Acta Anaesthesiol Scand 2014; 58: 525–533 Printed in Singapore. All rights reserved
© 2014 The Acta Anaesthesiologica Scandinavica Foundation. Published by John Wiley & Sons Ltd ACTA ANAESTHESIOLOGICA SCANDINAVICA
doi: 10.1111/aas.12290
Thromboelastometry as a supplementary tool for evaluation of hemostasis in severe sepsis and septic shock M. G. Andersen1, C. L. Hvas1, E. Tønnesen1 and A-M. Hvas2
1 Department of Anaesthesia and Intensive Care Medicine and 2Centre for Hemophilia and Thrombosis, Department of Clinical Biochemistry, Aarhus University Hospital, Aarhus, Denmark
Background: Sepsis leads to disruption of hemostasis, making early evaluation of coagulation essential. The aim of this study was to provide a detailed investigation of coagulation and the use of blood products in patients with severe sepsis or septic shock, admitted to a multidisciplinary intensive care unit. Methods: Thirty-six patients with severe sepsis or septic shock were included in this prospective observational study. Blood samples and information on transfusion of blood products were obtained for up to 3 consecutive days, and day 7 if the patient was still in the intensive care unit. Thromboelastometry (ROTEM®), analyses of thrombin generation, and conventional coagulation tests were performed. Results: ROTEM® revealed an overall normo-coagulable state among patients with severe sepsis or septic shock. Conventional coagulation analyses showed divergent results with hypercoagulable trends in terms of reduced antithrombin and acute phase response with increased fibrinogen and fibrin d-dimer, and on the other hand, coagulation disturbances with a decreased pro-
thrombin time and prolonged activated partial thromboplastin time. This hypocoagulabe state was supported by a delayed and reduced thrombin generation. Twelve patients experienced 21 independent transfusion episodes with fresh frozen plasma. Of these, only five (22%) transfusions were performed because of active bleeding. Conclusion: ROTEM® demonstrated an overall normocoagulation, whereas the conventional coagulation tests and thrombin generation analyses mainly reflected hypocoagulation. Given the dynamic and global features of ROTEM®, this analysis may be a relevant supplementary tool for the assessment of hemostasis in patients with severe sepsis or septic shock.
C
The complexity of pro- and anticoagulant changes hampers the hemostatic treatment of septic patients.4 Therapeutic approaches, such as transfusion with fresh frozen plasma (FFP), are widely based on conventional coagulation tests such as prothrombin time (PT) and activated partial thromboplastin time (aPTT). Yet, whether these tests predict hemorrhage has been intensively debated.5–7 Several studies suggest that abnormal test results do not necessarily represent clinically important coagulation factor deficiencies, and this may lead to inappropriate transfusions.5,8 Conventional coagulation tests as PT and aPTT only reflect limited parts of the coagulation system.9 To assess the overall balance in the hemostatic system and improve diagnosis as well as targeted therapeutic strategies, global tests of hemostasis could be valuable additional diagnostic tests.
oagulation abnormalities are common in critically ill patients1 and constitute a considerable therapeutic challenge in the intensive care unit (ICU).2 A frequent cause of coagulation disorders in the ICU is sepsis.3 Sepsis induces a procoagulant state, and the resultant hypercoagulability may in severe cases accelerate leading to disseminated intravascular coagulation (DIC).3 The excessive activation of coagulation involves consumption of platelets and coagulation factors, which may shift the hypercoagulant state into a hypocoagulant state, ultimately introducing the complication of hemorrhage.3 IRB:The study was approved by the Danish Data Protection Agency. The study was considered a validation of diagnostic methods; therefore, it was not necessary to notify the ethics committee, in accordance with Danish Committee Legislation
Accepted for publication 28 January 2014 © 2014 The Acta Anaesthesiologica Scandinavica Foundation. Published by John Wiley & Sons Ltd
525 bs_bs_banner
M. G. Andersen et al.
Thromboelastometry/graphy (ROTEM®/TEG®) are point-of-care devices that evaluate viscoelastic changes during coagulation,10 and may provide valuable additional information on coagulation in patients with severe sepsis or septic shock.11,12 Determination of thrombin generation is another method to analyze global hemostasis. Thus, useful information may be provided on hemostasis by assessing the velocity and extent of thrombin generation.13,14 The present study aimed to assess hemostasis during severe sepsis and septic shock, and to describe the use of transfusion of blood products in these patients.
Materials and methods The study was approved by the Danish Data Protection Agency (journal number 1-16-02-179-11). In accordance with Danish Committee Legislation, notification to the ethics committee was not required. The study was conducted as a prospective, observational study in a multidisciplinary ICU of Aarhus University Hospital; 36 patients with the diagnosis of severe sepsis or septic shock were enrolled from November 2011 to May 2012. The diagnosis of sepsis was based on the consensus criteria of the American College of Chest Physicians/Society of Critical Care Medicine.15 Severe sepsis was defined as sepsis associated with hypoperfusion, hypotension, or organ dysfunction; criteria for organ dysfunction were based on a modification of consensus criteria established by Bernard et al.16 Septic shock was defined as persistent hypotension, despite adequate fluid resuscitation, along with the presence of reduced perfusion. Diagnosis had to be made within the 24 h prior to study inclusion. Exclusion criteria were age less than 18 years, active cancer, or treatment with chemotherapy within the last 3 months, treatment with vitamin K antagonists, trauma, or pregnancy. The registration of clinical data and blood sampling was performed on the day of inclusion (day 1), the following 2 days (days 2 and 3), and on day 7 if the patient was still in the ICU. The primary outcome was changes in clotting time (CT) in the EXTEM-assay from day 1 to day 3. On the basis of previously published data on patients with sepsis, we estimated the mean CT to be 65 s with a standard deviation of 10 s.12 We chose the minimum relevant difference to be 7 s. With a significance level at 5% (2α) and a test power at 90% (1-β), we had to include 22 patients. We chose to
526
include 36 patients to be sure to obtain enough data on day 3.
Clinical data Demographic data and source of infection were obtained from medical records. We assessed comorbidity using Charlson’s comorbidity index.17 Type {FFP, erythrocyte-suspension [red blood cells (RBC)], and platelet concentrates} and volume of blood transfusions were noted from observation charts for all days of study. Treatment with heparin was also recorded. The simplified acute physiology score (SAPS II)18 and acute physiology and chronic health evaluation (APACHE II)19 were determined 24 h after inclusion. The sequential organ failure assessment (SOFA) score20 was determined daily.
Blood sampling Blood samples for ROTEM® analyses, conventional coagulation analyses, and thrombin generation analyses were drawn from an arterial cannula into 3.6 ml tubes anticoagulated with 3.2% sodium citrate. The first tube was discarded. Samples for thrombin generation were centrifuged at 3163g for 25 min at 18 °C, and platelet poor plasma was frozen at −80 °C until analysis.
Thromboelastometry Dynamic whole blood coagulation profiles were analyzed by thromboelastometry using a ROTEM® analyzer (Tem International GmbH, Munich, Germany). Blood samples were stored at room temperature until analysis 30 min after collection. Three hundred microliters of citrated, recalcified whole blood was used in four channels, and four standard assays were performed: INTEM, EXTEM, FIBTEM, and HEPTEM. We obtained the dynamic parameters of clot initiation (CT, s) and clot propagation (maximum velocity of clot formation: MaxVel, mm × 100/s, time to maximum velocity: t, MaxVel, s). Whole blood clot strength was assessed by maximum clot firmness (MCF, mm × 100). As an indicator for fibrinolysis the lysis index after 45 min (LI45) was recorded. Reference intervals for the ROTEM® parameters were established by Department of Clinical Biochemistry, Aarhus University Hospital, Denmark, based on blood samples from 75 healthy subjects with an equal gender distribution aged 20–60 years.
Laboratory analyses Conventional coagulation tests included PT, international normalized ratio, aPTT, fibrinogen
Hemostasis in sepsis
(functional), fibrin d-dimer (all performed on STA-R Evolution® analyzer, Diagnostica Stago SAS, Asniéres, France), and platelet count (Sysmex 2100XE, Sysmex Europe GmbH, Norderstedt, Germany). White blood cell count, plasmacreatinine, plasma-bilirubin, and blood-hemoglobin were measured daily.
Thrombin generation Thrombin generation was evaluated by calibrated automated thrombography (CAT; Thrombinoscope BV, Maastricht, the Netherlands) performed using platelet poor plasma. The assay was conducted according to the manufacturer’s instructions. The following parameters were analyzed: lag time until initial thrombin generation (min), maximum concentration of thrombin (peak, nM), time to peak (ttpeak, min), and the endogenous thrombin potential (ETP, nM × min). Reference intervals were obtained from a study including 32 healthy subjects.21
Statistical analyses Most data did not follow normal distribution. Therefore, values are presented as median and range. To test the difference in laboratory values between day 1 and day 3 we used the non-parametric paired test Wilcoxon signed rank test. Subgroup analyses were performed with a Wilcoxon rank-sum test for unpaired data. P-values below 0.05 were considered statistical significant. Statistical analyses were performed using Stata statistical software, version 11.2 (StataCorp LP, College Station, TX, USA) and figures performed using Sigmaplot 11.1 (Systat Software GmbH, Erhrath, Germany).
