Effectiveness of different antibiotic lock solutions on catheter-related infections of hemodialysis patients: a systematic review and meta-analysis Authors: Sogand Sarmadi, Neda Sanaie, Akbar Zare-Kaseb Published: 2025 ABSTRACT AbstractBackgroundAntibiotic lock therapy (ALT) delivers high‐concentration antimicrobials into catheter lumens and has been proposed to prevent infections in hemodialysis (HD) patients.MethodsWe conducted a systematic review and meta-analysis of randomized clinical trials (RCTs) and quasi-experimental studies comparing ALT with non-antibiotic or no-lock solutions in adults on HD. Six databases (PubMed, Embase, Scopus, Web of Science, Cochrane CENTRAL, CINAHL) were searched through 27 June 2025, identifying 17 studies involving 3,482 catheters. Two independent reviewers extracted data and assessed bias using the Cochrane RoB 2. Random‐effects models yielded pooled risk ratios (RR) and incidence rate ratios (IRR) with 95% confidence intervals (CI).ResultsALT reduced catheter‐related bloodstream infections (CRBSI) by 63% (RR= 0.37, 95% CI: 0.29–0.49; p < 0.001). Overall, catheter‐related infection incidence rates fell (IRR 0.24, 95% CI: 0.15–0.38; p < 0.001) despite potential for publication bias (Egger’s p = 0.003). No significant effects appeared for exit‐site infections (ESI) (RR= 0.90; p = 0.59), thrombosis (RR= 0.73; p = 0.23), or mortality (RR= 0.58; p = 0.09). GRADE assessed the evidence for CRBSI as having high certainty, while evidence for ESI was moderate, and evidence for other outcomes ranged from low to very low.ConclusionsALT significantly reduces CRBSI in HD patients and can be considered as a preventive adjunct. Introduction Chronic kidney disease (CKD) signifies the irreversible decline of renal function, eventually leading to end-stage renal disease (ESRD) and the subsequent requirement for renal replacement therapy (RRT) [1,2]. The global prevalence of CKD is substantial, estimated at 13%, resulting in heightened morbidity and healthcare resource consumption [3]. Hemodialysis (HD) constitutes the primary modality of RRT, supporting more than 90% of individuals in the ESRD stage [2,4]. The utilization of intravascular catheters has risen concomitantly with advancements in blood purification technologies, especially within the field of nephrology. The dependable vascular access essential for HD is typically established through endogenous arteriovenous fistulas (AVFs), arteriovenous grafts (AVGs), and peripherally inserted central catheters (PICCs) [5,6]. Catheter-associated complications—such as thrombosis, stenosis, mechanical failure, and bloodstream infections—remain significant challenges [7,8]. A considerable risk of serious complications, including catheter-related bloodstream infections (CRBSIs) and central line-associated bloodstream infections (CLABSI), is inherent in using central venous catheters (CVCs), contributing to elevated morbidity, mortality rates, and healthcare expenditures [9]. The global incidence of CRBSI varies, with rates estimated at approximately 0.18 per 1,000 catheter-days worldwide [10]. Iran reports an estimated incidence of 1.2–2.5 per 1000 patient-days, while data from developed countries show significantly higher rates [11,12]. In pediatric populations, CRBSI rates vary widely, ranging from 0.2 to 11.3 per 1000 line days, depending on underlying conditions and catheter type [13]. The pathogenesis of CRBSI is complexly associated with microbial biofilm development on catheter surfaces. Bacterial or fungal communities embed within an extracellular matrix, thus fostering adhesion, proliferation, and antibiotic resistance [14,15]. Prevalent pathogens encompass Gram-positive organisms—including Staphylococcus aureus, Coagulase-negative staphylococci, and Enterococci—as well as Gram-negative Bacilli and Candida species [16,17]. Patient-related factors, including immunosuppression, neutropenia, comorbidities like diabetes or malignancy, and treatment-related factors, such as catheter type, insertion method, and lock solution selection, represent key risk factors [11,18]. Conventional CRBSI management entails catheter extraction and systemic antibiotic administration [19,20]. However, removal may prove impractical or risky for patients dependent on CVCs for continuous HD, or those exhibiting limited peripheral venous access or a heightened risk of hemorrhage [6,21]. A conservative prophylactic approach is antibiotic lock therapy (ALT), which involves instilling high-concentration antimicrobial solutions (100–1000× minimum inhibitory concentration (MIC)), often with anticoagulants, into the catheter lumen during dialysis to directly target intraluminal biofilm, aiming for pathogen eradication and catheter salvage [1,4,22]. The suitability of several antibiotics for incorporation into locking solutions has been the subject of recent studies. Using gentamicin-citrate locks resulted in a significant decrease in CRBSI rates without inducing resistance. Using vancomycin-heparin locks demonstrated a statistically significant increase in catheter patency duration and a concomitant reduction in infection rates [23]. In vitro and in vivo studies showed that the Minocycline-EDTA combination yielded superior biofilm eradication against methicillin-resistant Staphylococcus aureus [24]. Also, Daptomycin demonstrated high efficacy in biofilm removal [25]. Data from adult cohorts with hospital-acquired infections show that prophylactic and therapeutic ALT, with systemic antibiotics, significantly decrease CRBSI incidence and catheter removal rates; reported cure rates surpass 70% in treatment-refractory cases [26]. Although the number of pediatric studies is limited and most are retrospective, they report similar success in catheter retention without safety concerns [13,27]. Despite this, established protocols typically recommend against antimicrobial lock solutions for infections caused by Staphylococcus aureus or Candida, when catheter replacement is feasible [1,4,28]. Effective ALT necessitates using agents compatible with anticoagulants (such as heparin), demonstrating stability across dwell time, and exhibiting tailored activity against identified pathogens [26,29]. In addition to ALT, non-antibiotic lock solutions, such as citrate, taurolidine, and tetrasodium EDTA, have been used to prevent CRBSIs in patients undergoing HD. These solutions are regarded as effective