If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Clinical Chemistry and Department of Biomedical and Clinical Sciences, Linköping University, Linköping, SwedenCardiovascular Research Centre, School of Medical Sciences, Faculty of Medicine and Health, Örebro University, Örebro, Sweden
Department of Cardiothoracic Surgery, Oslo University Hospital, Rikshospitalet, Oslo, NorwayInstitute of Clinical Medicine, University of Oslo, Oslo, Norway
Department of Cardiothoracic and Vascular Surgery and Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, SwedenDepartment of Clinical Chemistry and Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden
Department of Cardiothoracic and Vascular Surgery and Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, SwedenDepartment of Clinical Chemistry and Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden
Previous studies have described impaired platelet function after cardiopulmonary bypass (CPB). Whether this is still valid in contemporary cardiac surgery is unclear. The aim of this study was to quantify changes in function and numbers of platelets during CPB in a present-day cardiac surgery cohort.
Design
Prospective, controlled clinical study.
Setting
Singel centre university hospital.
Participants
Thirty-nine patients scheduled for coronary artery bypass graft (CABG) surgery with CPB.
Interventions
Platelet function and numbers was measured at seven time points in 39 patients during and after CABG surgery; at baseline before anaesthesia, after 30 minutes on CPB, at the end of CPB, after protamine administration, at ICU arrival, three hours after ICU arrival and on the morning after surgery.
Measurements and Main Results
Platelet function was assessed with impedance aggregometry and flow cytometry. Platelet numbers are expressed as actual concentration and numbers corrected for dilution using hemoglobin as a reference marker. There was no consistent impairment of platelet function during CPB with neither impedance aggregometry nor flow cytometry. After protamine administration, a decrease in platelet function was seen with impedance aggregometry and for some markers of activation with flow cytometry. Platelet function was restored three hours after arrival to the ICU. During CPB (85.0±21 mins). Corrected for dilution, the number of circulating platelets increased during CPB from 1.73±0.42*109/g to 1.91±0.51*109/g (p<0.001).
Conclusions
During cardiac surgery with moderate CPB times, platelet function was not impaired and no consumption of circulating platelets could be detected. Administration of protamine transiently affected platelet function.
PAR1 activating peptide (specific peptide activating the thrombin receptor PAR1, same as TRAP)
PAR4-AP
PAR4-activating peptide (specific peptide activating the thrombin receptor PAR4)
PS
Phosphatidylserine
TRAP
Thrombin receptor activating peptide (same as PAR1-AP)
U
Units.
INTRODUCTION
Cardiac surgery is associated with an increased risk of perioperative blood loss and need for transfusion of blood products1. Cardiopulmonary bypass (CPB) interacts with the patient in a complex way, eliciting an inflammatory response and activating coagulation. It is also reported to affect platelet function and consume circulating platelets2. These interactions, together with heparin administration, increase the risk of bleeding during and after surgery and contributes to mortality3. How and when different systems associated with hemostasis are affected during the perioperative process is intricate and not fully described nor understood. The general viewpoint has been that platelets are affected by CPB times, and that this may account for some of the coagulopathy associated with cardiac surgery4-8.
In recent decades, significant surgical and technical improvements have been introduced in cardiac surgery using CPB. Technical innovations such as improved membrane oxygenators and the incorporation of new biocompatible materials have reduced the damage to blood constituents during CPB9. It is therefore reasonable to re-evaluate the effects of CPB and surgical procedures on platelet function in present-day cardiac surgery procedures.
The understanding of platelet function has also deepened with new methods of analysis, improvements in analysis techniques and handling when studying platelets both in vitro and in vivo. Several methods have been used such as light transmission aggregometry 10, 11, bleeding time11, flow cytometry12, 13 and impedance aggregometry14-16. No consensus exists on which method best describes the clinically relevant aspects of platelet function in cardiac surgery.
It has previously been documented that protamine affects platelet function 17, 18, thus it is important to distinguish the effect of CPB from the effect of protamine per se. In some previous studies on platelet function during CPB sampling was done after protamine administration and in some it is unclear when sampling was done in relation to protamine administration 14, 16, 19-21. It is therefore imperative in studies describing platelet function and consumption in relation to CPB to clearly state whether sampling after CPB is done before or after administering protamine.
The aim of this study was to describe, in more detail, the changes in platelet function and numbers after CPB and protamine administration during contemporary cardiac surgery by using flow cytometry with a modern protocol together with impedance aggregometry.
METHODS
Study population
The study was approved by the regional ethics review board. After giving oral and written consent, 40 patients undergoing coronary artery bypass graft (CABG) surgery using CPB were included.
Exclusion criteria were emergency surgery, inclusion in interfering studies, prior diagnosis of bleeding disorders such as coagulopathy or platelet dysfunction or treatment with P2Y12 receptor antagonists not terminated a minimum of five days prior to surgery.
Blood sampling
Blood was drawn from the radial artery after insertion of a 20-gauge arterial cannula (Becton Dickinson and Company, Franklin Lakes, New Jersey, USA), combined with a Saf-T Holder® device (Smiths Medical ASD, London, Great Britain). EDTA tubes (Becton Dickinson and Company, Franklin Lakes, New Jersey, USA) were used for cell counting and hirudin tubes (Roche Diagnostics Rotkreuz, Switzerland) were used for impedance aggregometry and flow cytometry. Blood was collected immediately after insertion of lines before anesthesia induction (Baseline), immediately after decannulation after CPB (End CPB), five minutes after reversal of heparin with protamine (Protamine), on arrival to the ICU (ICU), three hours after arrival to ICU (ICU 3h) and the morning after surgery (ICU 1d). In a subset of 23 patients, we also collected blood 30 minutes after the start of CPB. Due to logistical reasons, it was not possible to use flow cytometry at this timepoint. Immediately after blood sampling, 100µL blood from each hirudin tube was also transferred to one Eppendorf tube with 34µL formaldehyde 4% for fixation for analysis of circulating platelets exposing the activation marker P-selectin and also for analysis of the percentage of leukocytes in conjugates with platelets in vivo.
