- Split View
-
Views
-
Cite
Cite
Stephen T. Chambers, Alan Pithie, Katrina Gallagher, Tina Liu, Christopher J. Charles, Lois Seaward, Treatment of Staphylococcus epidermidis central vascular catheter infection with 70% ethanol locks: efficacy in a sheep model, Journal of Antimicrobial Chemotherapy, Volume 59, Issue 4, April 2007, Pages 779–782, https://doi.org/10.1093/jac/dkl542
- Share Icon Share
Abstract
To determine whether a single treatment with ethanol/water (70:30) will sterilize infected vascular catheters.
A double-blinded, block-randomized trial was conducted in a sheep model comparing the efficacy of one 3 h treatment with ethanol/water (70:30) with heparinized saline for treatment of Hickman vascular catheters infected with Staphylococcus epidermidis. Catheters were cultured using endoluminal brushes, blood cultures, roll plates of catheter tip, broth flushed through the catheter and hub swabs.
There were significantly more sterile catheters in the ethanol treatment group than the saline treatment group (9/11 versus 0/11, P < 0.01, McNemar's χ2 test). The median number of positive cultures in the ethanol treatment group was less than in the saline treatment group (0 versus 5, P = 0.009, Wilcoxon signed-rank test).
A single treatment of 70% ethanol is effective and clinical trials are warranted for treatment of infected vascular catheters using this regimen.
Introduction
Vascular-catheter-related infections are common, difficult to treat and often lead to removal of the catheter. Coagulase-negative staphylococci are the most common cause of these infections, especially in immunocompromised patients in whom long-term central venous access is required.1 Contamination of the catheter hub and lumen contributes to the majority of these infections.2 Treatment with systemic antimicrobial therapy has a high failure rate, although this can be improved by the addition of antibiotic locks to the catheter lumen for 10–14 days.3
Ethanol (30–70%) has not been well studied in vascular catheter infections but is rapidly bactericidal against planktonic bacteria and against both polycarbonate-adherent and Teflon-adherent microorganisms with 2–3 h in vitro.4,5 There is a single report of an open, non-randomized study suggesting that 74% ethanol locked into the catheter lumen for 20–24 h over 3 days in addition to intravenous antimicrobial therapy reduced relapse rates.6 Prolonged ethanol locks have also been reported to be useful prophylactically.7
To test the efficacy of single short exposure of 70% ethanol against catheter infections we used a sheep model of a central venous catheter infected with Staphylococcus epidermidis.
Methods
Ethics
The experimental protocol was approved by the animal ethics committee of the Christchurch School of Medicine and Health Sciences (Number C9/04).
Study design
This study used 3-year-old female Coopworth sheep that were treated in accordance with animal care standards of the University of Otago.
Hickman central venous catheters (Broviac® 6.6 Fr. single lumen silicone catheters, 90 cm length, Bard Access Systems) were placed in using aseptic technique. The wool was clipped and the skin cleansed with 4% chlorhexidine gluconate (Les Enterprises Solumed Inc.) and then 2% chlorhexidine and 70% isopropyl alcohol (Les Enterprises Solumed Inc.). The catheters were placed in the internal jugular vein by percutaneous puncture using a peel away introducer and sutured to the skin. The free end of the catheter was closed with an access port (Interlink injection site, Baxter Healthcare) and kept covered under an elastic bandage (Tubigrip SSL International). After placement catheters were flushed with normal saline and locked with heparinized saline (100 IU/mL). Catheters were inoculated with 1.0 mL of S. epidermidis ATCC 49134 suspension made from an overnight culture on a sheep blood agar (SBA) plate and made up to a turbidity equivalent to that of a 0.5 McFarland standard (approx. 1.5 × 108 cfu/mL) in Mueller-Hinton broth (Oxoid). The exit site was swabbed daily with 2% chlorhexidine gluconate/70% isopropyl alcohol solution (Les Enterprises Solumed) and covered with a dry gauze dressing. The access port was swabbed with 0.5% chlorhexidine/70% isopropyl alcohol (Pharmacia) before use.
Each animal was catheterized twice with a catheter placed on day 0 of each study period and removed on day 5. The second catheter was inserted in the contralateral side and there were at least 8 catheter-free days between treatments. The infected catheters were studied in a cross-over design; one treated with ethanol/water (70:30, v/v) and one treated with heparinized normal saline (100 units/mL). Half of the sheep (two groups of three) received ethanol treatment first and half saline treatment first. The initial treatment was decided by lot. The treatments were prepared in the pharmacy and other staff were blinded as to the identity of the treatments.
