Intraoperative cell salvage (IOCS) allows the collection, treatment, and retransfusion of shed blood from the surgical field and cardiopulmonary bypass (CPB) collector during cardiac surgery. This technique has been shown to reduce the need for allogeneic blood products in many surgical specialties.
Centrifugation-based IOCS devices enable red blood cell (RBC) reinfusion, whereas platelets and other coagulation factors are largely eliminated during processing. Although this technique permits reducing RBC transfusion exposure, it does not reduce the need for allogeneic platelet transfusion.
Knowing the role of platelets in surgical bleeding and considering the risks associated with platelet transfusion the SAME device (i-SEP) was developed to save and wash both RBC and platelets with a recovery rate of 86.1% and 52.4%, respectively. The SAME device was developed based on a noncentrifugal recovery of cells from shedded blood, using membrane filtration technology. This technique preserves a higher number of platelets and avoids cell alteration due to centrifugation process.
The SAME Erythrocytes and Platelets Innovative Autotransfusion (SEPIA) registry was created to compare routine cell-salvage solutions with the same device in patients undergoing hemorrhagic surgeries throughout France. Herein we report our pilot experience from the first institution enrolling patients in the SEPIA registry.
METHODS
From January to June 2023, we prospectively gathered data from 104 patients who underwent cardiac surgical procedures with high-bleeding risk at the Infirmerie Protestante (Lyon). 52 were treated with a new filtration-based IOCS device (SAME, i-SEP), and 52 with a standard IOCS device (Xtra, LivaNova). Patients were included in the SEPIA registry, a postmarketing clinical registry gathering data from the IOCS devices (laboratory on reinfusion bags), in addition to allogeneic transfusion data, laboratory results, postoperative follow-up data, and adverse events. Sepia was registered in the French Health Data Hub and was considered nonhuman-subject research, as it requires no patient-impacting procedures. All patients consented to data collection after being informed.
The decision for which device to use was solely based on the availability of the respective devices. Perioperative management and transfusion guidelines did not differ between the 2 devices, and a standardized transfusion algorithm based on TEG (Haemonetics) was used.
Statistical Analysis
Data are expressed as means ± standard deviation (SD), or numbers and percentages (%). Continuous outcomes were compared with unpaired Student t test or Mann-Whitney U test. The alpha risk was set at 5%. Statistical analysis was performed with EasyMedStat (version 3.35).
RESULTS
Demographics, surgical characteristics, and autotransfusion data are reported in the Table. The nature of surgeries, procedures, and CPB durations were comparable between both groups. Almost all patients were treated with Tranexamic Acid (SAME: 96% vs Xtra: 100%). Although the treated blood volumes were lower in the same group, these patients had higher reinfused blood volumes compared to the Xtra group. Lower hematocrits were found in the SAME group. However, the SAME treated bags contained 4 times higher platelet content than the Xtra bags.
Table. – Preliminary Data on SEPIA Registry: Population Characteristics, Cell-Salvage Processing Performances, and Quality Control Results on Reinfusion Bags
| Parameters | Xtra (n = 52) |
SAME (n = 52) |
|---|---|---|
| Sex [female, n (%)] | 10 (19%) | 19 (37%) |
| Age (yr) | 68 ± 9 | 66 ± 12 |
| Height (cm) | 173 ± 9 | 171 ± 10 |
| Weight (kg) | 83 ± 17 | 80 ± 17 |
| BMI | 27.9 ± 5.4 | 27.4 ± 5.9 |
| Surgery characteristics | ||
| Isolated CABG | 7 | 6 |
| Single non-CABG | 7 | 16 |
| Combined procedures | 38 | 30 |
| Aortic procedures | 19 | 21 |
| Procedure duration (incision-closing, min) | 151 ± 45 | 157 ± 39 |
| Cardiopulmonary bypass time (min) | 100 ± 33 | 98 ± 33 |
| Aortic cross clamping time (min) | 77 ± 28 | 73 ± 25 |
| Autotransfusion | ||
| Number of cycles (n) | 3.2 ± 1.0 | 4.2 ± 2.0** |
| Treated blood volume (mL) | 2704 ± 804 | 2069 ± 831*** |
| Returned blood volume (mL) | 685 ± 244 | 794 ± 353* |
| Returned blood hematocrit (%) | 57.1 ± 6.5 | 40.7 ± 3.3*** |
| Quality control on reinfusion bags | ||
| Number of analyzed bags | 24 | 27 |
| Hemoglobin (g/dL) | 18.4 ± 2.8 | 15.3 ± 1.6*** |
| Red blood cells concentration (106/mcl) | 6.2 ± 0.9 | 5.0 ± 0.7*** |
| Platelet concentration (103/mcl) | 35 ± 31 | 139 ± 360*** |
| Leukocyte concentration (103/mcl) | 11 ± 6 | 12 ± 5 |
Hemoglobin levels were comparable between both groups (Figure A), whereas patients’ platelet counts tended to be higher in the SAME group, especially at the end of surgery and during the first 2 postoperative days (Figure B), without reaching significance.
Patients treated with SAME were exposed to allogeneic platelet transfusion, plasma, and RBC (Figure C) less frequently. Additionally, a smaller quantity of allogeneic blood products was used in SAME group compared to Xtra group (Figure D).
Fibrinogen concentrate was administered per-operatively in 4 patients in each group and was less frequently used postoperatively in SAME group than Xtra group (respectively in 7 (13%) and 11 (21%) patients). Patients received no prothrombin complex concentrates nor recombinant activated factor VII.
Postoperative drainage cumulated more volume in Xtra group than SAME group, respectively at 12, 24, and 48 hours postoperative: 541 ± 354, 718 ± 388, 904 ± 470 mL vs 424 ± 225, 613 ± 296, and 790 ± 398 mL.
DISCUSSION
Our pilot analysis of the SEPIA registry suggests improved hemostasis in the patients treated with SAME, both on single parameters such as allogeneic transfusions, and on a multistaged definition for perioperative bleeding (UDPB). Comparing these data with previously published analyses our study confirms the performance of the SAME device on hematocrit, RBC, platelets, and leukocyte salvage.
In highly hemorrhagic approaches, blood salvage has always been a major concern. Even if the cell saver devices allow salvage of RBCs, platelets are nearly completely lost due to the centrifugation-based recovery process. In dealing with a novel technology, we faced some issues in our early experience. The initial software was not intuitive and only a few nurses could manage the device. We also faced massive clotting in the reservoir with 1 patient due to an intentional pause of the recycling process for about 10 minutes. This experience helped to modify the software by introducing an alarm after 3 minutes of pause in the filtration cycle, avoiding any recurrence of accidental clotting. Finally, if the cycle of cell recovery is longer when compared to a cell saver (7–10 minutes for the SAME device vs 5min for the cell saver), a fast reinfusion is possible and already possible within the software in case of an immediate need for massive volume reinfusion.
Concerning low hematocrit levels in the SAME device, its reinfusion bags provided higher volume along with lower hemoglobin concentrations due to the device’s cell salvaging process, but the overall amount of reinfused hemoglobin was comparable between the 2 groups, whereas platelet concentrations were higher for the SAME device.
Although postoperative levels of hemoglobin and platelets were not significantly different between groups, the SAME device allowed for lower exposure to allogeneic blood products, and a notably lower number of allogeneic platelet transfusions.
These results encourage further data collection and they may be used to estimate the number of patients to be included in the multicentric SEPIA registry to demonstrate the advantage of SAME in terms of reducing platelet transfusions and improving hemostasis in patients undergoing high-bleeding-risk cardiac surgery while confirming the safety of SAME across a large number of institutions.