1 Dimethyl sulfoxide enhances the effectiveness of skin antiseptics
Transcript of 1 Dimethyl sulfoxide enhances the effectiveness of skin antiseptics
1
Dimethyl sulfoxide enhances the effectiveness of skin antiseptics and reduces the 2
contamination rates of blood cultures. 3
4
Jeffrey J. Tarrand1*, Paul R. LaSala1, Xiang-Yang Han1, Kenneth.V. Rolston2, Dimitrios. 5
P. Kontoyiannis2 6
7
Departments of Laboratory Medicine1 and Department of Infectious Diseases, Infection 8
Control, and Employee Health,2 The University of Texas M. D. Anderson Cancer Center, 9
Houston, Texas. 10
11
*Corresponding Author. Mailing Address: Department of Laboratory Medicine, Unit 12
084, 1515 Holcombe Blvd., The University of Texas M. D. Anderson Cancer Center, 13
Houston, TX. 77030. Phone: (713) 792 2932. FAX: (713) 792 0936. E-mail: 14
16
Running title: Dimethyl sulfoxide enhances antiseptic killing. 17
Key words: polar-aprotic-solvents, antiseptic, contamination 18
Word Count: 362819
Copyright © 2012, American Society for Microbiology. All Rights Reserved.J. Clin. Microbiol. doi:10.1128/JCM.05106-11 JCM Accepts, published online ahead of print on 29 February 2012
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
Abstract. 20
21
Effective skin antisepsis is of central importance in the prevention wound infections, 22
colonization of medical devices, and nosocomial transmission of microorganisms. 23
Current antiseptics have a suboptimal efficacy resulting in substantial infectious 24
morbidity, mortality, and increased health care costs. Here we introduce an in-vitro 25
method for antiseptic testing and a novel alcohol-based antiseptic containing 4-5% of the 26
polar aprotic solvent dimethyl sulfoxide (DMSO). The DMSO-containing antiseptic 27
resulted in 1 to 2 log enhanced killing of Staphylococcus epidermidis and other microbes 28
in vitro when compared to the same antiseptic without DMSO. In a prospective clinical 29
validation, blood culture contamination rates were reduced from 3.04% for 70% 30
isopropanol/1% iodine (control antiseptic), to 1.04% for 70% isopropanol/1% iodine/ 31
with 5% DMSO (P < 0.01). Our results predict that improved skin antisepsis is possible 32
using new formulations of antiseptics containing strongly polarized but non-ionizing 33
(polar-aprotic) solvents. 34
35
36
37
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
Antiseptics are crucial for the prevention of post-operative and device associated 38
infections; such infections result in substantial additional morbidity and health care costs 39
(1-6, 11, 14, 27). Recently, attention has focused on antiseptic hand washing to 40
decolonize bacteria from the skin of health-care workers (5). Bacterial skin-flora typically 41
become sequestered in layers of dead keratinized skin, sweat glands, and hair follicles, 42
making effective skin decontamination difficult. In addition to wound infections and 43
other true infections, antiseptic failure can cause blood specimen contamination and 44
contribute to inappropriate treatment and increased costs (27). Specifically, in one report 45
when blood cultures were analyzed alone, a single false positive blood culture from a 46
hospitalized inpatient costs the patient an additional $4,200 in unnecessary medication, 47
additional follow up testing, and increased length of stay (1). 48
Clinical studies of antiseptic efficacy usually employ blood culture contamination 49
with skin flora as the indicator system, due to the large number of samples screened and 50
the standardization of antiseptic practices. Currently used antiseptics have a significant 51
failure rate, resulting in inappropriate evaluations for sepsis, unnecessary antibiotics, and 52
increased days of hospitalization (1, 2, 9, 27). This has been shown for iodine/alcohol as 53
well as chlorhexidine/alcohol mixtures where blood culture contamination where rates 54
ranging from 3%-5% have been reported in a tertiary care setting (2, 14, 18, 20). Thus, 55
