PAPER
A comparison of microbial growth in alfaxalone, propofol and thiopental OBJECTIVES: To compare the growth of Staphylococcus aureus and Escherichia coli in alfaxalone with that in propofol and thiopental and to evaluate contaminant microbial growth in these agents under two different conditions of storage and handling. METHODS: Known quanta of S aureus and E coli were inoculated into separate 5 ml samples of propofol, thiopental and alfaxalone. Quantitative bacterial analysis was performed at intervals over a 14 day period. Commercial preparations of propofol, thiopental and alfaxalone were stored and handled using ‘‘dirty’’ or ‘‘clean’’ techniques. Microbial quantification and identification was performed over a 14 day period. RESULTS: S aureus and E coli grew rapidly in propofol after six hours. Both bacteria were killed by thiopental. S aureus numbers slowly declined in alfaxalone; E coli growth was rapid after 24 hours. In ‘‘dirty’’ and ‘‘clean’’ groups of intravenous anaesthetics, 9 3 and
7 4 per cent of samples, respectively, were positive for microbial
growth; none were considered to represent colonisation of bottles. CLINICAL SIGNIFICANCE: Alfaxalone supports growth of some microorganisms but less readily than propofol. Bacterial colonisation of intravenous anaesthetic bottles is uncommon, but contamination as syringes are prepared for injection occurs regardless of storage and handling technique. F. A. STRACHAN, J. C. MANSEL R. E. CLUTTON
INTRODUCTION
AND
Journal of Small Animal Practice (2008) 49, 186–190 DOI: 10.1111/j.1748-5827.2007.00473.x
Department of Veterinary Clinical Studies, Royal (Dick) School of Veterinary Studies, Easter Bush Veterinary Centre, University of Edinburgh, Roslin, Midlothian EH25 9RG
186
Propofol (2,6-diisopropylphenol) is a popular drug for intravenous induction and maintenance of anaesthesia in dogs and cats. Unfortunately, its formulation as a lipid emulsion with soya bean oil, glycerol and egg lecithin promotes growth of bacteria and yeasts following contamination. This capacity as an excellent culture medium has led to its implication in many outbreaks of postoperative surgical wound infections and septicaemia and has resulted in death of human beings (Veber and others 1994, Bennett and others 1995, Kuehnert and others 1997, Henry and others 2001). Propofol has also been associated with an increase in postoperative wound infections in dogs and cats Journal of Small Animal Practice
Vol 49
(Heldmann and others 1999). Current recommendations for its use in veterinary practice are to discard any drug remaining after withdrawal of the required dose, as the product contains no antimicrobial preservative. Sodium thiopental is still widely used as an anaesthetic induction agent in veterinary medicine. As solutions of barbiturates degrade relatively rapidly over time, thiopental is stored as a crystalline powder and dissolved in water before use. Current manufacturers’ recommendations are to store reconstituted solutions in the refrigerator and use within 24 hours, but storage for several weeks at room temperature is probably common clinical practice. Thiopental has been shown to remain sterile at room temperature for at least six days after reconstitution (Haws and others 1998) and kills inoculated Staphylococcus aureus, Escherichia coli and Pseudomonas aeruginosa within three to six hours (Crowther and others 1996). Alfaxalone (3a-hydroxy-5a-pregnane-11, 20-dione) is a synthetic neuroactive steroid, recently licensed in a new formulation for intravenous induction and maintenance of anaesthesia in dogs and cats (Alfaxan-CD RTU; Jurox Pty. Ltd). Alfaxalone has been available for many years for use in cats as Saffan (ScheringPlough Animal Health) combined with alphadolone and dissolved in a polyoxyethylated castor oil-based surfactant (Cremophor EL; BASF Fine Chemicals). Alfaxan-CD RTU contains a novel cyclodextrin (2-hydroxypropyl-beta cyclodextrin) as an excipient. This large oligosaccharide molecule encases the relatively hydrophobic alfaxalone molecule and has a sufficiently hydrophilic exterior to impart water solubility to the complex. As the product contains no antimicrobial preservative, the manufacturers recommend that the bottle is discarded following withdrawal of the required dose, but no information is available on its capacity to support microbial growth. The aims of this study were: (1) to quantify the growth of S aureus and E coli in alfaxalone compared with growth in propofol and thiopental and (2) to evaluate
April 2008 Ó 2008 British Small Animal Veterinary Association
Microbial growth in anaesthetic agents
contaminant microbial growth in these three anaesthetics under two different clinical conditions of storage and handling.