Experimental protocol
Day 0. Catheter inserted and fresh inoculum (0.5 mL) was instilled through the hub into the lumen of the catheter and locked in situ for 72 h.
Day 3. The inoculum was aspirated and discarded, and 2 × 10 mL of normal saline was flushed through the catheter. Next 1 mL of ethanol/water (70:30, v/v) or heparinized saline (100 units/mL) was instilled through the access port using an interlink needleless injection system and locked in the catheter for 3 h (catheter lumen volume 0.7 mL). The fluid within the catheter was then aspirated and discarded, the catheter flushed with 2 × 10 mL of normal saline and then locked with 1 mL of heparinized saline (100 units/mL).
Day 5. Samples for culture:
An endoluminal brush (FAS Medical International) was pushed to within 2 cm of the catheter tip, withdrawn, and the brush placed in a sterile container.
Blood (10 mL) aspirated through the catheter was cultured in standard blood culture bottles (two sets) (Biomerieux).
The catheter was withdrawn and the terminal 2 cm removed and placed in a sterile container.
The remaining catheter was placed in a sterile bag, and sealed for transport to the laboratory.
Laboratory procedures
The endoluminal brush tip was placed in 1 mL of phosphate buffered saline (PBS) and the contents were vortexed for 30 s. Then 100 µL of the sample was spread over SBA plates.
Blood culture bottles were incubated in a BacT/AlertR 3D system (Biomerieux).
The terminal 2 cm of the catheter was rolled onto an SBA plate.
The access port was removed from the hub and the inner surface of the hub swabbed and then streaked onto an SBA plate.
The catheter was flushed with 2.5 mL of Mueller-Hinton broth into a sterile plastic bottle and a 10 µL aliquot inoculated immediately onto an SBA plate. The flushed broth was incubated for 48–72 h and subcultured on SBA if turbid.
All cultures were incubated at 36°C and plates were read at 24 and 72 h. All microbiology procedures were carried out in the Canterbury Health Laboratories.
Typing of recovered strains
PFGE was carried out at the Department of Environmental and Scientific Research in Keneperu, Wellington. Chromosomal DNA fragments generated by SmaI digestion were prepared. The fragments were separated on a CHEF Mapper (Bio-Rad Laboratories) using a linear ramp of 5 to 35 s at 6 V/cm for 22 h.
Statistics
Culture positive rates were compared between the paired ethanol and control catheters using McNemar's χ2 test and number of positive cultures per catheter by the Wilcoxon signed-rank test.
Results
The results are presented in Table 1. All 12 sheep completed the saline treatment group of the trial and 11 completed the ethanol treatment group. The second catheter of sheep 7 became blocked and thus all results from this animal were excluded from the statistical analysis.
. | Controls . | Ethanol treated . | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Sheep . | catheter tip . | hub . | flush broth . | flush plate . | FAS brush . | blood culture . | catheter tip . | hub . | flush broth . | flush plate . | FAS brush . | blood culture . |
1 | 0 | 0 | + | 2+ | 1+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
2 | 1+ | 0 | + | 2+ | 0 | +a | 0 | 0 | 0 | 0 | 0 | 0 |
3 | 2+ | 2+ | + | 2+ | 3+ | +a | 0 | 0 | 0 | 0 | 0 | 0 |
4 | 1+ | 2+ | + | 1+ | 4+ | + | 1+ | 3+ | + | 1+ | 2+a | 1+a |
5 | 0 | 2+ | 0 | 1+ | 3+ | + | 0 | 0 | 0 | 0 | 0 | 0 |
6 | 0 | 0 | 0 | 0 | 0 | + | 0 | 0 | 0 | 0 | 0 | 0 |
7 | 0 | 1+ | + | 1+a | 0 | +a | ||||||
8 | 0 | 2+ | + | 1+ | 4+a | + | 0 | 0 | 0 | 0 | 0 | 0 |
9 | 0 | 0 | 0 | 0 | 0 | +a | 0 | 1+a | +a | 0 | 1+a | +a |
10 | 0 | 2+ | + | 3+ | 3+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
