- Email: [email protected]
Influence of two handwashing frequencies on reduction in colonizing flora with three handwashing products used by health care personnel
Influence of two handwashing fr ties on reduction in colonizing flura with three handwashing products used by health care personnel Elaine Larson, RN, PhD, FAAN Kumudlni Mayur, PhD Barbara A. Laughon, PhD Baltimore,
Maryland
Four handwashing products (containing either 2% chlorhexidine gluconate, 0.6% parachlorometaxylenol, 0.3% triclosan, or a nonantimicrobial control) at two handwashing frequencies (6 or 18 times/day) were compared with regard to their effectiveness in reducing colonizing hand flora. Eighty adult volunteers were assigned by block randomization to one of the four products and one of the two frequency schedules (n = lo/group) and washed their hands under supervision for 5 consecutive days. There were no significant differences between products in mean log,, colony-forming units after the initial wash (p = 0.61), nor were there significant differences in products after 5 days among subjects washing six times per day. For those who washed 18 times per day, however, the effectiveness of all three antimicrobial soaps was significantly better than that of the control soap (p < 0.05). Chlorhexidine gluconate produced significantly greater reductions than triclosan or parachlorometaxylenol (p < 0.05). which were not significantly different from each other. On the basis of these findings an antimicrobial soap is recommended when handwashing frequency is high and a long-term reduction in colonizing flora is desirable. When handwashing frequency is low (6 times/day), there seems to be less advantage of one product over another, although the use of chlorhexidine gluconate resulted in greater reductions at both high and low handwashing frequencies. (AM J INFECT CONTROL 1988;17:83-8)
Three distinct types of hand-wash products are used by health care professionals. Each product type has its own specific requirements and applications. The first is plain soap, which is used for removal of dirt and many transient From the Johns Hopkins University School of Nursing and the Division of Infectious Diseases, Johns Hopkins University School of Medicine. Supported in part by Huntington Labs, Inc., Huntington, Indiana. Reprint requests: Elaine Larson, RN, PhD, FAAN, Nutting Chair in Clinical Nursing, Johns Hopkins School of Nursing, 600 North Wolfe St., Baltimore, MD 21205.
contaminants. The second is the antisepticcontaining health care personnel hand wash (HCPHW). Repeated 15 to 30-second hand washes with such a product should produce significant reductions in transient flora after artificial contamination of the hands and forearms. The repeated use of a HCPHW also will have some reductive effect on the resident or colonizing flora of the hands. The third type of handwash product is the antiseptic-containing surgical scrub. Scrubbing duration usually is approximately 5 minutes. Repeated use of a surgical scrub should produce at least a 2-log reduction in resident flora. 83
American
84
Journai
cd
Larson, Mayur, and Laughon
Two important characteristics of antiseptic ingredients are their ability to effect a rapid reduction in microbial counts from the hands and to bind to stratum corneum, resulting in a sustained release of the agent. Some antiseptics such as the alcohols are rapid acting and are agents of choice when the goal is an immediate reduction in transient and colonizing flora.‘” The alcohols, however, have several disadvantages. First, when used frequently over prolonged periods of time as the primary handwashing product, they remove lipid from the skin, causing dryness. Newer products compensate for that problem by the addition of emollients to the alcohol base. Second, they have little or no residual effect on the skin. Third, they may be inactivated in the presence of organic matter.4 Other antiseptic ingredients such as chlorhexidine gluconate (CHG), parachlorometaxylenol (PCMX), and triclosan (TRI), although slower acting than alcohols, bind to the stratum corneum, resulting in a sustained chemical release of the agent. In the presence of this persistent effect, one can expect to see a continued reduction in the numbers of both transient and resident organisms on the skin with prolonged use of the product over time.5, 6 Such substantivity is desirable when health care personnel work in clinical settings where patients are at high risk for infection and where a sustained reduction in numbers of organisms on the hands of personnel may be desirable (e.g., surgery and neonatal or critical care areas). Health care personnel hand wash products that contain CHG, PCMX, or TRI are available in a number of commercial formulations. The purposes of this study were to compare the antimicrobial effects of (1) three such products (containing CHG, PCMX, or TRI) on the normal resident flora after a 30-second handwash procedure over a period of 5 days and (2) each of these products at two different handwashing frequencies (6 or 18 washes/day) over a period of 5 days. METHODS Protocol Eighty healthy adult nonmedical volunteers were selected who were not receiving systemic
