Journal of Pharmacy And Bioallied Sciences

: 2021  |  Volume : 13  |  Issue : 3  |  Page : 298--304

Physical compatibility, antimicrobial activity, and stability of cefazolin combined with gentamicin or ethanol in sodium citrate as a catheter lock solution

Rinda Devi Bachu1, Akshith Dass2, Emily To3, Sai H S Boddu4, Rose Jung5, Mariann D Churchwell1,  
1 Department of Pharmacy Practice, University of Toledo College of Pharmacy and Pharmaceutical Sciences, Toledo, OH, USA
2 Department of Pharmacy, University Hospitals Richmond Medical Center, Richmond Heights, OH, USA
3 Department of Pharmacy, AdventHealth Orlando, Orlando, FL, USA
4 Department of Pharmaceutical Sciences, College of Pharmacy and Health Sciences, Ajman University, Ajman, UAE
5 Department of Pharmacology, College of Medicine - Northern Kentucky Campus, University of Kentucky, Highland Heights, KY, USA

Correspondence Address:
Dr. Mariann D Churchwell
Department of Pharmacy Practice, University of Toledo College of Pharmacy and Pharmaceutical Sciences, Toledo, OH


Background: Catheters provide vascular access for patients requiring intravenous treatments, but frequently are a source of infection and/or thrombosis. Instilling a solution of an antimicrobial agent with an anticoagulant into the catheter lumen may salvage-infected catheters. Objective: The aim is to evaluate the physical compatibility, antibacterial activity, and stability of varying combinations of cefazolin (10 mg/mL), 40% ethanol, 4% sodium citrate with or without gentamicin (1 mg/mL) as a catheter lock solution over 48 h. Methods: Admixtures were prepared using aseptic technique and stored under four conditions with or without light at 25°C or 37°C. Prepared admixtures were assessed for physical compatibility, antimicrobial susceptibility, and chemical stability in triplicate at 0, 24 and 48 h. Admixture physical compatibility was determined by visual clarity, pH, and ultraviolet (UV) spectroscopy. Antibacterial activity was determined using the Kirby-Bauer disk diffusion method. The chemical stability of cefazolin and gentamicin were assessed using high performance liquid chromatography and UV spectroscopy, respectively. Results: All admixtures maintained clarity for 48 h. All admixtures stored at 25°C and the admixture containing 10 mg/mL cefazolin-4% sodium citrate stored at 37°C sustained antimicrobial activity and were chemically stable. A significant change in pH, antimicrobial activity, cefazolin concentration (<95% of baseline), were observed in admixtures containing ethanol stored at 37°C after 24 h. Gentamicin concentrations remained stable throughout the study. Conclusion: The admixture of 10 mg/mL cefazolin-4% sodium citrate sustained antimicrobial activity over 48 h and was chemically stable. However, admixtures containing ethanol stored at 37°C showed incompatibility with decreased antibacterial activity and cefazolin degradation after 24 h.

How to cite this article:
Bachu RD, Dass A, To E, Boddu SH, Jung R, Churchwell MD. Physical compatibility, antimicrobial activity, and stability of cefazolin combined with gentamicin or ethanol in sodium citrate as a catheter lock solution.J Pharm Bioall Sci 2021;13:298-304

How to cite this URL:
Bachu RD, Dass A, To E, Boddu SH, Jung R, Churchwell MD. Physical compatibility, antimicrobial activity, and stability of cefazolin combined with gentamicin or ethanol in sodium citrate as a catheter lock solution. J Pharm Bioall Sci [serial online] 2021 [cited 2021 Nov 27 ];13:298-304
Available from:

Full Text


The utilization of intravenous catheters for vascular access has increased over the past 30 years in both inpatient and outpatient settings for short- and long-term usage.[1] The 2018 United States Renal Data System Annual Report stated that 80.2% of patients initiating hemodialysis (HD) utilized a catheter for vascular access and 69% of patients continued to receive HD with a catheter after 90 days.[2] Unfortunately, catheters are highly susceptible to thrombosis and infection.[3] It is estimated that in-dwelling catheters cause approximately 250,000 catheter-related bloodstream infections (CRBSI) annually in the U. S.[1] In patients receiving HD, vascular access-related bloodstream infections were recently reported at a rate of 0.64/100 patient months.[4] When the average rate of CRBSIs was compared to patients with a catheter the rate was more than three times higher for a patient with a catheter (2.16 per 100 patient months).[4]

