Comparative sensitivity of 13 species of pathogenic bacteria to seven chemical germicides

Comparative sensitivity of 13 species of pathogenic bacteria to seven chemical germicides Jos~-Luis Sagripantil DSc a Cheryl A. Eklund, BA b Paula A. Trost, BA c Karen C. Jinneman, MS ° Carlos Abeyta, Jr., BS ° Charles A. Kaysner, MS, SM (AAM) c Walter E. Hill, PhD c Rockville, Maryland, and Bothell, Washington

Background: The relative resistance of diverse human bacterial pathogens to commonly used germicidal agents has not been established. Methods: We measured by titration the survival of thirteen different bacteria after exposure to glutaraldehyde, formaldehyde, hydrogen peroxide, peracetic acid, cupric ascorbate, sodium hypochlorite, or phenol. Results: Our comparative experiments allowed classification of the organisms' survival into four groups: (a) Pseudomonas aeruginosa and Staphylococcus aureus showed the most resistance, (b) Clostridium perfringens, Salmonella typhimurium, Staphylococcus epidermidis, and Escherichia coli O157:H7 showed intermediate resistance, (c) Listeria monocytogenes, Shigella sonnei, and Vibrio parahaemolyticus survived some treatments with chemical agents only in the presence of protecting protein (serum albumin), and (d) Vibrio cholerae, Vibrio vulnificus, Bacillus cereus, and Yersinia enterocolitica did not survive any of the treatments applied. Conclusion: We found species that more frequently survived exposure to germicidal agents were also those most commonly reported in association with hospital infections. Our findings suggest that resistance to disinfectants may be more important than pathogenicity in determining the relative prominence of an organism as an agent responsible for nosocomial infections. (AJIC Am J Infect Control 1997;25:335-9)

A b o u t 5% to 10% (1.75 to 3.5 million) of the 35 m i l l i o n p a t i e n t s a n n u a l l y a d m i t t e d in the U n i t e d States a c q u i r e a n i n f e c t i o n while hospitalized. 1-3 The p a t h o g e n s r e p o r t e d m o s t f r e q u e n t l y in hospital-wide surveillance and their a p p r o x i m a t e d p r e v a l e n c e w e r e S t a p h y l o c O c c u s a u r e u s (12%), c o a g u l a s e - n e g a t i v e s t a p h y l o c o c c i (11%), E n t e r o c o c c u s spp. (10%), E s c h e r i c h i a coli (12% to 18%), P s e u d o m o n a s spp. (9% to 14%), E n t e r o b a c t e r spp. From the Molecular Biology Branch, Division of Life Sciences, Office of Science and Technology, Center for Devices and Radiological Health, Food and Drug Administration, Rockville, a Science Branch b and Seafood Products Research Center,° Seattle District Office, Food and Drug Administration, Bothell. Reprint requests: Jose-Luis Sagripanti, DSc, (HFZ-113) Molecular Biology Branch, Center for Devices and Radiological Health, Food and Drug Administration, 5600 Fishers Ln., Rockvilie, MD 20857.

17146/75588

(6%), a n d C a n d i d a spp. (5% to 9%). 4, 5 W h e t h e r this p r e v a l e n c e is d u e to the relative a b u n d a n c e of these o r g a n i s m s in the clinical setting o r to t h e i r c o m p a r a t i v e r e s i s t a n c e to i n a c t i v a t i o n has n o t b e e n established. A relatively large b o d y of i n f o r m a t i o n o n inactiv a t i o n of m i c r o o r g a n i s m s b y liquid g e r m i c i d a l agents has a c c u m u l a t e d . 6 However, assessing the relative r e s i s t a n c e of different b a c t e r i a to c o m m o n disinfectants r e m a i n s difficult. M o s t studies are d e s i g n e d to establish effectiveness of disinfectants r a t h e r t h a n to c o m p a r e the differential sensitivity of m i c r o o r g a n i s m s . 7 B e c a u s e slight m e t h o d o l o g i c differences c a n s u b s t a n t i a l l y alter the effectiveness of biocides, 8, 9 c o m p a r i s o n of d a t a f r o m different studies w i t h p r e c i s i o n is u s u a l l y difficult. As a result, to c o n t r o l a n e w o u t b r e a k , i n f o r m a t i o n o n the effectiveness of a d i s i n f e c t a n t o b t a i n e d w i t h a s t a n d a r d bacterial s t r a i n m u s t s o m e t i m e s be ap335

AJIC

3 3 6 Sagripanti et al.