in environments where reducing antibiotic utilization is a priority, as they inhibit microbial adhesion and biofilm development [29–31]. Notwithstanding numerous studies on HD affirming the effectiveness of antimicrobial lock therapy in reducing CRBSI rates, the literature remains heterogeneous regarding the antimicrobial agents used, lock solution concentrations, and dwell times. Moreover, there is a notable absence of direct comparisons among various antibiotic lock formulations, particularly when contrasting PICCs with tunneled central venous catheters. The paucity of data from pediatric and immunocompromised populations impedes the ability to confidently extrapolate the findings from adult studies to these vulnerable groups. Additionally, a consensus regarding standardized ALT protocols, which include the selection of optimal antimicrobial-anticoagulant combinations based on pathogen susceptibility, continues to be elusive. An innovative aspect of this systematic review is its synthesis of all available interventional evidence comparing different antibiotic lock formulations across various catheter types. Previous reviews have not systematically evaluated the relative efficacy and safety profiles of these lock solutions. The development of evidence-based guidelines and guidance for future randomized clinical trials (RCTs) on antibiotic lock solutions in HD requires a comprehensive systematic review and critical appraisal of current data. Accordingly, this study aims to conduct a systematic review and meta-analysis to assess the efficacy of different ALTs in HD patients with catheter-related infections (CRI). Method The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Supplementary Material 1) guided this systematic review and meta-analysis [32]. Our method strictly adhered to the recommendations outlined in the Cochrane Handbook for Systematic Reviews of Interventions [33]. The review protocol was prospectively registered in the PROSPERO database under registration number CRD420251086155. Literature search strategy A medical information specialist developed and executed the comprehensive search strategy. We drew upon MeSH and Emtree terms, pertinent literature, and clinical expertise to identify keywords and their synonyms. The electronic search was then performed across MEDLINE (via PubMed), Embase (Elsevier), Scopus, Web of Science, Cochrane CENTRAL, and CINAHL (EBSCOhost) from database inception through 27 June 2025. We also hand‑searched the reference lists of all included studies to capture any additional records. Beyond our primary database queries, the first 200 results from Google Scholar and registrations listed on the International Clinical Trials Registry Platform (ICTRP) and ClinicalTrials.gov were examined to identify further eligible trials. We used the following keywords and their alternatives: “Anti-Infective Agents,” “Hemodialysis,” and “Catheter-Related Infection.” The full search syntax is presented in Supplementary Material 2. We also consulted subject‑matter experts to identify sources not captured by our database and gray‑literature searches. No language restrictions were applied to the studies considered for inclusion. Eligibility criteria The PICO(S) approach (population, outcome, intervention, comparisons, and study design) guided the classification of studies addressing the primary research question for interventional studies, respectively [34]. Participants In our review, studies were included based on the population, with inclusion criteria consisting of individuals undergoing chronic HD via either a tunneled or non-tunneled catheter. Interventions The study included all types of ALTs. Moreover, our study encompassed a broad range of antibiotics. Comparisons Any type of non-antibiotic lock techniques or no interventions were eligible for inclusion. In fact, control groups included patients receiving heparin-only locks, non-antibiotic antimicrobial/anticoagulant locks (citrate, taurolidine, sodium bicarbonate), or no lock intervention, depending on the original study design. Outcomes CRBSI is characterized as a clinical diagnosis in which, through concordant clinical and microbiological evidence, an intravascular catheter is identified as the source of a patient’s bloodstream infection [35]. Also, Exit-site infection (ESI), Catheter-related thrombosis, and mortality were among our secondary outcomes. Study design Eligible studies used an interventional design to assess ALT’s impact on CRI. The following studies were excluded: Observational studies, animal studies, systematic reviews, meta-analyses, studies with no accessible data, case reports, case series, and conference abstracts. Study selection Deduplications were done through HubMeta. Two authors (A.Z. and S.S.) independently screened the titles and abstracts to determine the eligibility of all studies identified by the earlier strategies. Articles with uncertain eligibility underwent full‑text review. Each study was evaluated against the predefined inclusion and exclusion criteria. Two authors conducted assessments independently, and any discrepancies were resolved through discussion between the two investigators. Persistent disagreements were adjudicated by a third reviewer (N.S.), who reassessed the data and facilitated further discussion until consensus was reached. Data extraction Data were extracted from the studies included in the systematic review by at least two (A.Z. and S.S.) independent reviewers using a data‐extraction sheet (Table 1). This form captured information on the study reference, targeted population, mean age of participants, geographical details, intervention details, comparison details, and outcome measured. We piloted the extraction form on 10% of the included studies and refined it as necessary. Any discrepancies between reviewers were resolved through discussion or, if required, with input from an additional reviewer (N.S.). When data were missing or unclear, the study authors were contacted to obtain or clarify the information. Table 1.Basic characteristics of the included interventional studies in this systematic review.No.First Author/yearStudy characteristics1. Design2. Sample Size (I/C)3. CountryParticipantsCatheter TypeMean Age (SD)Measured OutcomesInterventionComparison1.Al-Hwiesh et al. 2008 RCT69(36/33)Saudi Arabia HDTunneled cuffed central venous catheters (TCC)47.4 ± 11.5- Infection-free catheter