For most patients with CRBSIs, the removal of a catheter is indicated. However, in those patients with limited vascular access, catheter removal may result in the loss of necessary therapies including HD, nutrition, and antimicrobials. Treatments to salvage a catheter may be attempted using a catheter lock solution in addition to systemic antibiotics.[3],[5] The Infectious Disease Society of America (IDSA) recommends that catheter lock solutions contain a supratherapeutic antibiotic concentration as a possible method to eradicate bacteria related to CRBSIs.[3],[5]

Antibiotic lock solutions commonly utilize an anticoagulant with an antibiotic(s) depending on the catheter type and pathogen(s). Heparin is a commonly used anticoagulant in catheter lock solutions, however, heparin precipitates when mixed with ethanol.[6] Furthermore, patients with a heparin allergy or a history of heparin-induced thrombocytopenia will require an alternative anticoagulant. Sodium citrate may be an alternative anticoagulant to heparin for patients that are not able to receive heparin or if ethanol is used as an antiseptic or disinfectant in a lock solution. Previous studies have demonstrated in vitro compatibility of antimicrobial agents and sodium citrate with or without ethanol.[7],[8] Cefazolin, a first-generation cephalosporin is frequently administered systemically to patients with methicillin sensitive gram-positive infections. The national health-care safety network reports the most common bloodstream infection in patients receiving HD was Staphylococcus aureus and that approximately 60% of these infections were methicillin-sensitive organisms.[4] In this study, the physical compatibility, antibacterial activity were evaluated for cefazolin 10 mg/mL combined with sodium citrate 4%, with or without ethanol 40% and gentamicin 1 mg/mL, as catheter lock solutions over 48 h. As a supportive study to the antibacterial susceptibility testing, chemical stability of the lock solutions was investigated. The admixture concentrations for this in vitro study were selected to achieve supratherapeutic concentrations within a catheter lumen to increase the likelihood of bacterial eradication.[3]

 Materials and Methods

Preparation and storage of admixtures

Different combinations of admixtures were prepared using cefazolin (Lot #1760351, West Ward Pharmaceuticals, NJ), gentamicin sulphate (Lot # 74177DK, Hospira, IL), 4% sodium citrate (Lot # C727016 Baxter Healthcare Corporation, IL), ethanol (Lot # 223512, Decon Laboratories, PA) and sterile water as listed in [Table 1]. Briefly, cefazolin and sodium citrate (Admixture I), cefazolin and ethanol (Admixture II), cefazolin, sodium citrate and ethanol (Admixture III) and cefazolin, sodium citrate, ethanol, and gentamicin (Admixture IV). Four milliliters of each admixture were prepared under aseptic conditions in a laminar flow hood and were stored in sterile capped glass blood collection tubes (BD vacutainers). Vacutainers of each admixture were stored at room temperature 25°C and 37°C (in a water bath) with or without light. The dark conditions were maintained by completely covering the vacutainers with foil wrap. Admixtures were assessed in triplicate at 0, 24, and 48 h for physical compatibility, chemical stability, and antimicrobial susceptibility. After analyzing the physical and chemical stability, all the samples were passed through Millex® sterile nylon syringe filter (0.22 μm) and stored at-80°C for antibacterial susceptibility testing.{Table 1}

Evaluation of physical compatibility

Physical compatibility was determined by evaluating admixtures for clarity and pH. Clarity of admixtures was visually inspected against a white and black background by the same investigator. Each admixture was scored using a 0–4 scale: No precipitate and no color change (zero); change in color without precipitate (1); light haze (2); medium haze (3); and heavy visible precipitation (4). Clarity was further analyzed using an ultraviolet (UV)-visible spectrophotometer (Agilent 8453 UV-visible spectroscopy system, Waldbronn, Germany) at 566 nm, the wavelength that showed 100% transmittance of light for the blank solution. Samples with readings of >0.01 absorbance units were considered to be turbid. This value was selected based on our past experience and the instrument sensitivity.[8],[9] The pH of admixtures was measured using a pH meter (Accumet AB15, Fisher Scientific, Pittsburgh, PA). Before every use, the pH meter was calibrated using standard buffer solutions of pH 4.0, 7.0 and 10.0. A change in pH >0.1 was considered a significant change from baseline.[7],[8]