plied to an untested organism. Obviously, such a practice should be safe when a more resistant organism is used to test disinfectant effectiveness. However, the opposite situation could have disastrous consequences. In this study we answered a relatively straightforward question: what is the comparative order of bacterial resistance to treatment with a panel of c o m m o n l y used chemical agents? A matrix of quantitative data on the relative sensitivity of a spectrum of bacterial species to treatment with several disinfectants could provide the data necessary to reduce the risk of inadequate decontamination. MATERIAL AND METHODS Strains t e s t e d

S. aureus ATCC 25923, Staphylococcus epidermidis, ATCC 12228, Pseudomonas aeruginosa ATCC 27853, SalmOnella typhimurium ATCC 14028, Shigella sonnei (food isolate, lettuce), Listeria monocytogenes V-7 (food isolate, milk 1983), Bacillus cereus ATCC 14579, Yersinia enterocolitica 8081 serotype 0:8 harboring a virulence plasmid, and E. coli 0157:H7 EDL933 (food isolate, ground beef, obtained from the Centers for Disease Control and Prevention [CDC]) were grown in trypticase soy broth with 0.6% yeast extract (Difco, Detroit, Mich.) aerobically at 35 ° C for 24 hours. Cultures were plated onto trypticase soy agar with 0.6% yeast extract and incubated aerobically at 35 ° C for 24 hours. Vibrio cholerae C6706 (from CDC), Vibrio parahaemolyticus NY4771° and Vibrio vulnificus LA M624 (California State Department of Health Services) were grown in trypticase soy broth with 2% NaC1 aerobically at 35 ° C for 24 hours. Cultures were plated onto trypticase soy agar with 2% NaC1 and incubated aerobically at 35 ° C for 24 hours. Clostridium perfringens CDC 1861 was grown in trypticase-peptone-glucoseyeast extract broth aerobically at 37 ° C for 24 hours. Culture was plated onto Shahidi-Ferguson perfringens agar plates and anaerobically incubated for 48 hours at 37 ° C in a gas pack jar equipped with GasPack H ÷ CO2 generator (BBL, Baltimore, Md.). Before testing against biocides, cultures were stored in sterile vials with glycerol and frozen at -70 ° C. After cultures were transferred to their appropriate sterile enrichment/ selective broths, they were incubated for 24 hours and then plated onto trypticase soy agar with 0.6% yeast extract for purity. Isolates were examined by Gram staining and wet mounts and identity confirmed with conventional tests u and commercial bacterial identification procedures (API20E; bio-

August 1997

Merieux, Plainview, N.Y.) (Vitek; Vitek Systems, Hazelwood, Mo.) (MICRO-ID Listeria; Organon Teknica Corporation, Durham, N.C.). No additional tests were performed on plasmid-harboring Y. enterocolitica. Chemical agents

We exposed each organism to seven different substances at concentrations suggested to decontaminate medical devices. 12-15The optimal pH and storage conditions for each chemical agent have been previously established. 16 Glutaraldehyde (8% weight/volume [w/v], molecular biology grade I, no. 7526; Sigma Chemical Co., St. Louis, Mo.) was shipped in dry ice, maintained frozen, and used within a week of reception. Glutaraldehyde was adjusted to pH 9.3 with sodium bicarbonate before each experiment and used at a final concentration of 2%. Sodium hypochlorite (6.8% w/v available chlorine, no. 23,930-5; Aldrich Chemicals, Milwaukee, Wis.) was adjusted to pH 7.3 with hydrochloric acid before use at a final concentration of 0.05%. Formaldehyde (37% w/v solution in water, no. 25,254-9; Aldrich Chemicals) was used at 8%, pH 3.4. Peracetic acid (32% w/v solution in diluted acetic acid, no. 26,933-6, Aldrich Chemicals) was used at a final concentration of 0.03%, pH 2.6. Ascorbic acid (crystals) was obtained from Aldrich Chemicals (no. 25,556-4). Phenol "Ultra Pure" molecular biology grade (no. 5509UA; Life Technologies, Gaithersburg, Md.) was received frozen and maintained at -20 ° C until used at a concentration of 5% and p H 6.3. Cupric chloride (CuC12 • 2H20) and hydrogen peroxide (30% w/v) were purchased from Mallinckrodt Chemicals (nos. 4824 and 5239, respectively; Paris, Ky.). The copper ascorbate mixture consisted of 0.5% cupric ions (as cupric chloride), 0.1% ascorbic acid, and 0.003% hydrogen peroxide, p H 3.0. Hydrogen peroxide by itself was used at a final concentration of 10% at pH 3.9. T r e a t m e n t protocol