MATERIALS AND METHODS Part 1 The methodology was similar to that described by Sosis and Braverman (1993). A colony from a stock culture of S aureus (National Collection of Type Cultures No. 6571; Health Protection Agency) maintained by our hospitals’ clinical microbiology laboratory was transferred to a culture medium of ColumbiaÒ base agar with 5 per cent horse blood (Oxoid Ltd) and incubated at 37°C overnight. Ten colonies from this plate were suspended in 10 ml of 09 per cent sterile saline (Phosphate Buffered Saline; Oxoid Ltd) and serial dilutions made to achieve a concentration of 1:10,000. Inoculation of a nutrient agar plate (Tryptone Soya Agar; Oxoid Ltd) with 20 ll of this suspension resulted in growth of 415 colonies after incubation at 37°C for 24 hours, from which a bacterial concentration of 207104 colony forming units (CFUs)/ ml was calculated. A suspension of E coli (National Collection of Type Cultures No. 10418; Health Protection Agency) was prepared in the same way, which resulted in a bacterial concentration of 823103 CFUs/ml. Aliquots (50 ll) of the prepared S aureus and E coli suspensions were added to separate sterile vials (Universal Container 30 ml; Scientific Laboratory Supplies), containing 5 ml of anaesthetic from each of three bottles of alfaxalone, propofol (Rapinovet; Schering-Plough Animal Health) and thiopental (Thiovet 25 g; Novartis Animal Health UK Ltd) reconstituted to a 25 per cent solution with 100 ml sterile water (Water for injection; Novartis Animal Health UK Ltd). The suspensions were agitated between each aliquot removal. The initial concentration in the vials inoculated with S aureus was 207 CFUs/ml and in those inoculated with E coli was 82 CFUs/ml. After inoculation, the 18 sealed vials were kept at room temperature between sampling times. Each vial was agitated Journal of Small Animal Practice
Vol 49
and two 20 ll samples removed with a precision loop and plated on nutrient agar at zero, six, 24, 48 hours and seven and 14 days after inoculation. The plates were incubated at 37°C for 24 hours, and the number of colonies was counted; the mean of counts from the two samples from each vial was recorded. Part 2 Three commercially available rubbertopped glass bottles of each of alfaxalone (10 ml), propofol (20 ml) and 25 per cent thiopental (100 ml) were used for Part 2A, and three bottles of each for Part 2B. Strict aseptic technique was used to remove caps from bottles of propofol and alfaxalone and to reconstitute thiopental powder with sterile water using a transfer set. Aseptic technique was required as during our preliminary study, opening of alfaxalone bottles and reconstitution of thiopental using non-sterile methods resulted in microbial growth from time 0 samples (data not included). A 05 ml sample was removed from each of the 18 bottles using a 25 ml syringe and 22G needle immediately after opening and reconstitution and placed in a sterile vial. Twenty microlitres of the sample was spread evenly over a nutrient agar plate and incubated at 37°C for 24 hours. All plates were incubated for 24 hours before inoculation to exclude previous contamination. Colonies on the plate were counted and microorganisms identified using standard techniques (Quinn and others 1994). In Part 2A, the three bottles of each agent were stored in ‘‘dirty’’ conditions in an open-top box near the sink in a room where anaesthesia was induced and dogs and cats prepared for surgery. Throughout the study, these bottles were handled by final year veterinary students, without prior hand washing. The rubber seals on the bottles were broached using a 22G needle attached to a 25 ml syringe, and a 05 ml sample was drawn from each bottle three times on day 1, twice on days 2 to 5 and once on days 8, 9, 10, 13, 14 and 15 to simulate withdrawing multiple doses for clinical use. In Part 2B, bottles were stored in the same room, but on a ‘‘clean’’ shelf in a wall cabinet. These bottles were only handled by one investigator (F. A. S.)