Antimicrobial susceptibility testing

Antimicrobial activity was evaluated using Kirby-Bauer disk diffusion method according to the guidelines established by the Clinical Laboratory Standards Institute.[10] Overnight growth of Escherichia coli ATCC 25922 (Sigma Aldrich, St. Louis, MO) was adjusted to the 0.5 McFarland turbidity standard (1.5 × 108 CFU/mL) and the subsequent inoculum was used to seed the Muller-Hinton agar plates (Difco™). Antimicrobial susceptibility test disks containing cefazolin 30 μg ([Lot # 1814288], BBL™ Sensi-Discs, Becton Dickson Biosciences, USA) were used as controls. Aliquots of admixtures equivalent to cefazolin (30 μg) were impregnated onto a blank BD disks at 0, 24, and 48 h. The control and sample disks were applied to the surface of agar plates and plates were incubated at 37°C for 24 h. Following incubation, the diameters of inhibition zones were measured to determine antimicrobial activity.

Analysis of variance was used to compare physical, chemical, and antimicrobial stability between baseline, 24-, and 48-h samples. P < 0.05 was considered statistically significant. All statistics were performed with SAS (version 9.1; SAS Institute, Cary, NC, USA).

Evaluation of chemical stability

Chemical stability of cefazolin was evaluated using high-performance liquid chromatography (HPLC, Waters Alliance e2695 separation module, Milford, MA) equipped with a UV detector and a Hypersil C18 column. The drug content in the samples was determined as per our previously published method with minor changes.[11] A mobile phase composed of phosphate citrate buffer pH 3.6 and acetonitrile (87:13) was pumped at a flow rate of 1.0 mL/min. The reagents used for preparing mobile phase included acetonitrile (Lot # 57773) and sodium dibasic phosphate (Lot # 081498) were procured from Fischer Scientific, NJ. Citric acid (Lot # C106724) was obtained from PCCA, TX. The reference standard of cefazolin was procured from Fischer Scientific, PA. The absorbance of cefazolin was measured at 270 nm. A stock solution of 1000 μg/ml of cefazolin in water was prepared and calibration standards ranging from 1.5625 to 50 μg/ml were prepared. Each calibration standard was analyzed in triplicate and the average peak area was plotted against the amount of cefazolin required to obtain the calibration curve.

Gentamicin was analyzed using a UV-Visible Spectrophotometer. The reference standard of gentamicin was procured from Fischer Scientific, PA. Calibration standards (5 ml) ranging from 25 to 200 μg/ml were taken in 10 ml glass tubes to which 1 ml of 2% sodium bicarbonate solution (Lot # 116734, Fisher Scientific, NJ) was added and mixed. A 2 ml of freshly prepared 95% ethanolic solution of 2, 4-dinitroflurobenzene (Lot # A0351447, ACROS, NJ) was added (0.25 ml/100 ml) and mixed thoroughly. After 20 min, the solution was neutralized by adding 0.5 ml of 1M HCl (Lot # 091256, Fisher Scientific, NJ). The glass tubes were tapped gently to remove any carbon dioxide bubbles and the absorbance was determined at 415 nm using distilled water as a blank.[12] Each standard was analyzed in triplicate and the absorbance was plotted against the concentration of gentamicin to obtain the calibration curve. The chemical stability of the drugs was assessed by measuring the drug content at 0, 24, and 48 h. Drug content ranging from 95% to 105% of the initial concentration was considered stable.


