Timing of Antimicrobial Lock Replacement for Gram-Positive Port Infections: Results of a Randomized Trial

1. Background

Totally implantable venous access ports are widely used in cancer patients, but their use is frequently complicated by thrombosis and infection [1]. The risk of catheter-related bloodstream infection (BSI) is estimated at 0.1 per 1,000 port-days [2], and repeated device manipulation further increases this risk [3]. Overall, up to 10% of port carriers will experience device-related infectious complications [4], and more than half of these cases ultimately require port removal [5]. Staphylococci account for nearly 70% of all port-related BSIs [6].

Management of uncomplicated catheter-related bacteremia (CRB) typically combines systemic therapy with local antimicrobial administration through antimicrobial lock therapy (ALT). Current recommendations suggest 10–14 days of treatment [7]. ALT involves instilling a highly concentrated antimicrobial solution into the catheter lumen and maintaining it for a defined dwell time to eradicate intraluminal microorganisms [8]. According to the Infectious Diseases Society of America (IDSA), antimicrobial lock solutions (ALS) should generally be replaced every 24 hours, except in the case of hemodialysis catheters [7]. However, evidence is lacking to determine the optimal replacement interval for ALS.

Extending ALS dwell time could reduce port manipulation to only once or twice during the course of therapy, thereby lowering costs and minimizing procedure-related morbidity without compromising efficacy. Encouraged by previous reports and by our own clinical experience [9,10,11], we designed a randomized clinical trial to evaluate the optimal replacement intervals for five commonly used ALS—vancomycin, teicoplanin, linezolid, daptomycin, and tigecycline—instilled into noninfected ports and maintained in situ for variable dwell times.

2. Patients and Methods

Setting and study population. This clinical trial was conducted at the University Hospital, Clínica Universidad de Navarra (Pamplona, Spain) between May 2012 and January 2015. All patientteicoplanins with recently implanted venous access ports were eligible. Exclusion criteria were: clinical or microbiological evidence of infection (with or without systemic antimicrobial therapy); known allergy to heparin or any of the study antimicrobials; concomitant systemic anticoagulation; pregnancy; and inability to provide written informed consent. The study was approved by the local Institutional Review Board (79/2010) and the Spanish Agency of Medicines and Medical Devices (EudraCT 2010-023814-29). All participants gave written informed consent. The trial was registered at ClinicalTrials.gov (NCT01592032).

Trial design. Patients were randomly assigned, in a single-blind design, to receive one of five antimicrobial lock solutions (ALS). Each ALS contained 100 IU/mL of 1% sodium heparin plus one antimicrobial: vancomycin 2 mg/mL, teicoplanin 10 mg/mL, linezolid 1.8 mg/mL, daptomycin 5 mg/mL, or tigecycline 4.5 mg/mL, for a final volume of 10 mL. Antimicrobial concentrations were chosen based on the highest levels reported in the literature [12,13,14] and our clinical experience [9,10,15]. All antimicrobials were reconstituted in normal saline, except linezolid (ready-to-use) and daptomycin (lactated Ringer’s solution). Randomization was computer-generated, and the research team was blinded to antimicrobial assignment. Each patient could participate only once.

ALS dwell times were escalated sequentially (1, 3, 5, 7, and 10 days). Each dwell-time period required inclusion of five patients per antimicrobial group. Prior to ALS instillation, a 10 mL blood sample was drawn to confirm port sterility. After flushing with 10 mL of 0.9% sodium chloride, 5 mL of ALS was instilled into each port, which then remained untouched until the end of the dwell time. After completion, the first 2 mL withdrawn from the port was analyzed for antimicrobial concentration using high-performance liquid chromatography (HPLC; Agilent Technologies, Santa Clara, CA, USA). Urea concentrations were determined (Synchron® Clinical Systems, Beckman Coulter, Brea, CA, USA), and the systemic-to-sample urea ratio was used to correct antimicrobial concentrations for individual blood dilution [16].

If the median antimicrobial concentration remained above 1 mg/mL for a given ALS group, a new set of five patients was randomized to the next dwell-time interval. A median concentration <1 mg/mL terminated randomization for that ALS group. The primary endpoint was the time until intraluminal antimicrobial concentration fell below 1 mg/mL.

Statistical analysis. Antimicrobial concentrations were compared between groups using the Kruskal–Wallis test, and paired comparisons of concentrations at baseline versus end of dwell time were analyzed with the Wilcoxon test. Statistical significance was set at an alpha level of 0.05 (two-tailed). Analyses were performed using SPSS software, version 15.0.1 (SPSS Inc., Chicago, IL, USA).

