Catheter-related bloodstream infections (CRBSIs) are an important cause of nosocomial infection, with morbidity, mortality, and cost. Reference Gahlot, Nigam, Kumar, Yadav and Anupurba1 Pathogens may enter the circulation via an extraluminal route (ie, resulting from migration of microorganism from the skin at the catheter insertion site to the vein) or an intraluminal route. Reference De Cicco, Chiaradia and Veronesi2–Reference Helder, van Rosmalen and van Dalen4 In CRBSI related to the intraluminal route, the catheter hub has been identified as a major entry point for microorganisms Reference Koeppen, Weinert, Oehlschlaeger, Koerner, Rosenberger and Haeberle5,Reference Salzman, Isenberg, Shapiro, Lipsitz and Rubin6 because they can adhere to, migrate to, and colonize the internal lumen of the catheter as well as form a biofilm that allows them to disseminate into the bloodstream. Reference Frasca, Dahyot-Fizelier and Mimoz3,Reference Bond, Chadwick, Smith, Nightingale and Lal7
Because these contaminations occur when handling intravascular line connectors during infusion connection, drug injections or blood sampling, needleless connectors (NC) have been introduced into clinical practice to reduce handling of catheter connections and thus reduce the time during which microorganisms can contaminate the ports. Reference Koeppen, Weinert, Oehlschlaeger, Koerner, Rosenberger and Haeberle5 Their use has also eliminated the risk of needle-stick injuries, which prevents blood-exposure accidents and limits the use of needles on elastomer injection sites when accessing intravascular catheters. Reference Btaiche, Kovacevich, Khalidi and Papke8 Their effectiveness in reducing infections has been much debated in the literature. Reference Koeppen, Weinert, Oehlschlaeger, Koerner, Rosenberger and Haeberle5,Reference Slater, Fullerton, Cooke, Snell and Rickard9
To further reduce the risk of CRBSI, catheters can be disinfected by either active or passive disinfection. Active disinfection consists of 15–30 seconds mechanical scrubbing of the hub membrane (or NC) using an alcohol wipe, followed by a drying period before using the catheter. Reference O’Grady, Alexander and Burns10 Interestingly, passive disinfection through antiseptic barrier caps (ABCs) have also been proposed to reduce CRBSI. Reference Menyhay and Maki11–Reference Helder and Vos14 These ABCs containing a disinfectant (usually isopropanol 70%) are placed directly on the NC to continuously impregnate and disinfect the catheter access up to 7 days. Reference Inchingolo, Pasciuto and Magnini15,Reference Kamboj, Blair and Bell16 Overall, available evidence suggests that ABCs are effective, safe, easy to use, and cost-effective in reducing CRBSIs compared with isopropanol wipes in adults. Reference Flynn, Larsen, Keogh, Ullman and Rickard17,Reference Gillis, van Es, Wouters and Wanten18
To the best of our knowledge, no cases of isopropanol intoxication have been reported with the use of an ABC on a NC. However, in 2 recent studies, isopropanol contained in the ABCs leaked into the NC, raising safety questions about potential isopropanol exposure to patients during routine care, particularly in a pediatric and neonatal intensive care settings. Reference Hjalmarsson, Hagberg, Schollin and Ohlin19,Reference Sauron, Jouvet and Pinard20 To date, no publication has compared a wide range of NCs with respect to the parameters influencing the leakage of isopropanol from ABC to the NC.
In this study, we characterized the leakage of isopropanol from the ABC to the NC on all the products available on the European market, and we sought to determine which ABC and NC parameters influence the leakage of isopropanol through the infusion circuit.
Material and methods
Materials
We purchased 4 commercially available ABCs from their respective suppliers: 3M Curos (3M, St Paul, MN), DualCap (Catheter Connections, Salt Lake City, UT), BD PureHub (Becton-Dickinson, Franklin Lakes, NJ), Swabcap (B Braun, Melsungen, Germany) (Fig. 1). NC were purchased at the respective manufacturing laboratories, and the associated pictures and product brands are detailed in Figure 2. V-green extension lines (no. 71100.01) were purchased from Vygon (Ecouen, France). Versylène NaCl 0.9% was purchased from Fresenius (Homburg, Germany). Due to the availability of the Swabcap and the MaxPlus NC in our institution, these 2 brands were mainly used in this study.