survival- Incidence rate of infections- Rate of infection- Bacteremia rate- Rate of clinical sepsis- Rate of access site infectionAt the end of each dialysis session, a lock solution of vancomycin (25 mg/mL), gentamicin (40 mg/mL), and heparin (5,000 U/mL) was instilled (1.5–1.75 mL venous side; 1.25–1.5 mL arterial side) and withdrawn immediately before the next session. Catheter‐related infections were monitored via exit‐site and blood cultures, and vancomycin and gentamicin blood levels were measured at each session.Group II with routine TCC placement and management.2.Anil et al. 2012 RCT82(39/43)Saudi Arabia HDTunneled central venous catheters53.7 ± 17.2- Presence of MSSA or MRSA in nasal swabs- CRBSI incidence- TCC survival at 365 daysGroup I (Cefotaxime–Heparin Lock):After each 3–4-hour dialysis session, a syringe containing 10 mg/mL cefotaxime mixed with 5,000 IU/mL heparin—pre‑filled and coded for blinding—was instilled into both lumens of the tunneled catheter and withdrawn immediately before the next session. Catheters were placed under fluoroscopy with strict aseptic technique and verified by chest radiograph.Group II (Standard Heparin Lock):Following identical dialysis schedule and catheter placement protocols, patients received syringes containing only 5,000 IU/mL heparin, coded in the same manner. The lock solution was instilled and removed on the same timetable as in Group I, with no additional prophylactic antibiotics or decolonization measures.3.Arabzadeh et al. 2025 RCT60(30/30)Iran HDTunneled central venous cathetersNA-CRI (diagnosed using CDC criteria)- Repeat CRI Episodes (CDC criteria)- Exit‐Site Infections (Clinical assessment)- Catheter Dysfunction or Need for Exchange/Removal-Clinical Signs and SymptomsGroup B received a lock solution of 5,000 IU/mL heparin plus 500 mg/mL vancomycin and 500 mg/mL amikacin. After each dialysis session, the solutions were instilled into both lumens, left in situ during the interdialytic interval, and withdrawn before the next session. The procedure was repeated throughout the study under double‑blind conditions. Patients were followed for up to 6 months, assessed three times weekly by nephrologists and nurses for signs of infection (fever, redness, pain), and given written instructions and a contact number for interim concerns.Group A patients received 5,000 IU/mL heparin in each catheter lumen4.Bueloni et al. 2019 A quasi-experimental trial127(65/62)Brazil HDTunneled central venous catheters60.2 (15.7)- Catheter-related bloodstream infections (CR-BSI)- Exit site infections (ESI)- Presence of gram-positive oxacillin-resistant pathogens- Mechanical complications- Catheter removal complicationsPatients with an initial CR‑BSI and no hospitalization in the prior three months received 2 g IV cefazolin post‑dialysis for 21 days. Those with severe symptoms (e.g., septic shock), recent hospitalization, or prior broad‑spectrum antibiotic use received 15 mg/kg IV vancomycin plus either 500 mg levofloxacin or 2 g ceftazidime after each session, with dosing adjusted per antibiogram results.Group 1 (Antibiotic Lock):During the interdialytic interval, this group received a lock solution comprising cefazolin 12 mg/mL, gentamicin 7 mg/mL, and heparin 3,500 IU/mL. The solution volume exactly filled each catheter lumen according to its size.Group 2 (Tauro‑Lock‑Hep500):This group received a lock solution containing taurolidine, 4% citrate, and heparin 500 IU/mL (Tauro‑Lock‑Hep500). As in Group 1, the solution volume matched the lumen capacity precisely.5.Campos et al. 2011 RCT187(92/95)Brazil HDBoth tunneled and non-tunneled54.56 (16.86)- Catheter-related bacteremia- Rate of catheter-related bacteremia per 1000 catheter-days- Rate of catheter removal for dysfunction- Catheter-related bacteremia-free survivalGroup (Minocycline–EDTA Lock):Using the same timing and procedural steps, this group received a Minocycline–EDTA solution (3 mg/mL minocycline + 30 mg/mL EDTA; CATH‑SAFE®), prepared by diluting the lyophilized content with 3 mL saline. Lumens were locked in the operating room and relocked post‑dialysis after saline flushes, with volumes precisely matching lumen capacity. Identical monitoring of catheter function and exit‑site care was performed as in the other group.Group (Heparin Lock):Immediately after catheter implantation, both lumens were filled with unfractionated sodium heparin (5,000 U/mL) and locked before leaving the operating room. After each dialysis session, lumens were flushed with 10 mL isotonic saline and relocked with heparin in a volume matching the internal lumen capacity. Catheter blood flow, exit‑site appearance, and tunnel status were assessed and documented at every session, and exit‑site dressings were changed post‑dialysis following chlorhexidine‑alcohol cleansing by masked, gloved staff.6.Filiopoulos et al. 2011 RCT177 (225/226)Greece HDNot Defined.72 (50–80)-CRB episodes-cumulative CRB-free catheter survival at 90 days-Thrombosis ratesGroup A (Gentamicin–Heparin Lock):At the end of each dialysis session and continuously since catheter insertion, patients received a lock solution of gentamicin 40 mg/mL plus unfractionated heparin 5,000 U/mL (1:3 ratio), using 2 mL per lumen (0.5 mL gentamicin + 1.5 mL heparin).Group B (Taurolidine–Citrate Lock):Under the exact timing, patients received 2 mL per lumen of TauroLock™ (1.35 % taurolidine + 4 % sodium citrate).Group C (Heparin‑Only Historical Control):The historical control group received 2 mL per lumen of unfractionated heparin 5,000 U/mL after each session.7.Kim et al. 2006 RCT120(60/60)Korea HDNot Defined.53.68(15.20)-Catheter-associated bacteremia (per 1000 patient-days)- thrombosisAntibiotic–Heparin Lock Group (ALT Group):Patients received a catheter lock containing heparin (1,000 U/mL), cefazolin (10 mg/mL), and gentamicin (5 mg/mL) after each dialysis session, left in place until the next. Mahurkar catheters were inserted under imaging, with care per NKF‑K/DOQI guidelines.Heparin‑Only Control Group:Patients received a heparin-only lock solution (1,000 U/mL) after each dialysis session. Both patients and nurses were blinded to group assignment. The primary outcome was catheter-related bacteremia (CRB), defined by growth of the same organism from the catheter tip (≥ 15 CFU), peripheral, and catheter blood cultures.8.Mcintyre et al. 2004 RCT50(25/25)United Kingdom HDTunneled central venous catheters63.6 ± 