or topical antibiotics and who had no history of skin disease nor sensitivity to soaps. They were assigned by block randomization to one of four handwashing products, a liquid dctergent base containing (1) 2.0% CHG (CidaStat; Huntington Laboratories, Inc., Huntington, Ind.), (2) 0.6% PCMX (Medi-Scrub; Huntington), (3) 0.3% TRI (Bacti-Stat; Huntington), or (4) no antiseptic ingredient, called con& soap (Wash; Huntington). Three days before the beginning of the testing period, all subjects were provided with control soap for general bathing and handwashing and with disposable vinyl gloves for use during shampooing, dishwashing, or hand contact with other soaps, lotions, or cleaning products. They were instructed to use only this soap throughout a 3-day weaning period and the entire test period. The wearing of rings and nail polish and the use of hand lotions were prohibited during the study. After assignment to a treatment group each subject was instructed in a standardized handwashing procedure and practiced this technique in the presence of an investigator until competence was demonstrated. The standardized handwashing technique consisted of the following procedure. The subject rinsed the hands under warm running water and applied 3 ml of soap (premeasured in a syringe) into the palm. With the USC of a stop watch the subject vigorously rubbed all surfaces of the hands with the test product for 30 seconds, paying particular attention to the interdigital spaces and fingernails. During the wash period, the hands were kept out of running water to avoid rinsing off the test product. Hands were rinsed thoroughly with running water and dried with a paper towel. Handwashes were distributed throughout each day, with approximately half done in the morning and half in the afternoon. Washes were performed in the laboratory under the supervision of a laboratory assistant or an investigator for 5 consecutive days. Half the subjects (10 in each test product group) washed 6 times per day, and half washed 18 times per day. Microorganisms on the hand were harvested with the sterile bag technique. Each subject inserted the dominant hand into a sterile poly-
Volume
17 Number
2
Handwashing
April 1989
Table 1. Mean loglo counts (k standard
deviation)
by product Tricoslan
control
Baseline, total group Baseline, 6 wash/day Baseline, 18 wash/day After first wash, day 1, total group After final wash, day 5 6 wash/day 18 wash/day Log change After first wash, day 1, total group After final wash, day 5 6 wash/day 18 wash/day Percentage change* After first wash After final wash 6 wash/day 18 wash/day ‘Baseline
- Test count Baseline
x 100 (actual
5.73 5.71 5.75 5.81
and washing
(0.56) (0.68) (0.45) (0.59)
6.14 6.17 6.11 6.13
6.31 6.13 6.48 6.01
(0.84) (0.83) (0.86) (0.59)
Chlorhexidina gluconate
6.09 6.16 6.01 5.81
(0.61) (0.71) (0.52) (0.81)
5.25 (0.65) 5.45 (0.56)
5.42 (0.62) 4.56 (0.50)
5.48 (0.44) 4.74 (0.72)
5.03 (1.12) 3.62 (0.74)
+ 0.08
- 0.01
- 0.30
- 0.28
- 0.46 - 0.30
- 0.75 - 1.55
- 0.65 - 1.74
- 1.13 - 2.39
+ 20.2
- 2.3
- 49.9
- 47.5
- 65.3 - 49.9
- 82.2 - 97.2
- 77.6 - 98.2
- 92.6 - 99.6
rather than log counts
used)
counts
ethylene bag (Steri-Lok; 3M Co., St. Paul, Minn.) containing 50 ml sterile sampling solution (distilled water containing, per liter, lecithin, 20 gm; sodium thiosulfate, 6 gm; sodium oleate, 6 gm; protease peptone, 1 gm; tryptone, 1 gm; and Tween 80, 50 ml, pH: 7.2-7.4). The sampling solution contained neutralizers for the three antiseptic products7* 8 and in preliminary studies there was no significant increase nor decrease in CFUs in the solution within the first 2 hours after sampling. The entire hand surface was rubbed vigorously through the wall of the bag for 1 minute. All specimens were plated within 2 hours of sampling. A 0.2 ml volume of serial dilutions up to 1O-3 was placed on trypticase soy agar, which contained yeast extract, 5 gm/L, and Tween 80, 1 ml/L (TYT), and on MacConkey agar (BBL Microbiology Systems, Cockeysville, Md.). The TYT plates were incubated at 35” C for 48 hours followed by incubation at room temperature for an additional 48 hours. The MacConkey agar plates were incubated at 35” C for 24 hours. Colonyforming units were counted and calculated to provide a total CFU count per hand. Gramnegative bacteria were not identified to species
85
frequency
Parachlorometaxylenol
(0.76) (0.61) (0.91) (0.51)
frequency
level. Microbiologic sampling of the hands was performed at three times: immediately before the study after a brief wash with control soap (baseline), after the first wash on day 1, and after the last wash on day 5. Statistical
analysis
Analysis of covariance, that is, controlling for baseline counts, was used to test the significance of differences in microbial counts (counted as log,, CFU) among the four product groups and between the two frequencies within each product group. When multiple comparisons between groups were completed, the Tukey technique was used? A probability of p < 0.05 was considered to be statistically significant. With a sample size of 20 per product group a 75% difference in changes in log,, CFU could be detected between groups with a power of 8O%.‘O RESULTS
All 80 subjects completed the entire 5 days of testing. There were 60 women (75%) and 20 men, with a mean age of 3 1.9 years (range: 1855 years). There were no significant differences
kmefcan
86
Larson,Mayur,andLaughon
0
PCMX
CHG
INFECTIGK
TRI
Fig. 1. Log reductions In colonizing flora washing product and frequency, day 5.