there remains a significant need for improved antiseptics. One element of discovery of 56
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
new antiseptics is developing suitable models for screening new antiseptics. In addition 57
to the choice of antiseptic used, it is important to remember that effective technique and 58
dedicated phlebotomy teams have been shown to have major impacts on contamination 59
rates as well (2, 14, 27). For the clinical study presented here we attempt to study only 60
the effect of antiseptic type in isolation from these other technique and practice related 61
effects. 62
In this paper we describe a two-part study aimed at improving both the antiseptics 63
used in clinical practice, and the screening method used to determine candidate 64
antiseptics. First, we describe a method that allows rapid in vitro screening of antiseptic 65
agents. Second, we introduce a novel antiseptic containing the polar-aprotic solvent 66
dimethyl sulfoxide (DMSO), a biocompatible solvent with low toxicity (FDA directive 67
67/548/ec). Finally, we validate the performance of the DMSO-containing antiseptic in a 68
clinical trial involving 1,590 antiseptic application events from patients receiving blood 69
cultures at our institution. 70
71
MATERIALS AND METHODS 72
To determine the relative effectiveness of the new verses standard antiseptics both 73
in-vitro and in vivo approaches were used. The in vitro method directly compares split 74
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
samples using methods detailed below (Fig1). Figure 1 illustrates the simultaneous 75
transfer, mixing, dilution, and plating, of both antiseptics in a paired manner providing 76
the critical timing control necessary for this method. The in vivo studies achieve control 77
of variability by using a dedicated phlebotomy team trained to perform skin antisepsis in 78
an identical manner for both kit types, kits that are identical in appearance except for 79
coded labeling, and finally, phlebotomy and microbiology teams are blinded to the study 80
agents. All studies presented are acceptable under our institutional review processes and 81
as per our Institutional Review Board approved protocol for evaluation of skin antiseptics 82
(LAB01-321, Tarrand PI). 83
Strains. We performed in vitro testing using commercial and laboratory strains: 84
Staphylococcus epidermidis ATCC 12228, ATCC 29212, Escherichia coli ATCC 25922, 85
and Pseudomonas aeruginosa ATCC 27853 (American Type Culture Collection, 86
Manassas, VA), and laboratory strains of coagulase-negative staphylococci A1, B2, C3, 87
and D4 and Acinetobacter baumanni strains 1, 2, and 3. (Microbiology Laboratory, 88
Department of Laboratory Medicine, The University of Texas M. D. Anderson Cancer 89
Center, Houston, TX). 90
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
Reagents. Isopropanol, atomic iodine, dimethyl sulfoxide (DMSO), ethanol, and 91
chlorhexidine gluconate, all as ASC grade, were obtained from Sigma/Aldrich, Inc. 92
(DMSO Product # 472301, MSDS 1.17.12, Sigma/Aldrich, St. Louis, MO). 93
Procedure for in vitro antiseptic evaluation. Microbial organisms were grown 94
overnight at 35°C in three 35-ml blood culture bottles containing Columbia broth 95
(BACTEC Plus aerobic/F medium, Becton Dickinson, Sharpsburg, MD). The bottles 96
were cooled to 4°C, centrifuged at 3000g for 20 min, and the bacterial cells resuspended 97
in sterile 0.9% saline. The bacterial cells were allowed to incubate in saline at room 98
temperature for 24 to 48hr to mimic the low temperature and low nutrition environment 99
of human skin. These ‘aged’ cells presumably also mimic the thick walled sessile cells 100
seen in stationary bacterial cultures The cells were centrifuged as above, resuspended in 101
a minimal volume (~1.5 ml) of PBS or saline, placed in a pipetting trough, and 102
distributed in 100µl aliquots into the first 12 well column of a 96 well ‘U’ bottom plate 103
(Fig. 1). Thus, each well contained 100µl of a concentrated suspension, estimated to 104