Results Thirty-six patients were enrolled in the study and followed the flow chart presented in Fig. 1. Characteristics of the included patients are shown in Table 1. In the majority of patients sources of infection were abdominal [17 (47%) patients] or pulmo-
Fig. 1. Flow chart of 36 patients at an intensive care unit (ICU) with severe sepsis or septic shock.
527
M. G. Andersen et al.
nary [16 (44%) patients]. Remarkably, only six (17%) patients needed dialysis, and only eight (22%) patients died within the first 7 days (Table 1 and Fig. 1). However, the 30-day mortality was relatively high as 14 (36%) patients died within the first 30 days after inclusion. Overall, most ROTEM® results remained within the reference intervals indicating normo-
coagulability. However, MCF in FIBTEM, EXTEM, and INTEM were slightly increased or in the high end of the reference interval, as was MaxVel of the EXTEM assay (Table 2). Procoagulant tendencies were reflected by the conventional coagulation analyses in terms of increased fibrinogen the first three days and persistently reduced levels of antithrombin (Table 3). In contrast, PT and aPTT showed mild coagulation disturbances indicated by an early and persistent decrease in median PT and a slightly prolonged aPTT on days 1 and 2 (Table 3). While the majority of the patients (69–80%) had PT values below the reference interval at least 1 day, a prolonged EXTEM CT was found in just a few during their stay at ICU (3–5%). No results were affected by heparin. This was confirmed by the ROTEM® analyses as no differences were found between the CT in INTEM or HEPTEM (Table 2). No therapeutic anticoagulation was administered during the study period. Platelet count declined from the lower level of the reference interval on day 1, to values below the reference interval on day 3 (Table 3). All median parameters of thrombin generation assays demonstrated a reduced thrombin generation compared with the values from healthy controls (Table 4).21 The reduced thrombin generation was present at inclusion and persisted for all days analyzed. Delay of activation and propagation of hemostasis was indicated by prolonged lag time and
Table 1 Characteristics of 36 patients with severe sepsis or septic shock. Variables Age, years Gender, n (%) Type of ICU admission, n (%) Charlson’s comorbidity score Dialysis, n (%) SAPS II (Scale = 0–163 points) APACHE II score (Scale = 0–71 points) SOFA score (Scale = 0–24 points) Day 1 Day 2 Day 3 Day 7
65 (47–90) 17 females (47)/19 males (53) 24 medical (67)/12 surgical (33) 1 (0–5) 6 (17) 45 (29–96) 19 (6–41)
8 (2–21) 8 (2–21) 9 (2–23) 10 (5–21)
Data are presented as the median (range) or n (%) ICU, intensive care unit; SAPS, simplified acute physiology score; APACHE, acute physiology and chronic health evaluation; SOFA, sequential organ failure assessment.
Table 2 Results of thromboelastometry among 36 patients with severe sepsis or septic shock. Median and (range) are indicated for day 1, day 2, day 3, and day 7. Parameter EXTEM CT (s) MaxVel (mm x 100/s) t, MaxVel (s) MCF (mm) LI45 (%) INTEM CT (s) MaxVel (mm x 100/s) t, MaxVel (s) MCF (mm) FIBTEM MCF (mm) HEPTEM CT ratio
Reference interval
Day 1 n = 36
Day 2 n = 28
Day 3 n = 22
P-values*
Day 7 n=5
38–74 8–22 48–154 48–66 > 85%
53 (36–85) 20 (10–32) 74 (46–144) 67 (46–82) 96 (77–100)
52 (37–83) 17 (6–35) 64 (43–139) 64 (40–78) 97 (89–100)
51 (43–77) 19 (5–34) 64 (49–196) 63 (37–80) 99 (91–100)
0.85 0.63 0.97 0.06 0.04
51 (42–59) 14 (8–38) 57 (49–108) 52 (37–82) 98 (92–100)
129–181 11–25 147–223 51–67
165 (112–208) 21 (10–39) 196 (132–244) 66 (43–81)
174 (116–341) 18 (7–38) 199 (144–367) 62 (36–79)
163 (129–210) 18 (7–38) 184 (148–248) 62 (30–78)
0.43 0.31 0.10 0.01
183 (149–221) 14 (9–42) 185 (156–262) 52 (37–84)
0.25
17 (12–64)
8–20
21 (6–48) 1.0
23 (6–50) 1.0
24 (5–53) 1.0
0.98
The reference interval for LI45 is according to Brenner et al.22 *Wilcoxon signed rank test comparing day 1 and day 3. CT, clotting time; MaxVel, maximum velocity; t, MaxVel, time to maximum velocity; MCF, maximum clot firmness; LI45, Lysis index after 45 min.
528
Hemostasis in sepsis Table 3 Conventional coagulation tests and other laboratory investigations among 36 patients with severe sepsis or septic shock. Median and (range) are indicated for day 1, day 2, day 3, and day 7. Parameter
Reference interval
Day 1 n = 36
Day 2 n = 28
Day 3 n = 22
P-values*
Day 7 n=5
Platelet count (× 109/L) Prothrombin time aPTT (s) INR Antithrombin (U/mL) Fibrinogen (μmol/L) Fibrin d-dimer (mg/L) WBC count (× 109/L) CRP (mg/L) Hemoglobin (mmol/L)† – Males – Females
145–350 0.70–1.30 25–38 < 1.2 0.80–1.20 5.5–12.0 < 0.50 3.5–10 < 10 mg/L
196 (16–509) 0.50 (0.10–0.89) 39 (27–69) 1.3 (1.0–4.0) 0.57 (0.12–1.21) 12.5 (4.2–27.6) 3.65 (1.04–20.00) 15.6 (0.6–53.5) 137 (59–216)
149 (3–493) 0.46 (0.17–0.93) 39 (27–70) 1.4 (1.0–2.6) 0.61 (0.17–1.19) 13.5 (4.3–24.8) 4.35 (1.10–20.00) 14.3 (3.6–42.7) 109 (67–267)
120 (3–502) 0.49 (0.19–0.89) 37 (27–50) 1.3 (1.0–2.4) 0.49 (0.19–1.05) 12.1 (3.7–25.2) 3.55 (1.60–20.00) 13.4 (5.5–44.0) 131 (59–276)
0.02 0.66 0.43 0.97 0.22 0.05 0.11 0.25 0.83
91 (16–513) 0.54 (0.18–0.90) 29 (27–57) 1.3 (1.0–2.5) 0.68 (0.32–1.18) 9.2 (6.6–24.2) 7.70 (2.30–15.60) 13.5 (6.2–15.7) 165 (28–145)
8.3–10.5 7.3–9.5
7.0 (4.4–9.9) 6.7 (5.0–8.2)
6.3 (5.4–9.0) 6.0 (5.2–7.8)
5.7 (4.5–7.2) 5.8 (5.0–7.6)
0.01 0.41
5.3 (4.5–6.0) 6.5 (6.2–6.8)
*Wilcoxon signed rank test comparing day 1 and day 3. †Hemoglobin values were missing for two patients on day 1, day 2 and day 3, and for one patient on day 7. CRP, c-reactive protein; PT, prothrombin time; aPPT, activated partial thromboplastin time; INR, international normalized ratio; WBC, white blood cell.