CXJNT by hand-
in gender or age distribution between the product or frequency groups. Baseline mean log counts for each group ranged from 5.7 1 ( + 0.68 standard deviation) to 6.48 (kO.86) (Table 1). After the initial 30second wash, mean counts in the control group increased 0.08 logs and mean counts for subjects using one of the three antiseptics (TRI, PMCX, or CHG) decreased 0.01, 0.30, and 0.28 logs, respectively. There were no significant differences among the four test products (p = 0.61). Again, after six washes per day for 5 days there were no significant differences among products (p > OSO), although subjects using CHG had the greatest reductions. After 18 washes per day for 5 days, the control group mean count decreased 0.30 logs, and mean counts for TRI, PCMX, and CHG decreased 1.55, 1.74, and 2.39 logs, respectively (Fig. 1). All three antiseptics performed significantly better than control soap (p < 0.05). In addition, CHG produced significantly greater reductions than TRI or PCMX (p < 0.05); PCMX and TRI were not significantly different from each other (p > 0.50). Many subjects (51180, 64%) had gramnegative bacteria isolated from their hands at baseline, and subjects in all four product groups had substantial reductions in carriage of gramnegative bacteria by the end of the test period (Table 2). When gram-negative bacteria were present, they were also in greater numbers at baseline (baseline: mean log CFU = 3.95, range 2.9-4.8; end of test period; mean log CFU = 3.26, range 2.4-3.9).
Journal
of
CGNTRGIL
Although plain soap performs well for removing dirt and some transient contaminants from hands, it has little effect on colonizing flora.“,” It is not surprising, then, that in this study subjects using control soap had little change in log counts of normal flora after 5 days of handwashing. Health care personnel hand wash products, on the other hand, are chosen to minimize the numbers of transient contaminating flora and to aid in reducing the colonizing flora by means of their chemical bactericidal activity after a 15- to 30-second handwash. A high prevalence of gram-negative bacteria on hands of nonmedical personnel is not surprising and has been previously reported.13, l4 The dramatic reductions in gram-negative bacteria among all test subjects indicated that the hands were clean and relatively free of transient flora. Surgical scrub products containing 4% CHG frequently are used in the United States for health care personnel handwashing, surgical hand scrub, and occasionally for preoperative bathing or preparation of patient skin A 2% concentration of CHG has been shown to be slightly lower in antimicrobial effectiveness than 4% concentrations of CHG.“. I6 Several possible reasons explain the improved performance of CHG over that of the other test products. First, its concentration (2%) was greater than the HCPHW concentrations of either PCMX (0.6%) or TRI (0.3%). Perhaps the two latter agents at higher surgical scrub concentrations would have more antimicrobial activity. Indeed, others have used higher concentrations of each of these ingredients in comparative tests.*. ” PCMX, for example, is approved for topical use by the Food and Drug Administration for concentrations up to 3.75% and TRI up to 1 .O%. Second, CHG has excellent substantivity, binding to the stratum corneum to effect prolonged release of chemical activity on the skin. TRI also has been shown to have this same characteristic, whereas the persistence on skin of PCMX may be less ?s‘, ’ Third, CHG may give greater antimicrobial activity than the other two antiseptic ingredients tested. A recent comparison of 1.5% TRI and 4% CHG with the use of an HCPHW protocol showed that the two products were equally effective in removing artificially inoculated Serratia marcescens. How-
VOlUme
17 tvumber
2
Handwashing
April 1989
frequency
87
Table 2. Percentage (number) of subjects with gram-negative bacteria recovered from hands Baseline Product Parachlorometaxylenol Chlorhexidine gluconate Triclosan Control Total Chi-square ‘Percent
to test differences in reductions between Baseline - final x ,oo, = Baseline
FiMl
per-ntegs
Pr-
55 (11/20) 60 (12120) 80 (16/20) 60 (12/20) 63.8 @l/80) groups,
10 15 15 10
12.5
PWCWI redUctlOn’
( 2/20) ( 3/20) ( 3/20) ( 2/20) (10/80)
81.8 75.0 81.3 83.3
p = 0.86
reduction