contain approximately 1x 1011 bacterial cells per ml based on the original turbidly of 105
Bactec bottle growth (100ml multiplied by approx 109/ml). 50µl of cells from this first 106
column were then picked up using a 12 channel multichannel pipette and transferred to 107
column 2 containing 150µl of the test or control antiseptics (previously loaded into the 108
plate) and mixed immediately (Fig. 1). Six wells contained the control antiseptic and 6 109
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
wells contained the test antiseptic. The bacterial cells were allowed to interact with 110
antiseptic for 30 sec, and then a fresh 12 channel pipette was used to simultaneously 111
transfer antiseptic/bacterial mixtures from column 2 to corresponding dilution wells 112
containing Tryptic Soy Broth (TSB – preloaded into the plate). Three serial dilutions 113
were made rapidly by adding 50µl antiseptic/bacterial mixture to 100µl of TSB broth 114
media to yield 1:3, 1:9, and 1:27 dilutions. This dilution step was done to rapidly stop 115
antiseptic activity. Next a fresh multichannel pipette was again used to simultaneously 116
transfer 50µl samples from each of these dilution wells onto sheep-blood agar plates. The 117
two plates were opened prior to the procedure and positioned closely together to 118
simultaneously allow 6 channels tips to dispense to one plate for control wells, and 6 119
channel tips to dispense to a second plate corresponding to the test antiseptic wells (Fig. 120
1). Plates were spread simultaneously using two sterile ‘hockey-sticks’. Colony count 121
enumeration was performed after 18 to 24hr at 35°C. The starting alcohol concentration 122
was usually 93.3% and the starting DMSO concentration was 6.6% to result in a final 123
concentration of 70% alcohol and 5% DMSO following dilution of the 150µl antiseptic 124
well with the 50µl of bacterial suspension containing approximately 1010 bacteria final 125
per well.“ In some experiments, the isopropanol control antiseptic was supplemented 126
with iodine or chlorhexidine gluconate; this mixture was then tested with or without 127
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
DMSO (Table 1). For in vitro studies, the final antiseptic compositions and 128
concentrations are as stated in Table 1. 129
Clinical Evaluation of Antiseptic Effectiveness. To further demonstrate the 130
effectiveness of the DMSO-containing antiseptic, we performed a clinical validation 131
study using blood culture contamination rates as our measurable endpoint. All subjects 132
were enrolled under LAB01-321, an approved protocol at M. D. Anderson Cancer 133
Center. Patient Demographics are shown in Table 2. We compared the rate of skin flora 134
contamination of aseptically collected blood culture samples derived from patients 135
exposed to either test or standard antiseptic. For this study only coagulase-negative 136
staphylococci (CNS) and catalase positive coryneform bacteria were classified as 137
contaminants and cultures were further required to show less than or equal to 1 CFU/ ml 138
of blood sample. All subjects had blood cultures ordered, and our phlebotomy team 139
routinely disinfected ~50cm2 of antecubital skin using our standard application technique. 140
Samples were randomized 1:1 between kits containing control antiseptic and those 141
containing an experimental antiseptic. All kits were manufactured at the same time and 142
stored at 4°C until use. Exposure to either antiseptic was a two-step process. Step 1 was 143
to expose all antecubital sites to 1ml 60% isopropanol for 30 seconds. In step two, 144
randomized subjects were exposed for 3 minutes to either the standard antiseptic, 145
consisting of 1ml 70% isopropanol/1% iodine, 30% water (IPI) or the test antiseptic, 1ml 146
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
70% isopropanol/1% iodine, 25% water, with DMSO 5% (IPID). Antiseptics were 147
applied only by the Department of Laboratory Medicine’s phlebotomy team, who were 148