time to peak. A deficit in thrombin generation was indicated by reduced peak thrombin and EPT. No statistically significant differences were found in conventional coagulation tests, ROTEM® analysis or results of thrombin generation when comparing medical and surgical patients or when comparing patients with abdominal versus pulmonary focus (data not shown). Twenty-five patients had severe sepsis and 11 patients had septic shock. Comparing conventional coagulation tests, ROTEM®, and thrombin generation analyses among these two groups showed that only CT in the EXTEM assay differed, being significantly longer in the group with septic shock (P = 0.03). The test results did not differ substantially between the 14 (36%) patients who died within 30 days after inclusion and the survivors (data not shown). Further subgroup analyses were performed dividing the patients into a group of non-overt DIC and overt DIC according to the criteria made by the International Society of Thrombosis and Haemostasis,23 Table 5. Few patients had overt DIC. In the EXTEM and INTEM assays, patients with overt DIC showed a prolonged CT and a reduced MaxVel compared with patients with out DIC, although not statistically significant. The FIBTEM MCF was reduced in patients with overt DIC, though only significantly on day 3. Only one patient showed signs of hyperfibrinolysis, and this patient did not have DIC. Patients with overt DIC seemed to have a lower degree of fibrinolysis indicated by a lower LI45 than
patients with non-DIC (P-values = 0.01–0.06). Furthermore, the maximum concentration of thrombin (peak) was reduced in patients with overt DIC (P-values = 0.02–0.11). Twelve patients experienced 21 independent transfusion episodes with FFP. Patients treated with FFP had significantly lower levels of PT on admission compared with patients not receiving FFP [0.37 (0.10–0.84) vs. 0.61 (0.19–0.89) (P = 0.009)]. In contrast, median EXTEM CT among treated patients was in the middle of the reference interval [56 s (42–67 s)], not differing from EXTEM CT among the non-transfused patients [49 s (36– 85 s) P = 0.12]. Neither did median aPTT on admission differ between patients treated with FFP compared with the values of non-treated patients [40 s (28–69) vs. 38 (27–47) s, P = 0.16]. As expected, platelet counts and hemoglobin values were lower in patients receiving platelet concentrate and RBC (data not shown). The indications for administration of RBC, FFP, or platelet concentrate are shown in Fig. 2. Of the 21 transfusion episodes with FFP, five (22%) were motivated by overt hemorrhage. Eight (39%) transfusion episodes were a preventive measure prior to an invasive procedure. In the remaining eight (39%) episodes, the indication was unclear or seemed to be based on a reduced value of PT or prolonged aPTT. PT remained unchanged after transfusion with FFP. The median score of SAPS and SOFA on day 1 was significantly higher among patients transfused with
529
530 4.4 140 978 7.3
Median
(0.9) (100) (281) (1.2)
(SD)
Mean
2.4 454 1681 4.2
Day 1 n = 33
Control subjects n = 32 (2.3–24.7) (14–267) (173–2126) (4.6–27.9)
(Range) 4.9 164 1026 8.0
Median
Day 2 n = 27 (Range) (2.3–33.5) (1–275) (0–1946) (4.8–34.7)
4.7 177 1139 7.2
Median
Day 3 n = 21 (Range) (2.3–40.7) (5–260) (0–1996) (4.4–52.4)
P-values† 0.35 0.35 0.60 0.21
7.4 135 817 9.7
Median
Day 4 n=4 (6.7–8.4) (130–194) (609–873) (9.3–11)
(Range)
64 (61–67) 14 (10–17) 67 (65–69) 100 (99–100) 196 (185–206) 12 (11–12) 208 (202–214) 17 (13–20) 64 (14–114) 12.1 (5.6–18.6)
147 (33–314) 4.4 (2.3–24.7)
0.06 0.15
0.07 0.11 0.68 0.06 0.06 0.03 0.37 0.32
177 (8–300) 4.9 (2.3–11.3)
49 (37–61) 19 (6–35) 64 (43–139) 95 (89–100) 166 (116–197) 22 (7–38) 198 (144–259) 25 (6–50)
Non-DIC (n = 23)
53 (36–85) 21 (10–32) 77 (46–144) 96 (77–100) 163 (112–208) 21 (10–39) 196 (132–244) 22 (6–48)
Day 2 P-values*
Non-DIC (n = 34)
Overt DIC (n = 2)
Day 1
*Mann–Whitney test used for comparison of non-DIC and overt DIC. CT, clotting time; MaxVel, maximum velocity; MCF, maximum clot firmness.
ROTEM® EXTEM, CT (s) EXTEM, MaxVel (mm × 100/s) EXTEM, tMaxVel (s) EXTEM, Lysis45 (%) INTEM, CT (s) INTEM, MaxVel (mm × 100/s) INTEM, tMaxVel (s) FIBTEM, MCF (mm) Thrombin generation Peak thrombin (nM) Lag time (min)
Parameter
82 (1–206) 6.8 (3.8–33.5)
61 (53–83) 15 (9–17) 64 (55–103) 100 (98–100) 186 (163–341) 11 (9–13) 199 (175–367) 18 (12–30)
Overt DIC (n = 5)
0.11 0.14
0.004 0.07 0.98 0.01 0.13 0.005 0.47 0.20
P-values*
211 (5–300) 4.7 (3.0–40.7)
49 (43–59) 24 (8–34) 64 (49–153) 97 (91–100) 156 (129–210) 22 (7–38) 183 (148–248) 31 (10–53)
Non-DIC (n = 15)
Day 3
129 (18–177) 5.1 (2.3–15.3)
57 (48–77) 18 (5–19) 65 (52–196) 100 (99–100) 183 (161–206) 13 (9–16) 199 (169–244) 15 (5–31)
Overt DIC (n = 7)
0.02 0.59
0.06 0.06 0.46 0.01 0.02 0.003 0.31 0.02
P-values*
Results of thromboelastometry and thrombin generation in 36 patients with severe sepsis or septic shock comparing patients with no disseminated intravascular coagulation (DIC) and overt DIC. Median and (range) are indicated for day 1, day 2, and day 3.
Table 5
*Values from control subjects were obtained from a previous study on thrombin generation in which patients with severe sepsis are compared with healthy controls.21 †Wilcoxon signed rank test comparing day 1 and day 3. Data from one patient from days 1–7 and two patients on day 1 are missing because of error in the analyses. ETP, endogenous thrombin potential.
Lag time (min) Peak thrombin (nM) ETP (nM × min) Time to peak (min)
Parameter
Thrombin generation in 36 patients with severe sepsis or septic shock.
Table 4
M. G. Andersen et al.
Hemostasis in sepsis
Fig. 2. The indications for administration of erythrocyte suspension, fresh frozen plasma, or platelet concentrate among patients with severe sepsis or septic shock.
FFP compared with non-transfused patients [SAPS II: 60 (36–96) vs. 40 (29–86) (P = 0.02) and SOFA day 1: 14 (6–21) vs. 7 (2–17) (P = 0.0009)]. The median APACHE II score was not significantly higher in transfused compared with non-transfused patients [APACHE II: 23 (13–41) vs. 16 (6–34) (P = 0.12)].
Discussion In the present study, global and dynamic analysis of hemostasis performed by ROTEM® showed an overall normal hemostasis in patients with severe sepsis or septic shock. In contrast, a decreased PT, slightly prolonged aPTT, and a reduced platelet count and delayed thrombin generation indicated mild coagulation disturbances among the patients. Conventional coagulations analyses as PT and aPTT are routinely used in the clinic to assess hemostatic status. A decrease in PT and prolongation of aPTT may contribute to a clinician’s decision to treat with FFP.2 However, PT and aPTT only reflect the initiation of clot formation, and do not reflect propagation or termination of clot formation. Furthermore, previous studies indicate that abnormal PT and aPTT values are poor predictors of hemorrhage and may lead to overestimating the need for transfusion therapy.5,8 Analyses by ROTEM® may differ from the results obtained by the conventional coagulations analyses as the ROTEM® assays also reflect clot propagation and termination in whole blood. Therefore, analyses performed by ROTEM® may provide additional useful information on the coagulation status of the patient. In
accordance, other studies have shown a decrease in the amount of blood components transfused after implementation of ROTEM® in clinical situations such as surgery.24,25 As a consequence, some centers base perioperative transfusion of blood products on an algorithm using ROTEM® or TEG® (Haemonetics Corp, Braintree, MA, USA).26 In the present study, as well as seen elsewhere,27 more than one-third of the FFP transfusion episodes were without relation to active bleeding or invasive procedures. In these cases, the transfusions seemed to be given solely on the basis of abnormal coagulation test results, i.e. low values of PT, while ROTEM® parameters remained within the reference interval. In this context, the contradictory normocoagulability reflected by ROTEM® results may indicate inappropriate use of FFP transfusions. While ROTEM® parameters generally remained within reference interval indicating an overall normo-coagulable state, PT and aPTT showed mild coagulation abnormalities. This incongruence between these analyses has previously been demonstrated.12,28 EXTEM CT and the velocity parameters t, MaxVel and MaxVel are considered useful surrogate measures for thrombin generation when examined in healthy volunteers.29 The reduced and delayed thrombin generation with concomitant normocoagulable ROTEM® parameters in this and the study by Massion et al.11 indicate that this might not be the case in severe sepsis and septic shock. Two other studies reported fluorogenic thrombin generation assays in cases of severe sepsis.14,22 As opposed to our study, they demonstrated a delayed thrombin generation but not a decrease in peak thrombin, indicating that the amplification and total amount of thrombin generation were not affected in severe sepsis. The decreased PT, delayed and decreased thrombin generation, and the prolonged aPTT indicated a depletion of coagulation factors, which may be due to increased consumption, representing an overactivation of the coagulation system, induced by the severe sepsis. These findings are consistent with the study by Massion et al. including patients with septic shock admitted to the ICU.11 Furthermore, that study also revealed ROTEM® parameters within the reference interval concomitant with delayed and reduced thrombin generation. Massion et al. argued that increased fibrinogen might mask the decreased platelet count and other consumed factors in the ROTEM® measurements. As opposed to the present study, a previous study in patients with severe sepsis reported a
