Cell salvage in bacterially contaminated surgical fields – A scoping review

TL;DR

This review explores the safety of using blood collected during surgeries in contaminated environments, finding frequent bacterial presence but limited evidence linking it to infections.

Contribution

A systematic scoping review of cell salvage in contaminated surgical fields, analyzing contamination rates, decontamination methods, and infection risks.

Findings

Bacterial contamination of salvaged blood was common, with wide variation in positive culture rates.

Decontamination methods like filtration and antibiotics reduced contamination but with inconsistent effectiveness.

Only nine patients had a confirmed link between reinfused blood and postoperative infection.

Abstract

This scoping review systematically examines the use of intraoperative cell salvage in bacterially contaminated surgical settings, examining contamination levels, decontamination strategies, and clinical outcomes following the reinfusion of contaminated salvaged blood.

Following the Joanna Briggs Institute methodology and the PRISMA extension for scoping reviews, comprehensive searches were conducted in MEDLINE, Embase, Web of Science, and Cochrane CENTRAL. Eligible studies included primary empirical research reporting bacterial contamination of salvaged blood confirmed by culture, as well as studies reporting on decontamination strategies or clinical outcomes following reinfusion, across any surgical setting. Data was extracted and synthesized descriptively. Key findings were summarized in structured tables.

Thirty-nine studies, involving 1654 patient, met the inclusion criteria. Bacterial contamination was consistently reported, with positive culture rates varying widely. Methodological quality was assessed for all studies, except the ex-vivo studies. Decontamination strategies, including leukocyte depletion filtration (10 studies) and antibiotic additives (4 studies), reduced contamination, though effectiveness varied. Among 26 studies, involving 1203 patients, who received reinfused salvaged blood, postoperative infections were reported in 15 studies. A total of 55 cases of postoperative bacteraemia were identified. Only five studies described a plausible microbiological link between the reinfused blood and subsequent infection, involving nine patients in whom the isolated pathogen matched the organism cultured from the salvaged blood.

Bacterial contamination of salvaged blood occurred frequently, even in procedures not typically classified as contaminated. Decontamination strategies demonstrated variable effectiveness in reducing bacterial contamination. Despite contamination, a potential microbiological link between reinfused salvaged blood and infection was described in only nine patients across five studies. However, substantial heterogeneity in methodologies and a small sample size of most studies makes it difficult to draw definite conclusions about safety. Therefore, we would advise to use cell salvage in known bacterial contaminated areas on a case by case basis, after careful evaluation of the pros and cons, until future research defined safe contamination thresholds, evaluated the effectiveness of decontamination techniques, and assessed clinical outcomes in standardized and controlled settings.

Figures

PRISMA flowchart of study selection process.

Quality assessment of RCTs included.

Full text

Introduction

Intraoperative cell salvage (ICS) is a technique in which autologous, shed blood from the surgical field is collected, processed, and reinfused into the patient. Cell salvage was developed as a way to minimize allogeneic transfusion requirements. By reinfusing the patient’s own blood, ICS reduces exposure to allogeneic blood and thereby helps to avoid transfusion-related risks such as immunologic reactions [1–4]. Currently, ICS is widely utilized across different surgical specialties, including cardiac, orthopaedic and trauma surgery, as part of blood conservation strategies [5–10]. A recent Cochrane review confirms that in certain elective surgeries, the use of ICS significantly decreases the need for allogeneic blood transfusions without increasing adverse outcomes [10].

Despite its potential benefits, the application of ICS in bacterially contaminated surgical fields remains controversial. Both manufacturers and clinical guidelines suggest caution in contaminated settings. Manufacturers of ICS devices list active infection or gross contamination as contraindications due to the risk of bacteraemia [11,12]. Similarly, international guidelines, for example from the UK Cell Salvage Action Group, generally discourage ICS in infected settings [13]. Simultaneously, guidelines suggest that ICS may be considered in selected cases. For instance, the Association of Anaesthetists guidelines (ASA) state that the use of ICS in infected or contaminated surgical fields should be assessed on a case-by-case basis, as there is no conclusive evidence that ICS leads to poorer clinical outcomes [14].