blinded to the type of antiseptic kit used. The DMSO containing antiseptic had no 149
difference in odor, color, or other observable difference when compared to the standard 150
antiseptic. The clinical microbiology technologist team, who were also blinded to the 151
type of antiseptic kit, determined the level of sample contamination by using the 152
ISOLATOR 10 (Wampole Laboratories, Cranbury, NJ) lysis-centrifugation blood culture 153
tubes, and by applying previously published criteria (21). 154
Statistical evaluation. Data from in vitro experiments were collected into random 155
blocks (strata) and analyzed using a non-parametric, non-paired, two-tailed, signed-rank 156
test (Dr. Nebiyou Bekele, Department of Biostatistics, M D Anderson). Comparison of 157
continuous data was performed using Student t test. Clinical categorical data were 158
compared using Fishers exact test, two-tailed contingency statistics (Cytel Software Corp. 159
Cambridge, MA.). P values ≤ 0.05 were considered significant (7). 160
161
RESULTS 162
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
Table 1 shows the effect of adding DMSO to isopropanol based antiseptics versus 163
isopronol control mixtures on coagulase negative staphylococci (CNS), including S. 164
epidermidis ATCC strain 12228, and a variety of other bacteria. Inocula, dilutions, 165
alcohol concentrations, and exposure times were all closely paired (Fig. 1). Plating of 166
dilutions proceeded from the lowest dilution to the highest, and again, within a dilution 167
isopropanol and isopropanol/DMSO plates were planted (spread) together. The plates 168
derived following exposure to the DMSO containing antiseptic show reduced numbers of 169
bacterial colonies by our in vitro method at 24hrs when compared to paired control 170
antiseptic plates (Table 1). 171
Table 1 shows that the antiseptics effected E. coli plate counts in a similar manner 172
to CNS; 70% isopropanol/5% DMSO showed lower counts than 70% isopropanol, 173
(P=0.03). Inocula, dilutions, alcohol concentrations, and exposure times were all closely 174
paired. Plate counts at 24 hr indicate that more bacteria were killed by the solution 175
containing DMSO. Similar enhancement of killing was seen for Pseudomonas 176
aeruginosa. E. coli, Pseudomonas aeruginosa, Acenitobacter baumanni were tested 177
since all are associated with nosocomial to skin-skin transmission. Additional studies are 178
presented in figure 2 showing a DMSO effect using matched plates of S. epidermidis 179
exposed to different antiseptics; [50% ethanol solutions containing 1% chlorhexidine 180
gluconate, +/- 4% DMSO] or [50% ethanol containing 0.6% Brij-35, +/- 4% DMSO]. 181
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
Although most of our data concerns alcohol based antiseptics, the effect of polar- 182
aprotic solvents is also seen in water based solutions as well. When 18% DMSO vs. 183
18% water was added to 10% povidone iodine, enhanced killing activity was seen; 18% 184
DMSO vs. 18% water = 2 vs. 273 CFU, (6% DMSO vs. 6% water gave of 26 vs. 93 185
CFU). 186
Fig. 3A shows the effectiveness of 70% isopropanol with the addition of various 187
concentrations of DMSO. The addition of DMSO to isopropanol significantly increased 188
antiseptic killing of S. epidermidis ATCC 12228, even at the lowest concentration tested 189
(2.8% DMSO; t test P<0.001). However, DMSO in PBS without isopropanol had no 190
antibacterial activity; 20% DMSO + Phosphate buffered saline pH 7.4 resulted in mean 191
250.5 CFU, SD 44.7, n4 vs. 20% Water + PBS (211.5, SD13.3, n 4; not significant). 192
Fig. 3B, shows that when DMSO concentration were held constant at 5%, then increasing 193
concentrations of isopropanol increased the activity of the alcohol-DMSO mixtures 194
substantially; (P<0.0001). DMSO had no detectable effect at isopropanol concentrations 195
of 40% or below in this series of experiments. 196
Finally, in the clinical validation portion of this study, we enrolled 1,590 197