531
M. G. Andersen et al.
marked and general hypercoagulability assessed by TEG®.28 The study by Gonano et al. was not conducted in an ICU, and the time delay from fulfillment of criteria to inclusion was a maximum of 6 h. Thus, these patients were included at an earlier stage of severe sepsis than the patients in our study. Other thromboelastometric investigations of septic patients in the ICU have demonstrated normo-12 or hypocoagulation,30 with the latter being associated with more advanced stages of disease.31 A previous study21 demonstrated a delay in initiation of clotting (prolonged CT) and an enhanced hemostasis once initiated (increased MaxVel and MCF). These findings are supported by the present study. However, we did not show substantial differences between patients having severe sepsis and septic shock. In the present study, only a few patients had overt DIC, but our results supported the recent findings by Brenner et al.22 that patients with overt DIC seems to be in a more hypocoagulant state as compared with patients with non-overt DIC. A limitation of the present study was the delay between fulfillment of sepsis criteria and inclusion in the study. This made it hard to investigate hemostasis in the very early phase of severe sepsis and to take the timing between initiation of the sepsis and blood sampling into account. This is problematic because of the very dynamic course of sepsis. In the present study, the sepsis-induced coagulopathy might be underestimated as the disease severity was only moderately increased. Finally, the results related to thrombin generation should be interpreted with caution because of the lack of locally established reference intervals for thrombin generation measurements. In conclusion, the present study showed that ROTEM® analyses reflected an overall normocoagulation among patients with severe sepsis or septic shock, while PT and aPTT and the analyses of thrombin generation showed mild coagulation abnormalities. Given the dynamic and global features of ROTEM®, it may be a valuable supplementary tool for the assessment of hemostasis in patients with severe sepsis or septic shock. Conflicts of interest: The authors declare no conflicts of interest. Funding: The A.P. Møller Foundation for the Advancement of Medical Science (A.P. Møller and wife Chastine Mc-Kinney Møller’s Foundation for General Purposes). Aase and Ejnar Danielsen’s Foundation.
532
References 1. Levi M, Opal SM. Coagulation abnormalities in critically ill patients. Crit Care 2006; 10: 222. 2. Gajic O, Dzilk WH, Toy P. Fresh frozen plasma and platelet transfusion for nonbleeding patients in the intensive care unit: benefit or harm? Crit Care Med 2006; 34: S170–3. 3. Levi M. The coagulant response in sepsis. Clin Chest Med 2008; 29: 627–42. 4. Hook KM, Abrams CS. The loss of homeostasis in hemostasis: new approaches in treating and understanding acute disseminated intravascular coagulation in critically ill patients. Clin Transl Sci 2012; 5: 85–92. 5. Segal JB, Dzik WH. Paucity of studies to support that abnormal coagulation test results predict bleeding in the setting of invasive procedures: an evidence-based review. Transfusion 2005; 45: 1413–25. 6. Dzik WH. Predicting hemorrhage using preoperative coagulation screening assays. Curr Hematol Rep 2004; 3: 324–30. 7. Stanworth SJ, Walsh TS, Prescott RJ, Lee RJ, Watson DM, Wyncoll D. A national study of plasma use in critical care: clinical indications, dose and effect on prothrombin time. Crit Care 2011; 15: R108. 8. Chowdary P, Saayman AG, Paulus U, Findlay GP, Collins PW. Efficacy of standard dose and 30 ml/kg fresh frozen plasma in correcting laboratory parameters of haemostasis in critically ill patients. Br J Haematol 2004; 125: 69–73. 9. Brummel KE, Paradis SG, Butenas S, Mann KG. Thrombin functions during tissue factor-induced blood coagulation. Blood 2002; 100: 148–52. 10. Ganter MT, Hofer CK. Coagulation monitoring: current techniques and clinical use of viscoelastic point-ofcare coagulation devices. Anesth Analg 2008; 106: 1366– 75. 11. Massion PB, Peters P, Ledoux D, Zimermann V, Canivet JL, Massion PP, Damas P, Gothot A. Persistent hypocoagulability in patients with septic shock predicts greater hospital mortality: impact of impaired thrombin generation. Intensive Care Med 2012; 38: 1326–35. 12. Daudel F, Kessler U, Folly H, Lienert JS, Takala J, Jakob SM. Thromboelastometry for the assessment of coagulation abnormalities in early and established adult sepsis: a prospective cohort study. Crit Care 2009; 13: R42. 13. Hemker HC, Giesen P, AlDieri R, Regnault V, de Smed E, Wagenvoord R, Lecompte T, Béguin S. The calibrated automated thrombogram (CAT): a universal routine test for hyper- and hypocoagulability. Pathophysiol Haemost Thromb 2002; 32: 249–53. 14. Petros S, Kliem P, Siegemund T, Siegemund R. Thrombin generation in severe sepsis. Thromb Res 2012; 129: 797–800. 15. Bone RC, Sibbald WJ, Sprung CL. The ACCP-SCCM consensus conference on sepsis and organ failure. Chest 1992; 101: 1481–2. 16. Bernard GR, Vincent JL, Laterre P, Larosa SP, Dhainaut JF, Lopez-Rodriguez A, Steingrub JS, Gerber GE, Helterbrand JD, Ely EW, Fisher CJ Jr. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344: 699–709. 17. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies. Development and validation. J Chronic Dis 1987; 40: 373–83. 18. Le Grall J, Lemeshow S, Saunlier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/ North American multicenter study. JAMA 1993; 270: 2957– 63.
Hemostasis in sepsis 19. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. Apache-II – A severity of disease classification-system. Crit Care Med 1985; 13: 818–29. 20. Vincent JL, Moreno R, Takala J, Willatts S, DeMendonca A, Bruining H, Reinhart CK, Suter PM, Thijs LG. The SOFA (sepsis-related organ failure assessment) score to describe organ dysfunction/failure. Intensive Care Med 1996; 22: 707–10. 21. Collins PW, Macchiavello LI, Lewis SJ, Macartney NJ, Saayman AG, Luddington R, Baglin T, Findlay GP. Global tests of haemostasis in critically ill patients with severe sepsis syndrome compared to controls. Br J Haematol 2006; 135: 220–7. 22. Brenner T, Schmidt K, Delang M, Mehrabi A, Bruckner T, Lichtenstern C, Martin E, Weigand MA, Hofer S. Viscoelastic and aggregometric point-of-care testing in patients with septic shock – cross-links between inflammation and haemostasis. Acta Anaesthesiol Scand 2012; 56: 1277–90. 23. Taylor FB, Toh CH, Hoots WK, Wada H, Levi M, On behalf of the Scientific Subcommittee on Disseminated Intravascular Coagulation (DIC) of the International Society on Thrombosis and Haemostasis (ISTH). Towards definition, clinical and laboratory criteria, and a scoring system for disseminated intravascular coagulation. Thromb Haemost 2001; 86: 1327– 30. 24. Romlin BS, Wahlander H, Berggren H, Synnergren M, Baghaei F, Nilsson K, Jeppsson A. Intraoperative thromboelastometry is associated with reduced transfusion prevalence in pediatric cardiac surgery. Anesth Analg 2011; 112: 30–6. 25. Shore-Lesserson L, Manspeizer HE, DePerio M, Francis S, Vela-Cantos F, Ergin MA. Thromboelastography-guided transfusion algorithm reduces transfusions in complex cardiac surgery. Anesth Analg 1999; 88: 312–9. 26. Lier H, Vorweg A, Hanke A, Görlinger K. Thromboelastometry guided therapy of severe bleeding. Hamostaseologie 2013; 33: 51–61.
27. Vlaar APJ, in der Maur AL, Binnekade JM, Schultz MJ, Juffermans NP. A survey of physicians’ reasons to transfuse plasma and platelets in the critically ill: a prospective singlecentre cohort study. Transfus Med 2009; 19: 207–12. 28. Gonano C, Sitzwohl C, Meitner E, Weinstabl C, Kettner SC. Four-day antithrombin therapy does not seem to attenuate hypercoagulability in patients suffering from sepsis. Crit Care 2006; 10: R160. 29. Rivard GE, Brummel-Ziedins KE, Mann KG, Fan L, Hofer A, Cohen E. Evaluation of the profile of thrombin generation during the process of whole blood clotting as assessed by thrombelastography. J Thromb Haemost 2005; 3: 2039–43. 30. Adamzik M, Langemeier T, Frey UH, Gorlinger K, Saner F, Eggebrecht H, Peters J, Hartmann M. Comparison of thrombelastometry with simplified acute physiology score II and sequential organ failure assessment scores for the prediction of 30-day survival: a cohort study. Shock 2011; 35: 339–42. 31. Ostrowski SR, Windelov NA, Ibsen M, Haase N, Perner A, Johansson PI. Consecutive thrombelastography clot strength profiles in patients with severe sepsis and their association with 28-day mortality: a prospective study. J Crit Care 2013; 28: e1–11.