11 | 1+ | 2+a | + | 3+ | 0 | +a | 0 | 0 | 0 | 0 | 0 | 0 |
12 | 1+ | 2+ | + | 3+ | 3+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
Total | 5 | 8 | 9 | 10 | 7 | 12 | 1 | 2 | 2 | 1 | 2 | 2 |
. | Controls . | Ethanol treated . | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Sheep . | catheter tip . | hub . | flush broth . | flush plate . | FAS brush . | blood culture . | catheter tip . | hub . | flush broth . | flush plate . | FAS brush . | blood culture . |
1 | 0 | 0 | + | 2+ | 1+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
2 | 1+ | 0 | + | 2+ | 0 | +a | 0 | 0 | 0 | 0 | 0 | 0 |
3 | 2+ | 2+ | + | 2+ | 3+ | +a | 0 | 0 | 0 | 0 | 0 | 0 |
4 | 1+ | 2+ | + | 1+ | 4+ | + | 1+ | 3+ | + | 1+ | 2+a | 1+a |
5 | 0 | 2+ | 0 | 1+ | 3+ | + | 0 | 0 | 0 | 0 | 0 | 0 |
6 | 0 | 0 | 0 | 0 | 0 | + | 0 | 0 | 0 | 0 | 0 | 0 |
7 | 0 | 1+ | + | 1+a | 0 | +a | ||||||
8 | 0 | 2+ | + | 1+ | 4+a | + | 0 | 0 | 0 | 0 | 0 | 0 |
9 | 0 | 0 | 0 | 0 | 0 | +a | 0 | 1+a | +a | 0 | 1+a | +a |
10 | 0 | 2+ | + | 3+ | 3+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
11 | 1+ | 2+a | + | 3+ | 0 | +a | 0 | 0 | 0 | 0 | 0 | 0 |
12 | 1+ | 2+ | + | 3+ | 3+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
Total | 5 | 8 | 9 | 10 | 7 | 12 | 1 | 2 | 2 | 1 | 2 | 2 |
Results from sheep 7 were excluded from statistical analysis.
+ = growth in liquid medium, 1+ 1–10 colonies, 2+ 11–100 colonies, 3+ 101–1000 colonies and 4+ > 1000 colonies on an agar plate.
aStrain typed by PFGE.
. | Controls . | Ethanol treated . | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Sheep . | catheter tip . | hub . | flush broth . | flush plate . | FAS brush . | blood culture . | catheter tip . | hub . | flush broth . | flush plate . | FAS brush . | blood culture . |
1 | 0 | 0 | + | 2+ | 1+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
2 | 1+ | 0 | + | 2+ | 0 | +a | 0 | 0 | 0 | 0 | 0 | 0 |
3 | 2+ | 2+ | + | 2+ | 3+ | +a | 0 | 0 | 0 | 0 | 0 | 0 |
4 | 1+ | 2+ | + | 1+ | 4+ | + | 1+ | 3+ | + | 1+ | 2+a | 1+a |
5 | 0 | 2+ | 0 | 1+ | 3+ | + | 0 | 0 | 0 | 0 | 0 | 0 |
6 | 0 | 0 | 0 | 0 | 0 | + | 0 | 0 | 0 | 0 | 0 | 0 |
7 | 0 | 1+ | + | 1+a | 0 | +a | ||||||
8 | 0 | 2+ | + | 1+ | 4+a | + | 0 | 0 | 0 | 0 | 0 | 0 |
9 | 0 | 0 | 0 | 0 | 0 | +a | 0 | 1+a | +a | 0 | 1+a | +a |
10 | 0 | 2+ | + | 3+ | 3+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
11 | 1+ | 2+a | + | 3+ | 0 | +a | 0 | 0 | 0 | 0 | 0 | 0 |
12 | 1+ | 2+ | + | 3+ | 3+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
Total | 5 | 8 | 9 | 10 | 7 | 12 | 1 | 2 | 2 | 1 | 2 | 2 |
. | Controls . | Ethanol treated . | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Sheep . | catheter tip . | hub . | flush broth . | flush plate . | FAS brush . | blood culture . | catheter tip . | hub . | flush broth . | flush plate . | FAS brush . | blood culture . |
1 | 0 | 0 | + | 2+ | 1+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
2 | 1+ | 0 | + | 2+ | 0 | +a | 0 | 0 | 0 | 0 | 0 | 0 |
3 | 2+ | 2+ | + | 2+ | 3+ | +a | 0 | 0 | 0 | 0 | 0 | 0 |
4 | 1+ | 2+ | + | 1+ | 4+ | + | 1+ | 3+ | + | 1+ | 2+a | 1+a |
5 | 0 | 2+ | 0 | 1+ | 3+ | + | 0 | 0 | 0 | 0 | 0 | 0 |
6 | 0 | 0 | 0 | 0 | 0 | + | 0 | 0 | 0 | 0 | 0 | 0 |
7 | 0 | 1+ | + | 1+a | 0 | +a | ||||||
8 | 0 | 2+ | + | 1+ | 4+a | + | 0 | 0 | 0 | 0 | 0 | 0 |
9 | 0 | 0 | 0 | 0 | 0 | +a | 0 | 1+a | +a | 0 | 1+a | +a |
10 | 0 | 2+ | + | 3+ | 3+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
11 | 1+ | 2+a | + | 3+ | 0 | +a | 0 | 0 | 0 | 0 | 0 | 0 |
12 | 1+ | 2+ | + | 3+ | 3+a | +a | 0 | 0 | 0 | 0 | 0 | 0 |
Total | 5 | 8 | 9 | 10 | 7 | 12 | 1 | 2 | 2 | 1 | 2 | 2 |
Results from sheep 7 were excluded from statistical analysis.
