (HI), Anesthesiology, Emergency and Critical Care Center, Kariya Toyota General Hospital, Kariya (MM) and Department of Gastroenterological Surgery, Kochi Health Sciences Center, Kochi, Japan (YF) Correspondence to Gaku Takahashi, Department of Critical Care Medicine, Iwate Medical University, 19-1 Uchimaru, Morioka 020-8505, Japan E-mail: [email protected] 0265-0215 ß 2015 Copyright European Society of Anaesthesiology
DOI:10.1097/EJA.0000000000000178
Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
200 Takahashi et al.
confirmed the usefulness of presepsin as a marker for the diagnosis of sepsis,5–7 but the levels of presepsin have varied. For example, the mean presepsin levels in patients with systemic inflammatory response syndrome (SIRS), sepsis and severe sepsis/septic shock were reported to be 2516.4, 2866.7 and 3167.1 pg ml1, respectively,6 while the median presepsin levels in patients with SIRS, sepsis, severe sepsis and septic shock were described as 212, 325, 787 and 1084 pg ml1, respectively.7 Few prospective studies have measured presepsin levels at not only day 0 but also until day 7 when, in the early stages of illness, the patient may be unstable. In a previous study, we found that the presepsin levels were elevated in patients with local infection (infection without SIRS) as well as sepsis.8 Also, disseminated intravascular coagulation (DIC) is often associated with infection.9 Therefore, we prospectively evaluated the prognostic accuracy of presepsin measured from days 0 to 7 for 28-day mortality and its diagnostic accuracy for DIC. We compared presepsin in this role with infection markers such as procalcitonin (PCT), interleukin (IL)-6 and C-reactive protein (CRP), as well as markers of DIC such as fibrin degradation products (FDPs) and D-dimer, from days 0 to 7 at four medical institutions.
Materials and methods Ethics
Approval for this study was provided by the Ethical Committee of Iwate Medical University, Morioka, Japan on 6 May 2010 (Ethical Committee No. H24-42). Informed consent was obtained from all patients in accordance with the guidelines of each institution. Patients
Blood samples were collected from patients admitted to the emergency rooms at Iwate Medical University Hospital, Fukuoka University Hospital, Kochi Health Sciences Centre or the ICU at Kariya Toyota General Hospital between June 2010 and June 2011.10 We measured the presepsin, PCT, IL-6, CRP, FDP and D-dimer levels as well as the platelet count in all blood specimens collected upon admission (on arrival; day 0) and on days 1, 3, 5 and 7, at 7 : 00 a.m. The blood specimens were categorised into six groups according to the patient conditions (SIRS, non-SIRS, infection without SIRS, sepsis, severe sepsis and septic shock).11 The diagnoses of SIRS, sepsis, severe sepsis and septic shock were made according to the seven criteria described by the American College of Chest Physicians/Society of Critical Care Medicine (ACCP/ SCCM).10 Non-SIRS was defined as patients without infection and a SIRS score less than 2, while infection without SIRS was defined as patients with infection but a SIRS score less than 2. The Acute Physiology and Chronic Health Evaluation II (APACHE II) score12 and sequential organ failure assessment (SOFA) score13
were used as indices of severity. The diagnosis of DIC was made according to the Japanese Association for Acute Medicine (JAAM) DIC diagnostic criteria.14 The definitive diagnoses were deliberated and confirmed by two infection control doctors certified by the Japanese College of Infection Control Doctors, and the clinical research coordinator. The presepsin concentrations were measured using a compact automated immunoanalyser, PATHFAST, based on a chemi-luminescent enzyme immunoassay (Mitsubishi Chemical Medience Co., Tokyo, Japan). Briefly, whole blood was collected using a conventional blood collection tube (Terumo Co., Tokyo, Japan) containing EDTA-2K as an anticoagulant, and the sample was assayed within 4 h after collection using the PATHFAST presepsin assay. The PCT concentrations were measured using an Elecsys BRAHMS PCT assay (Roche Diagnostics, Tokyo, Japan) with EDTA-plasma as the sample. The IL-6 concentrations were measured using an Immulyze 2000 assay system (Siemens Healthcare Diagnostics K.K., Tokyo, Japan) with EDTA-plasma as the sample. The CRP concentrations were measured using CRP-LATEX(II)X2’SEIKEN’ (Denka Seiken Co., Tokyo, Japan) with EDTA-treated plasma as the sample. The FDP concentrations were measured using an LPIAS500 FDP-P assay (Mitsubishi Chemical Medience Co.) with EDTA-plasma as the sample. Statistical analysis
Results are expressed as mean SD or median (Q1 to Q3). Two-group comparisons were performed using the Mann–Whitney U-test with the Bonferroni correction, and multiple comparisons were performed using the Kruskal–Wallis H test or one-way repeated-measures analysis of variance (ANOVA) with Dunnett’s multiple comparison tests. We used the cut-off value obtained by a receiver-operating characteristic (ROC) analysis. The patterns of the changes were investigated using interactions and two-way ANOVA by log-transformed values.15,16 Values of P less than 0.05 were considered to indicate statistical significance. Group comparisons and ROC analyses were performed using JMP software (SAS Institute, Cary, North Carolina, USA). Analyses of Spearman correlation coefficients and interactions were performed using Dr SPSS II software (SPSS, Chicago, Illinois, USA).