antiseptic/blood-culture events and evaluated the efficacy of the control vs. DMSO-198
containing antiseptics. Enhanced antiseptic activity was seen when 5% DMSO, 70% 199
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
isopropanol, 1% iodine (DIPI) was compared to 70% isopropanol, 1% iodine (IPI). A 200
66% reduction in contamination was observed with DIPI vs IPI; 1.04% vs. 3.04%, 201
(Fisher exact, unpaired, two tailed P< 0.01), (Table 3). The control antiseptic showed 23 202
coagulase negative staphylococci (CNS) and 2 coryneform bacteria, and the test 203
antiseptic showed 3 CNS and 5 coryneform bacteria. On average each patient had 2.5 204
samples colleted during the study period, and no patient had more than one contaminant 205
detected during the study time period. Toxicity was monitored using an incident report 206
mechanism driven by the clinical team, phlebotomists, and patient reports. No time 207
constraints were made on the incident reports. No reports of rash, irritation, redness, or 208
other incident reports were made in association with the study patients. 209
210
DISCUSSION 211
The results of our in vitro and in vivo antiseptic studies showed that a new alcohol 212
based formulation containing DMSO resulted in increased antiseptic effectiveness. 213
Further, this study demonstrated a useful dilutional screening method for antiseptic 214
evaluation. 215
Preventing inappropriate treatment and increased length of hospital stay have 216
become a national quality goals (6, 8, 9, 17), in part because of data showing that the cost 217
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
of potentially preventable wound infections exceeds 3 billion annually in the United 218
States alone (10, 12, 13, 16, 24, 26, 29). 219
Dimethyl sulfoxide has been previously used for drug delivery in topical 220
therapeutic compositions. High concentration (50% or greater) DMSO are necessary in 221
this setting, where DMSO facilitates the gradual absorption of drugs such a nitroglycerin 222
directly through the dermis (28). We have used DMSO previously as an enhancement to 223
standard oxidase testing (22). However, in the current paper we show surprising 224
improvements in in vitro and in vivo antisepsis using only low concentrations of DMSO. 225
The mechanism is not entirely clear. Aqueous channels (pours) have been shown to 226
result from the swelling and increased mobility of phospholipid head-groups in lipid bi-227
layers using 27% DMSO (19). We hypothesize that this action of DMSO enhances 228
access of active agents (alcohol, iodine, chlorhexidine) into critical bacterial cell 229
structures such as bacterial pores. 230
Skin bacteria appear to exist in a sessile state, with low energy charge, possibly 231
due to low temperature, low water activity, and low nutrition availability to bacterial cells 232
(15, 25). Twenty four hr bacterial cell aging in vitro seemed to mimic the behavior of 233
bacteria from directly scraped skin cells exposed to test or control antiseptic in our 234
studies. The 24hr aging time period is entirely arbitrary, nonetheless, the in vitro 235
screening model using aged bacteria had the best agreement with skin scraping as well as 236
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
clinical validation findings. Finally, it should be emphasized that the killing rates with 237
modern alcohol based antiseptics are very rapid indeed and careful timing and pair-wise 238
sampling are critical to control variability in this model. We are not aware of other in 239
vitro models suitable for antiseptic screening. 240
In general, iodine or other antiseptic-adjuvant containing combinations are 241
superior to alcohol antiseptics alone (18). All of the antiseptics that we tested (alcohol, 242
alcohol/iodine, alcohol/chlorhexidine, alcohol/brij-35, and water/povidone iodine) have 243
shown a proportional enhancement of killing with the addition of small amounts of 244
DMSO. The addition of DMSO to water based povidone iodine was also superior to 245