April 2008 Ó 2008 British Small Animal Veterinary Association
after washing hands using an alcohol free antimicrobial foam (Foam Hand Wash; NewGenn Research Ltd) and disinfecting rubber seals on the bottles with chlorhexidine (Hibitane Concentrate 5 per cent; Regent Medical Ltd) mixed with distilled water and spirit (Industrial Methylated Spirit; W&J Dunlop Ltd) before withdrawing samples. The seals were broached and 05 ml samples withdrawn at the same times as Part 2A. Samples (05 ml) taken as part of the protocol described above from each bottle at six, 24 and 48 hours, seven and 14 days after first opening were submitted for bacteriological culture. The temperature in the ‘‘dirty’’ and ‘‘clean’’ areas was between 205 and 238°C for the duration of the study (Digital Thermo-Hygrometer; Harvard Apparatus Ltd). pH of three samples from recently opened bottles of propofol, thiopental and alfaxalone in clinical use was measured using a pH meter (pHepÒ1 HI96108; Hanna Instruments) calibrated using standard buffer solutions.
RESULTS Part 1 The number of CFUs/ml of S aureus and E coli, calculated from colonies counted on inoculated plates, are shown in Tables 1 and 2, respectively. When the number of colonies exceeded 500 per plate, counting was stopped because of overgrowth of the microorganism on the plate and difficulty in determining individual colonies. With S aureus, CFUs in propofol increased exponentially from 24 hours, and at seven days and thereafter, numbers were too high to count. In thiopental, bacteria were killed by six hours, although positive cultures from one vial at seven days and a different vial at 14 days were noted. Numbers of CFUs of S aureus in alfaxalone were maintained initially then slowly declined. Figure 1 shows S aureus growth over the first 48 hours, after which time numbers in propofol could not be plotted. With E coli, profuse growth was noted in propofol at 24 hours, with more CFUs than could be accurately counted at 48 187
F. A. Strachan and others
Agent
Zero hours
Six hours
24 hours
48 hours
Seven days
14 days
PROP THIO ALFAX
186104 2676144 206150
226104 006000 276029
1256427 006000 136029
156761155 006000 276104
.2506000 6061039 036029
.2506000 156260 036058
CFUs\ml x 100
Table 1. Growth of inoculated Staphylococcus aureus in intravenous anaesthetics: colony forming units/ml 102 (mean of three vials of each agent6SD)
PROP Propofol, THIO Thiopental, ALFAX Alfaxalone
hours and thereafter. No growth of E coli was noted in thiopental, except in one vial at 24 hours. In alfaxalone, E coli numbers were increased at 48 hours, and at seven days and 14 days, the numbers were too many to count. Figure 2 shows E coli growth over the first 24 hours, after which time numbers in propofol and later in alfaxalone could not be plotted. Part 2 Microorganisms identified in Part 2A and 2B are shown in Table 3. Out of 108 samples submitted for bacteriological culture, nine (83 per cent) were positive for microbial growth. Organisms cultured were DNAse negative Staphylococcus (n=6), Bacillus species (n=2) and a yeast. Five out of 54 samples (93 per cent) from the ‘‘dirty’’ group and four out of 54 (74 per cent) from the ‘‘clean’’ group were positive for microbial growth. Of the positive samples, two (222 per cent) were from propofol, five (555 per cent) from thiopental and three (333 per cent) from alfaxalone. Mean pH (6sd) of three samples of propofol was 796006, thiopental 1096 012 and alfaxalone 676006.