Address: Anne-Mette Hvas Centre for Haemophilia and Thrombosis, Department of Clinical Biochemistry Aarhus University Hospital Brendstrupgårdsvej 100 DK-8200 Aarhus N Denmark e-mail: [email protected]
533
© 2014 The Acta Anaesthesiologica Scandinavica Foundation. Published by John Wiley & Sons Ltd ACTA ANAESTHESIOLOGICA SCANDINAVICA
doi: 10.1111/aas.12290
Thromboelastometry as a supplementary tool for evaluation of hemostasis in severe sepsis and septic shock M. G. Andersen1, C. L. Hvas1, E. Tønnesen1 and A-M. Hvas2
1 Department of Anaesthesia and Intensive Care Medicine and 2Centre for Hemophilia and Thrombosis, Department of Clinical Biochemistry, Aarhus University Hospital, Aarhus, Denmark
Background: Sepsis leads to disruption of hemostasis, making early evaluation of coagulation essential. The aim of this study was to provide a detailed investigation of coagulation and the use of blood products in patients with severe sepsis or septic shock, admitted to a multidisciplinary intensive care unit. Methods: Thirty-six patients with severe sepsis or septic shock were included in this prospective observational study. Blood samples and information on transfusion of blood products were obtained for up to 3 consecutive days, and day 7 if the patient was still in the intensive care unit. Thromboelastometry (ROTEM®), analyses of thrombin generation, and conventional coagulation tests were performed. Results: ROTEM® revealed an overall normo-coagulable state among patients with severe sepsis or septic shock. Conventional coagulation analyses showed divergent results with hypercoagulable trends in terms of reduced antithrombin and acute phase response with increased fibrinogen and fibrin d-dimer, and on the other hand, coagulation disturbances with a decreased pro-
thrombin time and prolonged activated partial thromboplastin time. This hypocoagulabe state was supported by a delayed and reduced thrombin generation. Twelve patients experienced 21 independent transfusion episodes with fresh frozen plasma. Of these, only five (22%) transfusions were performed because of active bleeding. Conclusion: ROTEM® demonstrated an overall normocoagulation, whereas the conventional coagulation tests and thrombin generation analyses mainly reflected hypocoagulation. Given the dynamic and global features of ROTEM®, this analysis may be a relevant supplementary tool for the assessment of hemostasis in patients with severe sepsis or septic shock.
C
The complexity of pro- and anticoagulant changes hampers the hemostatic treatment of septic patients.4 Therapeutic approaches, such as transfusion with fresh frozen plasma (FFP), are widely based on conventional coagulation tests such as prothrombin time (PT) and activated partial thromboplastin time (aPTT). Yet, whether these tests predict hemorrhage has been intensively debated.5–7 Several studies suggest that abnormal test results do not necessarily represent clinically important coagulation factor deficiencies, and this may lead to inappropriate transfusions.5,8 Conventional coagulation tests as PT and aPTT only reflect limited parts of the coagulation system.9 To assess the overall balance in the hemostatic system and improve diagnosis as well as targeted therapeutic strategies, global tests of hemostasis could be valuable additional diagnostic tests.
oagulation abnormalities are common in critically ill patients1 and constitute a considerable therapeutic challenge in the intensive care unit (ICU).2 A frequent cause of coagulation disorders in the ICU is sepsis.3 Sepsis induces a procoagulant state, and the resultant hypercoagulability may in severe cases accelerate leading to disseminated intravascular coagulation (DIC).3 The excessive activation of coagulation involves consumption of platelets and coagulation factors, which may shift the hypercoagulant state into a hypocoagulant state, ultimately introducing the complication of hemorrhage.3 IRB:The study was approved by the Danish Data Protection Agency. The study was considered a validation of diagnostic methods; therefore, it was not necessary to notify the ethics committee, in accordance with Danish Committee Legislation
Accepted for publication 28 January 2014 © 2014 The Acta Anaesthesiologica Scandinavica Foundation. Published by John Wiley & Sons Ltd
525 bs_bs_banner
M. G. Andersen et al.
Thromboelastometry/graphy (ROTEM®/TEG®) are point-of-care devices that evaluate viscoelastic changes during coagulation,10 and may provide valuable additional information on coagulation in patients with severe sepsis or septic shock.11,12 Determination of thrombin generation is another method to analyze global hemostasis. Thus, useful information may be provided on hemostasis by assessing the velocity and extent of thrombin generation.13,14 The present study aimed to assess hemostasis during severe sepsis and septic shock, and to describe the use of transfusion of blood products in these patients.
Materials and methods The study was approved by the Danish Data Protection Agency (journal number 1-16-02-179-11). In accordance with Danish Committee Legislation, notification to the ethics committee was not required. The study was conducted as a prospective, observational study in a multidisciplinary ICU of Aarhus University Hospital; 36 patients with the diagnosis of severe sepsis or septic shock were enrolled from November 2011 to May 2012. The diagnosis of sepsis was based on the consensus criteria of the American College of Chest Physicians/Society of Critical Care Medicine.15 Severe sepsis was defined as sepsis associated with hypoperfusion, hypotension, or organ dysfunction; criteria for organ dysfunction were based on a modification of consensus criteria established by Bernard et al.16 Septic shock was defined as persistent hypotension, despite adequate fluid resuscitation, along with the presence of reduced perfusion. Diagnosis had to be made within the 24 h prior to study inclusion. Exclusion criteria were age less than 18 years, active cancer, or treatment with chemotherapy within the last 3 months, treatment with vitamin K antagonists, trauma, or pregnancy. The registration of clinical data and blood sampling was performed on the day of inclusion (day 1), the following 2 days (days 2 and 3), and on day 7 if the patient was still in the ICU. The primary outcome was changes in clotting time (CT) in the EXTEM-assay from day 1 to day 3. On the basis of previously published data on patients with sepsis, we estimated the mean CT to be 65 s with a standard deviation of 10 s.12 We chose the minimum relevant difference to be 7 s. With a significance level at 5% (2α) and a test power at 90% (1-β), we had to include 22 patients. We chose to
526
include 36 patients to be sure to obtain enough data on day 3.
Clinical data Demographic data and source of infection were obtained from medical records. We assessed comorbidity using Charlson’s comorbidity index.17 Type {FFP, erythrocyte-suspension [red blood cells (RBC)], and platelet concentrates} and volume of blood transfusions were noted from observation charts for all days of study. Treatment with heparin was also recorded. The simplified acute physiology score (SAPS II)18 and acute physiology and chronic health evaluation (APACHE II)19 were determined 24 h after inclusion. The sequential organ failure assessment (SOFA) score20 was determined daily.
Blood sampling Blood samples for ROTEM® analyses, conventional coagulation analyses, and thrombin generation analyses were drawn from an arterial cannula into 3.6 ml tubes anticoagulated with 3.2% sodium citrate. The first tube was discarded. Samples for thrombin generation were centrifuged at 3163g for 25 min at 18 °C, and platelet poor plasma was frozen at −80 °C until analysis.
Thromboelastometry Dynamic whole blood coagulation profiles were analyzed by thromboelastometry using a ROTEM® analyzer (Tem International GmbH, Munich, Germany). Blood samples were stored at room temperature until analysis 30 min after collection. Three hundred microliters of citrated, recalcified whole blood was used in four channels, and four standard assays were performed: INTEM, EXTEM, FIBTEM, and HEPTEM. We obtained the dynamic parameters of clot initiation (CT, s) and clot propagation (maximum velocity of clot formation: MaxVel, mm × 100/s, time to maximum velocity: t, MaxVel, s). Whole blood clot strength was assessed by maximum clot firmness (MCF, mm × 100). As an indicator for fibrinolysis the lysis index after 45 min (LI45) was recorded. Reference intervals for the ROTEM® parameters were established by Department of Clinical Biochemistry, Aarhus University Hospital, Denmark, based on blood samples from 75 healthy subjects with an equal gender distribution aged 20–60 years.
Laboratory analyses Conventional coagulation tests included PT, international normalized ratio, aPTT, fibrinogen
Hemostasis in sepsis
(functional), fibrin d-dimer (all performed on STA-R Evolution® analyzer, Diagnostica Stago SAS, Asniéres, France), and platelet count (Sysmex 2100XE, Sysmex Europe GmbH, Norderstedt, Germany). White blood cell count, plasmacreatinine, plasma-bilirubin, and blood-hemoglobin were measured daily.
Thrombin generation Thrombin generation was evaluated by calibrated automated thrombography (CAT; Thrombinoscope BV, Maastricht, the Netherlands) performed using platelet poor plasma. The assay was conducted according to the manufacturer’s instructions. The following parameters were analyzed: lag time until initial thrombin generation (min), maximum concentration of thrombin (peak, nM), time to peak (ttpeak, min), and the endogenous thrombin potential (ETP, nM × min). Reference intervals were obtained from a study including 32 healthy subjects.21
Statistical analyses Most data did not follow normal distribution. Therefore, values are presented as median and range. To test the difference in laboratory values between day 1 and day 3 we used the non-parametric paired test Wilcoxon signed rank test. Subgroup analyses were performed with a Wilcoxon rank-sum test for unpaired data. P-values below 0.05 were considered statistical significant. Statistical analyses were performed using Stata statistical software, version 11.2 (StataCorp LP, College Station, TX, USA) and figures performed using Sigmaplot 11.1 (Systat Software GmbH, Erhrath, Germany).