povidone alone. This is important since water based antiseptics have a lower fire risk and 246
are used extensively in the surgical setting. Other polar aprotic solvents such as 247
dimethylacetimide showed a moderate to weak enhancing effect; however, this agent has 248
additional toxicity concerns. 249
In our clinical validation involving 1,590 antiseptic/blood-culture procedures we 250
found a 1.04% contamination rate for DMSO containing tincture of iodine. This is below 251
typical blood culture contamination rates (20, 23, 24, 29), and 1/3 of the rate seen with 252
the standard iodine tincture in our validation trial. Interestingly, this rate although low, 253
does not approach the 100 fold reduction seen in the in vitro model. Perhaps the model is 254
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
failing is some way, or perhaps this may relate to bacteria hidden in sebaceous glands, 255
sweat glands, hair follicles, or sequestered in lecuna of the stratum corneum, suggesting 256
an irreducible level of contamination for living skin. We performed earlier studies that 257
showed similar trends but had to be stopped due to slow accrual (Iodine / DMSO stopped 258
with 52 samples at 3 years: IPI contaminants 5 in 30, DIPI contaminants 0 in 22; 259
Chlorhexidine / DMSO stopped with 365 samples at 2 years: IPC contaminants 4 in 175, 260
DIPC contaminants 1 in 190 patients). 261
Rapid in vitro screening may facilitate further antiseptic development. Here we 262
demonstrate that the inclusion of small amounts of the polar-aprotic solvent DMSO can 263
improve the effectiveness of several currently used skin antiseptics. A new class of 264
antiseptics based on inclusion of polar-aprotic solvents may offer general improvements 265
in skin antisepsis, including lower rates of wound infection, catheter infection, and blood 266
culture contamination, and nosocomial infections derived from health care worker hands. 267
268
269
270
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
Acknowledgment: The authors wish to thank Dr. Nebiyou Bekele, Department of 271 Biostatistics, M D Anderson, for assistance with our statistical analysis. 272 We also thank Kimberly J. Herrick in the Department of Scientific Publications. 273
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
References: 274 275
1. Bates D.W., L. Goldman, and T.H. Lee. 1991. Contaminant blood cultures 276 and resource utilization: The true consequences of false-positive results. 277 JAMA 265:365-369. 278
279 2. Bekeris L.G., Tworek J.A., Walsh M.K., Valenstein P.N. 2005. Trends in 280
blood culture contamination: a College of American Pathologists Q-Tracks 281 study of 356 institutions. Arch. Pathol. Lab. Med. 129:122201225. 282
283 3. Boyce, J.M. 2004. New insights into improving hand hygiene practices. 284
Infection Control & Hospital Epidemiology. 25:187 -8. 285 286 4. Boyce, J.M. 2001. Antiseptic technology: access, affordability, and 287
acceptance. Emerging Infectious Diseases. 7:231-3. 288 289 5. Boyce, J.M., D. Pittet, Healthcare Infection Control Practices Advisory 290
Committee Hand Hygiene Task Force. 2002 Guideline for hand hygiene in 291 Health-Care Settings. Recommendations of the Healthcare Infection Control 292 Practices Advisory Committee and the HIPAC/SHES/APIC/IDSA Hand 293 Hygiene Task Force. American Journal of Infection Control. 30:S1-46. 294
295 6. Centers for Medicare and Medicaid Services. 2002. Quality of care 296
measure specifications (Statement 7) :Surgical infection prevention. Centers 297 for Medicare and Medicaid Services. Baltimore, Maryland. 298 (www.JCAHO.org) 299
300 7. Coltron, T.. 1974. Statistics in Medicine, Little Brown & Co., Boston, 301
Massachusetts. 302 303 8. Cruse, P.J.E., R. Foord. 1980. The epidemiology of wound infection: a 10 304
year prospective study of 62,939 wounds. Surgical Clin North Am. 60:27-40. 305 306 9. Culver, D.H., T.C. Horan, R.P. Gaynes. 1991. Surgical wound infection 307