Clinical management
Anesthesia was induced with sodium pentothal 2–6 mg/kg and 0.3–0.5 mg fentanyl. Muscle relaxation was achieved with 50 mg rocuronium. Anesthesia was maintained with isoflurane before, during and after CPB. Supplemental fentanyl was given on demand according to the attending anesthesiologist. After leaving the operating theater, anesthesia was maintained with propofol until the patient was extubated in the ICU. A heart-lung machine (LivaNova Sorin Stockert S5, London, United Kingdom) with reservoir (Inspire HVR N/S, LivaNova, London, United Kingdom), tubing (Sorin IN00325H, LivaNova, London, United Kingdom) and membrane oxygenator (Inspire 8F M, LivaNova, London, United Kingdom) primed with 1400 ml Ringer's acetate, 200 ml mannitol 150 mg/ml (Fresenius Kabi AB, Uppsala, Sweden) and 5000 U of heparin (LEO Pharma, Malmö Sweden) was used for CPB. In 31 patients a roller pump (Sorin S5 Roller Pump, LivaNova, London, United Kingdom) was used, in 8 patients a centrifugal pump (CP5, LivaNova, London, United Kingdom) was used. The initial heparin dose was calculated using the Hepcon hemostasis management system (HMS Plus, Medtronic, Minneapolis, USA) to achieve an activated clotting time (ACT) of 480 seconds (s) according to instructions from the manufacturer. A supplemental dose was given if needed to achieve an activated clotting time (ACT) of >480 s before starting CPB. Additional doses of heparin were given during bypass to keep a therapeutic ACT level (>480 s). Antifibrinolytic treatment was given according to the attending surgeon with tranexamic acid 2g (Pfizer AB, Sollentuna, Sweden) given to all but four patients after heparin administration before the start of CPB. Two patients also received an additional dose of 2g tranexamic acid after CPB. When given post-CPB, sampling for platelet aggregation was done before administration of tranexamic acid. During CPB, the temperature of the patient was kept between 36 and 37 °C.
After CPB, heparin anticoagulation was reversed with protamine sulfate. The protamine dose was calculated using the HMS Plus according to instructions from the manufacturer. The dose was split in an initial dose, that was given immediately after decanulation and a supplemental dose corresponding to the heparin in the blood remaining in the heart lung machine and infused back to the patient after decannulation. Before leaving the operating theater, the HMS plus was used to detect any residual heparin. Residual heparin was reversed with an additional third dose protamine if needed. All anesthetics, heparin and protamine treatment were given according to clinical routine and the attending physician was blinded to the aggregometry and flow cytometry results. Transfusion of blood products was given according to clinical preference of the attending physician. Analysis of hemoglobin concentration and platelet count was performed by the department of clinical chemistry at the hospital. To assess the actual number of circulating platelets a correction for dilution was made using hemoglobin as a marker. The concentration of platelets was divided by the concentration of hemoglobin and is henceforth called dilution-corrected platelet numbers
Flow cytometry
Flow cytometry was performed using a flow cytometry protocol enabling simultaneous usage of multiple activators and markers to study platelet function22. In short, agonists used to activate the platelets were adenosine diphosphate (ADP, final concentration 5µM), specific peptides activating the thrombin receptors PAR1 (same as TRAP-6 used in Multiplate®, final concentration 10µM) and PAR4 (final concentration 100µM). As GPVI agonist, we used cross-linked collagen-related peptide (CRP-XL 23, final concentration of 2µg/mL). To differentiate between primary activation from thrombin receptors and secondary activation from endogenously released ADP, we used apyrase, final concentration 0.2U/mL, to degrade ADP released from activated platelets.
Markers used for platelet function evaluation were PAC-1 (binds activated fibrinogen receptor [GPIIb/IIIa]), anti-P-selectin (CD62P, used as marker for platelet alpha granule release), anti-LAMP-1 (marker for platelet lysosome release), annexin V-V450 (binds phosphatidylserine), the mitochondrial dye DilC1(5) and its positive control CCCP (that disrupt the mitochondrial membrane integrity).
Tubes were also prepared to evaluate P-selectin exposure on circulating platelets and the percentage of circulating leukocyte-platelet conjugates from the fixed blood samples. Platelets were identified by anti-GPIIb (CD41, binds the fibrinogen receptor regardless of activation), and anti-P-selectin was used to evaluate P-selectin expression.
The staining and activation of platelets was started 60-120 minutes after blood sampling, and platelets were activated for 10 minutes before dilution and immediate analysis by flow cytometry.
Flow cytometry data were collected with a GalliosTM (Beckman Coulter Inc., Brea, USA) flow cytometer using the ultra-wide angle of detection (submicron particle setting) for forward scatter (FSC) and a fluorescence threshold on FL3 (CD41-ECD) to maximize detection of small particles, as previously described24. Data were processed and analysed with Kaluza® Analysis Software (Beckman Coulter Inc., Brea, USA) and Microsoft Excel (Microsoft Corporation, Redmond, Washington, USA).