Bacterial strains were recultured overnight and exposed to eight parallel treatments consisting of each of seven chemical agents described above, diluted daily just before each experiment, or to sterile distilled water as a control. Fifty gl of bacterial suspension, containing 1 to 2 x 108 organisms/ml were added to the b o t t o m of a 1.5 ml polypropylene microcentrifuge tube. Each chemical agent, 50 gl, at twice the concentration indicated in the previous section, was added to the tube with 50 gl of the bacterial suspension, and

AJIC Volume 25, Number 4

Sagripanti et al. 3 3 7

Table 1. Bacterial inactivation by common biocidal agents* Bacterium

Glutaraldehyde

Formaldehyde

Phenol

Cu/Asc

Hypochlorite

Peroxide

Peracetic

B. cereus C. perfringens E. coff L. monocytogenes tR aeruginosa S. typhimurium S. sonnei S, aureus S. epidermidis V. cholerae V, parahaemolyticus V. vulnificus Y, enterocolitica

>5.0 (2) >6.3 (2) >6.9 (2) >6.1 (2) 3.8 + 0.2 (2) >6.4 (3) >6.3 (2) >6.5 (3) >6.3 (3) >6.4 (2) >6.2 (1) >6.3 (2) >6.8 (2)

>5.0 (2) >6.3 (2) >6.9 (2) >6.1 (1) >6.1 (3) >6.2 (3) >6.3 (2) >6.5 (3) 5.9 _+ 1.1 (3) >6.4 (2) >6.2 (2) >6.3 (2) >6.8 (2)

>5.0 (2) >6.3 (2) >6.9 (2) >6.1 (2) 5.8 -+ 0.6 (3) >6,4 (3) >6.3 (2) >6.5 (3) >6.3 (3) >6.4 (2) >6.2 (2) >6.3 (2) >6.8 (2)

>5.0 (2) >6.3 (2) 6.3 + 0.8 (2) >6.1 (1) 5,6 + 0,9 (3) >6.4 (3) >6.1 (1) 5.5 +_ 1.2 (3) 5.1 + 0.1 (2) >6.4 (2) >6.2 (2) >6.3 (2) >6.8 (2)

>5.0 (2) 0.14 _+ 0.05 (2) 6.2 + 0,9 (2) >6,1 (2) 1,3 + 0.1 (2) 4.1 + 1.3 (2) >6.3 (2) 4.8 + 1,8 (2) 6.3 + 0.4 (3) >6.4 (2) >6.2 (2) >6.3 (2) >6.8 (2)

>5.0 (2) >6.3 (2) >6,9 (2) >6.1 (2) >6.1 (3) >6.4 (3) >6.3 (2) 5.6 + 0,7 (3) >6.3 (3) >6.4 (2) >6.2 (2) >6.3 (2) >6.8 (2)

>5.0 (2) 4.1 _+ 0.1 (2) >6.9 (2) >6.1 (1) 5.0 + 1.6 (3) >6.4 (3) >6.3 (2) 6.6 _+ 0.3 (3) >6.3 (3) >6,4 (2) >6.2 (2 >6.3 (2) >6.8 (2)

*Calculated as -Iog(TJTw) where Te is the titer of bacteria surviving 30 minutes exposure at 20 ° C to a given disinfectant, and T,~ is the titer of bacteria exposed under the same conditions to water. Results are expressed either as the limit of detection when no surviving colonies were obtained or as x _+s (n) where n is the number of replicate experiments.

timing was started. Incubation proceeded for 30 minutes at 20 ° C before adding 0.9 ml ice-cold m e d i u m compatible with the bacterium being tested. To remove chemical agents, reaction tubes were centrifuged for 5 minutes at 15,000 rpm (Microfuge model 5414; B r i n k m a n Instruments Inc., Westbury, N.Y.), the supernatant was carefully removed and replaced by new growth m e d i u m before the bacterial pellet was resuspended. Cells were serially diluted in a m e d i u m suitable for that bacterium and titrated by plating 0.1 ml of each dilution in duplicate onto the appropriate agar plate. Plates were incubated at 35 ° C for 24 hours, and typical colonies were counted. The n u m b e r of independently replicated experiments is indicated in the legend of Table 1. Titration of C. perfringens was done on agar plates containing ShahidiFerguson perfringens agar plates and incubated anaerobically for 48 hours at 37 ° C. Bacteria that failed to survive any of the treatments indicated in Table 1 were assayed for survival with the same chemical agents in the presence of albumin. Bovine a l b u m i n grade V (Gibco Diagnostics, Madison, Wis.) was dissolved in sterile distilled water at a concentration of 10 gin/L, sterilized by filtration through 0.22 ~tm filters, and used at 0.1 dilution. To m a i n t a i n experimental conditions and the activity of the chemical agents as constant and comparable as feasible, the study was completed within 10 consecutive days. Statistical methods