rates by wound class, operative procedure, and patient risk index. National 308 Nosocomial infections Survaeillance System. Am J Med. 91:S3B,152-7. 309
310 10. Darouiche, R.O., M.J. Wall, K.M. Itani, M.F. Otterson, A.L. Webb, M.M. 311
Carrik, H.J. Miller, S.S. Awad, C.T. Crosby, M.C. Mosier, A. Alsharif, 312 D.H. Berger. 2010. Chlorhexidine-alcohol versus povidone-iodine for 313 surgical-site antisepsis. N Engl J Med. 362:18-26. 314
315 11. Dhillon R.H., Clark J., Azadian B.S. 2009. Reducing blood culture 316
contamination. J. Hosp. Infect. 73:97-99. 317 318
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
12. DiPiro, J.T., R.G. Martindale, A. Bakst, P.F. Vacani, P. Watson, M.T. 319 Miller. 1998. Infections in surgical patients: effects on mortality, hospital stay 320 and postdischarge care. American Journal Health System Pharmacology. 321 55:777-81. 322
323 13. Fry, DE. 2002. The economic costs of surgical site infections. Surgical 324
Infections. 3:S37-43. 325 326 14. Gander R.M., Byrd L., DeCrescenzo M., Hirany S., Bowen M., 327
Baughman J. 2009. Impact of blood cultures drawn by phlebotomy on 328 contamination rates and health care costs in a hospital emergency department. 329 J Clin Microbiol. 47:1021-1024. 330
331 15. Gessey G.G., White D.C. 1990. Determination of bacterial growth and 332
activity at Solid-Liquid interfaces. Ann Rev Microbiol. 44:579-602. 333 334 16. Green, J.W., R.P. Wensel. 1997. Postoperative wound infection. A 335
controlled study of the increased duration of hospital stay and direct cost of 336 hospitalization. Ann Surg. 185:264-8. 337
338 17. Hall, M.J., M.F. Owings. 2002. National Hospital Discharge Survey. CDC 339
Advance 329: 1-19. 340 341 18. Joklik, W.J., Willett, H.P. 1992. Sterilization and Disinfection p. 188-199. 342
W. J. Joklik, H.P. Willett, D.B. Amos, and C.M. Wilfert. (ed.)., In Zinsser 343 Microbiology, 20th ed. Appleton & Lange, Norwalk, Connecticut. 344
345 19. Notman R., Noro M., O’Malley B., Anwar J. 2006. Molecular Basis for 346
Dimethylsulfoxide (DMSO) antion on lipid membranes. J Am Chem Soc. 347 128:13982-13982. 348
349 20. Strand, C.L. and Shulman J.A. Bloodstream Infections: Laboratory 350
Detection and Clinical Considerations. ASCP Press. Chicago. 1988. 351 352 21. Tarrand, J.J., C. Guillot, M. Wenglar, J. Jackson, J.D. Lajeunesse, K.V. 353
Rolston. 1991. Clinical comparison of the resin-containing Bactec 26 Plus 354 and the Isolator 10 blood culturing systems. Journal of Clinical Microbiology. 355 29:2245-9. 356
357 22. Tarrand JJ, Groschel DHM. Rapid modified oxidase test for oxidase-358
variable bacterial isolates. J Clin Microbiol 16:4, 772-774, 1982. 359 360 23. Trautner, B.W., J.E. Clarridge, R.O. Darovich. 2002. Skin antisepsis kits 361
containing alcohol and chlorhexidine gluconate or tincture of iodine are 362 associated with low rates of blood culture contamination. Infection Control & 363 Hospital Epidemiology. 25:397-401. 364
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
365 24. Urban, J.A. 2006. Cost analysis of surgical site infections. Surgical 366
Infections. 7:S19-22. 367 368 25. Vlassova N., Han A., Zenilman J.M., James G., Lazarus G.S. 2011. New 369
horizons for cutanious microbiology: the role of biofilms in dermatological 370 disease. Brithsh J. Dermatol. 165:756-759. 371
372 26. Weinstein, M., and Carroll, K.C. 2007. Manual and automated systems for 373
detection and identification of microorganisms. p. 194. In P. R. Murray, E. J. 374 Baron, M. A.. Phaller, J. H. Jorgensen, and R. H. Yolken (ed.). Manual of 375 Clinical Microbiology, 9th ed. ASM Press. Washington, DC. 376
377 27. Weinstein M.P. 2003. Blood culture contamination: persisting problems and 378
partial progress. J. Clin. Microbiol. 41:2275-2278. 379 380 28. Willson, J.E., D.E. Brown, and E.K. Timmens. 1965. A toxicologic study 381
of dimethyl sulfoxide. Toxicology Applied Pharmacol. 7:104-112. 382 383 29. Zoutman, D., S. McDonald, D. Velhanayagan. 1998. Total and attributable 384