Flow cytometry data describing platelet activation can be expressed in two different ways, as percentage positive platelets, or as median fluorescence intensity (MFI). MFI is the more discriminative way to express flow cytometry data if the activation is high25. Therefore, in this study, MFI was used in situations where the median percentage of positive platelets exceeded 90%.
Since previous studies have shown that platelets only transform to a procoagulant state with phosphatidylserine exposure and mitochondrial membrane disrupture when strongly activated24, 26, we only present results for annexin V and DilC1(5) after exposure to CRP-XL in high concentration and in combination with a PAR-activator.
Impedance aggregometry
Platelet aggregability was analyzed by the coagulation lab at the university hospital using the Multiplate® platelet function analyzer (Verum Diagnostica GmbH, Munich, Germany) according to the manufacturer's instructions, with adenosine diphosphate (ADP-test) and thrombin receptor activating peptide 6 (TRAP-test) as agonists. Results were presented as area under the curve (AUC) in units (U). To evaluate the effect of platelet concentration on aggregation, we used simple linear regression. Using the slope from the regression line together with the change in platelet concentration between baseline and 30 minutes on CPB, we calculated an expected aggregative response. We then compared the expected response with the measured response using Student's T-test.
Statistics
Statistical analysis and graphs were made using GraphPad Prism 9.3.1 (GraphPad Software, La Jolla, California, USA). Data are presented as mean and standard deviation, 95% confidence intervals or median and interquartile range (IQR), as appropriate. Repeated measures ANOVA or mixed effects model was used (depending on the presence of missing values) together with Šídák's multiple comparisons test. In the longitudinal data, all timepoints were compared with baseline. In addition, the timepoint after protamine (Protamine) was also compared to the timepoint immediately before protamine administration (End CPB). For correlation analysis, Pearson's test or simple linear regression was used. Expected and actual aggregative response at 30 minutes of CPB was compared with Student's T-test. A p-value of <0.05 was considered statistically significant. When studying platelet function, all results after receiving platelet concentrates were excluded. When studying platelet numbers and platelet concentration, all results after receiving platelet or erythrocyte concentrates were excluded.
RESULTS
We included 40 patients who were scheduled for CABG surgery. One patient received tirofiban up until eight hours before surgery, and since platelet function was affected at baseline, the patient was excluded, resulting in 39 patients in this analysis. Two were women. All patients received acetylsalicylic acid 75 mg daily until the day before surgery. Mean CPB time was 85.0 ± 21 minutes (range 42 to 138 minutes) and mean postoperative bleeding was 683 ± 320 mL (range 190 to 2090 mL). Mean heparin dose was 30721 ± 6340 U and mean dose per body weight was 348 ± 101 U/kg. Mean total protamine dose was 161 ± 45 mg and the protamine/heparin ratio was 0.539 ± 0.14 mg/ 100 U. For patient characteristics, see table 1. Four patients received one platelet concentrate each, all after arrival to the ICU. All results after receiving platelets were excluded. Six patients received in total 12 erythrocyte concentrates (one during surgery and five after arrival to the ICU). Results regarding platelet numbers in relation to hemoglobin after erythrocyte concentrates had been given were excluded from further analysis.
When using ADP as activator and the binding of PAC-1 as a marker of activation of the fibrinogen receptor, there was an increase in activation ability from baseline to the end of CPB (figure 1A). Mean MFI increased from 2.17 ± 1.57 to 2.55 ± 1.59 (p=0.001). After administration of protamine, the ability for activation of the fibrinogen receptor decreased to 2.01 ± 1.29 (p<0.001). Activation ability then continuously increased to 2.38 ± 1.38 in the morning the day after surgery. With the collagen peptide CRP-XL as activator (Figure 1B), there was no difference in activation ability between baseline and end of CPB. After protamine, the binding of PAC-1 decreased from 1.44 ± 1.4 to 0.95 ± 0.85 (MFI, p<0.001). The activation ability thereafter increased to 1.10 ± 0.88 three hours after arriving in the ICU.
Figure 1Activation of the fibrinogen receptor on platelets after stimulation. Binding of PAC-1 was used as indicator of activation of the fibrinogen receptor. In panel A, Adenosine diphosphate (ADP) 5 μM, in B, Collagen-related peptide (CRP-XL) 0.15 μg/mL, in C, PAR1-AP 10 µM and in panel D, PAR4-AP 100 µM was used as activators. The grey boxes in C had apyrase combined with PAR1-AP to degrade endogenously released ADP. Results are presented as median fluorescence intensity (MFI). The box indicates quartiles with the line as median and whiskers the 10 – 90 percentile range. ** denotes p<0.01 and *** denotes p<0.001 compared with baseline. ### denotes p<0.001 when compared with end of CPB using mixed-effects analysis with Šídák's multiple comparisons test. § denotes p<0.05 compared with end of CPB for apyrase-treated samples.
A similar pattern was seen when using the thrombin receptor agonists PAR1-AP or PAR4-AP as activators (Figure 1C, 1D). Similar to the results with ADP, the ability for activation of the fibrinogen receptor was unchanged at the end of CPB compared to baseline but decreased after protamine. With PAR1-AP as activator, PAC-1 MFI decreased after protamine from 1.03±0.56 to 0.72±0.30 (p<0.001) and with PAR4-AP from 2.60±1.3 to 1.81±0.97 (p<0.001). Both thereafter increased and peaked three hours after arrival in the ICU.