2.8-Catheter-related infection (CRI) rates (infective episodes per 1000 catheter days)- Epoetin requirements (mean dose in IU/week)The intervention arm received an antibiotic lock comprising gentamicin (5 mg/mL) combined with heparin (5,000 IU/mL) before catheter insertion—using concentrations optimized for heparin stability throughout the interdialytic periodreceived the standard catheter lock solution of unfractionated heparin (5,000 IU/mL in saline).9.Moghaddas et al. 2015 RCT86 (45/41)Iran HDTunneled cuffed central venous catheters (TCC)63.63 ± 10.63-CRBSI rates per 1000 catheter-days-cumulative infection-free catheter survivalAntibiotic‑Lock Group:After each hemodialysis session, the catheter lumen was flushed with 10 mL saline and then locked for the entire interdialytic period with 10 mg/mL cotrimoxazole (trimethoprim-based) plus 2,500 U/mL heparin, in a volume matching the lumen’s internal capacityControl Group (Heparin‑Only):Following each session, lumens were similarly flushed with 10 mL saline and locked with 2,500 U/mL unfractionated heparin alone.10.Moran et al. 2012 RCT303(155/148)USA Adult maintenance hemodialysis patients using a tunneled cuffed CVC for vascular accessTunneled cuffed central venous catheters (TCC)63.4(15.6)-catheter-related bloodstream infection (episodes/1,000 catheter-days)- catheter clotting (events/1,000 catheter-days)Gentamicin–Citrate Lock Group (Intervention):Following the same flush and blood‑return procedure, a gentamicin lock solution (320 µg/mL) in 4 % sodium citrate was instilled to fill each lumen. Exit‑site care and triple‑antibiotic ointment application were similar to the control arm.Heparin Lock Group (Control):After each dialysis session, blood was returned per protocol, and each catheter lumen was flushed with 10 mL of 0.9 % saline. Then, 1,000 U/mL heparin solution was instilled to fill each lumen. Triple‑antibiotic ointment was applied to the exit site at each dressing change, and exit‑site dressings were replaced every session.11.Nassiri et al. 2023 RCT86(44/42)Iran HDTunneled cuffed central venous catheters (TCC)63.36 ± 13.2infection rate (positive peripheral blood or catheter cultures)Antibacterial Lock Group:An in‑house vancomycin–heparin lock was prepared by reconstituting 500 mg vancomycin in 10 mL diluent (50 mg/mL), then diluting 1 mL of this stock into 9 mL saline to yield 10 mL at 5 mg/mL vancomycin. For each port, 1.5 mL (7.5 mg vancomycin) was instilled, with the remaining lumen volume filled with 5,000 U heparin.TauroLock Group:At the end of each dialysis session, pre‑made commercial TauroLock™ solution was instilled into each catheter lumen and left in place until the next session, following the manufacturer’s instructions for volume and dwell time.12.Zhang et al. 2009 RCT140 (71/69)China Patients with a newly inserted permanent tunneled cuffed hemodialysiscatheterTunneled cuffed central venous catheters (TCC)52.0 ± 16.31- Incidences of Catheter-Related Infection2- Infection at the catheter exit site3- Catheter thrombosis4- CRI-related admissionsAntibiotic-heparin lock solution (gentamicin 4 mg/ml and unfractioned sodium heparin 5,500 IU/ml,G group)Catheter locked with either unfractioned sodium heparin 5,500 IU/ml alone13.Tao et al. 2012 RCT348 (168/180)China Hemodialysis PatientsNot Defined.55.2 ± 14.7Incidents of catheter-related infection, catheter blockage, local bleeding, and unexpected catheter loss.Each lumen was locked with a mixture of 2 mL heparin (100 mg), 2 mL normal saline, and 0.5 g ceftazidime.Each lumen was locked with 2 mL heparin (100 mg) and 2 mL normal saline.14.Sofroniadou et al. 2012 RCT131 (51/49/52)Greece Hemodialysis Patients with nontunneledcatheter (NTC)Nontunneled CathetersIG1: 72 (62–80)IG2: 67.5 (47–75)1- Diagnosis of CRBSI2- Exit site infections, bacterial colonization, thrombosis, andbleeding episodes.IG1: vancomycin (5 mg ⁄ml) + unfractionated heparin (2000 U⁄ml)IG2: linezolid (2 mg ⁄ ml) + unfractionated heparin (2000 U⁄ml)unfractionated heparin (2000 U⁄ ml) alone15.Saxena et al. 2006 RCT109 (51/58)Saudi Arabia Diabetic ESRD patients were eligible for the study if theyrequired insertion of a TCC (SC or IJC) for the maintenance orinitiation of HD.Tunneled cuffed central venous catheters (TCC)59.7 ± 171- Symptomatic CRBSI and/or catheter thrombosis2- Events of electiveremoval of the catheter and CRBSI-related deathCefotaxime–heparin lock (cefotaxime, 10 mg/ml and heparin,5000 u/ml)Standard heparin lock (5000 u/ml)16.Salman et al. 2023 RCT160 (80/80)India Patients initiated on HD through a non-tunneledCVC inserted in the internal jugular vein (IJV)Internal jugular non-tunneled CVC49.91 (11.46)1- Safety through comparing catheter loss to CD or to CRT and CRBSI in the two study groups.2- Efficacy was assessed by the adequacy of blood flow (>300 mL/min). Acceptable empirical evidence of CRBSIThe catheter was locked with a solution of 0.2 mL gentamycin (1 mL = 20 mg) plus 3 mL heparin (1 mL = 1000 units)The catheter was locked with a 7.5% sodium bicarbonate solution17.Pervez et al. 2002 RCT55 (14/22/19)USA Patients who underwent tunnel catheter placement or changeTunneled cuffed central venous catheters (TCC)53.7 ± 4.01- Catheter-associated bacteremia and thrombosis2- Catheter loss due to access maturation, transplant, ortransferAntibiotic lock with tricitrasol (46.7%), gentamicin (40 mg/ml), and saline in a ratio of 1:5:5 and catheter hub covered with a sterile plastic bag after cleaning with a 10% povidone iodine solution.CG1: Heparin lock and sterile plastic bag over the catheter hub after cleaning with povidone iodineCG2: Heparin lock alone Basic characteristics of the included interventional studies in this systematic review. RCT RCT 69(36/33) 69(36/33) Saudi Arabia Saudi Arabia RCT RCT 82(39/43) 82(39/43) Saudi Arabia Saudi Arabia RCT RCT 60(30/30) 60(30/30) Iran Iran A quasi-experimental trial A quasi-experimental trial 127(65/62) 127(65/62) Brazil Brazil RCT RCT 187(92/95) 187(92/95) Brazil Brazil RCT RCT 177 (225/226) 177 (225/226) Greece Greece RCT RCT 120(60/60) 120(60/60) Korea Korea RCT RCT 50(25/25) 50(25/25) United Kingdom United Kingdom RCT RCT 86 (45/41) 86 (45/41) Iran Iran RCT RCT 303(155/148) 303(155/148) USA USA RCT RCT 86(44/42) 86(44/42) Iran Iran RCT RCT 140 (71/69) 140 (71/69) China China RCT RCT 348 (168/180) 348 (168/180) China China RCT RCT 131 (51/49/52) 131 (51/49/52) Greece Greece RCT RCT 109 (51/58) 