Bacteremia in the red blood cells obtained from the cell saver in patients submitted to heart surgery

Objective: to determine the microbiological characteristics of the red blood cells obtained with the cell saver in heart surgery patients on an extra-body circuit. Method: a cross-sectional and descriptive study conducted with 358 patients scheduled for heart surgery where the saver was used. Sociodemographic variables were collected, as well as from the saver and of the microbial identification in the re-infusion bag proceeding from the cell saver. Informed consent performed. Results: of the 170 GRAM+ bacteria isolations, the most frequent species were Staphylococcus epidermidis in 69% (n=138) of the cases and Streptococcus sanguinis with a report of 10% (n=20). Significant differences were found in the Staphylococcus epidermidis strain in patients with a Body Mass Index ≥25 (p=0.002) submitted to valve surgery (p=0.001). Vancomycin was the antimicrobial which resisted the Staphylococcus epidermidis strain with a minimum inhibitory concentration of >16 µg/ml. Conclusion: the microbiological characteristics of the red blood cells obtained after processing autologic blood recovered with the cell saver during heart surgery are of GRAM+ bacterial origin, the most isolated species being Staphylococcus epidermidis. Consequently, in order to reduce the presence of these GRAM+ cocci, an antibiotic should be added to the cell saver reservoir, according to a previously established protocol.


Introduction
Invasive infections during heart surgery have been reported internationally (1) ; in fact, bacterial contamination outbreaks have been described recently in the cooling-heating devices used to regulate the patient's blood temperature during the extra-body flow through the closed water circuits (2) . New studies indicate the presence of other bacterial colonizations in the membrane oxygenators of the extra-body circuit itself, which compromises bacterial re-circulation in the patients during the time the extra-body machine is used (3)(4)(5)(6)(7)(8) . Additionally, knowing that heart surgery is the medical specialty which consumes the most blood products (9) , the cell saver (CS) is used, a device which recovers blood losses from the surgical field and from the residual volume of the extra-body circuit to submit it to a "cleaning" process, obtaining a certain volume of red blood cells which is reinfused into the patient (reinfusion bag). In this device, bacterial growth was detected in 85% of the samples of recovered blood (10) and in the samples of red blood cells to reinfuse GRAM+ bacterial commensals (11) , and even in other surgical procedures like traumatologic ones (scoliose in Pediatrics), where the CS was also used and the same commensals were detected in the re-infusion bag, but of the staphylococcus genus (12) .
The presence of bacteria in the saver, both in its recovery process and in its reinfusion phase, supposes a potential infection risk for the patient during the surgical procedure since these devices lack anti-bacterial protection (13) . This fact gains importance because the presence of bacteremia in the patient is associated with vascular accesses, prolonged hospitalizations, treatments in intensive care units, and administration of multiple broad-range antibiotics where immunosuppresion is the most important comorbidity and where mortality is significant (14) .
Currently, considering the weight of the existing evidence and in the context of heart surgery, there is little evidence on the different types of bacteremia that can be contained in the red blood cell bag which is reinfused into the patient, proceeding from the cell saver, reason why we set ourselves the objective of determining the microbiological characteristics of the red blood cells obtained with the cell saver in the heart surgery patient on an extra-body circuit.   Streptococcus sanguínis (S. viridans) with a report of 10% (n=20) according to Table 2.

*Fr (%) = Frequency (percentages)
In the relation of the Staphylacoccus epidermidis strain according to the profile of the studied patient, statistically significant differences were found with respect to the Body Mass Index (p=0.002) and to the surgical procedure performed (p=0.001), according to Table 3.  Currently, in developed countries the vast majority of intra-hospital microbiological isolations in blood are due to GRAM+ cocci, including those due to coagulase-negative Staphylococcus (16) and, in some studies, it is the main cause of hospital bacteremia (17) . In a study, coagulasenegative Staphylococcus was identified in the concentrate obtained from the saver in 54.2% of the cases, reason why it was pointed out that the use of this device is not the best option in surgery. However, this study lacks the inclusion of an extra-body circuit and does not even expose the different types of isolated micro-organisms in the saver, since it only focused on isolating one of them (12) .
Likewise, in other studies GRAM+ bacteria were detected by blood extraction from the patient, concluding the hetero-resistance to certain antibiotics (14,18) , but its results include only hospitalized (not surgical) patients, reason why its prevalence is not demonstrated in the context of heart surgery for not using the saver.
It is to be recalled that bacterial infections are the leading cause of mortality due to blood transfusions (19) , as confirmed by the data from a research study (20) , which In other studies (21) , in 83 samples the presence of Staphylococcus epidermidis was identified, thus concluding that this strain could be susceptible to causing infections in the bloodstream. However, the isolations were performed in patients with neoplastic pathologies, and not in the context of heart surgery.
On the other hand, the report of a research study (22) states that Staphylococcus epidermidis is a well-known etiology of the endocarditis of prosthetic valves, and the one emerging from native valve endocarditis where they have been successfully treated with antibiotic therapy like Vancomycin. We disagree with this author because, in the study, Vancomycin showed resistance to the same strain >16 μg/ml.