Exposure of P-selectin as a marker of alpha granule release showed varying results depending on which agonist was used (Figure 2). With ADP (Figure 2A) as agonist, the ability for exposure of P-selectin MFI increased between baseline and end of CPB from 2.47±0.72 to 2.81±0.74 (p=0.003). The response continued to be higher than baseline until the day after surgery. With PAR1-AP as agonist (Figure 2C), there was a decrease in P-selectin exposure after stimulation at the end of CPB from 4.16±2.5 to 3.35±2.5 (p=0.007). The ability for exposure of P-selectin then decreased further after protamine and arrival at the ICU, whereafter it increased and peaked three hours after arrival to the ICU. When using the collagen peptide CRP-XL or PAR4-AP as activators, no significant differences were found for exposure of P-selectin (Figure 2B and 2D).
Figure 2P-selectin exposure on platelets after activation. Exposure of P-selectin was used as an indicator of release of alpha granule. In panel A, Adenosine diphosphate (ADP) 5 μM, B, Collagen-related peptide (CRP-XL) 0.15 μg/mL, C, PAR1-AP 10 µM and in panel D, PAR4-AP 100 µM was used as activator. The grey boxes in C had apyrase combined with PAR1-AP to degrade endogenously released ADP. Results are presented as median fluorescence intensity (MFI). The box indicates quartiles with the line as median and whiskers the 10 – 90 percentile range. * denotes p<0.05, ** denotes p<0.01 and *** denotes p<0.001 when compared with baseline using mixed-effects analysis with Šídák's multiple comparisons test.
As with exposure of P-selectin, exposure of LAMP-1 (as marker of lysosomal exocytosis) showed different findings depending on the agonist used (Figure 3). With ADP as activator (Figure 3A), we found a steady increase in the percentage of platelets able to expose LAMP-1 that peaked three hours after arrival to the ICU. With PAR1-AP (Figure 3C) as activator, we found a decrease after baseline with the lowest value at arrival to the ICU. When we used the collagen peptide CRP-XL (Figure 5B) or PAR4-AP (Figure 3D) as agonists, we found no differences in ability for exposure of LAMP-1 at the different sampling points.
Figure 3Lysosomal release in platelets. Exposure of LAMP-1 was used as an indicator of lysosomal exocytosis. Activators used were in panel A, Adenosine diphosphate (ADP) 5 µM, B, Collagen-related peptide (CRP-XL) 0.15 μg/mL, C, PAR1-AP 10 µM and in panel D, PAR4-AP 100 µM. The grey boxes in C had apyrase combined with PAR1-AP to degrade endogenously released ADP. Results are presented as percentage of positive platelets (%). The box indicates quartiles with the line as median and whiskers the 10 – 90 percentile range. * denotes p<0.05, ** denotes p<0.01 and *** denotes p<0.001 when compared with baseline using mixed-effects analysis with Šídák's multiple comparisons test.
To differentiate between the activation directly through PAR1 and the secondary activation by endogenously released ADP from activated platelets, we used apyrase to degrade ADP in the samples activated with PAR1-AP. In this way we could study the ADP-independent response from PAR1-AP. With all three markers of platelet activation (PAC-1, P-selectin and LAMP-1), all responses were markedly lower, and the statistically significant differences disappeared except for the decrease in PAC-1 binding ability after protamine (Figure 1C, 2C and 3C). When we subtracted the ADP-independent response from the total response, we found a decrease in the ADP dependent response during CPB for P-selectin and LAMP-1 and after protamine for all three markers (PAC-1, P-selectin, and LAMP-1). Data shown in supplement figure 1.
Finally, we investigated platelet ability to activate to a procoagulant state with binding of annexin V as a marker of exposure of phosphatidylserine and DilC1(5) as marker of intact mitochondrial membranes. We used the collagen peptide CRP-XL alone in a high concentration or together with either PAR1-AP or PAR4-AP. The differences we found were small and inconsistent. Data are shown in supplement figure 2.
We also fixed samples immediately after sampling to measure in vivo-activation with exposure of P-selectin in circulating platelets. These immediately fixed samples would reflect the level of activation of the circulating platelets at time of sampling. Before anesthesia, 7.6 ± 3.4 % of the circulating platelets exposed P-selectin. No differences in the percentage of P-selectin positive platelets at the different sampling points were observed (Figure 4). The mean percentage of monocyte-platelet conjugates (MPC) at baseline was 30.2 ± 27 %. During CPB the percentage increased to 40.2 ± 19% (p=0.05) and then increased further 5 minutes after protamine to 45.8 ± 23 (p<0.001 compared to baseline). The percentage of MPC then decreased to 22.2 ± 15 % at arrival to the ICU. The percentage of neutrophil- and lymphocyte- platelet conjugates decreased during CPB (neuthrophils from 18.6 ± 20 % at baseline to 9.91 ± 9.8 % at the end of CPB, p<0.001 and lymphocytes from 7.97 ± 5.3 % to 5.49 ± 2.3 %, p=0.007).
Figure 4Exposure of P-selectin in circulating platelets immediately fixed after sampling measured with flow cytometry. Results are presented as percentage of positive platelets (%). The box indicates quartiles with the line as median and whiskers the 10 – 90 percentile range.
There was no correlation between time on CPB and impairment in markers of platelet activation at the end of CPB. In contrary, we found a weak correlation between longer CPB duration and increased ability for exposure of P-selectin when activated by PAR4-AP (R2 0.13, p=0.04).
Impedance aggregometry
Platelet function measured with impedance aggregometry, using ADP and TRAP (PAR1-AP) as agonists; was unchanged at the end of CPB compared with baseline. After protamine, the aggregative response decreased. The platelet function returned to baseline levels at the arrival to ICU using TRAP-test and three hours after arrival to the ICU using the ADP-test (Figure 5 A and B).