109 (51/58) Saudi Arabia Saudi Arabia RCT RCT 160 (80/80) 160 (80/80) India India RCT RCT 55 (14/22/19) 55 (14/22/19) USA USA Risk of bias assessments Two reviewers (A.Z. and S.S.) independently assessed the risk of bias (RoB) for all included studies, employing assessment tools tailored to each study design [36]. The RoB in randomized and non-randomized controlled trials was evaluated using the Cochrane Collaboration’s RoB 2.0 tool. The RoB 2.0 tool was organized into five domains through which bias may be introduced [1]: bias arising from the randomization process [2], bias due to deviations from intended interventions [3], bias attributable to missing outcome data [4], bias in outcome measurement, and [37] bias in the selection of reported results. For each signaling question within these domains, reviewers assigned one of five response options: ‘yes’ (Y), ‘probably yes’ (PY), ‘probably no’ (PN), ‘no’ (N), or ‘no information’ (NI). These responses informed domain‑level RoB judgments, which were then synthesized into an overall RoB rating for each study outcome. Overall judgments were classified as “low risk of bias,” “some concerns,” or “high risk of bias” [38]. Where consensus could not be reached, a third reviewer (N.S.) was engaged to reassess study quality and facilitate resolution of any remaining disagreements until a final consensus was achieved. Statistical analysis Risk Ratios (RRs), computed using a random-effects model with restricted maximum likelihood (REML) and 95% confidence intervals (CIs), served as the summary statistics for the outcomes of interest. REML’s selection was based on its ability to generate superior estimates of between-study variance (τ2) compared to conventional methodologies, especially in meta-analyses with few heterogeneous studies. This enhances the precision of our pooled effect estimates [39]. Effect estimates were interpreted using conventional RR thresholds adapted from the GRADE Working Group’s partially contextualized approach, which defines ranges for trivial (RR 0.90–1.10), small (RR 0.75–0.89 or 1.11–1.49), moderate (RR 0.50–0.74 or 1.50–1.99), and large effects (RR < 0.50 or ≥ 2.00) [40]. Also, to determine the infection rate, an Incidence rate ratio (IRR) was derived by dividing the Incidence Rate (IR) in the intervention groups by that in the control groups for each study. IRRs and Standard Errors (SEs) were used to estimate the pooled IRR from the included studies in the meta-analysis. Cochran’s Q test (χ2) and the I2 statistic were used to assess variance between studies and heterogeneity in estimates. Cochran’s Q was reported as χ2, while I2 was reported as percentages. Statistical heterogeneity was quantified by measuring I2 statistic. Cochrane cut-off points were used to interpret these measures: 0%-40% not important; 30%-60% moderate heterogeneity; 50%-90% substantial heterogeneity; 75%-100% considerable heterogeneity [41]. Several subgroup analyses were conducted to address heterogeneity across geographical, sociodemographic, and study-related factors. To evaluate and ascertain the statistical significance of publication bias evident in the funnel plot, we performed Begg’s and Egger’s regression intercept tests. The Egger’s test is prone to type 2 errors in small samples. The Duval and Tweedie trim-and-fill method was applied to address the missing publications identified in the funnel plot [42]. We conducted a sensitivity analysis using the leave-one-out-remove method to evaluate the consequences of excluding each study from the overall estimates. The analysis was performed using Stata version 17. GRADE certainty of evidence An assessment of the outcomes’ importance was conducted per-protocol (Supplementary Material 3). We examined the certainty of evidence for the primary outcomes using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach [43]. The implemented criteria are fully presented in Supplementary Material 4. Results Systematic review Systematic searches yielded 2114 records. Upon importing all citations into HubMeta and eliminating 975 duplicates, 1139 unique records were available for title and abstract screening. At this stage, 1098 records were excluded; 695 did not align with the research parameters, and 403 were deemed unsuitable due to inappropriate study design or non-clinical context. Full texts of 39 potentially eligible articles were retrieved and independently assessed in duplicate; attempts to retrieve two records were unsuccessful. Following a comprehensive full-text review, 22 studies were excluded; the rationale for each exclusion is provided in Supplemental Material 5. Subsequently, 17 studies met all inclusion criteria and were included in the meta-analysis [23,24,44–57]. The complete study selection process is illustrated in the PRISMA 2020 flow diagram (Figure 1). Of the included studies, 16 were RCTs and one employed a quasi-experimental design [46]. In this meta-analysis, Saudi Arabia and Iran each contributed three studies; the United States, Brazil, China, and Greece each contributed two; and India, South Korea, and the United Kingdom each contributed one. The most frequently used antibiotic agents were gentamicin (used in six studies, including combinations with heparin, citrate, or other agents) and vancomycin (in four studies, either alone with heparin or combined with other antibiotics such as amikacin, gentamicin, levofloxacin, or ceftazidime). Other antibiotics included cefotaxime (two studies), cefazolin (one study), ceftazidime (one study), minocycline–EDTA (one study), and cotrimoxazole–heparin (one study). Of the 17 included studies, the majority focused on tunneled catheters, with tunneled cuffed central venous catheters (TCC) reported in seven studies (41.2%) and tunneled catheters without a cuff in four studies (23.5%). Non-tunneled catheters were included in two studies (11.8%); one study (5.9%) assessed both tunneled and non-tunneled catheters; and three studies (17.6%) did not define the catheter type. Regarding the control groups, most studies used Heparin-only locks (n = 14; 82.35%) as the comparator. Other non-antibiotic antimicrobial or anticoagulant lock solutions included Sodium Bicarbonate (n = 1; 5.88%), TauroLock (n = 1; 5.88%), and a combination of Taurolidine, 4% Citrate, and Heparin (n = 1; 5.88%). The detailed characteristics of the included studies are presented in Table 1. Results of RoB assessments