DISCUSSION To our knowledge, this study is the first to quantify growth of S aureus and E coli in alfaxalone compared with growth in pro-
pofol and thiopental and to attempt evaluation of contaminant microbial growth in these three agents in a veterinary clinical setting. Propofol readily supports the growth of various microorganisms (Arduino and others 1991, Sosis and Braverman 1993, Sosis and others 1995, Crowther and others 1996, Wachowski and others 1998, Sakuragi and others 1999). Our results confirm that propofol is an excellent medium for the growth of S aureus and E coli, both of which replicate rapidly in the emulsion at room temperature. Although more frequent sampling would be required to plot precise bacterial growth curves, our data demonstrates a lag phase of at least six hours after inoculation into propofol for both organisms studied. This finding is similar to previous work and has formed the basis of recommendations in human hospitals that propofol should be used within six to 12 hours after the vial is first pierced (Tre´panier and Lessard 2003, DiprivanÔ Data Sheet 2006) after which time exponential growth of contaminant bacteria occurs. In contrast, thiopental rapidly kills many pathogenic microorganisms. In our study, S aureus and E coli were killed within six and zero hours respectively, after inoculation into 25 per cent thiopental solution. This is comparable with work by Young and others (1958) who demonstrated a 107 per cent survival of
FIG 1. Growth of inoculated Staphylococcus aureus in anaesthetic agents. PROP Propofol, THIO Thiopental, ALFAX Alfaxalone, CFUs Colony forming units
S aureus at two hours and 0 per cent survival at 24 hours after inoculation into thiopental. Highsmith and others (1982) found that of 13 common nosocomial pathogens, 11 were killed within 24 hours of inoculation into thiopental solution, with only Enterococcus and Candida able to survive up to 24 and 48 hours respectively. Three positive cultures obtained from different thiopental vials in our study at 24 hours, seven days and 14 days were considered to be the result of survival of organisms washed up on the side or lid of the vial after agitation, or were due to laboratory contamination. Thiopental is frequently used for longer after reconstitution than recommended by the manufactures in many human hospitals (Wong and others 1992). Our study demonstrates that alfaxalone supports the growth of some microorganisms but less readily than propofol. Conditions in alfaxalone were not suitable for the growth of S aureus, which maintained
Table 2. Growth of inoculated Escherichia coli in intravenous anaesthetics: colony forming units/ml 102 (mean of three vials of each agent6SD) Agent
Zero hours
Six hours
24 hours
48 hours
Seven days
14 days
PROP THIO ALFAX
236029 006000 076029
256100 006000 106005
206766429 026029 176076
.2506000 006000 636293
.2506000 006000 .2506000
.2506000 006000 .2506000
PROP Propofol, THIO Thiopental, ALFAX Alfaxalone
188
Journal of Small Animal Practice
Vol 49
FIG 2. Growth of inoculated Escherichia coli in anaesthetic agents. PROP Propofol, THIO Thiopental, ALFAX Alfaxalone, CFUs Colony forming units
April 2008 Ó 2008 British Small Animal Veterinary Association
Microbial growth in anaesthetic agents
Table 3. Contaminant microbial growth in intravenous anaesthetic agents subject to ‘‘dirty’’ and ‘‘clean’’ storage and handling techniques: colony forming units/ml 310 2 Drug
Zero hours
Part 2A
‘‘dirty’’
PROP 1 PROP 2 PROP 3 THIO 1 THIO 2 THIO 3 ALFAX 1 ALFAX 2 ALFAX 3
0 0 0 0 0 0 0 0 0
Part 2B
‘‘clean’’
PROP 1 PROP 2 PROP 3 THIO 1 THIO 2 THIO 3 ALFAX 1 ALFAX 2 ALFAX 3
0 0 0 0 0 0 0 0 0
Six hours
24 hours
48 hours
Seven days
14 days
0 0 0 0 0 0 0 .250 Staph spp 0
0 0 0 0 05 Staph spp 35 Staph spp 0 0 0
0 0 0 0 .250 Staph spp 0 0 0 0
0 0 0 0 0 0 0 0 0
0 0 0 0 0 .250 yeast 0 0 0
0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 115 Bacillus species 0 0
0 0 05 Staph spp 0 0 0 0 05 Bacillus species 0
0 0 0 0 0 0 0 0 0
0 .250 Staph spp 0 0 0 0 0 0 0
PROP Propofol, THIO Thiopental, Staph spp Staphylococcus species, ALFAX Alfaxalone