Figure 5Platelet function measured with impedance aggregometry during surgery. In panel A Adenosine diphosphate (ADP) 6.5 μM and in panel B PAR1-AP (TRAP) 32 μM was used as activator. Results are shown as area under the curve (AUC) in units (U). The box indicates quartiles with the line as median and whiskers the 10 – 90 percentile range. *** denotes p<0.001 compared to baseline. # denotes p<0.05 compared to end of CPB, using mixed-effects analysis with Šídák's correction for multiple comparisons.
In the subset of 23 patients where we collected blood 30 minutes after the start of CPB, we found a reduced platelet aggregation when activated with ADP at 30 minutes of CPB compared with baseline (from baseline 77.1 ± 26 U to 49.3 ± 25 U, p<0.001). This decrease in platelet function coincided with the decrease in platelet concentration among these patients at that time (from baseline 227 ± 48 to 161 ± 32*109/L, p<0.001). To evaluate the effect of platelet concentration on aggregation, we used simple linear regression. At baseline, we found a weak correlation between aggregation and platelet concentration for both ADP (R2 = 0.35; p<0.001) and TRAP (R2 = 0.16; p<0.001) as agonists. We then calculated an expected aggregative response at 30 minutes of CPB in relation to the lower platelet concentration at that timepoint. There was no difference between expected aggregative response and measured aggregative response using ADP (expected aggregative response 53.6 ± 30 U, measured aggregative response 49.3 ± 25 U, p=0.61) or TRAP (expected aggregative response 92.7 ± 32 U, measured aggregative response 95.2 ± 32 U, p=0.80), suggesting that the reduction in aggregation at 30 minutes of CPB 30 could be explained by the hemodilution.
Using impedance aggregometry, there was no correlation between CPB times and aggregometry data.
Platelet numbers
Platelet concentration before anaesthesia was 227 ± 48.3 *109/L. At the end of CPB it had decreased to 179 ± 38*109/L (p<0.001). Platelet concentration then continued to be below baseline (Figure 6A).
Figure 6Platelet count and corrected numbers during cardiac surgery. Panel A shows platelet concentration at different timepoints during and after cardiac surgery. In panel B, platelet numbers are corrected for dilution by dividing platelet concentration by hemoglobin concentration at the same timepoint. The box indicates quartiles with the line as median and whiskers the 10 – 90 percentile range. ** denotes p<0.01 and *** p<0.001 compared to baseline. ## denotes p<0.01 compared to end of CPB, using mixed-effects analysis with Šídák's correction for multiple comparisons.
To correct for the dilution that occurs at the start of CPB, we calculated a corrected number of platelets by dividing by the actual hemoglobin concentration. When correcting for dilution, we did not see any decrease in dilution-corrected platelet numbers during CPB. Instead, dilution-corrected platelet numbers had increased at the end of CPB from 1.73 ± 0.43*109/g Hb at baseline to 1.90 ± 0.49 *109/g Hb (p<0.001) (Figure 6B). After protamine, there was a reduction in dilution-corrected platelet numbers to 1.75 ± 0.51*109/g Hb (p=0.0012), whereafter platelets increased again and peaked in the morning the day after surgery at 1.83 ± 0.46*109/g Hb (p<0.001 compared with baseline).
DISCUSSION
In this longitudinal study of platelet function during CABG surgery, platelets had a preserved ability to respond to agonist stimulation ex vivo at the end of CPB compared to baseline. Thus no impairment of platelet function after CPB was seen. There were neither any signs of in vivo activation in circulating platelets. Data confirmed that protamine administration is followed by a reduction in dilution-corrected platelet numbers and a transient impairment of platelet reactivity to exogenous agonists, as previously described17, 18. There were no signs of platelet consumption during CPB but rather preserved numbers of platelets when corrected for dilution.
To study the effect of CPB on platelet function, we used flow cytometry and impedance aggregometry. Flow cytometry enables the simultaneous study of many different aspects of platelet activation. We measured activation of the fibrinogen receptor as a marker of aggregation, release of alpha granule and lysosomes as markers of secretion, and lastly exposure of phosphatidylserine and rupture of mitochondrial membrane as markers of platelet transition to a procoagulant state. We did not find any consistent results indicating impaired platelet function after CPB compared to baseline. This absence of deterioration in platelet function during CPB is in contrast with previous studies10-14, 16, 27-30. We also found no increase in circulating platelets exposing P-selectin during CPB. Rinder et al.13 fixed blood immediately after sampling, in a similar manner as done herein, to study the percentage of circulating platelets expressing P-selectin as a marker of prior in vivo activation. They describe an increase from 7% before CPB to around 30% when CPB was terminated. In our study, the percentage of circulating platelets exposing P-selectin at baseline was around 7%, but in contrast to Rinder, it did not increase during CPB. In another study by Rinder et al.31 an increase of monocyte-platelet-conjugates (MPC) from 18 to 44 %, a slight increase in neutrophil-platelet-conjugates (NPC) and a decrease in lymphocyte-platelet-conjugates is demonstrated. In our study the percentage of MPC increased from 30 % at baseline to 40 % at the end of CPB. Our NPC and lymphocyte-platelet conjugates decreased during CPB. It is hard to interpret our results in relation to Rinder et al. since our baseline differed. This could be due to different methods and the absolute values depends partly on how you choose to set your gate to identify positive conjugates (the standardized method for gating we use is relatively strict). The lesser increase in MPC and the absence in increase of circulating platelets expressing P-selectin could indicate that modern normothermic CPB causes less platelet activation than older techniques.