ever, the effectiveness of CHG was consistently, although not statistically, significantly better after 4,7, and 10 washes.‘* The immediate effect (after a single 30-second wash) of products studied was not much different from that of plain, nonantimicrobial soap. It is clear that a 30second HCPHW with such products does not meet surgical hand scrub criteria. Others have reported a similar result for removal of transient as well as colonizing flora.18 When the most rapid reductions in flora are needed, a surgical scrub product and a surgical scrub procedure are required. Products containing CHG, PCMX, or TRI at HCPHW levels are most effective when used over a sustained period of time. For this reason such products are appropriate for health care personnel handwashing when a reduction in colonizing flora is desirable and when handwashing frequency is high, certainly more than six times per work period. It also would appear that in situations when frequency of handwashing is very low, there is less advantage of one product over another and probably little advantage of an antiseptic over a plain soap. In most clinical situations, however, frequent handwashing is common and the antimicrobial-containing products, particularly CHG, result in a significantly greater reduction in colonizing flora as compared with plain soap. This study examined the influence of two variables on the quality of handwashing: frequency of washing and product used. Both of these variables were clearly important. Other variables not examined in this protocol, inasmuch as the procedure was standardized to prevent confounding, include such factors as the duration
of handwashing and the amount of soap used. Many of these factors probably have independent and combined influences on the effectiveness of handwashing.lg Refemnces 1. Ayliffe GAJ. Surgical scrub and skin disinfection. Infect Control 1984;5:18-22. 2. Rotter ML. Hygiene hand disinfection. Infect Control 1984;5:18-22. 3. Larson EL, Eke PI, Laughon BE. Efficacy of alcoholbased hand rinses under frequent-use conditions. Antimicrob Agents Chemother 1986;30:542-4. 4. Morton ME. Alcohols. In: Block SS, ed. Disinfection, sterilization and preservation. 3rd ed. Philadelphia: Lea & Febiger, 1983:225-39. 5. Bruch M. Newer germicides: what they offer. In: Maibath HI, Aly R, eds. Skin microbiology: relevance to clinical infection. New York: Springer-Verlag, 1981: 103-12. 6. Larson E, Leyden J, McGinley K, Grove G, Talbot GH. Physiologic and microbiologic changes in skin related to frequent handwashing. Infect Control 1986;7:59-63. 7. Russell AD, Ahonkhai I, Rogers DT. Microbiological applications of the inactivation of antibiotics and other antimicrobial agents. J Appl Bacterial 1979;46: 207-45. 8.
9.
10. 11. 12. 13. 14. 15.
Bartzokas CA, Corkill JE, Makin T, Pinder DC. Assessment of the remanent antibacterial effect of a 2% triclosan-detergent preparation on the skin. J Hyg 1983;91:521-8. SPSS, Inc. SPSS-X User’s Guide. 2nd ed. New York: McGraw-Hill, 1984:451-64, 831-4. Lachin JM. Introduction to sample size determination and power analysis for clinical trials. Controlled Clin Trials 1981;2:93-113. Sprunt K, Redman W, Leidy G. Antibacterial effectiveness of routine handwashing. Pediatrics 1973;52:264-9. Reybrouck G. Handwashing and hand disinfection. J Hosp Infect 1986;8:15-23. Larson E. Persistent carriage of gram-negative bacteria on hands. AM J INFEST CONTROL 1981;9:112-9. Horn WA, Larson EL, McGinley KJ, Leyden JJ. Microbial flora on the hands of health care personnel. Infect Control Hosp Epidemiol 1988;9:189-93. Larson EL, Laughon BE. Comparison of four antiseptics
American
88
Larson,
Mayur,
and Laughon
INFECTION
containing chlorhexidine gluconate. Antimicrob Agents Chemother 1987;31:1572-4. 16. Mizuba S, Sheikh W. Microbiological evaluation of a new 2% aqueous chlorhexidine gluconate patient skin preparation on the axilla and abdomen. Curr Therap Res 1986;40:364-368. 17. Soulsby ME, Barnett JB, Maddox S. Antiseptic efficacy of chloxylenol-containing versus chlorhexidine gluconate-containing surgical scrub preparations. Infect Control 1986;7:223-6.
Bound vdumes
avnihble
Journai
of
CONTROI.
18. Bartzokas CA, Corkill JE, Makin T. Evaluation of the skin disinfecting activity and cumulative effect of chlorhexidine and triclosan handwash preparations on hands artificially contaminated with Sewatia mwcestens. Infect Control 1987;8:163-7. 19. Larson EL, Eke PI, Wilder MP, Laughon BE. Quantity of soap as a variable in handwashing. Infect Control 1987;8:371-5.
to sulrsuibus
Bound volumes of AMERICAN JOURNAL OF INFECTION CONTROL are available to subscribers (only) for the 1989 issues from the Publisher, at a cost of $19.00 ($25.00 international) for Vol. 17 (January-December). Shipping charges are included. Each bound volume contains a subject and an author index. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accmzpany all orders. Contact The C.V. h4osby Company, Circulation Department, 11830 Wes&ie Industrial Drive, St. Louis, Missouri 63146-3318, USA; phone 800-325-4177, ext. 351. Subsc&ption8 mu6t be ia force to quaitfy. Bosnd vdumcs are not av bm ofa~jownaIshtcrlption.