Figure 1. Different types of antiseptic barrier cap used in the experiments.
Figure 2. Product references of bidirectional needleless connector used in the experiments. Classification was adapted from bibliographic reference. Reference Lurton21 Note. NC, needleless connector.
Setting up in vitro circuits
Circuits consisting of an isopropanol ABC, a NC, an 11-cm polyethylene catheter line (no. 71100.01, batch 211222EH, Vygon) and an obturator (no. 9888.00, batch 241122FD, Vygon) were created (Fig. 3). In all circuits, the catheter line was rinsed and prefilled with NaCl 0.9% before the ABC was placed on the NC. At the end of the application time of the ABC, the circuits were rinsed with 5 mL NaCl 0.9%, which was collected for analysis.
Figure 3. Circuit to connect (1) the isopropanol antiseptic barrier cap, (2) needleless connector, (3) catheter line, and (4) obturator. The example is shown here with a MaxPlus needleless connector and a Swabcap.
In vitro experiments
The circuits were left in place for 1–7 days to assess the kinetics of isopropanol leakage from the ABC to the NC at room temperature (measured at 20°C). Experiments were carried out with or without a 30-second drying time before rinsing the circuit to evaluate the proportion of isopropanol passing through the Luer tip syringe. The drying time of 30 seconds after removal of the ABC from the NC was used as previously described, Reference Hjalmarsson, Hagberg, Schollin and Ohlin19 although the instructions for the use of these devices do not specify this period.
We used the following methods to evaluate the isopropanol content in the 4 commercially available ABCs. (1) The ABCs were weighed before and after evaporation of their contents in an oven at 37°C for 24 hours, and (2) the ABCs were left in a closed container containing 5 mL NaCl 0.9% for 24 hours to extract isopropanol.
Isopropanol quantitation by headspace gas chromatography with flame ionization detection
We used gas chromatography to simultaneously quantify methanol, acetone, ethanol, and isopropanol (propan-2-ol) using propan-1-ol as the internal standard. Gas chromatography analysis was conducted using a ThermoScientific GC Trace 1300 system (Thermo Fisher Scientific, Waltham, MA) with a flame ionization detector (FID) and a headspace Thermoscientific TriPlus sampler. Chromatographic separation was achieved on a 30-m × 0.25-mm × 0.25-µm DB-WAX fused silica column (Agilent Technologies, Santa Clara, CA) using nitrogen as a carrier gas. SSL Injection port temperature was set to 200°C; the injection volume was 1 mL with split flow; and the oven temperature program was held at 50°C for a 7-minute run time. The FID detector was set at 250°C. Data were recorded and analyzed using Chromeleon 7.2 software (Thermo Fisher Scientific) using peak area ratios of analyte to internal standard with comparison to a 6-point standard curve for quantitative analysis of each analyte.
In a sealed vial, 200 µL internal standard (propan-1-ol) was added to the 10-µL sample and incubated 20 minutes at 80°C before injection. Routine quality-control samples consisted of Medidrug ALC VB 030 and 110 congener alcohols were analyzed before each sample analysis. The lower limit of quantification of isopropanol is 10 mg/L, and this method is linear until 750 mg/L.
Statistical analysis
Data were expressed as the mean ± standard error of the mean (SEM). Intergroup differences as a function of the treatment were probed in a 1-way analysis of variance (ANOVA), with a Tukey post hoc test for group comparisons. All analyses were performed using Prism software (GraphPad Software version 8.0, La Jolla, CA). All tests were 2-sided, and the threshold for statistical significance was set to P < .05.