Data were analyzed by the NPARlWAY nonparametric SAS procedure 17 that calculates simple linear rank statistics on the basis of Van der

Waerden scoreslS: a(Rj) = qb-I [R](n +1)] where

a(R 9 is the rank score R i is the rank of the jth observation, and ~b is the distribution function for the normal distribution. These statistics were used to test whether the distribution of the variable "bacterial inactivation" had the same location parameters across the different bacterial species. RESULTS

The log of the reduction in viability of the bacterial strains after exposure to each disinfectant is presented in Table l. P. aeruginosa was the most resistant organism tested; less than 4 logs inactivation were observed after treatment with glutaraldehyde (2%, pH 9.3) or hypochlorite (0.05%, pH 7.3). Growth of P. aeruginosa colonies was obtained after treatment with five of the seven germicidal substances tested indicating that a longer contact time is required for effective inactivation. Sodium hypochlorite showed partial effectiveness in killing S. typhimurium and S. aureus; it had only a slight effect on P. aeruginosa and C. perfringens. Because most of the data in Table 1 have openended limits of detection, these survival results were analyzed by a nonparametric, computerassisted test on the basis of ranks, that is, Van der Waerden scores. 18 The NPARlWAY procedure 17 was used to calculate the scores, and the obtained results are displayed in Table 2. Our test methods and statistical analyses allowed ranking the six most resistant pathogens in our study (mean scores between -0.998 and +0.036). The low resistance of the seven remaining organisms (all with

AJIC

3 3 8 Sagripanti et al.

August 1997

Table 2, Relative bacterial resistance to liquid germicides Van der Waerden scores-[ Bacterium

Survival score*

Sum of scores~:

Mean score

Relative resistance

R aeruginosa S. aureus

9/21 = 0.43 9/21 = 0.43

-6.989 -3.956

43.988 43.565

High

C, S. S, E.

4/14 4/21 4/21 2/21

---. 0.29 = 0.19 = 0.19 = 0.10

-2.363 +0.250 -2.123 43.398

-0.338 +0.036 43.303 43.057

Medium

3/7 (BSA){} 1/7 (BSA) 1/7 (BSA)

+2.225 +2.225 +2.225

+0.318 +0.318 +0.318

Low

011 0 0

+2.225 +2.225 +2.225 +2.225

+0.318 +0.318 +0,318 +0.318

None d e t e c t e d

perfringens typhimurium epidermidis coil O157:H7

L. monocytogenes S. sonnei V, parahaemolyticus B. cereus V. cholerae V. vulnificus Y, enterocolitica

0

*Number of test results showing bacterial survival over the number of tests done with all disinfectants. 1-Van der Waerden scores (normal) for bacterial inactivation classified by variable bacterium. The null hypothesis that all 13 bacterial species have identical population distribution functions is rejected by the Van der Waerden one-way approximate chi-square (Z212= 33.030) with p value = 0.0010. :I:From seven independent data sets (one per germicide) for each bacterium. §No survivors were observed in the absence of BSA. IITests conducted with BSA also yielded no survivors.

m e a n scores of +0.318) precluded further differentiation and ranking, based solely on the data in Table 1. Presence of organic matter, particularly proteins, can decrease the biocidal activity of disinfectants. 19 The presence of 1 gm/1 bovine serum albumin protected L. monocytogenes against formaldehyde, cupric ascorbate, and peracetic acid; S. sonnei was protected against cupric ascorbate; and E parahaemolyticus survived in glutaraldehyde (data not shown). B. cereus, V. cholerae, E vulnificus, and Y. enterocolitica exhibited a complete loss of colony-forming ability in all of the chemical agents, even when albumin was present. DISCUSSION