ASA also advises the use of a leukocyte depletion filter (LDF) when ICS is applied in a surgical field that might be contaminated. LDFs have been shown to significantly reduce, and in some cases even eliminate bacterial contamination [15,16]. However, the exact mechanism by which LDFs reduce bacterial load is not fully understood, as bacteria are typically small enough to pass the filter pores. It is thought that bacterial removal occurs through adhesive interactions between bacterial cell surfaces and the filter material [17].

While these findings are encouraging, conclusive evidence regarding the safety of ICS in contaminated surgical fields is still lacking. There is no consensus on what level of bacterial contamination in salvaged blood is safe for reinfusion, or which decontamination protocols are most effective. Most published studies are small or context-specific, and their methodologies vary widely. As a result, clinicians are often required to make case-by-case decisions on using ICS in contaminated surgical fields, weighing the benefits of autologous transfusion against potential infection risks. To date, no review has comprehensively mapped the literature on ICS use in the context of bacterial contamination. Given the broad scope and methodological heterogeneity of the current evidence base, we conducted a scoping review to systematically assess existing evidence.

Methods

This review aims to assess: 1) bacterial contamination: characterize the types and levels of bacterial contamination encountered during ICS; 2) reduction strategies: describe the techniques used to reduce or manage contamination during salvage and reinfusion; 3) clinical outcome: document the reported clinical outcomes following reinfusion of bacterially contaminated salvaged blood; and 4) knowledge gaps: identify key uncertainties to inform future clinical studies or guideline development.

This scoping review was conducted in accordance with the methodological framework developed by the Joanna Briggs Institute (JBI) [18] and is reported in line with the PRISMA extension for Scoping Reviews [19]. An a priori protocol was developed following JBI guidelines [20], structured using the JBI template for scoping reviews [21], and registered in PROSPERO database (CRD420251023866).

Search strategy

A comprehensive search strategy was developed in collaboration with an experienced information specialist from the Erasmus MC Medical Library. Four electronic databases were systematically searched: MEDLINE, Embase, Web of Science Core Collection, and Cochrane CENTRAL (S1 Appendix). Search strings were tailored to each database and included terms related to intraoperative cell salvage, bacterial contamination, and clinical outcomes. No date restrictions were applied, and only studies published in English or Dutch were considered. Reference lists of relevant reviews identified during the screening process and of included articles were also checked to identify additional eligible publications.

Eligibility criteria

Eligibility criteria were defined using the Population-Concept-Context (PCC) framework [18]. Eligible studies included studies involving human subjects of any age or patient population, or animal models undergoing surgical procedures with the use of ICS. Additionally, preclinical models that simulate clinical use of ICS and assess bacterial contamination and/or reduction techniques were included.

Studies had to examine or describe at least one of the following: measurement of bacterial contamination in salvaged blood, confirmed by blood cultures; bacterial reduction strategies implemented during the cell salvage process, such as leukocyte depletion filtration or antibiotic treatment of salvaged blood; or evaluation of clinical outcomes following reinfusion of bacterially contaminated salvaged blood.

Studies were excluded if they did not quantify or confirm bacterial contamination of salvaged blood or failed to provide sufficient detail to determine whether such contamination was assessed. Studies focusing exclusively on non-bacterial contamination (e.g., viral or fungal) or solely on the salvage or transfusion of blood components other than red blood cells were also excluded.

All surgical settings were eligible for inclusion, provided that bacterial contamination was confirmed by blood cultures and clearly reported. Only primary empirical research was included. Case reports, reviews, editorials, commentaries, clinical guidelines, protocols, and conference abstracts were excluded from this review.

Study selection

Search results were imported into Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia). Two independent reviewers (PH, JS) screened titles and abstracts against the eligibility criteria. Full texts of potentially relevant studies were then assessed independently by the same reviewers. Discrepancies were resolved through discussion or consultation with a third reviewer (AK). If full-text articles were not readily accessible, attempts were made to obtain them through institutional library services. Reasons for exclusion at the full-text stage were recorded. The selection process was documented in a PRISMA-ScR flow diagram.