Results
Influence of antiseptic barrier cap parameters on isopropanol leakage through needleless connectors
To investigate the influence of ABC parameters on isopropanol leakage through NCs, we first analyzed the isopropanol contents in the 4 commercially available ABCs (Fig. 4A and B). In the first method, we placed the ABCs in a 37°C oven for 24 hours to evaporate their 70% isopropanol content. The isopropanol contents were significantly different among the ABCs (Fig. 4A). These results were confirmed by infusing the ABCs for 24 hours in 5 mL NaCl 0.9% and assaying for isopropanol in the resulting solution. The BD PureHub ABC contained the most isopropanol (250.8 ± 123.6 mg), followed by the Merit DualCap (181.5 ± 86.45 mg), SwabCap (114.8 ± 14.5 mg), and 3M Curos (72.1 ± 4.3 mg). However, these results did not reach statistical significance. (Fig. 4B).
Figure 4. Impact of antiseptic barrier cap parameters on isopropanol leakage through needleless connectors. (A) Difference in mass of antiseptic barrier caps before and after 24 hours in an oven at 37°C and (B) amount of isopropanol extracted during incubation of the antiseptic barrier caps for 24 hours in 5 mL NaCl 0.9%. (C) Influence of antiseptic barrier cap types on isopropanol leakage through needleless connector. The 4 antiseptic barrier caps were placed on a MaxPlus needleless connector for 1–7 days. (D) Influence of the drying time (30 seconds) of isopropanol before rinsing with 5 mL NaCl 0.9%. The SwabCap was placed on the MaxPlus needleless connector for 24 hours. These results are compared with a simple rubbing of an isopropanol wipe for 30 seconds before rinsing with 5 mL NaCl 0.9%. The data are quoted as the mean ± SEM from 3 measurements. **** P < .0001; ** P < .01. Note. NS, nonsignificant.
Next, we analyzed isopropanol leakage through a MaxPlus NC for these 4 ABCs left on for 1–7 days. Isopropanol leakage through the NC was not significantly different among the ABCs, either at 24 hours or 7 days (Fig. 4C). Lastly, we showed that a drying time of 30 seconds between ABC removal and rinsing with 5 mL NaCl 0.9% did not change the amount of isopropanol passing through the NC (Fig. 4D). However, rubbing with an isopropanol wipe caused very little isopropanol to pass through the NC. Taken together, these results suggest that isopropanol leakage from the ABC to the NC is not ABC dependent. Because the SwabCap is used in our institution and the drying time has no influence on the isopropanol leakage through the NC, we chose to use this product without drying time for further experimentation.
Influence of needleless connector parameters on isopropanol leakage
To investigate the influence of NC parameters on isopropanol leakage from the ABC to the NC, we left an ABC (SwabCap) on 13 different NCs for 24 hours or 7 days (Fig. 5). Our results showed that isopropanol leakage through the NC changes significantly depending on the NC (Fig. 5). Supplementary Table S1 (online) describes the results of overall difference testing for isopropanol leakage after 24 hours and 7 days. Complete statistical differences in isopropanol leakage between each NC after 24 hours and 7 days of experiments are described in Supplementary Tables S2 and S3 (online), respectively.
Figure 5. Impact of different needleless connector types on isopropanol leakage. (A) Isopropanol leakage from each antiseptic barrier cap–needleless connector pair was evaluated after 24 hours or (B) 7 days or (C) from 1 to 7 days. The data are quoted as the mean ± SEM from 3 measurements. Compared to Q-syte: & P < .05, && P < .01, &&& P < .001, &&&& P < .0001.
Moreover, 24 hours of ABC attachment to the NC was sufficient to observe differences in isopropanol leakage among the different NCs. The BD Q-Syte, Didactic, and BD MaxPlus NCs were the most permeable to isopropanol in a statistically significant way, passing 7.01 ± 1.03 mg, 6.27 ± 0.87 mg and 4.22 ± 0.65 mg, respectively. All other products did not appear to be significantly different in terms of isopropanol leakage despite the trends observed (Fig. 5A). The Bionector NC (Vygon) emerged as the least isopropanol-leaching NC, with an average of 0.66 ± 0.27 mg isopropanol leaked after 24 hours of use (Fig. 5A).