Results Thirty-six patients were enrolled in the study and followed the flow chart presented in Fig. 1. Characteristics of the included patients are shown in Table 1. In the majority of patients sources of infection were abdominal [17 (47%) patients] or pulmo-
Fig. 1. Flow chart of 36 patients at an intensive care unit (ICU) with severe sepsis or septic shock.
527
M. G. Andersen et al.
nary [16 (44%) patients]. Remarkably, only six (17%) patients needed dialysis, and only eight (22%) patients died within the first 7 days (Table 1 and Fig. 1). However, the 30-day mortality was relatively high as 14 (36%) patients died within the first 30 days after inclusion. Overall, most ROTEM® results remained within the reference intervals indicating normo-
coagulability. However, MCF in FIBTEM, EXTEM, and INTEM were slightly increased or in the high end of the reference interval, as was MaxVel of the EXTEM assay (Table 2). Procoagulant tendencies were reflected by the conventional coagulation analyses in terms of increased fibrinogen the first three days and persistently reduced levels of antithrombin (Table 3). In contrast, PT and aPTT showed mild coagulation disturbances indicated by an early and persistent decrease in median PT and a slightly prolonged aPTT on days 1 and 2 (Table 3). While the majority of the patients (69–80%) had PT values below the reference interval at least 1 day, a prolonged EXTEM CT was found in just a few during their stay at ICU (3–5%). No results were affected by heparin. This was confirmed by the ROTEM® analyses as no differences were found between the CT in INTEM or HEPTEM (Table 2). No therapeutic anticoagulation was administered during the study period. Platelet count declined from the lower level of the reference interval on day 1, to values below the reference interval on day 3 (Table 3). All median parameters of thrombin generation assays demonstrated a reduced thrombin generation compared with the values from healthy controls (Table 4).21 The reduced thrombin generation was present at inclusion and persisted for all days analyzed. Delay of activation and propagation of hemostasis was indicated by prolonged lag time and
Table 1 Characteristics of 36 patients with severe sepsis or septic shock. Variables Age, years Gender, n (%) Type of ICU admission, n (%) Charlson’s comorbidity score Dialysis, n (%) SAPS II (Scale = 0–163 points) APACHE II score (Scale = 0–71 points) SOFA score (Scale = 0–24 points) Day 1 Day 2 Day 3 Day 7
65 (47–90) 17 females (47)/19 males (53) 24 medical (67)/12 surgical (33) 1 (0–5) 6 (17) 45 (29–96) 19 (6–41)
8 (2–21) 8 (2–21) 9 (2–23) 10 (5–21)
Data are presented as the median (range) or n (%) ICU, intensive care unit; SAPS, simplified acute physiology score; APACHE, acute physiology and chronic health evaluation; SOFA, sequential organ failure assessment.
Table 2 Results of thromboelastometry among 36 patients with severe sepsis or septic shock. Median and (range) are indicated for day 1, day 2, day 3, and day 7. Parameter EXTEM CT (s) MaxVel (mm x 100/s) t, MaxVel (s) MCF (mm) LI45 (%) INTEM CT (s) MaxVel (mm x 100/s) t, MaxVel (s) MCF (mm) FIBTEM MCF (mm) HEPTEM CT ratio
Reference interval
Day 1 n = 36
Day 2 n = 28
Day 3 n = 22
P-values*
Day 7 n=5
38–74 8–22 48–154 48–66 > 85%
53 (36–85) 20 (10–32) 74 (46–144) 67 (46–82) 96 (77–100)
52 (37–83) 17 (6–35) 64 (43–139) 64 (40–78) 97 (89–100)
51 (43–77) 19 (5–34) 64 (49–196) 63 (37–80) 99 (91–100)
0.85 0.63 0.97 0.06 0.04
51 (42–59) 14 (8–38) 57 (49–108) 52 (37–82) 98 (92–100)
129–181 11–25 147–223 51–67
165 (112–208) 21 (10–39) 196 (132–244) 66 (43–81)
174 (116–341) 18 (7–38) 199 (144–367) 62 (36–79)
163 (129–210) 18 (7–38) 184 (148–248) 62 (30–78)
0.43 0.31 0.10 0.01
183 (149–221) 14 (9–42) 185 (156–262) 52 (37–84)
0.25
17 (12–64)
8–20
21 (6–48) 1.0
23 (6–50) 1.0
24 (5–53) 1.0
0.98
The reference interval for LI45 is according to Brenner et al.22 *Wilcoxon signed rank test comparing day 1 and day 3. CT, clotting time; MaxVel, maximum velocity; t, MaxVel, time to maximum velocity; MCF, maximum clot firmness; LI45, Lysis index after 45 min.
528
Hemostasis in sepsis Table 3 Conventional coagulation tests and other laboratory investigations among 36 patients with severe sepsis or septic shock. Median and (range) are indicated for day 1, day 2, day 3, and day 7. Parameter
Reference interval
Day 1 n = 36
Day 2 n = 28
Day 3 n = 22
P-values*
Day 7 n=5
Platelet count (× 109/L) Prothrombin time aPTT (s) INR Antithrombin (U/mL) Fibrinogen (μmol/L) Fibrin d-dimer (mg/L) WBC count (× 109/L) CRP (mg/L) Hemoglobin (mmol/L)† – Males – Females
145–350 0.70–1.30 25–38 < 1.2 0.80–1.20 5.5–12.0 < 0.50 3.5–10 < 10 mg/L
196 (16–509) 0.50 (0.10–0.89) 39 (27–69) 1.3 (1.0–4.0) 0.57 (0.12–1.21) 12.5 (4.2–27.6) 3.65 (1.04–20.00) 15.6 (0.6–53.5) 137 (59–216)
149 (3–493) 0.46 (0.17–0.93) 39 (27–70) 1.4 (1.0–2.6) 0.61 (0.17–1.19) 13.5 (4.3–24.8) 4.35 (1.10–20.00) 14.3 (3.6–42.7) 109 (67–267)
120 (3–502) 0.49 (0.19–0.89) 37 (27–50) 1.3 (1.0–2.4) 0.49 (0.19–1.05) 12.1 (3.7–25.2) 3.55 (1.60–20.00) 13.4 (5.5–44.0) 131 (59–276)
0.02 0.66 0.43 0.97 0.22 0.05 0.11 0.25 0.83
91 (16–513) 0.54 (0.18–0.90) 29 (27–57) 1.3 (1.0–2.5) 0.68 (0.32–1.18) 9.2 (6.6–24.2) 7.70 (2.30–15.60) 13.5 (6.2–15.7) 165 (28–145)
8.3–10.5 7.3–9.5
7.0 (4.4–9.9) 6.7 (5.0–8.2)
6.3 (5.4–9.0) 6.0 (5.2–7.8)
5.7 (4.5–7.2) 5.8 (5.0–7.6)
0.01 0.41
5.3 (4.5–6.0) 6.5 (6.2–6.8)
*Wilcoxon signed rank test comparing day 1 and day 3. †Hemoglobin values were missing for two patients on day 1, day 2 and day 3, and for one patient on day 7. CRP, c-reactive protein; PT, prothrombin time; aPPT, activated partial thromboplastin time; INR, international normalized ratio; WBC, white blood cell.