Physical compatibility of admixtures

Admixture I was physically compatible based on visual clarity and spectrophotometric measurements. The admixture also showed stable pH measurements for the entire study period, of 48 h. irrespective of temperature and light exposure [Table 2]. Admixtures II and III were clear without any physical incompatibility at 25°C for 48 h. However, a significant change was observed in pH (>0.1) at 24 and 48 h with Admixtures II and III stored at 37°C [Table 2]. Admixture IV containing gentamicin also remained clear without any physical incompatibility over 48 h when stored in light or dark conditions at 25°C and 37°C. This admixture did not show any haze or particulates and the pH measurements remained stable when stored at 25°C. However, when Admixture IV was stored at 37°C under light and dark conditions a significant change in pH at 24 and 48 h was observed when compared to baseline.{Table 2}

Antimicrobial susceptibility testing

The zone of inhibition values for the control and all admixtures at 0 h were within acceptable limits for E. coli ATCC 25922 and were maintained thoughout the 48 h period. The antimicrobial activity results are presented in [Table 3]. Admixture I stored at 25°C and 37°C with or without light maintained similar zone sizes at 24 and 48 h compared to baseline. Admixtures containing ethanol (II, III and IV) at 25°C also showed similar zone sizes for 48 h. However, admixtures stored at 37°C with or without light showed significantly smaller zones of inhibition compared to baseline (P < 0.05).{Table 3}

Chemical stability of admixtures

The HPLC analysis of samples was carried out to verify the results from physical compatibility and antimicrobial susceptibility testing. The chromatogram of cefazolin presented a sharp peak with a retention time of 4.6 min. A linear calibration curve (y = 1574.2x + 547.76) was obtained, with a regression coefficient (R2) of 0.9999. The developed HPLC method was also validated according to ICH Q2(R1) guidelines. Limit of detection (LOD) and limit of quantification (LOQ) were found to be 1.36 and 4.12 ng, respectively. The percentage recovery of cefazolin ranged from 100.51% to 104.96%. The method also showed good inter-day precision and suitability with a relative standard deviation (RSD) of <2%. Similarly, for gentamicin, a linear calibration curve (y = 0.0044x + 0.0112) was obtained with a regression coefficient (R2) of 0.9991. The LOD and LOQ were found to be 8.64 and 26.21 ng respectively. The percentage recovery of gentamicin ranged from 95.27% to 103.36%. The method also showed good inter-day precision and suitability with a RSD of <2%.

Admixture I was chemically stable and maintained at least 95% of the original cefazolin concentration up to 48 h under the four conditions tested. Admixtures containing ethanol (II, III and IV) maintained at least 95% of the original cefazolin concentration up to 48 h when stored at 25°C under light or dark conditions. However, when these admixtures were stored at 37°C the cefazolin concentration reduced to <95% by 24 h and continue to reduce further at the 48 h assessment. Sample HPLC chromatograms of Admixture IV at 0 and 48 h are presented in [Figure 1]. A minor shift in the cefazolin peak was observed in the samples stored at 25°C and 37°C with and without light. The gentamicin in Admixture IV showed minimal drug degradation under all four conditions over the entire study period [Table 4].{Figure 1}{Table 4}


Studies examining physical compatibility, efficacy, or antimicrobial activity and stability of admixtures containing ethanol and sodium citrate are limited. Our decision to study the admixture combinations, concentrations, and dwell times were based on our previous studies, a review of published literature and the IDSA clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update.[3],[7],[8] We chose to study cefazolin with sodium citrate as previous studies have examined cefazolin with heparin.[13],[14],[15] Moreover, citrate-based solutions have demonstrated improved clinical outcomes compared to heparin-based solutions.[16] The higher concentration of cefazolin with sodium citrate 4% was selected to attain supratherapeutic antimicrobial concentrations within the catheter. Supratherapeutic drug concentrations within a catheter have been recommended by the IDSA to attain higher bacterial kill rates and decreased bacterial resistance.[3] Instillation of higher drug concentrations within the catheter in addition to systemic antimicrobial therapy may provide a therapeutic option to combat increasing bacterial resistance.[3] Dwell time of antibiotic lock solutions vary, but we decided to study the admixtures over 48 h based on IDSA recommendations for lock solution dwell times not to exceed 48 h; although, it is preferred to change catheter lock solutions every 24 h.[3]

Our previous studies incorporated ethanol into admixtures with antimicrobial agents and sodium citrate 4% to determine if this additive altered compatibility, antimicrobial susceptibility or stability.[7],[8] Instillation of alternative solutions as a catheter lock solution may offer an option for clinicians combatting limited vascular access and increasing antimicrobial resistance. In addition to systemic antimicrobials, catheter lock solutions may extend catheter life for patients reliant on a catheter for their treatments. Based on our previous studies, examining ethanol and commercially available sodium citrate 4% compatibility, we hypothesized the maximum concentration for ethanol was 40% when used with an antimicrobial agent with or without sodium citrate 4%.[7],[8]