After 7 days of use, the Caresite NC (B Braun) emerged as the least isopropanol-leaching NC, with an average of 1.69 ± 0.01 mg isopropanol leaked. On the other hand, the Q-Syte NC allowed the most isopropanol to pass through, with an average of 28.32±2.62 mg of isopropanol leaked (Fig. 5B). The trends observed at 24 hours were also found at 7 days and were confirmed by the study of the kinetics of isopropanol leakage through the NC (Fig. 5C).
Discussion
Despite their vital importance, catheters may cause potentially life-threatening complications including CRBSI. To reduce this comorbidity, the most recent clinical evidence supports the use of NCs and ABCs for passive and continuous disinfection of NCs. Reference Gillis, van Es, Wouters and Wanten18 In this study, we showed that isopropanol can leak from the ABCs to the NC as a function of time, posing concerns about their use. We determined that the brand of NCs used was the most important parameter in this phenomenon.
First, we showed that the different types of ABCs did not influence the leakage of isopropanol through the NC, despite different isopropanol content in each ABC. The NC mechanism did not predict isopropanol leakage, as shown with the Q-syte and Safeflow NCs, as well as the Caresite and the MaxPlus NCs, which let through very variable amounts of isopropanol despite common mechanisms (Figs. 2 and 5). Overall, Caresite and Bionector NCs were safer choices when used with isopropanol ABCs, from an isopropanol leakage point of view. The brand and the quality of the NC seal on which the ABC is placed appeared to be the main factors influencing the leakage of isopropanol, as suggested by previous reports. Reference Hjalmarsson, Hagberg, Schollin and Ohlin19,Reference Sauron, Jouvet and Pinard20 Hjalmarsson et al Reference Hjalmarsson, Hagberg, Schollin and Ohlin19 reported isopropanol leakage ranging from 0.154 to 0.405 mg after 24 hours depending on the NC–ABC combination used (ie, Swan-lock or Bionector coupled to SwabCap or Curos). Reference Hjalmarsson, Hagberg, Schollin and Ohlin19 Particularly, the combination of Bionector and SwabCap showed a maximum isopropanol leakage of 0.372 mg after 24 hours, compared to 1.220 mg in our study. The combination Safeflow and Swabcap showed a maximum isopropanol leakage of 1.755 mg after 24 hours, compared to 2.066 mg in our study. Due to the different methodologies used in the literature (which aimed to mimic the real-world conditions Reference Hjalmarsson, Hagberg, Schollin and Ohlin19,Reference Sauron, Jouvet and Pinard20 ) and in our study (which aimed to determine how much isopropanol passes from the ABC to downstream of the NC), the comparison of the results is difficult. However, the results in the literature and in our study are of the same order of magnitude: Safeflow leaks more isopropanol than Bionector Reference Hjalmarsson, Hagberg, Schollin and Ohlin19 and Smartsite leaks more isopropanol than Caresite. Reference Sauron, Jouvet and Pinard20
Interestingly, the leakage hypotheses are also supported by the findings of Rickard et al, Reference Rickard, Flynn and Larsen22 who reported that 2 NCs (Smartsite or MaxPlus, unspecified) became opaque during their study of ABCs, suggesting that isopropanol appears to have seeped between the inner rubber and outer plastic, denaturing the plastic. Reference Rickard, Flynn and Larsen22 Similarly, Sauron et al Reference Sauron, Jouvet and Pinard20 found that the appearance of Smartsite and Caresite NCs was modified after ABC connexion, including loss of transparency and inflation of the NC’s fanfold piece. Reference Sauron, Jouvet and Pinard20 In our experiments, we did not observe such denaturation of the plastic.