Results Patient backgrounds
A total of 191 patients (111 men and 80 women) were enrolled in this study. The mean ages of the men and women were 87 19 and 71 16 years, respectively. The levels of the biomarkers did not correlate with age, and there was no difference in the biomarker levels according to sex (P > 0.05). The diagnoses were as follows: gastroenterological disease (n ¼ 45); respiratory disease
Eur J Anaesthesiol 2015; 32:199–206 Copyright © European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
Presepsin for infection prognosis and disseminated intravascular coagulation diagnosis 201
(n ¼ 30); trauma (n ¼ 26); circulatory disease (n ¼ 13); renal disease (n ¼ 9); burns (n ¼ 8); cerebral haemorrhage (n ¼ 8); malignant disease (n ¼ 7); hepatobiliary disease (n ¼ 3); cellulitis/phlegmon (n ¼ 3); drug poisoning (n ¼ 3); and others (n ¼ 36). Nine patients had died by day 7 (one at day 1, three at day 3, one at day 5 and four at day 7), and 11 died between day 8 and day 28. Overall, 20 patients died between day 0 and day 28. A total of 848 blood specimens were collected at different time points and categorised into six groups according to the patients’ illness (non-SIRS, SIRS, infection without SIRS, sepsis, severe sepsis and septic shock). The characteristics and diseases of the patients are summarised in Table 1. Of the 848 blood specimens, 489 and 359 samples were diagnosed as nonsepsis (non-SIRS, SIRS and infection without SIRS) and sepsis (sepsis, severe sepsis and septic shock), respectively. In addition, 292 and 556 samples were diagnosed as noninfection (non-SIRS and SIRS) and infection (infection without SIRS, sepsis, severe sepsis and septic shock), respectively. Median presepsin, procalcitonin, interleukin-6 and C-reactive protein levels in systemic inflammatory response syndrome, sepsis, severe sepsis and septic shock
The presepsin levels were measured in 456 samples, and the values of the measurements indicated the presence of SIRS, sepsis, severe sepsis and septic shock (Fig. 1). There was a significant correlation between presepsin levels and severity. The presepsin levels in patients with SIRS were significantly lower than those with sepsis (P ¼ 0.0004), severe sepsis (P < 0.0001) and septic shock (P < 0.0001). Conversely, the PCT levels in patients with SIRS were significantly lower than those in patients with severe sepsis (P ¼ 0.0002) or septic shock (P < 0.0001), although no difference in the PCT levels was observed between patients with SIRS and those with sepsis (P ¼ 0.3654). The IL-6 levels in patients with SIRS were Table 1
significantly lower than those in patients with septic shock (P < 0.0001), although no differences in the IL-6 levels were observed between patients with SIRS and those with sepsis or between patients with sepsis and those with severe sepsis. Although the CRP levels in patients with SIRS were significantly lower than those in patients with sepsis (P < 0.0001), severe sepsis (P < 0.0001) or septic shock (P < 0.0001), the levels did not correlate with severity. To discriminate between patients with SIRS and sepsis (including sepsis, severe sepsis and septic shock), the diagnostic accuracies of presepsin, PCT, IL-6 and CRP were analysed using ROC curves (Table 2). The area under the curve (AUC) and likelihood ratio for presepsin (0.89 and 4.75, respectively) were higher than those for PCT, IL-6 and CRP. The AUC and likelihood ratio generated by combining presepsin and PCT (0.91 and 4.96, respectively) were higher than those for presepsin or PCT alone. Relationships of time-course changes between presepsin, procalcitonin, interleukin-6 and C-reactive protein levels and severity in patients with infection
Figure 2a shows the relationships of the time-course changes between the presepsin, PCT, IL-6 and CRP levels and the SOFA scores indicating severity in 100 survivors with infection. The presepsin levels and SOFA scores seemed to decrease slightly from days 0 to 7. There was no significantly different pattern for the changes between the presepsin level and the SOFA score (P ¼ 0.841 for interaction). The PCT, IL-6 and CRP levels at day 7 tended to be much lower than those at day 0. There were significantly different patterns for the changes between the PCT, IL-6 and CRP levels and the SOFA score (P < 0.01 for interaction). To confirm the relationships of the time-course changes between the presepsin, PCT, IL-6 and CRP levels and the SOFA score, we investigated the Spearman correlation coefficients between these biomarkers and the SOFA score.
Patient characteristics
APACHE II SOFA Diagnosis GD Respiratory disease Trauma CD Renal disease Burns CH MD HD CP Drug poisoning Others Total of points (%)
Non-SIRS
SIRS
Infection without SIRS
Sepsis
Severe sepsis
Septic shock
P
11 3
15 5
12 4
15 5
19 10
21 10