time to peak. A deficit in thrombin generation was indicated by reduced peak thrombin and EPT. No statistically significant differences were found in conventional coagulation tests, ROTEM® analysis or results of thrombin generation when comparing medical and surgical patients or when comparing patients with abdominal versus pulmonary focus (data not shown). Twenty-five patients had severe sepsis and 11 patients had septic shock. Comparing conventional coagulation tests, ROTEM®, and thrombin generation analyses among these two groups showed that only CT in the EXTEM assay differed, being significantly longer in the group with septic shock (P = 0.03). The test results did not differ substantially between the 14 (36%) patients who died within 30 days after inclusion and the survivors (data not shown). Further subgroup analyses were performed dividing the patients into a group of non-overt DIC and overt DIC according to the criteria made by the International Society of Thrombosis and Haemostasis,23 Table 5. Few patients had overt DIC. In the EXTEM and INTEM assays, patients with overt DIC showed a prolonged CT and a reduced MaxVel compared with patients with out DIC, although not statistically significant. The FIBTEM MCF was reduced in patients with overt DIC, though only significantly on day 3. Only one patient showed signs of hyperfibrinolysis, and this patient did not have DIC. Patients with overt DIC seemed to have a lower degree of fibrinolysis indicated by a lower LI45 than
patients with non-DIC (P-values = 0.01–0.06). Furthermore, the maximum concentration of thrombin (peak) was reduced in patients with overt DIC (P-values = 0.02–0.11). Twelve patients experienced 21 independent transfusion episodes with FFP. Patients treated with FFP had significantly lower levels of PT on admission compared with patients not receiving FFP [0.37 (0.10–0.84) vs. 0.61 (0.19–0.89) (P = 0.009)]. In contrast, median EXTEM CT among treated patients was in the middle of the reference interval [56 s (42–67 s)], not differing from EXTEM CT among the non-transfused patients [49 s (36– 85 s) P = 0.12]. Neither did median aPTT on admission differ between patients treated with FFP compared with the values of non-treated patients [40 s (28–69) vs. 38 (27–47) s, P = 0.16]. As expected, platelet counts and hemoglobin values were lower in patients receiving platelet concentrate and RBC (data not shown). The indications for administration of RBC, FFP, or platelet concentrate are shown in Fig. 2. Of the 21 transfusion episodes with FFP, five (22%) were motivated by overt hemorrhage. Eight (39%) transfusion episodes were a preventive measure prior to an invasive procedure. In the remaining eight (39%) episodes, the indication was unclear or seemed to be based on a reduced value of PT or prolonged aPTT. PT remained unchanged after transfusion with FFP. The median score of SAPS and SOFA on day 1 was significantly higher among patients transfused with
529
530 4.4 140 978 7.3
Median
(0.9) (100) (281) (1.2)
(SD)
Mean
2.4 454 1681 4.2
Day 1 n = 33
Control subjects n = 32 (2.3–24.7) (14–267) (173–2126) (4.6–27.9)
(Range) 4.9 164 1026 8.0
Median
Day 2 n = 27 (Range) (2.3–33.5) (1–275) (0–1946) (4.8–34.7)
4.7 177 1139 7.2
Median
Day 3 n = 21 (Range) (2.3–40.7) (5–260) (0–1996) (4.4–52.4)
P-values† 0.35 0.35 0.60 0.21
7.4 135 817 9.7
Median
Day 4 n=4 (6.7–8.4) (130–194) (609–873) (9.3–11)
(Range)
64 (61–67) 14 (10–17) 67 (65–69) 100 (99–100) 196 (185–206) 12 (11–12) 208 (202–214) 17 (13–20) 64 (14–114) 12.1 (5.6–18.6)
147 (33–314) 4.4 (2.3–24.7)
0.06 0.15
0.07 0.11 0.68 0.06 0.06 0.03 0.37 0.32
177 (8–300) 4.9 (2.3–11.3)
49 (37–61) 19 (6–35) 64 (43–139) 95 (89–100) 166 (116–197) 22 (7–38) 198 (144–259) 25 (6–50)
Non-DIC (n = 23)
53 (36–85) 21 (10–32) 77 (46–144) 96 (77–100) 163 (112–208) 21 (10–39) 196 (132–244) 22 (6–48)
Day 2 P-values*
Non-DIC (n = 34)
Overt DIC (n = 2)
Day 1
*Mann–Whitney test used for comparison of non-DIC and overt DIC. CT, clotting time; MaxVel, maximum velocity; MCF, maximum clot firmness.
ROTEM® EXTEM, CT (s) EXTEM, MaxVel (mm × 100/s) EXTEM, tMaxVel (s) EXTEM, Lysis45 (%) INTEM, CT (s) INTEM, MaxVel (mm × 100/s) INTEM, tMaxVel (s) FIBTEM, MCF (mm) Thrombin generation Peak thrombin (nM) Lag time (min)
Parameter
82 (1–206) 6.8 (3.8–33.5)
61 (53–83) 15 (9–17) 64 (55–103) 100 (98–100) 186 (163–341) 11 (9–13) 199 (175–367) 18 (12–30)
Overt DIC (n = 5)
0.11 0.14
0.004 0.07 0.98 0.01 0.13 0.005 0.47 0.20
P-values*
211 (5–300) 4.7 (3.0–40.7)
49 (43–59) 24 (8–34) 64 (49–153) 97 (91–100) 156 (129–210) 22 (7–38) 183 (148–248) 31 (10–53)
Non-DIC (n = 15)
Day 3
129 (18–177) 5.1 (2.3–15.3)
57 (48–77) 18 (5–19) 65 (52–196) 100 (99–100) 183 (161–206) 13 (9–16) 199 (169–244) 15 (5–31)
Overt DIC (n = 7)
0.02 0.59
0.06 0.06 0.46 0.01 0.02 0.003 0.31 0.02
P-values*
Results of thromboelastometry and thrombin generation in 36 patients with severe sepsis or septic shock comparing patients with no disseminated intravascular coagulation (DIC) and overt DIC. Median and (range) are indicated for day 1, day 2, and day 3.
Table 5
*Values from control subjects were obtained from a previous study on thrombin generation in which patients with severe sepsis are compared with healthy controls.21 †Wilcoxon signed rank test comparing day 1 and day 3. Data from one patient from days 1–7 and two patients on day 1 are missing because of error in the analyses. ETP, endogenous thrombin potential.
Lag time (min) Peak thrombin (nM) ETP (nM × min) Time to peak (min)
Parameter
Thrombin generation in 36 patients with severe sepsis or septic shock.
Table 4
M. G. Andersen et al.
Hemostasis in sepsis
Fig. 2. The indications for administration of erythrocyte suspension, fresh frozen plasma, or platelet concentrate among patients with severe sepsis or septic shock.
FFP compared with non-transfused patients [SAPS II: 60 (36–96) vs. 40 (29–86) (P = 0.02) and SOFA day 1: 14 (6–21) vs. 7 (2–17) (P = 0.0009)]. The median APACHE II score was not significantly higher in transfused compared with non-transfused patients [APACHE II: 23 (13–41) vs. 16 (6–34) (P = 0.12)].
Discussion In the present study, global and dynamic analysis of hemostasis performed by ROTEM® showed an overall normal hemostasis in patients with severe sepsis or septic shock. In contrast, a decreased PT, slightly prolonged aPTT, and a reduced platelet count and delayed thrombin generation indicated mild coagulation disturbances among the patients. Conventional coagulations analyses as PT and aPTT are routinely used in the clinic to assess hemostatic status. A decrease in PT and prolongation of aPTT may contribute to a clinician’s decision to treat with FFP.2 However, PT and aPTT only reflect the initiation of clot formation, and do not reflect propagation or termination of clot formation. Furthermore, previous studies indicate that abnormal PT and aPTT values are poor predictors of hemorrhage and may lead to overestimating the need for transfusion therapy.5,8 Analyses by ROTEM® may differ from the results obtained by the conventional coagulations analyses as the ROTEM® assays also reflect clot propagation and termination in whole blood. Therefore, analyses performed by ROTEM® may provide additional useful information on the coagulation status of the patient. In
accordance, other studies have shown a decrease in the amount of blood components transfused after implementation of ROTEM® in clinical situations such as surgery.24,25 As a consequence, some centers base perioperative transfusion of blood products on an algorithm using ROTEM® or TEG® (Haemonetics Corp, Braintree, MA, USA).26 In the present study, as well as seen elsewhere,27 more than one-third of the FFP transfusion episodes were without relation to active bleeding or invasive procedures. In these cases, the transfusions seemed to be given solely on the basis of abnormal coagulation test results, i.e. low values of PT, while ROTEM® parameters remained within the reference interval. In this context, the contradictory normocoagulability reflected by ROTEM® results may indicate inappropriate use of FFP transfusions. While ROTEM® parameters generally remained within reference interval indicating an overall normo-coagulable state, PT and aPTT showed mild coagulation abnormalities. This incongruence between these analyses has previously been demonstrated.12,28 EXTEM CT and the velocity parameters t, MaxVel and MaxVel are considered useful surrogate measures for thrombin generation when examined in healthy volunteers.29 The reduced and delayed thrombin generation with concomitant normocoagulable ROTEM® parameters in this and the study by Massion et al.11 indicate that this might not be the case in severe sepsis and septic shock. Two other studies reported fluorogenic thrombin generation assays in cases of severe sepsis.14,22 As opposed to our study, they demonstrated a delayed thrombin generation but not a decrease in peak thrombin, indicating that the amplification and total amount of thrombin generation were not affected in severe sepsis. The decreased PT, delayed and decreased thrombin generation, and the prolonged aPTT indicated a depletion of coagulation factors, which may be due to increased consumption, representing an overactivation of the coagulation system, induced by the severe sepsis. These findings are consistent with the study by Massion et al. including patients with septic shock admitted to the ICU.11 Furthermore, that study also revealed ROTEM® parameters within the reference interval concomitant with delayed and reduced thrombin generation. Massion et al. argued that increased fibrinogen might mask the decreased platelet count and other consumed factors in the ROTEM® measurements. As opposed to the present study, a previous study in patients with severe sepsis reported a