Data extraction

A standardized and piloted data extraction form was used to chart data from included studies. One reviewer (PH) extracted the data, and a second reviewer (JS) verified the accuracy and completeness of the extracted data.

Methodological quality

The quality assessment of articles included was carried out by JS and PH independently. Whenever applicable a third reviewer (AK) arbitrated. The revised Cochrane risk of bias 2 tool for randomised controlled trials (RoB 2) was used for the included randomised controlled trials [22]. This tool uses stratification into five domains to detect potential bias. For included articles other than randomised controlled trials the Newcastle-Ottawa Score was used [23]. This tool uses stratification into three domains to score quality. Ex-vivo studies were not assessed for methodological quality.

Data analysis and presentation

Data were synthesized descriptively, with a focus on bacterial contamination, reduction strategies, and clinical outcomes. Findings were summarized in structured tables corresponding to these domains. A narrative synthesis was used to contextualize the data, highlight key patterns, and identify gaps in the current evidence. To enhance clarity and comparability, results were organized by surgical specialty.

Results

The initial database search (April 2025) yielded a total of 2139 records. Of these, 988 duplicates were automatically removed during the import process, and 5 more duplicates were excluded manually or by Covidence systematic review software. An additional 6 records were identified through reference list checking and expert consultation. Subsequently, 1152 titles and abstracts were screened for eligibility based on previously stated inclusion and exclusion criteria. Following this initial screening phase, 83 full-text articles were retrieved and assessed for inclusion. Of these, 39 studies met the eligibility criteria and were included in the final scoping review. The complete overview of the study identification and selection process is shown in Fig 1.

PRISMA flowchart of study selection process.

Table 1 presents the main characteristics of all 39 included studies, with a total of 1654 patients who underwent ICS, grouped by surgical specialty. Table 2 summarizes 14 studies that applied additional decontamination techniques beyond standard ICS processing, including leukocyte depletion filtration (n = 10), antibiotic treatment of salvaged blood (n = 4), and use of a customised antibacterial membrane (n = 1). Table 3 outlines 26 studies reporting reinfusion of salvaged blood in 1203 patients, along with corresponding clinical outcomes. Table 4 and Fig 2 summarize the quality assessments of RCTs (n = 4) and cohort studies (n = 31), except for the ex-vivo studies. The RCTs were assessed as some concerns (n = 4). The cohort studies were assessed as poor (n = 28) or good (n = 3). S1 Table lists the bacterial species cultured from the final processed blood product per study. S2 Table summarises studies that assessed bacterial contamination both before and after washing. S3 Table provides quantitative contamination data.

Table 1: Main characteristics and reported contamination rates of included studies.

Table 2: Additional decontamination strategies during cell salvage process.

Table 3: Clinical outcomes following reinfusion of salvaged blood.

Table 4: Quality assessment of cohort studies included.

Quality assessment of RCTs included.

Cardiovascular surgery

Eleven studies reported on the use of ICS during cardiothoracic or vascular procedures [24–34].

Bacterial contamination.

All studies reported bacterial contamination of salvaged blood, with positive culture rates ranging from 12.7% to 96.8% [24–34].

Reduction strategies.

One study routinely added antibiotics to the anticoagulant solution as part of the ICS protocol, without a comparator group. Bacterial contamination persisted in 12.7% of processed samples [32]. No other studies applied additional decontamination strategies.

Clinical outcomes.

Eight studies reinfused salvaged blood [25–28,30–33]. One study did not report clinical outcomes [31]. In four studies no significant infectious complications attributable to ICS were observed [26–28,33]. One study reported a single case of postoperative bacteraemia without clinical signs of sepsis. The organism isolated from the patient was not detected in the salvaged blood sample [25]. Another study reported infectious complications in 6% of patients (n = 20), but only in one patient this could be directly linked to the bacteria cultured from the ICS blood sample [32]. Another study found a significantly higher postoperative infection rate in patients who received culture-positive salvaged blood compared to those who did not (22.0% vs. 9.6%, p = 0.02), identifying bacterial contamination as an independent risk factor for infection [30]. The study does not report whether the organisms isolated from patients matched those cultured from the salvaged blood.