Maryland
Four handwashing products (containing either 2% chlorhexidine gluconate, 0.6% parachlorometaxylenol, 0.3% triclosan, or a nonantimicrobial control) at two handwashing frequencies (6 or 18 times/day) were compared with regard to their effectiveness in reducing colonizing hand flora. Eighty adult volunteers were assigned by block randomization to one of the four products and one of the two frequency schedules (n = lo/group) and washed their hands under supervision for 5 consecutive days. There were no significant differences between products in mean log,, colony-forming units after the initial wash (p = 0.61), nor were there significant differences in products after 5 days among subjects washing six times per day. For those who washed 18 times per day, however, the effectiveness of all three antimicrobial soaps was significantly better than that of the control soap (p < 0.05). Chlorhexidine gluconate produced significantly greater reductions than triclosan or parachlorometaxylenol (p < 0.05). which were not significantly different from each other. On the basis of these findings an antimicrobial soap is recommended when handwashing frequency is high and a long-term reduction in colonizing flora is desirable. When handwashing frequency is low (6 times/day), there seems to be less advantage of one product over another, although the use of chlorhexidine gluconate resulted in greater reductions at both high and low handwashing frequencies. (AM J INFECT CONTROL 1988;17:83-8)
Three distinct types of hand-wash products are used by health care professionals. Each product type has its own specific requirements and applications. The first is plain soap, which is used for removal of dirt and many transient From the Johns Hopkins University School of Nursing and the Division of Infectious Diseases, Johns Hopkins University School of Medicine. Supported in part by Huntington Labs, Inc., Huntington, Indiana. Reprint requests: Elaine Larson, RN, PhD, FAAN, Nutting Chair in Clinical Nursing, Johns Hopkins School of Nursing, 600 North Wolfe St., Baltimore, MD 21205.
contaminants. The second is the antisepticcontaining health care personnel hand wash (HCPHW). Repeated 15 to 30-second hand washes with such a product should produce significant reductions in transient flora after artificial contamination of the hands and forearms. The repeated use of a HCPHW also will have some reductive effect on the resident or colonizing flora of the hands. The third type of handwash product is the antiseptic-containing surgical scrub. Scrubbing duration usually is approximately 5 minutes. Repeated use of a surgical scrub should produce at least a 2-log reduction in resident flora. 83
American
84
Journai
cd
Larson, Mayur, and Laughon
Two important characteristics of antiseptic ingredients are their ability to effect a rapid reduction in microbial counts from the hands and to bind to stratum corneum, resulting in a sustained release of the agent. Some antiseptics such as the alcohols are rapid acting and are agents of choice when the goal is an immediate reduction in transient and colonizing flora.‘” The alcohols, however, have several disadvantages. First, when used frequently over prolonged periods of time as the primary handwashing product, they remove lipid from the skin, causing dryness. Newer products compensate for that problem by the addition of emollients to the alcohol base. Second, they have little or no residual effect on the skin. Third, they may be inactivated in the presence of organic matter.4 Other antiseptic ingredients such as chlorhexidine gluconate (CHG), parachlorometaxylenol (PCMX), and triclosan (TRI), although slower acting than alcohols, bind to the stratum corneum, resulting in a sustained chemical release of the agent. In the presence of this persistent effect, one can expect to see a continued reduction in the numbers of both transient and resident organisms on the skin with prolonged use of the product over time.5, 6 Such substantivity is desirable when health care personnel work in clinical settings where patients are at high risk for infection and where a sustained reduction in numbers of organisms on the hands of personnel may be desirable (e.g., surgery and neonatal or critical care areas). Health care personnel hand wash products that contain CHG, PCMX, or TRI are available in a number of commercial formulations. The purposes of this study were to compare the antimicrobial effects of (1) three such products (containing CHG, PCMX, or TRI) on the normal resident flora after a 30-second handwash procedure over a period of 5 days and (2) each of these products at two different handwashing frequencies (6 or 18 washes/day) over a period of 5 days. METHODS Protocol Eighty healthy adult nonmedical volunteers were selected who were not receiving systemic