3. Results

Over the 32-month study period, 484 consecutive patients were screened for eligibility (Figure 1). A total of 93 patients were randomized into the five ALS groups: 11 to vancomycin, 24 to teicoplanin, 10 to linezolid, 26 to daptomycin, and 22 to tigecycline. Baseline characteristics, including sex, age, underlying malignancy, and mean time from catheter placement to study entry, were comparable across groups (Table 1). No port-related infections occurred in any of the patients during the study, and no serious drug-related adverse events were observed.

After 1 day of dwell time, all five ALS maintained antimicrobial concentrations above 1 mg/mL. At 3 days, the median concentrations of vancomycin and linezolid had fallen below 1 mg/mL, leading to termination of further randomization in these groups. The decrease in concentrations of both drugs compared with baseline was statistically significant (P = .043).

Patients continued to be enrolled in the teicoplanin, daptomycin, and tigecycline groups for dwell times of 5 and 7 days. Teicoplanin median concentrations remained above 1 mg/mL at both time points, with no significant difference compared with the instilled concentration (P = .893). In contrast, daptomycin and tigecycline concentrations decreased significantly after 7 days (P = .043), but still remained above 1 mg/mL (Table 2).

Due to limited enrollment, the 10-day dwell time group could not be completed, and the study concluded without data for this interval.

4. Discussion

Antimicrobial lock replacement should be tailored to the agent: vancomycin and linezolid require daily changes, while teicoplanin, daptomycin, and tigecycline remain effective for up to 7 days, reducing interventions without compromising efficacy.

The Infectious Diseases Society of America (IDSA) guidelines recommend that the dwell time of an antimicrobial lock solution (ALS) should not exceed 24 hours, except for hemodialysis catheters [7]. Nevertheless, the optimal replacement interval remains undefined, with reported recommendations ranging between 8 and 24 hours per day [17]. Our trial demonstrates that the decline in antimicrobial concentration within ALS instilled into ports is critical for determining replacement time. Consequently, the general recommendation to replace ALS every 24 hours should be tailored according to the antimicrobial agent employed. Specifically, ALS containing teicoplanin (10 mg/mL), daptomycin (5 mg/mL), or tigecycline (4.5 mg/mL) could be safely replaced every 7 days, whereas ALS with vancomycin or linezolid should be replaced daily.

The duration of ALS exposure to the catheter lumen is pivotal for biofilm eradication. Our study was designed to establish the longest effective dwell time without compromising intraluminal concentrations. Extrapolating findings from hemodialysis catheter–related bacteremia (CRB) to port-related infections is problematic, as hemodialysis catheters are accessed every 48–72 hours, whereas ports are generally used daily or intermittently for chemotherapy administration.

Several factors influence antimicrobial concentrations and activity [14], including catheter design [18], intrinsic drug properties [19], protein-binding capacity [20], free drug fraction [21], and degradation rate [22]. Fluid dynamics also contribute: the Hagen–Poiseuille velocity distribution for flow [23], passive diffusion into the bloodstream [24], density differences between the lock solution and blood [25], and exchange driven by postural changes [26].

Soriano et al. described a gradient of vancomycin concentrations along hemodialysis catheters [27], inferred from sequential aspiration of three blood samples. However, this indirect approach is limited by assumptions regarding catheter segment representation. By contrast, our study analyzed a single 2 mL aspirate from five independent patients per ALS and dwell time, thereby reducing reliance on sequential sampling. Still, we recognize that recommendations based on the median of only five samples per group are limited.

Deposits accumulating in the port reservoir may contribute to the gradual decline of intraluminal concentrations. To date, no in vivo studies have characterized antimicrobial loss in port-locked solutions. This knowledge gap has reinforced the practice of daily ALS replacement, rather than exploring the benefits of instilling maximal concentrations and allowing prolonged dwell times, provided safety and efficacy are maintained. Key unanswered questions remain: would increasing antimicrobial concentrations be safer than extending dwell time, or could higher concentrations enable longer dwell intervals?

Previous studies have reported success with dwell times beyond 24 hours. Sánchez-Muñoz et al. described an 85.7% clinical success rate in 14 patients treated with a 3-day lock combining heparin and vancomycin or amikacin [28], though without measuring residual concentrations. Conversely, Haimi-Cohen et al. found a residual vancomycin concentration of 0.13 mg/mL even 28 days after instillation [12]. While maintaining residual drug levels is relevant, avoiding prolonged exposure to subtherapeutic concentrations is paramount, given the risk of resistance development [29]. Indeed, low vancomycin concentrations can enhance staphylococcal biofilm density [30]. In our trial, we established a conservative threshold of 1 mg/mL, representing at least 1000-fold the MIC for planktonic staphylococci [31] and safely above the mutant prevention concentration [19].