Liver surgery

Five studies reported on ICS during liver transplantation [16,35–38] and one during hemihepatectomy [31]. Liver surgery is particularly prone to contamination due to ascitic fluid, bile spillage, and bacterial translocation from the intestine, especially in patients with end-stage liver disease.

Bacterial contamination.

Bacterial contamination of salvaged blood was consistently reported across all six studies [16,31,35–38]. In one study, no contamination was detected in the final processed blood after leukocyte filtration while earlier processing stages did demonstrate positive cultures [16]. The contamination rates of the final ICS blood ranged from 0% to 68.4%.

Reduction strategies.

Three studies assessed the efficacy of LDFs [16,37,38], reporting reductions in contamination rates from 36.4% to 100%. Two studies demonstrated statistically significant conversion rates [16,37]. Residual contamination varied between 0% to 23.3%.

Clinical outcomes.

Salvaged blood was reinfused in five studies [16,35–38]. Three reported no cases of postoperative bacteraemia [16,35,36], while two other studies reported bacteraemia in up to 16.7% of patients [37,38]. In one of these studies, none of the transfused patients received culture-positive ICS blood [37]. In the other, 75% of bacteraemia cases involved organisms that matched those cultured from the salvaged blood [38].

Orthopaedic surgery

Five studies addressed the use of ICS in orthopaedic procedures [39–43]

Bacterial contamination.

Bacterial contamination of salvaged blood was consistently reported, with contamination rates ranging from 15 to 75% [39–43].

Reduction strategies.

One study demonstrated that the addition of vancomycin to the wash solution completely eliminated bacterial contamination in all samples, compared to a 50% contamination rate in the control group (p = 0.016) [42]. No additional bacterial reduction techniques were employed in the other studies.

Clinical outcomes.

All five studies reinfused salvaged blood [39–43]. One study described a single case of bacteraemia, with the isolated organism matching that cultured from the ICS blood sample [41]. None of the studies reported postoperative sepsis or clinically significant infections attributable to the reinfusion of ICS blood.

Trauma surgery

Three studies reported on ICS in cases of penetrating abdominal trauma [44–46]. In this setting, contamination of salvaged blood with enteric contents is a major concern due to frequent involvement of hollow viscus injuries.

Bacterial contamination.

Bacterial contamination of salvaged blood was consistently reported in all studies [44–46]. Contamination varied, with one study reporting 16% of units of ICS blood being contaminated, while another study reported contamination in 100% of patients.

Reduction strategies.

No additional bacterial reduction techniques were applied

Clinical outcomes.

In all three studies ICS blood was reinfused [44–46]. No consistent association was found between positive cultures of reinfused blood and post-operative bacteraemia. Mortality in these studies was primarily attributed to injury severity and exsanguination.

Gynaecology

Three studies reported on the use of ICS in gynaecological procedures [15,47,48], which are prone to contamination due to proximity to vaginal or gastrointestinal flora.

Bacterial contamination.

Bacterial contamination of ICS blood was observed in all studies. In abdominal hysterectomy for uterine myoma, 66.7% of processed blood samples were culture-positive [47]. During caesarean section, all shed blood samples were culture-positive. After washing, 93% remained culture-positive, and 50% of samples were still positive after final processing [15]. In vaginal delivery cases, contamination was present in all samples, with median bacterial concentrations of 8 colony-forming units per millilitre (CFU/mL) (range: 1–84) before processing and 2 CFU/mL (range: 1–25) after washing [48].

Reduction strategies.

LDFs were used in two studies [15,48], with variable reductions in bacterial contamination reported. One study additionally investigated the use of antibiotics added to the heparinized saline solution during processing, resulting in a 50% reduction in the amount of culture-positive ICS blood [47].

Clinical outcomes.

Only one study reinfused ICS blood and no significant postoperative infections or adverse effects related to reinfusion were observed [47].

Ear, nose and throat (ENT) and oral and maxillofacial (OMS) surgery

Three studies reported on the use of ICS in ENT and OMS surgical procedures [49–51]. These procedures typically involved contaminated surgical fields due to proximity to the oral or nasal cavity, which are colonized by commensal bacteria.