Recent studies show that adding Vancomycin to
the wash serum of the CS itself is effective on bacterial contamination, allowing for its elimination (2) but, in these studies, the participants were scheduled for back spinal fusion surgery, did not use an extra-body circuit, and the study only determines the bacterial elimination in the saver reservoir without subsequently analyzing bacteremia in the red blood cell bag, as shown in our study.

We do agree with the studies which identified that
Staphylococcus epidermidis was the most frequent micro-organism in heart surgery, where the unsaturation of different diagnostic and therapeutic measures taken by a multidisciplinary team had a favorable influence on the patients' prognosis (23) . Other studies (24) show the same results. However, they did not determine the isolation source or the antimicrobial sensitive to the strain.
Despite the improvements in the surgical techniques, in the materials, and in the design of the devices, the associated infection is still a relatively frequent and serious complication. Another research study (25) shows that the infection is generally produced during the surgery starting from the patient's cutaneous microbiota, when the micro-organisms colonize a device, grow on its surface forming a bio-layer which is determinant in the pathogenia of these infections, reason why we emphasize the importance of identifying the species in any isolation which is considered significant (26) because GRAM+ cocci are increasingly involved in intra-hospital infections (27) .
Additionally, a recent multi-centric study conducted in 26 hospitals (PROBAC cohort) detected bacteremia episodes in the intra-cardiac devices due to GRAM+ cocci in 30.4% of the patients, a fact which undoubtedly evidences that the infection risk due to hematogenous seeding is present in these patients (28) , even with the possibility of triggering an infectious endocarditis (IE).
According to a new study, the origin of the IE episode could be attributed to diagnostic procedures in 10.52% of the cases, to intercurrent infectious processes in 42.11% of the cases, to other surgical procedures in 26.3% of the cases, and to unknown causes in 21.05% of the cases, with the need for surgical treatment in all the cases (29) . The fact that an IE is produced due to any of the aforementioned causes generates a considerable increase in the risk of the bacterial layer settling on the prosthetic materials, or of residual injuries and frequently on right cavities (30) , which becomes a serious form of the disease with high global mortality (29) . Nevertheless, these researchers did not determine the source of the bacteremia, nor did they establish the elements or components of the possible infection focuses, which is necessary to establish a clinical suspicion prior to providing an early diagnosis and treatment to the patients.
Up to the present day, conventional therapies with antimicrobial agents are used to treat and prevent the associated infections; however, the professionals disagree on the origin of the cardiac patient's infections during the post-operative period, based on the existing data.
Furthermore, recent research studies assert that, under certain conditions, the effectiveness and benefits for the patients from using the cell saver under extra-body circulation are ambiguous, with discrepancies among the studies or the patients (31) . Not to mention the current safety problems in cases of severe bacterial contamination where future studies will be needed to better determine how and when the cell saver is to be used, alongside with new blood conservation measures (32) .
In this context, our study adds additional information on a possible infection focus to the scientific knowledge, incorporating isolation frequencies of bacterial strains in the red blood cell bag from the saver, which, in turn, allows acting accordingly with the available media.
This study has limitations: a) in its development because, for being conducted in a reference center, it could underestimate the prevalence of the isolation of the Staphylococcus epidermidis strain in the saver, and b) in its context, because in terms of clinical meaning, many times it is difficult to establish which strains can be harmless commensals and which invasive pathogens.

Conclusion
The microbiological characteristics of the red blood cells obtained after processing autologic blood recovered with the cell saver during heart surgery are of GRAM+ bacterial origin, the most isolated species being Staphylococcus epidermidis. Consequently, in order to reduce the presence of these GRAM+ cocci, an antibiotic should be added to the cell saver reservoir, according to a previously established protocol.