With impedance aggregometry using ADP and TRAP as activators we found similar dynamics. After 30 minutes on CPB, there was a decline in aggregability compared to baseline. The reduction in aggregability was, however, in the same order as what could be expected from the concomitant reduction in platelet concentration at 30 minutes of CPB. This could indicate that the reduction in aggregability mainly was due to hemodilution and not an impairment of platelet function per se. This is supported by studies showing a decrease in aggregability with lower platelet counts using impedance aggregometry32-35. Hanke and colleagues presented reduction in ADP aggregative response after a reduction in platelet concentration comparable to our data. At the end of CPB, we found no difference in aggregability compared to baseline. Considering the difference in platelet concentration, this could imply a possible increase in the platelet's individual response to activation, which corresponds to the increase in response we found with flow cytometry between baseline and the end of CPB using ADP or CRP-XL as activators and PAC-1 as marker of activation of the fibrinogen receptor.
After administration of protamine, the platelet response to activation decreased, in accordance with previous studies17, 18, 36. Olsson et al.18 reported a decrease in platelet response to stimulation by almost 50% compared to only 15% in our data. A probable reason for this discrepancy is that less protamine was used in the present study; 160.9 ± 45.2 mg compared to 402 ± 72 mg in the study by Olsson and colleagues18. Platelet function, in our study, returned to baseline values at arrival to the ICU and aggregability using TRAP exceeded the baseline values three hours after arrival to the ICU.
In two recent studies14, 16, investigating platelet function with aggregometry after CPB, the sampling of platelets after CPB was done after protamine administration and not before. The reduction in platelet aggregation reported in these studies exceeds our, but matches the finding by Olsson et al.18. They also used the same heparin: protamine ratio (1:1) as Olsson et al. A dose dependent impairment of platelet function after protamine exposure is supported by in vitro data17, where protamine in lower concentrations had a stimulating effect on platelet aggregation with activation of the fibrinogen receptor, but at higher concentrations inhibited secondary activation of thrombin receptors. The mechanism behind this dual, concentration dependent effect on platelet activation is not clear. A possible mechanism could be electrical interaction with surface charges as discussed by Tanaka et al37, or interaction with surface receptors. Another explanation could be that protamine affects the platelet's ability to promote and increase platelet activation through release of endogenous activating substances such as ADP.
To differentiate whether the impairment from protamine depends on an alternation in the direct response to receptor interaction, or is an effect on the response to activation from the endogenously released ADP38, we performed tests combining PAR1-AP with apyrase. Apyrase degrades ADP and removes the secondary activation from ADP released from dense granule in activated platelets. Without the released ADP, the impairing effect of protamine was blunted. This indicates that part of the negative effect of protamine on platelet activation is mediated through alteration in the release/response to endogenous ADP. Considering the direct stimulating effect of protamine previously described17, one possible explanation could be that protamine induces a partial release of ADP resulting in a diminished ADP response to further activation. We found a similar decrease in the stimulated release of alpha- and dense granule after CPB in the calculated ADP-dependent part of the PAR1-AP response indicating that also CPB may induce release of endogenous ADP and therefore impair the endogenous ADP activation.
During CPB, the platelet concentration decreased by 21% and dropped another five percent after protamine administration. Considering the hemodilution caused by the CPB priming volume and using hemoglobin concentration as a marker of dilution, there was no decline but rather a small increase in dilution-corrected platelet numbers during CPB. This is in contrast to many authors who describe consumption of platelets during CPB10, 13, 27.
The decline in platelet concentration of 21% during CPB in the present study is lower than previous data. Zilla et al.10, Van Poucke et al.14 and two reviews on coagulopathy after CPB 7, 8, all describe a decline in platelet count of 40% or more during CPB. The factors behind this decrease are probably hemodilution due to the CPB priming volume and the use of cardioplegia. In our data, dilution-corrected platelet numbers decrease after protamine administration. Many studies have sampled blood after protamine administration39, 40 or did not specify whether the sample was drawn before or after the administration of protamine41. Consumption of platelets after protamine administration could contribute to the reduction in platelet count reported if the sampling of the platelets after CPB was done after the administration of protamine instead of before.
To determine whether CPB results in a true decrease in number of platelets, the effect of dilution must be corrected for. We did not find any decrease in the number of platelets after correcting for the dilution. Zilla et al.10 and Rinder et al.13 corrected their platelet count for hemodilution (Zilla et al. by calculating the dilution from the fluids given and the patient blood volume and Rinder et al. by the change in hematocrit), but found a decrease during CPB.
With modern perfusion circuits and pumps, it is possible to circulate and oxygenate blood outside the body without profound impairment of platelet function. A study on extracorporeal membrane oxygenation treatment reported a difference in platelet function day one compared with healthy controls, but no further impairment between day one and three42. This could support the hypothesis that the difference between our findings and previous studies regarding less impairment of platelet function during CPB may be explained by improved perfusion circuits and techniques7, 43, 44.
Less activation and less consumption could explain why platelets in our study did not decrease in the same way as observed in previous studies. Since we did not directly measure platelet turnover, only the number of circulating platelets, we cannot exclude an exchange with a noncirculating pool of platelets. Platelets recruited from a noncirculating pool could mask consumption of platelets during CPB. In our data, platelet numbers increased at the end of CPB before protamine administration. One explanation for this could be autotransfusion of noncirculating platelets from the spleen. The spleen is said to hold as much as a third of the total amount of platelets in an exchangeable and recruitable pool45-47.