Bacterial contamination.

Bacterial contamination of salvaged blood was observed in all studies. One study reported complete bacterial clearance in the final product, but contamination was detected in earlier processing stages [50]. Final contamination rates of ICS blood ranged from 0% to 100%.

Reduction strategies.

Two studies examined the use of LDFs [50,51]. In one study, no bacterial contamination was detected after leukoreduction [51]. The other study reported that LDF alone led to partial bacterial reduction, but complete elimination was achieved through a multistep protocol that included antimicrobial treatment of the salvaged red blood cells [50].

Clinical outcomes.

In two studies, salvaged blood was reinfused [49,51]. No major infectious complications were reported. Transient bacteraemia, caused by bacteria also cultured from ICS blood, was observed in two patients where no LDF was used [51].

Other

Two studies evaluated ICS in burn excisional surgery [52,53], one assessed ICS in neurosurgical procedures [54], and one investigated contamination across different types of autologous transfusion without specifying the type of surgery [55].

Bacterial contamination.

In burn excisional surgery, bacterial contamination was detected in all ICS blood samples [52,53]. During neurosurgical procedures, contamination was observed in 46.7% of samples [54]. In the study comparing autologous transfusion methods, contamination was reported in 33.3% of ICS samples, compared to 4.8% in samples from preoperatively donated autologous blood and 2.6% in haemodilution samples [55].

Reduction strategies.

No additional reduction strategies were employed

Clinical outcomes.

Two studies reinfused ICS blood and reported no adverse events following reinfusion [54,55].

Ex-vivo studies

Three ex-vivo experimental studies investigated the efficacy of ICS systems combined with LDFs for bacterial decontamination of blood products [56–58]. One study investigated the effect of washing and centrifugation on contamination alone [59]. One study reported on the efficacy of an antibacterial membrane to remove fecal contamination from a RBC and plasma suspension and compared this to the efficacy of a cell washer [60].

Discussion

This scoping review studied the existing evidence regarding the use of ICS in bacterially contaminated surgical fields. A total of 39 studies were included, encompassing 1654 patients who underwent ICS. Overall, the findings indicate that bacterial contamination of salvaged blood is common across a wide range of surgical contexts, including those not typically classified as contaminated, such as cardiovascular or orthopaedic surgery. Most common contaminants were skin commensals. Other contaminants included organisms typically originating from the oral, gastrointestinal, or urogenital tract, depending on the anatomical site and type of procedure performed. In some cases, environmental bacteria were also identified, suggesting possible contamination from surgical instruments, suction systems or the operating room environment.

Several studies evaluated techniques aimed at reducing bacterial contamination during ICS, primarily the use of LDFs (10 studies) and, the addition of antibiotics to the anticoagulant or wash solution (4 studies). LDFs were found to effectively reduce bacterial contamination in several studies, in some cases even to undetectable levels. However, the extent of bacterial reduction varied widely. Similarly, the addition of antibiotics to the cell salvage process reduced the number of culture-positive samples, suggesting its potential as a bacterial reduction strategy. While current evidence does not allow for a definitive recommendation of any single decontamination strategy, the routine use of LDFs appears justifiable in cases where bacterial contamination is suspected or confirmed.

Salvaged blood was reinfused in 26 of the 39 included studies, involving a total of 1203 patients. Postoperative infectious complications of any kind were reported in 15 of these studies. A total of 55 cases of postoperative bacteraemia were documented across these studies. However, only five studies described a plausible microbiological link between the reinfused blood and subsequent infection, identifying a total of nine patients in whom the causative pathogen matched the organism cultured from the salvaged blood. In cardiovascular surgery, only one study identified bacterial contamination of ICS blood as an independent predictor for postoperative infection [30]. However, this study did not report whether the isolated pathogen matched the organism cultured from the salvaged blood. On the other hand, it has to be mentioned that the studies included in this review lacked the statistical power to properly detect an increased risk of infection.