The results of our investigation indicate that cefazolin when combined with sodium citrate 4% (Admixture I) is compatible, stable and maintains antimicrobial activity up to 48 h when stored at any of the four conditions examined in this study. Similar results have been reported by Bornstein et al. The study investigated cefazolin stability (5 mg/ml) in commonly used intravenous fluids and found that cefazolin was stable for up to 1 week at 5°C and 25°C.[17] Cefazolin admixtures containing ethanol (II, III and IV) were unstable when stored at 37°C with or without light. Furthermore, a significant change in pH was observed at 24 h with these admixtures stored at 37°C. The possible reason behind cefazolin degradation in admixtures could be due to the beta-lactam ring hydrolysis in aqueous solutions.[18] The literature also indicates that the degradation is more pronounced in the presence of metal ions, nucleophiles, oxidizing agents, and solvents such as water and alcohol.[19] This could be attributed to the susceptibility of the β-lactam ring. The stability of gentamicin was retained when combined with ethanol and sodium citrate over 48 h as we have previously reported.[8] A similar compatibility behavior was reported by Battistella et al. when gentamicin (2.5 mg/ml) was combined with sodium citrate solution (4%) and stored at 37°C in dialysis catheters over 96 h.[20]

The quality control strain of E. coli ATCC 25922 was selected to evaluate activity of cefazolin and gentamicin because this strain is susceptible to both antimicrobial agents used in the study. The zones of inhibition of the control and admixtures were within the acceptable limits for this quality control strain indicating accuracy of the disk diffusion method. Retention of antimicrobial activity was demonstrated with Admixture I as the zone of inhibition remained stable at 24 and 48 h independent of storage condition. Antimicrobial activity was also retained for the admixtures containing ethanol (II, III, and IV) when stored at 25°C for up to 48 h. When the zone of inhibition for these admixtures stored at 37°C were compared to baseline, a significant reduction in zones of inhibitions was observed at 24 and 48 h. The chemical stability data obtained using HPLC analysis, indicated the incompatibility of cefazolin in admixtures containing ethanol when stored at 37°C thus corroborated our antibacterial susceptibility study results.

Our findings therefore indicate that, cefazolin and sodium citrate 4% are compatible when mixed together in lock solutions, but caution should be exercised when combining cefazolin and ethanol 40%. Although we controlled for several variables in this study including temperature and light, this study has limitations that have to be revealed. In this study, we have not assessed the stability of admixtures at lower concentrations of ethanol (<40% v/v) and at smaller time intervals (< 24 hrs). Further investigation is needed to better resolve cefazolin degradation peak at 4.2 min from the parent drug peak. While physical compatibility and antimicrobial susceptibility testing clearly indicate the incompatibility of cefazolin and ethanol (40% v/v) with and without sodium citrate 4% and/or gentamicin at 37°C, chemical compatibility results are preliminary in nature and should only be considered as supportive data. Patient variables that can affect catheter patency including fluctuations in body temperature and daily exercise or movements were not studied. Lastly, this was an in vitro study and institutions should perform their own independent testing of an admixture prior to patient administration according to USP 795 and 797 standards.


All antibiotic-anticoagulant admixtures considered were physically compatible, maintained antibacterial efficacy and were chemically stable until 48 h when stored at 25°C. Cefazolin combined with sodium citrate 4% remained stable throughout the study period independent of storage conditions. Admixtures stored at 37°C containing cefazolin and ethanol (40% v/v) with and without sodium citrate 4% and/or gentamicin showed incompatibility, decreased antibacterial efficacy and degradation of cefazolin at 24 h. Our findings demonstrate that the combination of cefazolin 10 mg/mL with ethanol 40% (v/v) should be avoided as a catheter solution when utilized at 37°C. Further research should be considered to determine if admixtures containing cefazolin with concentrations of ethanol <40% would improve chemical stability and antimicrobial efficacy.