or topical antibiotics and who had no history of skin disease nor sensitivity to soaps. They were assigned by block randomization to one of four handwashing products, a liquid dctergent base containing (1) 2.0% CHG (CidaStat; Huntington Laboratories, Inc., Huntington, Ind.), (2) 0.6% PCMX (Medi-Scrub; Huntington), (3) 0.3% TRI (Bacti-Stat; Huntington), or (4) no antiseptic ingredient, called con& soap (Wash; Huntington). Three days before the beginning of the testing period, all subjects were provided with control soap for general bathing and handwashing and with disposable vinyl gloves for use during shampooing, dishwashing, or hand contact with other soaps, lotions, or cleaning products. They were instructed to use only this soap throughout a 3-day weaning period and the entire test period. The wearing of rings and nail polish and the use of hand lotions were prohibited during the study. After assignment to a treatment group each subject was instructed in a standardized handwashing procedure and practiced this technique in the presence of an investigator until competence was demonstrated. The standardized handwashing technique consisted of the following procedure. The subject rinsed the hands under warm running water and applied 3 ml of soap (premeasured in a syringe) into the palm. With the USC of a stop watch the subject vigorously rubbed all surfaces of the hands with the test product for 30 seconds, paying particular attention to the interdigital spaces and fingernails. During the wash period, the hands were kept out of running water to avoid rinsing off the test product. Hands were rinsed thoroughly with running water and dried with a paper towel. Handwashes were distributed throughout each day, with approximately half done in the morning and half in the afternoon. Washes were performed in the laboratory under the supervision of a laboratory assistant or an investigator for 5 consecutive days. Half the subjects (10 in each test product group) washed 6 times per day, and half washed 18 times per day. Microorganisms on the hand were harvested with the sterile bag technique. Each subject inserted the dominant hand into a sterile poly-
Volume
17 Number
2
Handwashing
April 1989
Table 1. Mean loglo counts (k standard
deviation)
by product Tricoslan
control
Baseline, total group Baseline, 6 wash/day Baseline, 18 wash/day After first wash, day 1, total group After final wash, day 5 6 wash/day 18 wash/day Log change After first wash, day 1, total group After final wash, day 5 6 wash/day 18 wash/day Percentage change* After first wash After final wash 6 wash/day 18 wash/day ‘Baseline
- Test count Baseline
x 100 (actual
5.73 5.71 5.75 5.81
and washing
(0.56) (0.68) (0.45) (0.59)
6.14 6.17 6.11 6.13
6.31 6.13 6.48 6.01
(0.84) (0.83) (0.86) (0.59)
Chlorhexidina gluconate
6.09 6.16 6.01 5.81
(0.61) (0.71) (0.52) (0.81)
5.25 (0.65) 5.45 (0.56)
5.42 (0.62) 4.56 (0.50)
5.48 (0.44) 4.74 (0.72)
5.03 (1.12) 3.62 (0.74)
+ 0.08
- 0.01
- 0.30
- 0.28
- 0.46 - 0.30
- 0.75 - 1.55
- 0.65 - 1.74
- 1.13 - 2.39
+ 20.2
- 2.3
- 49.9
- 47.5
- 65.3 - 49.9
- 82.2 - 97.2
- 77.6 - 98.2
- 92.6 - 99.6
rather than log counts
used)
counts
ethylene bag (Steri-Lok; 3M Co., St. Paul, Minn.) containing 50 ml sterile sampling solution (distilled water containing, per liter, lecithin, 20 gm; sodium thiosulfate, 6 gm; sodium oleate, 6 gm; protease peptone, 1 gm; tryptone, 1 gm; and Tween 80, 50 ml, pH: 7.2-7.4). The sampling solution contained neutralizers for the three antiseptic products7* 8 and in preliminary studies there was no significant increase nor decrease in CFUs in the solution within the first 2 hours after sampling. The entire hand surface was rubbed vigorously through the wall of the bag for 1 minute. All specimens were plated within 2 hours of sampling. A 0.2 ml volume of serial dilutions up to 1O-3 was placed on trypticase soy agar, which contained yeast extract, 5 gm/L, and Tween 80, 1 ml/L (TYT), and on MacConkey agar (BBL Microbiology Systems, Cockeysville, Md.). The TYT plates were incubated at 35” C for 48 hours followed by incubation at room temperature for an additional 48 hours. The MacConkey agar plates were incubated at 35” C for 24 hours. Colonyforming units were counted and calculated to provide a total CFU count per hand. Gramnegative bacteria were not identified to species
85
frequency
Parachlorometaxylenol
(0.76) (0.61) (0.91) (0.51)
frequency
level. Microbiologic sampling of the hands was performed at three times: immediately before the study after a brief wash with control soap (baseline), after the first wash on day 1, and after the last wash on day 5. Statistical
analysis
Analysis of covariance, that is, controlling for baseline counts, was used to test the significance of differences in microbial counts (counted as log,, CFU) among the four product groups and between the two frequencies within each product group. When multiple comparisons between groups were completed, the Tukey technique was used? A probability of p < 0.05 was considered to be statistically significant. With a sample size of 20 per product group a 75% difference in changes in log,, CFU could be detected between groups with a power of 8O%.‘O RESULTS
All 80 subjects completed the entire 5 days of testing. There were 60 women (75%) and 20 men, with a mean age of 3 1.9 years (range: 1855 years). There were no significant differences
kmefcan
86
Larson,Mayur,andLaughon
0
PCMX
CHG
INFECTIGK
TRI
Fig. 1. Log reductions In colonizing flora washing product and frequency, day 5.