Second, toxicokinetic data available in humans indicate that absorption of isopropanol is greater and more rapid through the lungs and gastrointestinal tract and less through the skin. However, no data are available on intravenous isopropanol administrations, raising questions about the safety of such exposure. Although isopropanol has a half-life of 3–7 hours, its metabolite acetone has a half-life of 22 hours, and it is primarily excreted by the kidneys. Reference Ashurst and Nappe23 Toxicological data in the literature are mainly related to intoxication by ingestion with ketosis without acidosis and pseudo-renal failure as hallmarks. Because isopropanol penetrates the central nervous system better than ethanol, isopropanol is more intoxicating than ethanol and can produce sensorium alteration, hypotension, hypothermia, and even cardiopulmonary collapse. Reference Ashurst and Nappe23
Previous reports mention that isopropanol toxic blood concentrations vary between 250 and 5,200 mg/L. Reference Hong, Morrow, Sandora and Priebe24–Reference Zaman, Pervez and Abreo28 Because the smallest weights of newborns can be ∼500 g (and the total blood volume of a child is ∼75–80 mL/kg), these issues cause greater concern in the pediatric and neonatal population in view of the possibility of substantial isopropanol exposure. Reference Howie29 Patients may have multiple infusions per day and multiple catheter lines and NC–ABC combinations, thus increasing the risk of isopropanol accumulation. In addition, a reported case of fatal isopropanol poisoning by inhalation in a 1,500-gram male infant suggests that the elimination half-life would be higher in infants than in adults (9.6 hours vs 3–7 hours, respectively). Reference Ashurst and Nappe23,Reference Vicas and Beck30 Lastly, cases of transcutaneous alcohol intoxication have been described with isopropanol in adults Reference Wolfshohl, Jenkins and Phillips31,Reference Chavez, Sweeney and Akpunonu32 and ethanol in children, Reference Püschel33–Reference Dalt36 raising the possibility of significant alcohol exposure outside the oral route. Exposure to isopropanol during care requiring the use of central nervous system depressants such as anesthetics could also result in drug interactions (ie, synergistic central nervous system depressant effect). The metabolism of isopropanol to acetone could also interfere with the interpretation of biological tests, Reference Wolfshohl, Jenkins and Phillips31 and a risk of venous toxicity cannot be excluded. For these reasons, we believe that this issue could be of major concern in neonatal and pediatric intensive care units. To the best of our knowledge, no cases of isopropanol intoxication have been reported following the use of ABCs on NC. However, we believe it is necessary to draw attention to the potential leakage of isopropanol through the NC. This factor should be considered when weighing the benefits of ABCs in reducing the risk of CRBSI against the risk of isopropanol entering the patient’s bloodstream.
Lastly, we have shown that the drying time between removal of the isopropanol ABC and injection of solute through the NC did not change the amount of isopropanol passed. These results suggest that the amount of isopropanol present on the NC septum is negligible and is not a major parameter in patient exposure to this alcohol. These results are interesting regarding the uses of isopropanol-soaked wipes, which constitute a simple solution for reducing patient exposure to intravenous isopropanol (Fig. 4). Reference Gillis, van Es, Wouters and Wanten18 However, the choice of the disinfection method used, whether active or passive, must also take into account the superiority of ABC in terms of reducing the risk of CRBSI compared to isopropanol-soaked wipes. Reference Gillis, van Es, Wouters and Wanten18
This study had several limitations. First, this was an in vitro study, and it would be interesting to confirm these results on patients exposed to ABCs. In this context, the contact of these devices with the patient’s skin may lead to an increase in the temperature of these devices (ABCs, NC, and catheter), causing an increase in isopropanol leakage, as suggested in a previous report. Reference Sauron, Jouvet and Pinard20 Second, we were interested in the NCs available on the European market, so further studies are needed to evaluate isopropanol leakage with other product brands used outside Europe.
In conclusion, the use of isopropanol ABCs on NCs can cause isopropanol leakage into the catheter circuit and bloodstream. This leakage is influenced by the NC parameters and not the ABCs. In view of the lack of toxicity data for isopropanol by intravenous administration, caution should be exercised when using these devices, especially in the pediatric and neonatal population. Further studies are needed to assess isopropanol and/or acetone exposure in patients using ABCs on NCs.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/ice.2023.285
Acknowledgments
Financial support
No financial support was provided relevant to this article.
Competing interests
All authors report no conflicts of interest relevant to this article.
Open access