These findings are in line with previous studies suggesting that the presence of bacteria in ICS blood does not necessarily translate into postoperative infection [25,36,43]. These observations support the hypothesis that low levels of bacterial contamination, especially by low-virulence organisms, may be cleared by the immune system without clinical consequences [61,62]. Other clinical fields have similar methodological challenges, where advanced statistical models have been applied to physiological signals to identify subtle but clinically relevant features [63]. This might be a solution to deal with the interaction of the immune system and reinfusion of bacterially contaminated blood. Although, in some contexts such as cardiac surgery, ICS was associated with higher postoperative infection rates, this effect was counterbalanced by the reduced exposure to allogeneic transfusion. As a result, the overall infection risk did not increase [64].

Furthermore, bacterial load alone does not appear to be a reliable predictor of clinical outcome. Previous research has shown that a higher bacterial load does not always translate to symptomatic infection. In a study where eight patients received bacterially contaminated platelets, five patients remained asymptomatic despite bacterial concentrations ranging from 10² to 10¹¹ CFU/mL. In contrast, symptomatic cases had bacterial loads ranging from 10⁶ to 10⁸ CFU/mL [65].

In addition, appropriate perioperative antibiotic prophylaxis may further reduce the risk of infection following reinfusion of bacterially contaminated salvaged blood. Antibiotics that are routinely administered during surgery to prevent surgical site infections may also provide effective coverage against potential contaminants introduced through ICS [39,66].

Lastly, intraoperative environmental and procedural factors appear to play a crucial role in the bacterial contamination of salvaged blood. One study reported that a longer operative duration, a greater number of OR staff present, and a higher surgical case order were all independent risk factors for positive bacterial cultures in salvaged blood [30]. These factors are also well-established risk factors for surgical site infections [67]. This overlap suggests that factors known to increase the risk of surgical site infections, such as extended operative time, frequent door openings, increased OR room traffic, and potentially other perioperative variables like inspired oxygen concentration (FiO₂), may likewise contribute to the risk of bacterial contamination in salvaged autologous blood.

Strengths and limitations of the evidence

A key strength of this scoping review is the inclusion of studies from a wide range of surgical specialties, providing a comprehensive overview of ICS in the setting of bacterial contamination. To our knowledge, this is the first review to systematically study the evidence on ICS use in this context. Another methodological strength is the unrestricted time frame applied during study selection. This approach was particularly important because, although advances in surgical technique and sterility may have reduced contamination rates over time, the clinical consequences of reinfusing contaminated blood remain fundamentally unchanged. Excluding older studies solely based on publication date would therefore risk disregarding valuable data that remains relevant to the objectives of this review. Furthermore, the consistency in reported contamination rates, alongside the generally low incidence of adverse clinical outcomes, adds to the credibility of the findings. Nonetheless, several important limitations must also be acknowledged.

Most studies were small, single-centred, and observational. There was substantial heterogeneity in study design, ICS protocols and devices used, bacterial detection methods, and clinical outcome definitions. Differences in bacterial detection methods included different culture media, volumes, incubation times and reporting of outcomes (e.g., only positivity, or also bacterial counts). Variations in patient characteristics, use of antimicrobial prophylaxis and perioperative protocols across institutions further complicate the interpretation and generalisability of the findings. Furthermore, several studies used outdated ICS devices, which may have inferior processing capabilities compared to modern systems. Many of the included studies were conducted decades ago, when intraoperative sterility standards differed from current practice. Improvements such as stricter infection control protocols and refined surgical techniques have reduced (environmental) contamination risk considerably [68–70]. As a result, direct comparison of older and more recent study outcomes is challenging. However, despite advancements in technology and surgical protocols, more recent studies do not consistently report lower contamination rates. This suggests that factors beyond equipment generation and surgical era may continue to contribute to ICS-related contamination.

This scoping review itself has limitations. The methodological quality or risk of bias was conducted for the studies included, except for the ex-vivo studies. However, due to the heterogeneity regarding population, design and outcome measurements, the methodological quality of the studies was not synthesized. The methodological quality is only presented to contextualize the findings. Secondly, the scope and heterogeneity of the topic precluded quantitative synthesis or meta-analysis, which limits the ability to compare findings across settings or interventions in a standardized way. However, due to the broad scope and methodological heterogeneity of the current evidence base, a systematic review was not considered appropriate.