costs of surgical wound infections at a Canadian tertiary-care center. 385 Infection Control & Hospital Epidemiology. 19:254-9. 386
387 388
389 390 391
392
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
393 394 395 396
397 TABLE 1. DMSO effect on Antiseptic activity 398 Organism I# IP - CFU IPD - CFU S. epi∞ 12228* - 2360 76 1730 58 S. epi 12228* - 400 5 384 3 CNS A1 - 250 30 CNS B2 - 15 0 CNS C3 - 200 0 CNS D4 - 147 10 S. epi 12228# + 313 1 399 5 S. epi 12228# + 146 1 S. epi 12228# + 116 0 S. epi 12228# + 51 0 A. baumannii1 - 1500 8 A. baumannii2 - 47 2 A. baumannii3 - ~10,000
~10,000 188 168
P. aeruginosa - 190 80
8 1
P. aeruginosa - 188 248
0 2
E. coli - 192 128
0 0
E. coli - 53 43
6 3
E. coli - 1248 1174
0 0
IP = 70% isopropyl alcohol solution, IPD = 70% isopropyl 399
alcohol + 4% DMSO solution. CFU indicates Colony Forming 400
Unites. ∞Indicates S. epidermidis. *Indicates experiments 401
performed using 50% ethanol with or without 4% DMSO. 402 #Indicates isopropyl + 2% iodine with or without 4% DMSO. 403
Plates were counted at 24hr. Overall signed rank P=<0.0001. 404
405
406
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
407
TABLE 2. Patient Demographics 408 Antiseptic IPI* DIPI** 409 Number of Patients 331 326 410 Number of Samples 822 768 411 Average Patient age (SD) 55.3 (14.1) 55.2(14.7) 412 Male % 56.5 59.8 413 Leukemia % 38.6 42.6 414 Lymphoma / Myeloma % 19.8 16.7 415 Solid Tumor % 41.6 40.7 416 *IPI= 70% Isopropanol, 1% iodine. **DIPI= 5% DMSO, 70% Isopropanol, 1% iodine. 417 418
419 420 421
422
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
423 424 425 TABLE 3. Control vs. DMSO containing antiseptic applied prior to blood culture. 426
Antiseptic IPI DIPI* 427
Contaminated blood cultures 25 8 428
Number of samples tested 822 768 429
IPI= 70% Isopropanol, 1% iodine. DIPI= 5% DMSO, 70% Isopropanol, 1% iodine. *P < 430
0.01, Fisher exact test. 431
432 433 434 435 436 437
438 439 440 441
442 443 444 445 446
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
447 448 449 450
451 452 Aaa (Fig. 1. Legend) 453 454 Fig 1. Method illustration. Samples of concentrated microorganism (12 wells) are 455
simultaneously transferred using a multi-channel pipettor into test and control antiseptics. 456 These wells are then further diluted in tryptic-soy broth before planting for colony 457 counting; 458
459 460 461 462
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
463 (Fig. 1) 464 465
466 467
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488
(Fig. 2. Legend) 489 490 Fig. 2. Top - S. epidermidis killing in 50% ethanol, 1% chlorhexidine gluconate (EC) compared to 491
50% ethanol, 1% chlorhexidine gluconate and 4% DMSO (ECD). Bottom - S. epidermidis killing in 492 70% isopropanol, 0.6% Brig-35 (IB) compared to 70% isopropanol, 0.6% Brig-35, and 5% DMSO 493 (IBD). Paired plates shown at the1:3 dilution. Starting bacterial cell well contains ~1011 CFU/ml. 494
495 496 497 498 499 500 501 502
503 504 505 506
507
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
508 509 510 511 512 (Fig. 2) 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527
528
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
529 530 531 532
533 534 535 536 537 538 (Fig. 3. Legend) 539 540
Fig. 2. Effect of DMSO concentrations on the activity of 70% isopropanol (A). Effect of 541 various isopropanol concentrations on antiseptic activity with 5% DMSO or water (B). S. 542 epidermidis ATCC 12228 was used as the test organism. Note: 10,000 CFU is an estimate 543 based on comparison to a dilution series of know standards. 544
545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 Fi F 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
575 576 577 578 579
(Fig. 3) 580 581 582
583
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from
1/3 1/9 1/27 1/3 1/9
Column # 1 2 3 4 5
1/27
Test vs control antiseptic
- Serial dilution wells (TSB broth)
50 µL dilution plates
- Microorganism suspension
- Test vs. control antiseptic 6 wells each, (30 seconds)
on March 27, 2019 by guest
http://jcm.asm
.org/D
ownloaded from