showed that all included studies were “low-risk” according to RoB 2.0 (Figure 2). PRISMA Flowchart. Risk of bias and summary graph of the included studies. Meta-analysis Catheter-related bloodstream infection Our analysis showed that there was a statistically significant difference between the ALT and control groups in terms of the risk of CRBSI (RR= 0.37, 95% CI: 0.29– 0.49; I2 =0.00%; p < 0.001) (Figure 3). Studies employing solely heparin as a control group showed a larger effect size, suggesting a more substantial intervention impact within this subset (RR= 0.34 vs 0.71 and 0.60). However, this difference was not statistically significant (p = 0.27). The effects of the intervention were non-significantly greater in studies designed and conducted in Greece and China (p = 0.70). The study combining vancomycin, heparin, and amikacin as the ALT solution showed enhanced effects. The type of ALT solution used showed no statistically significant difference between subgroups (p = 0.95). Furthermore, the subgroup analysis based on catheter type indicated that whether the catheter was tunneled or non-tunneled, and whether it was cuffed or non-cuffed, did not significantly affect the observed effect size (p = 0.99). Sensitivity analysis showed that excluding Nasiri et al.’s study resulted in the largest change in the aggregate effect size, a shift of 0.02 units [52]. The funnel plot analysis also showed no publication bias within the selected studies. Begg’s and Egger’s tests further supported this finding (p = 0.47) (Supplementary Material 6). Estimated pooled RR for CRBI. Exit-site infection The meta‐analysis demonstrated that the ALT intervention was not associated with a significantly lower incidence of ESI compared to the control group (RR = 0.90; 95 % CI: 0.63–1.30; I2 = 0.00 %; p = 0.59) (Figure 4). Subgroup analysis showed that the effectiveness of the interventions was considerably higher in studies conducted in China (RR= 0.67; 95 % CI: 0.26–1.74; I2 = 0.00 %; p = 0.55) and Iran (RR= 0.79; 95 % CI: 0.04–17.30; I2 = 83 %; p = 0.02). Notably, in studies from Iran, the intervention’s effectiveness was associated with a statistically significant difference in the RR for ESI between the groups. Additionally, subgroup analysis by catheter type showed that neither tunneling nor cuff presence significantly influenced the observed effect size (p = 0.51). The sensitivity analysis demonstrated that omitting the Arabzadeh et al. study produced the most pronounced change in the pooled effect estimate, shifting it by 0.08 units [23]. Furthermore, assessment of publication bias via the funnel plot revealed no discernible asymmetry, which was corroborated by both Begg’s and Egger’s tests (p = 0.72) (Supplementary Material 7). Estimated pooled RR for ESI. Thrombosis The meta‐analysis found that the ALT intervention did not produce a statistically significant reduction in thrombosis risk compared with the control group (RR = 0.73; 95 % CI: 0.44–1.21; I2 = 42.8 %; p = 0.23) (Figure 5). The sensitivity analysis indicated that removing the Saxena et al. trial led to the most significant deviation in the overall effect estimate, amounting to a 0.22‑unit shift [55] (Supplementary Material 8). Furthermore, assessment of publication bias via funnel plot analysis revealed no apparent asymmetry, a finding corroborated by Begg’s and Egger’s tests (p = 0.42). Estimated pooled RR for Thrombosis. Mortality Meta-analysis revealed a decreased mortality risk associated with the ALT intervention relative to the control group (RR = 0.58; 95% CI: 0.30–1.09; I2 = 0.00%; p = 0.09), although this reduction did not reach statistical significance (Figure 6). The sensitivity analysis indicated that excluding the Saxena et al. trial resulted in the most significant deviation in the overall effect estimate, with a shift of 0.17 units [55] (Supplementary Material 8). This finding was supported by both Begg’s and Egger’s tests (p = 0.54), indicating no evidence of publication bias. Estimated pooled RR for Mortality. Catheter-related infections (CRI) incidence rate Ten studies reported the incidence rate of catheter-related infections (CRIs), defined as the number of new CRIs events per 1,000 catheter-days. This metric facilitates standardization across studies with varying follow-up periods by considering both the frequency of infections and the total duration of catheter exposure. Pooled analysis showed a statistically significant reduction in CRIs with ALT (IRR = 0.24; 95% CI: 0.13–0.45; p < 0.001; I2 = 79.27%) (Figure 7). Moreover, visual inspection of the funnel plot revealed notable asymmetry, indicating potential publication bias, which was further supported by Egger’s test results (p = 0.003). Additionally, the trim-and-fill method suggested imputing four studies that were potentially missing (Supplementary Material 9). Estimated pooled incidence rate ratio (IRR) for CRI. Certainty of evidence Results of the certainty assessment using GRADE criteria showed that the reported findings for CRBSI were of high certainty. This means that our findings are an excellent indication of the likely effect of CRBSI. The likelihood that the impact will be substantially different is low. For ESI, our findings were of moderate certainty. This means that this research provides a good indication of the likely effect of the mentioned outcome. The likelihood that the impact will be substantially different is moderate. The evidence for thrombosis, mortality, and CRI incidence rate was of low to very low certainty. The results of the certainty assessment are presented in Table 2. Table 2.Certainty assessment of evidence for each outcome.No of studiesDesignRisk of biasInconsistencyIndirectnessImprecisionPublication BiasStrong associationDose responseAll plausible confounding & biasCertainty (overall score)The use of ALT compared to usual care.Outcome: Catheter‐Related Bloodstream Infections10RCT[4]No Serious Risk of bias (0)No Serious inconsistency (0)Serious indirectness(−1)No serious imprecision (0)None(0)Strong(+1)No(0)No(0)⊕⊕⊕⊕HighThe use of ALT compared to usual care.Outcome: Exit-Site Infection7RCT[4]No Serious Risk of bias (0)No Serious inconsistency (0)Serious indirectness(−1)Serious imprecision (−1)None(0)No(0)Evidence of a gradient (+1)No(0)⊕⊕⊕⊖ModerateThe use of ALT compared to usual care.Outcome: Thrombosis5RCT[4]No Serious Risk of bias (0)No Serious inconsistency (0)Serious indirectness(−1)Very Serious imprecision(−2)None(0)No(0)No(0)No(0)⊕⊖⊖⊖Very lowThe use of ALT compared to usual care.Outcome: Mortality4RCT[4]No Serious Risk of bias (0)No Serious inconsistency (0)Serious indirectness(−1)Very Serious imprecision(−2)None(0)Strong(+1)No(0)No(0)⊕⊕⊖⊖LowThe use of ALT compared to usual care.Outcome: Catheter-related incidence rate10Interventional[3]No Serious Risk of bias (0)No Serious inconsistency (0)Serious indirectness(−1)No serious imprecision (0)Very serious(−2)Strong(+1)No(0)No(0)⊕⊖⊖⊖Very low Certainty assessment of evidence for each outcome. Discussion This systematic review and meta-analysis revealed that ALT consistently and significantly reduced CRBSI; however, its effects on ESIs, thrombosis, and mortality were limited and statistically insignificant. Studies consistently showed a protective effect of the intervention on overall CRI incidence rates. Notably, the lack of significant heterogeneity in key outcome measures supports the robustness of our findings; however, potential publication bias in the broader CRI incidence rate analyses causes cautious interpretation and further research. The study showed a statistically significant association between using ALT and a lower risk of CRBSIs. This aligns with the conclusions of Sheng et al.’s network meta-analysis, which evaluated the efficacy and safety of various lock solutions for preventing catheter-related complications in adult HD patients. Results showed a significant association between ALT use and a decreased risk of CRBSI. According to established effect-size interpretations, the intervention showed a moderate to large impact [58]. Although prior meta-analyses, including those conducted by Zacharioudakis et al. [59] and Sheng et al. [58], have demonstrated the overall effectiveness of antimicrobial lock solutions in reducing CRBSI, our investigation expands and refines the existing body of evidence in several significant aspects. Firstly, in contrast to earlier reviews that encompassed diverse patient populations—such as oncology, parenteral nutrition, and intensive care—this meta-analysis concentrates solely on adult HD patients, thereby providing a targeted, disease-specific synthesis. Secondly, by incorporating 17 RCTs and quasi-experimental studies published up to June 2025, our analysis includes the latest clinical data not previously considered in earlier reviews. Thirdly, the utilization of advanced statistical methodologies—including IRR estimation, publication bias correction via trim-and-fill, and GRADE certainty assessment—bolsters the methodological robustness and clarity of the results. Finally, by differentiating ALT from non-antibiotic alternatives, this study offers a more clinically relevant and current synthesis that directly supports evidence-based medical practice. In addition, a 2008 study investigated the efficacy of antimicrobial lock solutions in preventing infections associated with intravascular catheters during HD. The results showed ALT contributed to a substantial decrease in CRBSIs. The study, as mentioned earlier, reported a large effect size. Our findings corroborate this [60]. Analysis from a 2021 systematic review revealed a statistically significant decrease in CRBSI cases associated with antimicrobial lock solutions, including antibiotic and non-antibiotic types [61]. A recent systematic review and meta-analysis evaluating antibiotic and antimicrobial lock solutions for CRBSI prevention showed significantly reduced CRBSI incidence. Moreover, the observed effect size was substantial and of clinical importance, highlighting the considerable effectiveness of these interventions in reducing the incidence of infection among HD patients [62]. In 2008, a study by James et al. assessed the prophylactic effect of antibiotics against CRBSI in patients undergoing HD. Findings indicated a statistically significant decrease in CRBSI among HD patients following intraluminal antibiotic administration. Consistent with our study, this research also revealed a large effect size, confirming the high efficacy of this intervention in preventing CRBSI [63]. The subgroup analysis in our review, stratified by solution type, revealed no significant impact of solution type on the intervention’s efficacy in reducing CRBSI. This result is entirely consistent with the findings reported by Zacharioudakis et al. [60]. Our study findings regarding the effectiveness of ALT in reducing the incidence of CRBSI are consistent with those of previous research. The alignment of these results reinforces the established importance of ALT in preventing CRIs, especially in HD. Syntheses emphasize that non-antibiotic antimicrobial locks—particularly taurolidine-containing and trisodium citrate solutions—serve as effective alternatives to heparin and, in numerous head-to-head comparisons, demonstrate performance comparable to that of antibiotic-containing locks. For instance, a targeted meta-analysis of taurolidine-containing locks reported a 70% lower incidence rate in favor of taurolidine across various patient populations. Nonetheless, the authors determined that the overall quality of evidence was constrained by RoB and issues of indirectness [64]. A comprehensive systematic review of antibiotic and antimicrobial catheter locks in HD populations similarly concluded that antimicrobial strategies—such as antibiotic combined with low-dose heparin, citrate, or taurolidine formulations—are effective in reducing CRBSI compared with heparin alone. The review also emphasized heterogeneity in formulations and raised concerns regarding the emergence of antimicrobial resistance [30]. Notably, several RCTs and quasi-experimental head-to-head studies have failed to demonstrate a definitive superiority of pure antibiotic-only locks over high-quality non-antibiotic antimicrobial locks, indicating that non-antibiotic options may be clinically competitive. However, the number of direct comparisons remains limited, and these studies are often characterized by clinical heterogeneity—differing in carrier solutions, antibiotic concentrations, catheter types, and outcome definitions—which diminishes the statistical power of subgroup analyses and complicates assertions of definitive superiority. It is important to note that in some trials, antibiotics were administered in a citrate carrier (gentamicin in 4% citrate), blurring the distinction