CXJNT by hand-
in gender or age distribution between the product or frequency groups. Baseline mean log counts for each group ranged from 5.7 1 ( + 0.68 standard deviation) to 6.48 (kO.86) (Table 1). After the initial 30second wash, mean counts in the control group increased 0.08 logs and mean counts for subjects using one of the three antiseptics (TRI, PMCX, or CHG) decreased 0.01, 0.30, and 0.28 logs, respectively. There were no significant differences among the four test products (p = 0.61). Again, after six washes per day for 5 days there were no significant differences among products (p > OSO), although subjects using CHG had the greatest reductions. After 18 washes per day for 5 days, the control group mean count decreased 0.30 logs, and mean counts for TRI, PCMX, and CHG decreased 1.55, 1.74, and 2.39 logs, respectively (Fig. 1). All three antiseptics performed significantly better than control soap (p < 0.05). In addition, CHG produced significantly greater reductions than TRI or PCMX (p < 0.05); PCMX and TRI were not significantly different from each other (p > 0.50). Many subjects (51180, 64%) had gramnegative bacteria isolated from their hands at baseline, and subjects in all four product groups had substantial reductions in carriage of gramnegative bacteria by the end of the test period (Table 2). When gram-negative bacteria were present, they were also in greater numbers at baseline (baseline: mean log CFU = 3.95, range 2.9-4.8; end of test period; mean log CFU = 3.26, range 2.4-3.9).
Journal
of
CGNTRGIL
Although plain soap performs well for removing dirt and some transient contaminants from hands, it has little effect on colonizing flora.“,” It is not surprising, then, that in this study subjects using control soap had little change in log counts of normal flora after 5 days of handwashing. Health care personnel hand wash products, on the other hand, are chosen to minimize the numbers of transient contaminating flora and to aid in reducing the colonizing flora by means of their chemical bactericidal activity after a 15- to 30-second handwash. A high prevalence of gram-negative bacteria on hands of nonmedical personnel is not surprising and has been previously reported.13, l4 The dramatic reductions in gram-negative bacteria among all test subjects indicated that the hands were clean and relatively free of transient flora. Surgical scrub products containing 4% CHG frequently are used in the United States for health care personnel handwashing, surgical hand scrub, and occasionally for preoperative bathing or preparation of patient skin A 2% concentration of CHG has been shown to be slightly lower in antimicrobial effectiveness than 4% concentrations of CHG.“. I6 Several possible reasons explain the improved performance of CHG over that of the other test products. First, its concentration (2%) was greater than the HCPHW concentrations of either PCMX (0.6%) or TRI (0.3%). Perhaps the two latter agents at higher surgical scrub concentrations would have more antimicrobial activity. Indeed, others have used higher concentrations of each of these ingredients in comparative tests.*. ” PCMX, for example, is approved for topical use by the Food and Drug Administration for concentrations up to 3.75% and TRI up to 1 .O%. Second, CHG has excellent substantivity, binding to the stratum corneum to effect prolonged release of chemical activity on the skin. TRI also has been shown to have this same characteristic, whereas the persistence on skin of PCMX may be less ?s‘, ’ Third, CHG may give greater antimicrobial activity than the other two antiseptic ingredients tested. A recent comparison of 1.5% TRI and 4% CHG with the use of an HCPHW protocol showed that the two products were equally effective in removing artificially inoculated Serratia marcescens. How-
VOlUme
17 tvumber
2
Handwashing
April 1989
frequency
87
Table 2. Percentage (number) of subjects with gram-negative bacteria recovered from hands Baseline Product Parachlorometaxylenol Chlorhexidine gluconate Triclosan Control Total Chi-square ‘Percent
to test differences in reductions between Baseline - final x ,oo, = Baseline
FiMl
per-ntegs
Pr-
55 (11/20) 60 (12120) 80 (16/20) 60 (12/20) 63.8 @l/80) groups,
10 15 15 10
12.5
PWCWI redUctlOn’
( 2/20) ( 3/20) ( 3/20) ( 2/20) (10/80)
81.8 75.0 81.3 83.3
p = 0.86
reduction