Limitations
This is a small study of CABG patients with relatively short normothermic CPB times. Our data should therefore be interpreted with caution regarding other situations with longer CPB-times, complex cardiac surgery, and the use of hypothermia. We did not, however, find any correlation in our material between CPB times and effects on platelet function. Another limitation is that the methods used do not cover all aspects of platelet function or possible effects on primary hemostasis. We did not study platelet turnover and thus cannot determine if an influx of platelets from a spleen pool occurred and affected their numbers and characteristics. Nevertheless, platelet turnover in our data is comparable to other studies of platelets during CPB. Also, our study does not explain the mechanisms behind the effect of protamine on platelet function.
CONCLUSIONS
In this longitudinal study of CABG patients, our data could not support prior studies describing a major impairment of platelet function after CPB. There was no detectable platelet activation during CPB and platelet ability to be activated after stimulation with exogenous agonists was preserved. With flow cytometry we could confirm and further characterize previous in vitro and ex vivo impedance aggregometry findings of impairment of platelet function by protamine exposure. Corrected for dilution, the number of circulating platelets was increased at end of CPB.
REFERENCES
1
Boer C, Meesters MI, Milojevic M, et al. 2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. J Cardiothorac Vasc Anesth 2018;32(1):88-120.
2
Sniecinski RM, Chandler WL. Activation of the hemostatic system during cardiopulmonary bypass. Anesth Analg 2011;113(6):1319-1333.
3
Karkouti K, Wijeysundera DN, Yau TM, et al. The independent association of massive blood loss with mortality in cardiac surgery. Transfusion 2004;44(10):1453-1462.
4
Spiess BD; Armour S; Horrow J KJ. Transfusion Medicine and Coagulation Disorders. in Kaplan JA RD, Konstadt SN, (ed): Kaplan's cardiac anesthesia: for cardiac and noncardiac surgery. Philadelphia: St. Louis: Elsevier Saunders, 2017, pp 1248-1290.
5
Panigrahi A; Liu L. Patient Blood Management: Coagulation. in Gropper MA, Miller RD, (eds): Miller's anesthesia. New York: Elsevier/Churchill Livingstone, 2020, pp 1579-1603.
6
Thiele RH, Raphael J. A 2014 Update on Coagulation Management for Cardiopulmonary Bypass. Semin Cardiothorac Vasc Anesth 2014;18(2):177-189.
7
Ranucci M. Hemostatic and thrombotic issues in cardiac surgery. Semin Thromb Hemost 2015;41(1):84-90.
8
Höfer J, Fries D, Solomon C, et al. A Snapshot of Coagulopathy After Cardiopulmonary Bypass. Clin Appl Thromb Hemost 2016;22(6):505-511.
9
Stoney WS. Evolution of cardiopulmonary bypass. Circulation 2009;119(21):2844-2853.
10
Zilla P, Fasol R, Groscurth P, et al. Blood platelets in cardiopulmonary bypass operations. J Thorac Cardiovasc Surg 1989;97(3):379-388.
11
Harker LA, Malpass TW, Branson HE, et al. Mechanism of abnormal bleeding in patients undergoing cardiopulmonary bypass: acquired transient platelet dysfunction associated with selective alpha-granule release. Blood 1980;56(5):824-834.
12
Rinder CS, Mathew JP, Rinder HM, et al. Modulation of platelet surface adhesion receptors during cardiopulmonary bypass. Anesthesiology 1991;75(4):563-570.
13
Rinder CS, Bohnert J, Rinder HM, et al. Platelet activation and aggregation during cardiopulmonary bypass. Anesthesiology 1991;75(3):388-393.
14
Van Poucke S, Stevens K, Wetzels R, et al. Early platelet recovery following cardiac surgery with cardiopulmonary bypass. Platelets 2016;27(8):751-757.
15
Görlinger K, Shore-Lesserson L, Dirkmann D, et al. Management of Hemorrhage in Cardiothoracic Surgery. YJCAN 2013;27(Supplement):S20-S34.
16
Kjellberg G, Holm M, Lindvall G, et al. Platelet function analysed by ROTEM platelet in cardiac surgery after cardiopulmonary bypass and platelet transfusion. Transfus Med 2020;30(5):369-376.
17
Törnudd M, Ramström S, Kvitting JP, et al. Protamine stimulates platelet aggregation in vitro with activation of the fibrinogen receptor and alpha-granule release, but impairs secondary activation via ADP and thrombin receptors. Platelets 2021;32(1):90-96.
18
Olsson A, Alfredsson J, Håkansson E, et al. Protamine reduces whole blood platelet aggregation after cardiopulmonary bypass. Scand Cardiovasc J 2015;50(1):58-63.
19
Varghese SJ, Unni MK, Mukundan N, et al. Platelet Functions in Cardiopulmonary Bypass Surgery. Med J Armed Forces India 2005;61(4):316-321.
20
Solomon C, Hagl C, Rahe-Meyer N. Time course of haemostatic effects of fibrinogen concentrate administration in aortic surgery. Br J Anaesth 2013;110(6):947-956.
21
Rahe-Meyer N, Solomon C, Tokuno ML, et al. Comparative assessment of coagulation changes induced by two different types of heart-lung machine. Artif Organs 2010;34(1):3-12.
22
Törnudd M, Rodwan Al Ghraoui M, Wahlgren S, et al. Quantification of platelet function - a comparative study of venous and arterial blood using a novel flow cytometry protocol. Platelets 2022:1-9.