between an “antibiotic” strategy and a “citrate” approach and thereby impeding straightforward head-to-head comparisons. In our review, the limited number of RCTs directly comparing pure antibiotic-only locks with single non-antibiotic antimicrobial locks constrains the ability of subgroup analyses to yield conclusive results. Consequently, although ALT reduces CRBSI overall, clinicians must weigh the potential benefits against the risks of antimicrobial resistance. Future research necessitates adequately powered, well-designed RCTs that directly compare pure antibiotic locks with non-antibiotic alternatives (taurolidine or citrate at clinically relevant concentrations). Meta-analysis revealed no statistically significant reduction in ESI incidence for the ALT intervention compared to the control. The results suggest that the impact of ALT on ESI prevention is uncertain, despite potential benefits in other contexts, thus requiring further study. A comparable research found that antimicrobial solutions were ineffective in reducing catheter ESI; citrate solutions alone were highly effective [65]. An additional study of HD patients showed the limited efficacy of ALT solutions in ESI reduction [63]. Network meta-analysis indicated no significant effect of antibiotic lock solutions for reducing ESI, except for trisodium citrate, which showed a significant decrease [58]. A 2014 meta-analysis by Zacharioudakis et al. evaluated the effectiveness of antimicrobial lock solutions in reducing CLABSI. The study’s results revealed a 32% reduction in catheter-associated ESI risk, suggesting a substantial protective effect from ALT. While this study yielded a marginally larger effect size than ours, consistent with our results, no statistically significant intergroup difference was detected. In contrast to our study, the cited research encompassed all catheter types [59]. Aslam et al. conducted a meta-analysis to assess the efficacy of ALT in the management of CRI among HD patients. Analysis demonstrated that ALT was more effective than systemic antibiotics alone in CRI treatment. This research shows that ALT may offer a valuable supplementary approach to managing CRI in HD [66]. A study encompassing diverse patient populations and catheter types found that combining ALT with intravenous antibiotics yielded significantly superior outcomes for treating CRIs compared with intravenous antibiotics alone. This statistically significant difference suggests a potentially more effective strategy for managing CRIs across diverse patient populations by combining these two approaches [67]. Our investigation revealed a moderate effect of ALT intervention on CRI-related mortality reduction. Our findings are consistent with those of a recent systematic review and meta-analysis, showing only a negligible difference in effect sizes [63]. A 2014 study yielded findings consistent with ours, showing an association between ALTs and reduced mortality. The results, however, did not reach statistical significance, a limitation possibly because of the small number of studies and participants [59]. Regarding the non-significant effect of ALTs on mortality reduction, a network meta-analysis found no statistically significant reduction in all-cause mortality [58]. Our study shows a correlation between ALTs and a lower incidence of thrombosis. The observed decrease was not statistically significant. Further research by Jaffer et al. demonstrated that ALT therapy had an insignificant impact on thrombosis incidence [68]. In addition, an analysis of pooled data from fourteen studies evaluating ALT’s efficacy against thrombosis revealed no statistically significant difference in thrombosis risk per 1000 catheter days [69]. The 2016 study, “Anticoagulants and antiplatelet agents for preventing central venous hemodialysis catheter malfunction in patients with end-stage kidney disease,” found no statistically significant evidence of effectiveness of antibiotic lock solutions in preventing catheter malfunction, including thrombosis, findings that align with our own research. This study’s conclusions diverged from ours; it reported an association between ALT utilization and increased catheter malfunction, whereas our study showed that ALTs had a protective effect against thrombosis [70]. The absence of independent studies reporting the IRR for CRBSI currently limits direct comparison with previous work. The lack of supplementary data in many studies that report only incidence rates prevents the calculation of IRRs for thorough analysis, thereby limiting the inclusion of all relevant research. Alternatively, the absence of comparable data shows a need for further clinical trials employing rigorous methodologies and comprehensive incidence rate reporting to validate and generalize our findings. Pending further substantiation, interpreting our present findings causes caution and acknowledgment of inherent limitations. To enhance the clinical utility of antibiotic lock solutions in HD, future studies should conduct large-scale RCTs and network meta-analyses to optimize protocol parameters, such as formulation, dwell time, and administration frequency. Limitations Inconsistencies in incidence-rate analyses and evidence of publication bias temper our conclusions and may cause an overestimation of ALT’s efficacy. The preponderance of small, single-center trials and inconsistencies in lock-solution composition, dwell times, and control protocols complicate direct comparisons. Furthermore, the limited pediatric and immunocompromised patient data restricts the generalizability of these findings. In contrast, no comprehensive long-term safety data prevent definitive conclusions on antimicrobial resistance and catheter lifespan. Conclusion This meta-analysis shows that integrating ALT into HD catheter management demonstrably reduces CRBSI, providing a focused prophylactic strategy. The consistent protective effects observed across various antibiotic formulations and clinical settings highlight the practical utility of ALT. The lack of substantial reductions in ESIs, thrombosis, and mortality suggests that ALT should be integrated with, rather than substitute for, comprehensive infection control bundles. Our results highlight the importance of consistent implementation and practical evaluation of these recommendations across diverse HD settings. Supplementary Material Supplemental Material