+ = growth in liquid medium, 1+ 1–10 colonies, 2+ 11–100 colonies, 3+ 101–1000 colonies and 4+ > 1000 colonies on an agar plate.
aStrain typed by PFGE.
The primary endpoint was sterilization of the catheter and there were significantly more sterile catheters in the ethanol treated group than the heparinized saline treated group (9/11 versus 0/11, P < 0.01).
The median number of positive culture sites from catheters in the ethanol treatment group was also less than in the saline treatment group (0 versus 5, P = 0.009).
Blood cultures were the most sensitive measure and were positive in all 11 of the saline treated catheters included in the analysis and 2 of the 11 ethanol treated catheters (P < 0.01). Both the flush plate cultures (9/11 versus 2/11, P = 0.008) and flush broth cultures (8/11 versus 2/11, P = 0.016) were positive significantly more often in the saline treatment group than in the ethanol treatment group.
Twenty isolates were recovered from storage for typing as shown in Table 1. All strains had an identical PFGE profile to the type strain S. epidermidis ATCC 49134.
Discussion
The study showed that a single 3 h exposure of ethanol/water (70:30, v/v) was effective in 9/11 infected catheters compared with 0/11 treated with heparinized saline. This result is consistent with the results of two in vitro studies that show ethanol is bactericidal against adherent S. epidermidis and other organisms after a single short exposure.4,5 It extends these observations by demonstrating that ethanol is active in vivo and that prolonged repeated doses that have been used clinically are not necessary.6
Ethanol was not effective in two of the treated catheters. The most likely source of these organisms is the junction between the access port and catheter hub that was protected from the ethanol while the access port was in place. This area was exposed when the access port was removed to introduce the endoluminal brush through the hub, potentially contaminating the brush and subsequent cultures. Disinfecting the internal part of the catheter hub prior to brushing and taking cultures might have prevented this. It is also possible that some organisms persisted within the track made in the membranous septum of the access port by the plastic needle used to inoculate the catheter, remove the alcohol and instil saline.8 We cannot distinguish between these possibilities.
This study is limited in that a single catheter type, access port and material were tested, however ethanol is active on polycarbonate and Teflon surfaces and should be active on all materials used to make catheters and their accessories. Extensive testing has shown that ethanol does not appreciably harm polyurethane or silicone catheters and should be further considered for clinical use.9 One catheter became blocked in the trial but the ethanol could not be implicated as this occurred after the bacterial suspension was put into the catheter and prior to the ethanol lock.
Ethanol has an attractive profile for clinical use. The rapid bactericidal activity offers a much more convenient regimen than the 10–14 days needed for antibiotic locks. It has a broad spectrum of activity although the dwell time needed for organisms such as Candida albicans may need to be defined further.5 There is a low risk of cross-resistance to antibiotics as ethanol acts by denaturation rather than at a specific target. The adverse effects are well known and minor with minimal exposure required to disinfect the lines.6 In this model 0.4 g of ethanol (1 mL of 70% ethanol) was injected into the circulation which is trivial compared with a standard drink of 10 g. Given the proven activity in this model clinical trials of short-term single dose treatment of infected vascular catheters with 70% ethanol are warranted.
Acknowledgements
We thank Helen Heffernan for doing the PFGE. FAS Medical International kindly donated the brushes that were used in the study. This research was funded by a summer studentship given to Tina Liu by the Canterbury Medical Research Foundation (November 2004–January 2005), the Infectious Diseases Trust Fund of Christchurch Hospital and by the Haematology Trust Fund of Christchurch Hospital.
Transparency declarations
None to declare.