fairly constant numbers initially then slowly declined. E coli had a similar exponential growth curve in alfaxalone to that in propofol but with a longer lag phase of at least 24 hours. This suggests that although alfaxalone will support the growth of certain microorganisms, it is a less favourable medium than propofol, and bacteria require a longer period of adaptation before exponential growth occurs. Alfaxalone does not possess the bactericidal properties of thiopental. In the absence of a specific antimicrobial preservative, differences between bacterial growth in intravenous solutions depend on the nutritional components present (Jarvis and Highsmith 1984) and also on pH and temperature. None of the three agents in our study contain an antimicrobial preservative but vary greatly in composition. Propofol is formulated as a lipid emulsion containing soya bean oil, which has nutrients capable of supporting massive growth of a number of nosocomial pathogens (McHugh and Roper 1995). Most pathogenic bacteria prefer a narrow pH range of 60 to 80. The pH of propofol and alfaxalone provides a suitable environment for bacterial replication, whereas the strongly alkaline pH of thiopental creates a bactericidal environment for most pathogenic microorganisms. Although reconstituted barbiJournal of Small Animal Practice
Vol 49
turates are unstable in solution, Wong and others (1992) showed that even uncapped bottles of 25 per cent thiopental maintained their pH for at least four weeks at room temperature. In our study, the three anaesthetic agents were kept at room temperature, although the manufactures of thiopental recommend storage at 2 to 8°C after reconstitution. In Part 2 of this study, we tried to replicate poor (‘‘dirty’’) and good (‘‘clean’’) clinical practice in the storage and handling of intravenous anaesthetic agents. Rubber seals on the bottles were pierced on multiple occasions to simulate frequent use over a two week period. The samples withdrawn were necessarily smaller in volume than for clinical use as each bottle was required to last two weeks, but we made the assumption that the action of piercing the vial was more important than the volume withdrawn in terms of causing contamination. From the samples cultured over the two week study period, 83 per cent were positive for microbial growth. The microorganisms cultured, although not identified to species level, were considered to be common non-pathogenic commensals or saprophytes. None of the positive cultures between 6 and 48 hours were considered to represent colonisation of the bottles, as, with one exception in the ‘‘dirty’’ thio-
April 2008 Ó 2008 British Small Animal Veterinary Association
pental group, the same organism was not present in subsequent samples from the same bottle. Also, more than half of the positive cultures were from the thiopental samples, in which microorganisms would be least likely to survive. Contamination must therefore have occurred during the process of preparing the syringe and needle and withdrawing drug from the bottle, or because of laboratory contamination. The use of ‘‘dirty’’ or ‘‘clean’’ storage and handling techniques in our study did not affect the chances of getting a positive culture from the samples. This finding is similar to that of Lorenz and others (2002) who demonstrated 82 per cent contamination of samples subject to routine handling of propofol and 88 per cent where there was strict adherence to the manufacturers’ recommendations. However, in both studies, neither of the handling methods completely prevented contamination, and in view of the potential for rapid replication of contaminants in some intravenous anaesthetics, these results must not be used to excuse poor storage and handling techniques. As demonstrated by our pilot study, use of a sterile technique when opening bottles of intravenous anaesthetic, and when using transfer sets for reconstitution of thiopental, is important for avoidance of contamination. 189
F. A. Strachan and others
Conclusions This study demonstrates that there is a low risk of introducing contamination into rubber-topped bottles of anaesthetic agent during routine use, even if these are stored and handled inappropriately. However, it appears to be fairly easy to cause contamination of the drug as syringes are prepared for injection, even after washing hands and disinfecting rubber seals. As microorganisms have potential to replicate very rapidly in propofol, it would appear that the veterinary manufacturers’ recommendation to discard bottles after withdrawal of the required dose is appropriate, and in addition, samples drawn into syringes should be rapidly used or discarded. As there is a longer delay before bacteria multiply rapidly in alfaxalone, it may be safe to use for a longer period after opening. However, further investigation using a wider range of microorganisms and larger sample numbers would be required to state this definitively. Thiopental is likely to remain safe for use for much longer after reconstitution than the manufacturers recommend.