Our findings confirmed a progressive decline in ALS concentrations over time. After 3 days, vancomycin concentrations decreased by 77%, falling below bactericidal thresholds [31]. Adequate antimicrobial levels are essential for targeting persister cells, which tolerate high concentrations despite intrinsic susceptibility, thereby contributing to therapeutic failure and biofilm persistence [32]. Similarly, linezolid concentrations decreased by 60% after 3 days. Clinical data on linezolid as lock therapy remain scarce [10,27], though Sofroniadou et al. reported that linezolid locks prevented CRB in hemodialysis patients [35]. While linezolid may be valuable in specific scenarios—such as prosthetic infections [36] or endocarditis [34,35]—evidence for its use in lock solutions remains limited.

Teicoplanin emerged as the most stable ALS in our study. Median concentrations remained unchanged after 7 days (P = 0.89), possibly due to its affinity for catheter materials [39]. Clinical success rates with daily teicoplanin locks for CRB range from 88% to 100% [11,37]. Our data suggest that this efficacy may derive from the remarkable stability of the 10 mg/mL solution over extended dwell times.

There is increasing evidence for daptomycin in CRB treatment, both in vivo [12,30] and in vitro [31,32]. Experimental catheter infection models with daptomycin 2–5 mg/mL have shown significant eradication of staphylococci [33,34]. Raad et al. and Meije et al. demonstrated superior outcomes with daptomycin locks compared to glycopeptides [45]. In our trial, daptomycin maintained concentrations >2000 times the staphylococcal MIC for up to 7 days, supporting extension of replacement intervals to once or twice during the 10–14 days recommended by IDSA [7].

Tigecycline has been rarely used in ALT. Aslam et al. reported an 83% success rate with a lock containing N-acetylcysteine, heparin, and tigecycline 1 mg/mL in hemodialysis CRB [11]. However, tigecycline monotherapy for BSI remains controversial [46]. Minocycline–EDTA locks have demonstrated efficacy in hemodialysis [47] and pediatric oncology patients [48], with dwell times up to 7 days. Our findings suggest that tigecycline may be considered in select scenarios when alternatives are lacking.

Overall, we observed that teicoplanin (10 mg/mL), tigecycline (4.5 mg/mL), and daptomycin (5 mg/mL) maintained intraluminal concentrations above 1 mg/mL for 7 days, whereas vancomycin and linezolid did not. We acknowledge the limitations of our study, including small sample sizes and heterogeneity in antimicrobial concentrations. Furthermore, maintaining high intraluminal concentrations for 7 days does not guarantee therapeutic success, given the variability in antimicrobial biofilm penetration [49].

In conclusion, our findings suggest that replacement intervals for ALS in ports could be extended from daily to once every 7 days when using teicoplanin, daptomycin, or tigecycline against coagulase-negative staphylococci (CoNS). By contrast, vancomycin and linezolid locks should continue to be replaced every 24 hours.

5. Strengths and Limitations

Strengths

• This is the first randomized clinical trial specifically designed to evaluate the optimal dwell time of antimicrobial lock solutions (ALS) in totally implantable venous access ports.
• The study directly quantified intraluminal antimicrobial concentrations in vivo, using standardized high-performance liquid chromatography with urea correction, thereby providing robust pharmacokinetic data under clinical conditions.
• Multiple antimicrobials commonly used in lock therapy (vancomycin, teicoplanin, linezolid, daptomycin, tigecycline) were compared simultaneously, allowing drug-specific recommendations for replacement intervals.
• The trial design with stepwise dwell-time escalation enabled the identification of thresholds for loss of efficacy without exposing patients to unnecessary risk.

Limitations

• The sample size within each antimicrobial–dwell time subgroup was limited (n=5 per time point), reducing statistical power and generalizability.
• Port use in the study was restricted to noninfected, newly implanted devices, which may not fully reflect clinical practice in infected or long-term ports.
• Concentration thresholds were based on pharmacokinetic/pharmacodynamic principles (≥1 mg/mL) rather than direct clinical outcomes, so translation to real-world effectiveness requires caution.
• The 10-day dwell time arm could not be completed due to recruitment constraints, leaving the upper limit of safe dwell times undetermined.
• Findings apply to the antimicrobial concentrations tested, which may not account for alternative formulations, catheter materials, or patient populations.

Clinical Implications

• Replacement intervals for antimicrobial lock solutions (ALS) should be individualized according to the specific agent used.
• Vancomycin and linezolid locks require daily replacement, while teicoplanin, daptomycin, and tigecycline maintain effective concentrations for up to 7 days.
• Extending ALS dwell times could reduce port manipulations, lower healthcare costs, and minimize patient discomfort without compromising efficacy.