Implications for clinical practice

Given the current body of evidence, ICS may be cautiously considered in selected cases, particularly where expected blood loss is high or allogeneic donor blood is not widely available. The consistent finding that reinfusion of bacterially contaminated blood is not directly associated with postoperative infection suggests that ICS may be feasible under appropriate conditions. Moreover, ICS may offer protective benefits by avoiding allogeneic transfusion and its associated immunomodulatory effects.

As earlier mentioned manufacturers of ICS devices list active infection or gross contamination as contraindications due to the risk of bacteraemia and the UK Cell Salvage Action Group, generally discourage ICS in infected settings [11–13]. We believe these recommendations remain robust. However, this review highlights that bacterial contamination is present in specialties generally not considered as contaminated, like cardiovascular and orthopaedic surgery. Therefore, each clinical decision should be guided by a case-by-case assessment that considers contamination severity, contamination source, patient immune status, and the availability of decontamination techniques such as leukocyte depletion filters and antibiotic prophylaxis. In situations with gross contamination, such as visible enteric contents, reinfusion may carry a higher risk and should be approached with particular caution. Conversely, in cases of minimal contamination involving low-virulence organisms such as skin flora the risk may be more acceptable, especially when decontamination techniques are applied.

Use of ICS in contaminated settings should ideally occur within standardized protocols or clinical trials to ensure patient safety and contribute to the evidence base. Until higher-quality data become available, clinical discretion remains key.

Implications for future research

This review highlights several critical knowledge gaps that should guide future investigations. Firstly, the identification of safe contamination thresholds is needed. Future studies should aim to correlate quantitative bacterial loads with clinical outcomes, to determine whether there are tolerable levels of contamination below which reinfusion is safe. Importantly, such assessments should also account for the type and virulence of the contaminating organisms, as some bacteria are more likely to cause clinical infection than others.

Secondly, the effectiveness of specific decontamination techniques, such LDFs or the addition of antibiotics, requires more rigorous evaluation. Comparative trials could help identify which methods are most effective in reducing bacterial contamination.

Thirdly, prospective, large-scale clinical studies are needed to determine safety of ICS in contaminated fields. Ideally, these would include multicentre cohort studies or controlled trials comparing ICS versus no ICS (or versus allogeneic transfusion) across contaminated surgical scenarios. These studies should systematically document intraoperative factors that may influence contamination risk, such as case order, duration of surgery, frequency of door openings, and the number of personnel present in the operating room. In addition to infection rates, broader clinical outcomes, such as ICU and hospital length of stay, need for allogeneic transfusion, mortality, and cost-effectiveness, should also be assessed.

Establishing the determinants that influence the safety of ICS in contaminated surgical settings is critical. The influence of factors such as microbial resistance patterns, the degree of contamination, and patient-specific variables must be systematically evaluated to better understand when ICS can be safely used. These efforts will be essential for the development of context-specific, evidence-based guidelines for ICS in contaminated surgical fields.

Conclusion

Bacterial contamination of salvaged blood was frequently observed across a wide range of surgical disciplines, including procedures not typically classified as contaminated. Decontamination strategies such as leukocyte depletion filtration and antibiotic additives were implemented in a subset of studies and demonstrated variable effectiveness in reducing bacterial contamination. Among 1203 patients who were reinfused salvaged blood, 55 cases of postoperative bacteraemia were identified. A potential microbiological link between salvaged blood and infection was described in only nine patients across five studies. These findings suggest that reinfusion of bacterially contaminated salvaged blood is not consistently associated with adverse clinical outcomes. However, substantial heterogeneity in methodologies and generally insufficient sample sized to detect rare adverse events limits the ability to draw definitive conclusions. Future research should aim to define safe contamination thresholds, evaluate the effectiveness of decontamination techniques, and assess clinical outcomes in standardized and controlled settings with sufficient sample size to detect adverse events.

Supporting information

S1 Table(Species).(DOCX)

S2 Table(Pre- and post-wash).(DOCX)

S3 Table(Bacterial load).(DOCX)

S4 Table(Prisma checklist filled).(DOCX)

S1 AppendixSearch strategies.(DOCX)