Acknowledgements The authors wish to thank Lorna Hume and Jennifer Harris for their help with the bacteriological analysis in this study,
190
and Schering-Plough Animal Health and Vetoquinol for their financial assistance. References ARDUINO, M. J., BLAND, L. A., MCALLISTER, S. K., AGUERO, S. M., VILLARINO, M. E., MCNEIL, M. M., JARVIS, W. R. & FAVERO, M. S. (1991) Microbial growth and endotoxin production in the intravenous anesthetic propofol. Infection Control and Hospital Epidemiology 12, 535-539 BENNETT, S. N., MCNEIL, M. M., BLAND, L. A., ARDUINO, M. J., VILLARINO, M. E., PERROTTA, D. M., BURWEN, D. R., WELBEL, S. F., PEGUES, D. A., STROUD, L., ZEITZ, P. S. & JARVIS, W. R. (1995) Postoperative infections traced to contamination of an intravenous anesthetic, propofol. New England Journal of Medicine 333, 147-154 CROWTHER, J., HRAZDIL, J., JOLLY, D. T., GALBRAITH, J. C., GREACEN, M. & GRACE, M. (1996) Growth of microorganisms in propofol, thiopental and a 1:1 mixture of propofol and thiopental. Anesthesia and Analgesia 82, 475-478 DiprivanÔ 10 mg/mL (1%) and 20 mg/mL (2%) Emulsion for Injection or Infusion Core Data Sheet. (2006). London, UK: AstraZeneca HAWS, J. L., HERMAN, N., CLARK, Y., BJORAKER, R. & JONES, D. (1998) The chemical stability and sterility of sodium thiopental after preparation. Anesthesia and Analgesia 86, 208-213 HELDMANN, E., BROWN, D. C. & SHOFER, F. (1999) The association of propofol usage with postoperative wound infection rate in clean wounds: a retrospective study. Veterinary Surgery 28, 256-259 HENRY, B., PLANTE-JENKINS, C. & OSTROWSKA, K. (2001) An outbreak of Serratia marcescens associated with the anesthetic agent propofol. American Journal of Infection Control 29, 312-315 HIGHSMITH, A. K., GREENHOOD, G. P. & ALIEN, J. R. (1982) Growth of nosocomial pathogens in multiple-dose parenteral medication vials. Journal of Clinical Microbiology 15, 1024-1028 JARVIS, W. R. & HIGHSMITH, A. K. (1984) Bacterial growth and endotoxin production in lipid emulsion. Journal of Clinical Microbiology 19, 17-20 KUEHNERT, M. J., WEBB, R. M., JOCHIMSEN, E. M., HANCOCK, G. A., ARDUINO, M. J., HAND, S., CURRIER, M. & JARVIS, W. R. (1997) Staphylococcus aureus
Journal of Small Animal Practice
Vol 49
bloodstream infections among patients undergoing electroconvulsive therapy traced to breaks in infection control and possible extrinsic contamination by propofol. Anesthesia and Analgesia 85, 420-425 LORENZ, I. H., KOLBITSCH, C., LASS-FLORL, C., GRITZNIG, I., VOLLERT, B., LINGNAU, W., MOSER, P. L. & BENZER, A. (2002) Routine handling of propofol prevents contamination as effectively as does strict adherence to the manufacturer’s recommendations. Canadian Journal of Anesthesia 49, 347-352 MCHUGH, G. J. & ROPER, G. M. (1995) Propofol emulsion and bacterial contamination. Canadian Journal of Anesthesia 42, 801-804 QUINN, P. J., CARTER, M. E., MARKEY, B. & CARTER, G. R. (1994) Clinical Veterinary Microbiology. London, UK: Mosby International Ltd SAKURAGI, T., YANAGISAWA, K., SHIRAI, Y. & DAN, K. (1999) Growth of Escherichia coli in propofol, lidocaine, and mixtures of propofol and lidocaine. Acta Anaesthesiologica Scandinavica 43, 476479 SOSIS, M. B. & BRAVERMAN, B. (1993) Growth of Staphylococcus aureus in four intravenous anesthetics. Anesthesia and Analgesia 77, 766-768 SOSIS, M. B., BRAVERMAN, B. & VILLAFLOR, E. (1995) Propofol, but not thiopental, supports the growth of Candida albicans. Anesthesia and Analgesia 81, 132-134 TRE´PANIER, C. A. & LESSARD, M. R. (2003) Propofol and the risk of transmission of infection. Canadian Journal of Anesthesia 50, 533-537 VEBER, B., GACHOT, B., BEDOS, J. P. & WOLFF, M. (1994) Severe sepsis after intravenous injection of contaminated propofol. Anesthesiology 80, 712-713 WACHOWSKI, I., JOLLY, D. T., HRAZDIL, J., GALBRAITH, J. C., GREACEN, M. & CLANACHAN, A. S. (1998) The growth of microorganisms in propofol and mixtures of propofol and lidocaine. Anesthesia and Analgesia 88, 209-212 WONG, C. L., WARRINER, C. B., MCCORMACK, J. P. & CLARKE, A. M. (1992) Reconstituted thiopentone retains its alkalinity without bacterial contamination for up to four weeks. Canadian Journal of Anesthesia 39, 504-508 YOUNG, J., COLLETE, T. S. & BREHM, W. F. (1958) Sterility of multiple-dose vials after repeated use. American Surgeon 24, 811-814
April 2008 Ó 2008 British Small Animal Veterinary Association