23
Morton LF, Hargreaves PG, Farndale RW, et al. Integrin alpha 2 beta 1-independent activation of platelets by simple collagen-like peptides: collagen tertiary (triple-helical) and quaternary (polymeric) structures are sufficient alone for alpha 2 beta 1-independent platelet reactivity. Biochem J 1995;306 (Pt 2)(Pt 2):337-344.
24
Södergren AL, Ramström S. Platelet subpopulations remain despite strong dual agonist stimulation and can be characterised using a novel six-colour flow cytometry protocol. Sci Rep 2018;8(1):1441.
25
Ramström S, Södergren AL, Tynngård N, et al. Platelet Function Determined by Flow Cytometry: New Perspectives? Semin Thromb Hemost 2016;42(3):268-281.
26
Ramström S, O'Neill S, Dunne E, et al. Annexin V binding to platelets is agonist, time and temperature dependent. Platelets 2010;21(4):289-296.
27
Khuri SF, Alan Wolfe J, Josa M, et al. Hematologic changes during and after cardiopulmonary bypass and their relationship to the bleeding time and nonsurgical blood loss. J Thorac Cardiovasc Surg 1992;104(1):94-107.
28
Wenger RK, Lukasiewicz H, Mikuta BS, et al. Loss of platelet fibrinogen receptors during clinical cardiopulmonary bypass. J Thorac Cardiovasc Surg 1989;97(2):235-239.
29
Kestin AS, Valeri CR, Khuri SF, et al. The platelet function defect of cardiopulmonary bypass [see comments]. Blood 1993;82(1):107-117.
30
Greilich PE, Brouse CF, Beckham J, et al. Reductions in platelet contractile force correlate with duration of cardiopulmonary bypass and blood loss in patients undergoing cardiac surgery. Thromb Res 2002;105(6):523-529.
Hanke AA, Roberg K, Monaca E, et al. Impact of platelet count on results obtained from multiple electrode platelet aggregometry (Multiplate). Eur J Med Res 2010;15(5):214-219.
33
Stissing T, Dridi NP, Ostrowski SR, et al. The influence of low platelet count on whole blood aggregometry assessed by Multiplate. Clin Appl Thromb Hemost 2011;17(6):E211-217.
34
Würtz M, Hvas AM, Kristensen SD, et al. Platelet aggregation is dependent on platelet count in patients with coronary artery disease. Thromb Res 2012;129(1):56-61.
35
Boknäs N, Macwan AS, Södergren AL, et al. Platelet function testing at low platelet counts: When can you trust your analysis? Res Pract Thromb Haemost 2019;3(2):285-290.
36
Ortmann E, Klein AA, Sharples LD, et al. Point-of-Care Assessment of Hypothermia and Protamine-Induced Platelet Dysfunction with Multiple Electrode Aggregometry (Multiplate®) in Patients Undergoing Cardiopulmonary Bypass. Anesth Analg 2013;116(3):533-540.
37
Tanaka KA, Butt AL, Abuelkasem E, et al. Protamine and coagulopathy after cardiopulmonary Bypass – may the perfect electrostatic force be with you. J Cardiothorac Vasc Anesth 2023; doi:https://doi.org/10.1053/j.jvca.2023.01.029
38
Golebiewska EM, Poole AW. Platelet secretion: From haemostasis to wound healing and beyond. Blood Rev 2015;29(3):153-162.
39
Blome M, Isgro F, Kiessling AH, et al. Relationship between factor XIII activity, fibrinogen, haemostasis screening tests and postoperative bleeding in cardiopulmonary bypass surgery. Thromb Haemost 2005;93(6):1101-1107.
40
Coakley M, Hall JE, Evans C, et al. Assessment of thrombin generation measured before and after cardiopulmonary bypass surgery and its association with postoperative bleeding. J Thromb Haemost 2011;9(2):282-292.
41
Rahe-Meyer N, Pichlmaier M, Haverich A, et al. Bleeding management with fibrinogen concentrate targeting a high-normal plasma fibrinogen level: a pilot study. Br J Anaesth 2009;102(6):785-792.
42
Balle CM, Jeppesen AN, Christensen S, et al. Platelet Function During Extracorporeal Membrane Oxygenation in Adult Patients. Front Cardiovasc Med 2019;6:114.
43
Hofer A, Kozek-Langenecker S, Schaden E, et al. Point-of-care assessment of platelet aggregation in paediatric open heart surgery. Br J Anaesth 2011;107(4):587-592.
44
Besser MW, Klein AA. The coagulopathy of cardiopulmonary bypass. Crit Rev Clin Lab Sci 2010;47(5-6):197-212.
45
Aster RH. Pooling of platelets in the spleen: role in the pathogenesis of "hypersplenic" thrombocytopenia. J Clin Invest 1966;45(5):645-657.
46
Harker LA, Finch CA. Thrombokinetics in man. J Clin Invest 1969;48(6):963-974.
47
Wadenvik H, Kutti J. The spleen and pooling of blood cells. Eur J Haematol 1988;41(1):1-5.
Author contributions
Mattias Törnudd, designed the study, performed experiments, interpreted the data, and wrote the manuscript; Linnea Nyberg and Erik Björkman performed experiments; John-Peder Escobar Kvitting and Joakim Alfredsson interpreted the data, and wrote the manuscript; Sofia Ramström and Sören Berg designed the study, interpreted the data and wrote the manuscript.
00180-5&id=gr1.jpg){kind=link}
00180-5&id=gr2.jpg){kind=link}
00180-5&id=gr3.jpg){kind=link}
00180-5&id=gr4.jpg){kind=link}
00180-5&id=gr5.jpg){kind=link}
00180-5&id=gr6.jpg){kind=link}