ever, the effectiveness of CHG was consistently, although not statistically, significantly better after 4,7, and 10 washes.‘* The immediate effect (after a single 30-second wash) of products studied was not much different from that of plain, nonantimicrobial soap. It is clear that a 30second HCPHW with such products does not meet surgical hand scrub criteria. Others have reported a similar result for removal of transient as well as colonizing flora.18 When the most rapid reductions in flora are needed, a surgical scrub product and a surgical scrub procedure are required. Products containing CHG, PCMX, or TRI at HCPHW levels are most effective when used over a sustained period of time. For this reason such products are appropriate for health care personnel handwashing when a reduction in colonizing flora is desirable and when handwashing frequency is high, certainly more than six times per work period. It also would appear that in situations when frequency of handwashing is very low, there is less advantage of one product over another and probably little advantage of an antiseptic over a plain soap. In most clinical situations, however, frequent handwashing is common and the antimicrobial-containing products, particularly CHG, result in a significantly greater reduction in colonizing flora as compared with plain soap. This study examined the influence of two variables on the quality of handwashing: frequency of washing and product used. Both of these variables were clearly important. Other variables not examined in this protocol, inasmuch as the procedure was standardized to prevent confounding, include such factors as the duration
of handwashing and the amount of soap used. Many of these factors probably have independent and combined influences on the effectiveness of handwashing.lg Refemnces 1. Ayliffe GAJ. Surgical scrub and skin disinfection. Infect Control 1984;5:18-22. 2. Rotter ML. Hygiene hand disinfection. Infect Control 1984;5:18-22. 3. Larson EL, Eke PI, Laughon BE. Efficacy of alcoholbased hand rinses under frequent-use conditions. Antimicrob Agents Chemother 1986;30:542-4. 4. Morton ME. Alcohols. In: Block SS, ed. Disinfection, sterilization and preservation. 3rd ed. Philadelphia: Lea & Febiger, 1983:225-39. 5. Bruch M. Newer germicides: what they offer. In: Maibath HI, Aly R, eds. Skin microbiology: relevance to clinical infection. New York: Springer-Verlag, 1981: 103-12. 6. Larson E, Leyden J, McGinley K, Grove G, Talbot GH. Physiologic and microbiologic changes in skin related to frequent handwashing. Infect Control 1986;7:59-63. 7. Russell AD, Ahonkhai I, Rogers DT. Microbiological applications of the inactivation of antibiotics and other antimicrobial agents. J Appl Bacterial 1979;46: 207-45. 8.
9.
10. 11. 12. 13. 14. 15.
Bartzokas CA, Corkill JE, Makin T, Pinder DC. Assessment of the remanent antibacterial effect of a 2% triclosan-detergent preparation on the skin. J Hyg 1983;91:521-8. SPSS, Inc. SPSS-X User’s Guide. 2nd ed. New York: McGraw-Hill, 1984:451-64, 831-4. Lachin JM. Introduction to sample size determination and power analysis for clinical trials. Controlled Clin Trials 1981;2:93-113. Sprunt K, Redman W, Leidy G. Antibacterial effectiveness of routine handwashing. Pediatrics 1973;52:264-9. Reybrouck G. Handwashing and hand disinfection. J Hosp Infect 1986;8:15-23. Larson E. Persistent carriage of gram-negative bacteria on hands. AM J INFEST CONTROL 1981;9:112-9. Horn WA, Larson EL, McGinley KJ, Leyden JJ. Microbial flora on the hands of health care personnel. Infect Control Hosp Epidemiol 1988;9:189-93. Larson EL, Laughon BE. Comparison of four antiseptics
American
88
Larson,
Mayur,
and Laughon
INFECTION
containing chlorhexidine gluconate. Antimicrob Agents Chemother 1987;31:1572-4. 16. Mizuba S, Sheikh W. Microbiological evaluation of a new 2% aqueous chlorhexidine gluconate patient skin preparation on the axilla and abdomen. Curr Therap Res 1986;40:364-368. 17. Soulsby ME, Barnett JB, Maddox S. Antiseptic efficacy of chloxylenol-containing versus chlorhexidine gluconate-containing surgical scrub preparations. Infect Control 1986;7:223-6.
Bound vdumes
avnihble
Journai
of
CONTROI.
18. Bartzokas CA, Corkill JE, Makin T. Evaluation of the skin disinfecting activity and cumulative effect of chlorhexidine and triclosan handwash preparations on hands artificially contaminated with Sewatia mwcestens. Infect Control 1987;8:163-7. 19. Larson EL, Eke PI, Wilder MP, Laughon BE. Quantity of soap as a variable in handwashing. Infect Control 1987;8:371-5.
to sulrsuibus
Bound volumes of AMERICAN JOURNAL OF INFECTION CONTROL are available to subscribers (only) for the 1989 issues from the Publisher, at a cost of $19.00 ($25.00 international) for Vol. 17 (January-December). Shipping charges are included. Each bound volume contains a subject and an author index. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accmzpany all orders. Contact The C.V. h4osby Company, Circulation Department, 11830 Wes&ie Industrial Drive, St. Louis, Missouri 63146-3318, USA; phone 800-325-4177, ext. 351. Subsc&ption8 mu6t be ia force to quaitfy. Bosnd vdumcs are not av bm ofa~jownaIshtcrlption.