531
M. G. Andersen et al.
marked and general hypercoagulability assessed by TEG®.28 The study by Gonano et al. was not conducted in an ICU, and the time delay from fulfillment of criteria to inclusion was a maximum of 6 h. Thus, these patients were included at an earlier stage of severe sepsis than the patients in our study. Other thromboelastometric investigations of septic patients in the ICU have demonstrated normo-12 or hypocoagulation,30 with the latter being associated with more advanced stages of disease.31 A previous study21 demonstrated a delay in initiation of clotting (prolonged CT) and an enhanced hemostasis once initiated (increased MaxVel and MCF). These findings are supported by the present study. However, we did not show substantial differences between patients having severe sepsis and septic shock. In the present study, only a few patients had overt DIC, but our results supported the recent findings by Brenner et al.22 that patients with overt DIC seems to be in a more hypocoagulant state as compared with patients with non-overt DIC. A limitation of the present study was the delay between fulfillment of sepsis criteria and inclusion in the study. This made it hard to investigate hemostasis in the very early phase of severe sepsis and to take the timing between initiation of the sepsis and blood sampling into account. This is problematic because of the very dynamic course of sepsis. In the present study, the sepsis-induced coagulopathy might be underestimated as the disease severity was only moderately increased. Finally, the results related to thrombin generation should be interpreted with caution because of the lack of locally established reference intervals for thrombin generation measurements. In conclusion, the present study showed that ROTEM® analyses reflected an overall normocoagulation among patients with severe sepsis or septic shock, while PT and aPTT and the analyses of thrombin generation showed mild coagulation abnormalities. Given the dynamic and global features of ROTEM®, it may be a valuable supplementary tool for the assessment of hemostasis in patients with severe sepsis or septic shock. Conflicts of interest: The authors declare no conflicts of interest. Funding: The A.P. Møller Foundation for the Advancement of Medical Science (A.P. Møller and wife Chastine Mc-Kinney Møller’s Foundation for General Purposes). Aase and Ejnar Danielsen’s Foundation.
532
References 1. Levi M, Opal SM. Coagulation abnormalities in critically ill patients. Crit Care 2006; 10: 222. 2. Gajic O, Dzilk WH, Toy P. Fresh frozen plasma and platelet transfusion for nonbleeding patients in the intensive care unit: benefit or harm? Crit Care Med 2006; 34: S170–3. 3. Levi M. The coagulant response in sepsis. Clin Chest Med 2008; 29: 627–42. 4. Hook KM, Abrams CS. The loss of homeostasis in hemostasis: new approaches in treating and understanding acute disseminated intravascular coagulation in critically ill patients. Clin Transl Sci 2012; 5: 85–92. 5. Segal JB, Dzik WH. Paucity of studies to support that abnormal coagulation test results predict bleeding in the setting of invasive procedures: an evidence-based review. Transfusion 2005; 45: 1413–25. 6. Dzik WH. Predicting hemorrhage using preoperative coagulation screening assays. Curr Hematol Rep 2004; 3: 324–30. 7. Stanworth SJ, Walsh TS, Prescott RJ, Lee RJ, Watson DM, Wyncoll D. A national study of plasma use in critical care: clinical indications, dose and effect on prothrombin time. Crit Care 2011; 15: R108. 8. Chowdary P, Saayman AG, Paulus U, Findlay GP, Collins PW. Efficacy of standard dose and 30 ml/kg fresh frozen plasma in correcting laboratory parameters of haemostasis in critically ill patients. Br J Haematol 2004; 125: 69–73. 9. Brummel KE, Paradis SG, Butenas S, Mann KG. Thrombin functions during tissue factor-induced blood coagulation. Blood 2002; 100: 148–52. 10. Ganter MT, Hofer CK. Coagulation monitoring: current techniques and clinical use of viscoelastic point-ofcare coagulation devices. Anesth Analg 2008; 106: 1366– 75. 11. Massion PB, Peters P, Ledoux D, Zimermann V, Canivet JL, Massion PP, Damas P, Gothot A. Persistent hypocoagulability in patients with septic shock predicts greater hospital mortality: impact of impaired thrombin generation. Intensive Care Med 2012; 38: 1326–35. 12. Daudel F, Kessler U, Folly H, Lienert JS, Takala J, Jakob SM. Thromboelastometry for the assessment of coagulation abnormalities in early and established adult sepsis: a prospective cohort study. Crit Care 2009; 13: R42. 13. Hemker HC, Giesen P, AlDieri R, Regnault V, de Smed E, Wagenvoord R, Lecompte T, Béguin S. The calibrated automated thrombogram (CAT): a universal routine test for hyper- and hypocoagulability. Pathophysiol Haemost Thromb 2002; 32: 249–53. 14. Petros S, Kliem P, Siegemund T, Siegemund R. Thrombin generation in severe sepsis. Thromb Res 2012; 129: 797–800. 15. Bone RC, Sibbald WJ, Sprung CL. The ACCP-SCCM consensus conference on sepsis and organ failure. Chest 1992; 101: 1481–2. 16. Bernard GR, Vincent JL, Laterre P, Larosa SP, Dhainaut JF, Lopez-Rodriguez A, Steingrub JS, Gerber GE, Helterbrand JD, Ely EW, Fisher CJ Jr. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344: 699–709. 17. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies. Development and validation. J Chronic Dis 1987; 40: 373–83. 18. Le Grall J, Lemeshow S, Saunlier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/ North American multicenter study. JAMA 1993; 270: 2957– 63.
Hemostasis in sepsis 19. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. Apache-II – A severity of disease classification-system. Crit Care Med 1985; 13: 818–29. 20. Vincent JL, Moreno R, Takala J, Willatts S, DeMendonca A, Bruining H, Reinhart CK, Suter PM, Thijs LG. The SOFA (sepsis-related organ failure assessment) score to describe organ dysfunction/failure. Intensive Care Med 1996; 22: 707–10. 21. Collins PW, Macchiavello LI, Lewis SJ, Macartney NJ, Saayman AG, Luddington R, Baglin T, Findlay GP. Global tests of haemostasis in critically ill patients with severe sepsis syndrome compared to controls. Br J Haematol 2006; 135: 220–7. 22. Brenner T, Schmidt K, Delang M, Mehrabi A, Bruckner T, Lichtenstern C, Martin E, Weigand MA, Hofer S. Viscoelastic and aggregometric point-of-care testing in patients with septic shock – cross-links between inflammation and haemostasis. Acta Anaesthesiol Scand 2012; 56: 1277–90. 23. Taylor FB, Toh CH, Hoots WK, Wada H, Levi M, On behalf of the Scientific Subcommittee on Disseminated Intravascular Coagulation (DIC) of the International Society on Thrombosis and Haemostasis (ISTH). Towards definition, clinical and laboratory criteria, and a scoring system for disseminated intravascular coagulation. Thromb Haemost 2001; 86: 1327– 30. 24. Romlin BS, Wahlander H, Berggren H, Synnergren M, Baghaei F, Nilsson K, Jeppsson A. Intraoperative thromboelastometry is associated with reduced transfusion prevalence in pediatric cardiac surgery. Anesth Analg 2011; 112: 30–6. 25. Shore-Lesserson L, Manspeizer HE, DePerio M, Francis S, Vela-Cantos F, Ergin MA. Thromboelastography-guided transfusion algorithm reduces transfusions in complex cardiac surgery. Anesth Analg 1999; 88: 312–9. 26. Lier H, Vorweg A, Hanke A, Görlinger K. Thromboelastometry guided therapy of severe bleeding. Hamostaseologie 2013; 33: 51–61.
27. Vlaar APJ, in der Maur AL, Binnekade JM, Schultz MJ, Juffermans NP. A survey of physicians’ reasons to transfuse plasma and platelets in the critically ill: a prospective singlecentre cohort study. Transfus Med 2009; 19: 207–12. 28. Gonano C, Sitzwohl C, Meitner E, Weinstabl C, Kettner SC. Four-day antithrombin therapy does not seem to attenuate hypercoagulability in patients suffering from sepsis. Crit Care 2006; 10: R160. 29. Rivard GE, Brummel-Ziedins KE, Mann KG, Fan L, Hofer A, Cohen E. Evaluation of the profile of thrombin generation during the process of whole blood clotting as assessed by thrombelastography. J Thromb Haemost 2005; 3: 2039–43. 30. Adamzik M, Langemeier T, Frey UH, Gorlinger K, Saner F, Eggebrecht H, Peters J, Hartmann M. Comparison of thrombelastometry with simplified acute physiology score II and sequential organ failure assessment scores for the prediction of 30-day survival: a cohort study. Shock 2011; 35: 339–42. 31. Ostrowski SR, Windelov NA, Ibsen M, Haase N, Perner A, Johansson PI. Consecutive thrombelastography clot strength profiles in patients with severe sepsis and their association with 28-day mortality: a prospective study. J Crit Care 2013; 28: e1–11.
Address: Anne-Mette Hvas Centre for Haemophilia and Thrombosis, Department of Clinical Biochemistry Aarhus University Hospital Brendstrupgårdsvej 100 DK-8200 Aarhus N Denmark e-mail: [email protected]
533