The authors are grateful for the support of this research provided by the Department of Pharmacy Practice, University of Toledo College of Pharmacy and Pharmaceutical Sciences.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Maki DG, Kluger DM, Crnich CJ. The risk of bloodstream infection in adults with different intravascular devices: A systematic review of 200 published prospective studies. Mayo Clin Proc 2006;81:1159-71.
2United States Renal Data System. 2018 USRDS Annual Data Report. Epidemiology of Kidney Disease in the United States. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases. Bethesda, MD; 2018. Available from: [Last accessed on 2021 Feb 23].
3Mermel LA, Allon M, Bouza E, Craven DE, Flynn P, O'Grady NP, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis 2009;49:1-45.
4Nguyen DB, Shugart A, Lines C, Shah AB, Edwards J, Pollock D, et al. National healthcare safety network (NHSN) dialysis event surveillance report for 2014. Clin J Am Soc Nephrol 2017;12:1139-46.
5O'Grady NP, Alexander M, Burns LA, Dellinger EP, Garland J, Heard SO, et al. Guidelines for the prevention of intravascular catheter-related infections. Am J Infect Control 2011;39:S1-34.
6Schilcher G, Schlagenhauf A, Schneditz D, Scharnagl H, Ribitsch W, Krause R, et al. Ethanol causes protein precipitation-New safety issues for catheter locking techniques. PLoS One 2013;8:e84869.
7Martha C, Boddu SH, Churchwell MD. Physical compatibility of sodium citrate with alcohol and cefepime. Am J Health Syst Pharm 2013;70:932-7.
8Wei Y, Yang JW, Boddu SH, Jung R, Churchwell MD. Compatibility, stability, and efficacy of vancomycin combined with gentamicin or ethanol in sodium citrate as a catheter lock solution. Hosp Pharm 2017;52:685-90.
9Howard MS, Boddu SH, Mauro VF, Churchwell MD. Compatibility of conivaptan injection with select cardiovascular medications. Am J Health Syst Pharm 2014;71:1534-5.
10Clinical and Laboratory Standards Institute. Performance standards for antimicrobial disk susceptibility tests. 12th ed. CLSI document M02-A12. Wayne, PA: Clinical and Laboratory Standards Institute; 2015.
11Rao PS, Goodwani S, Bell RL, Wei Y. Boddu SH, Sari Y. Effects of ampicillin, cefazolin and cefoperazone treatments on GLT1 expressions in the mesocorticolimbic system and ethanol intake in alcohol-preferring rats. Neuroscience 2015;295:164-74.
12Ryan JA. Colorimetric determination of gentamicin, kanamycin, tobramycin, and amikacin aminoglycosides with 2,4-dinitrofluorobenzene. J Pharm Sci 1984;73:1301-2.
13Krishnasami Z, Carlton D, Bimbo L, Taylor ME, Balkovetz DF, Barker J, et al. Management of hemodialysis catheter-related bacteremia with an adjunctive antibiotic lock solution. Kidney Int 2002;61:1136-42.
14Vercaigne LM, Sitar DS, Penner SB, Bernstein K, Wang GQ, Burczynski FJ. Antibiotic-heparin lock: In vitro antibiotic stability combined with heparin in a central venous catheter. Pharmacotherapy 2000;20:394-9.
15Silva TN, Mendes ML, Abrão JM, Caramori JT, Ponce D. Successful prevention of tunneled central catheter infection by antibiotic lock therapy using cefazolin and gentamicin. Int Urol Nephrol 2013;45:1405-13.
16Zhao Y, Li Z, Zhang L, Yang J, Yang Y, Tang Y, et al. Citrate versus heparin lock for hemodialysis catheters: A systematic review and meta-analysis of randomized controlled trials. Am J Kidney Dis 2014;63:479-90.
17Bornstein M, Thomas PN, Coleman DL, Boylan JC. Stability of parenteral solutions of cefazolin sodium Am J Hosp Pharm 1974;31:296-302.
18Rattie ES, Guttman DE, Ravin LJ. Kinetics of degradation of cefazolin and cephalexin in aqueous solution. Arzneimittelforschung 1978;28:944-8.
19Deshpande AD, Baheti KG, Chatterjee NR. Degradation of β-lactam antibiotics. Curr Sci 2004;87:1684-95.
20Battistella M, Vercaigne LM, Cote D, Lok CE. Antibiotic lock: In vitro stability of gentamicin and sodium citrate stored in dialysis catheters at 37°C. Hemodial Int 2010;14:322-6.