On the basis of the relative resistance to all chemical agents, the bacteria we studied can be clustered into the four groups shown in Table 2. The criteria to place an organism into a particular g r o u p according to its relative resistance to inactivation are as follows: (1) high, when the Van der Waerden m e a n survival scores were below -0.500; (2) medium, with survival m e a n scores between -0.500 and +0.200; (3) low, with scores above +0.200 and survival to treatment only in the presence of protective protein; and (4) none detected, with scores above +0.200, w h e n inactivation occurred under every tested condition, even in the presence of protective protein. We believe this

scheme is useful in classifying microorganisms according to their relative risk of incomplete decontamination. Desire to avoid the exposure of laboratory personnel to pathogenic organisms could promote the use of S. epiderrnidis as a surrogate of S. aureus for testing germicidal agents. The results in Tables 1 and 2 demonstrate that S. aureus and S. epidermidis differ in their relative resistance to disinfectants. Hence, data on the inactivation of S. epidermidis cannot be safely applied to S. aureus. The bacterial species most resistant in our comparative test (P. aeruginosa and S. aureus) are also two of the organisms most c o m m o n l y reported as responsible for nosocomial infections. 4, 5 The pathogenic V. cholerae strain responsible for the deadly epidemic in Peru, which is not usually associated with nosocomial infections, was highly sensitive to chemical agents. These facts suggest that resistance tO disinfectants m a y be more important than pathogenicity in determining the relative prominence of an organism as an agent responsible for nosocomial infections. Overall, the data in Tables 1 and 2 are at the same time reassuring and cautionary. We observed that commonly used liquid germicidal substances killed bacteria in most instances. This agrees with the observation that disease transmission by contaminated medical devices, food, and other inanimate vectors appears to be relatively

AJIC Volume 25, Number 4

Sagripanti et al. 3 3 9

uncommon. However, the fact that germicidal agents did not inactivate every bacterial species in every instance may help explain cases of nosocomial infections caused by improperly decontaminated medical devices, disease outbreaks caused by insufficiently processed food, and epidemics related to inadequate sanitation practices. We thank Harry Bushar, PhD, Divisionof Biostatistics, Officeof Surveillance and Biometrics, FDA,Rockville,Md., for performing the statistical analysis of our data. We thank Steve Weagant and June Wetherington, FDA, Bothell, Wash., for providing some of the strains tested. References 1. Hughes JM, Jarvis WR. Epidemiology of nosocomial infections. In: Lennette EH, Belows A, Hausler WG Jr, Shedomy H J, editors. Manual of clinical microbiology. 4th ed. Washington, DC: American Association for Microbiology; 1985. p. 99-104. 2. United States General Accounting Office. Hospital sterilants: insufficient FDA regulation may pose a public health risk. Federal document GAO/HRD-93-79; 1993. 3. Herwaldt LA, Wenzel RE Dynamics of hospital-acquired infection. In: Murray PR, Baron E J, Pfaller MA, Tenover FC, Yolken RH, editors, Manual of clinical microbiology. 6th ed. Washington, DC: American Association for Microbiology; 1995. p. 69-181. 4. Beck-Sagu6 C, Jarvis WR. The epidemiology and prevention of nosocomial infections. In: Block SS, editor. Disinfection, sterilization, and preservation. 4th ed. Philadelphia: Lea & Febiger; 1991. p. 655-62. 5. Emori GT, Gaynes RR An overview of nosocomial infections, including the role of the microbiology laboratory. Clin Microbiol Rev 1993;6:428-42. 6. Block SS, editor. Disinfection, sterilization, and preservation. 4th ed. Philadelphia: Lea & Febiger; 1991. 7. Beloian A. Disinfectants. In: AOAC official methods of analysis. Arlington (VA): Association of Official Analytical Chemists; 1990. p. 133-46.

8. Gorman SP, Scott EM. Effect of alkalinination on the bacterial cell and glutaraldehyde molecule. Microbios 1977;6:39-44. 9. Hodges NA, Melling J, Paker SJ. A comparison of chemically defined and complex media for the production of Bacillus subtilis spores having reproducible resistance and germination characteristics. J Pharm Pharmacol 1980; 32:126-30. 10. Spite GT, Twedt RM. Isolation of an enteropathogenic, Kanagawa positive strain of Vibrio parahaemoIyticus from seafood implicated in acute gastroenteritis. Appl Environ Microbiol 1978;35:1226-7. 11. Bacteriological Analytical Manual. Arlington (VA): Association of Official Analytical Chemists; 1993. 12. Weller IVD. Cleaning and disinfection of equipment for gastrointestinal flexible endoscopy: interim recommendations of a working party of the British society of gastroenterology. Gut 1988;29:1134-51. 13. World Health Organization. The World Health Organization guidelines on sterilization and disinfection methods against HIV. WHO AIDS, series 2.2nd ed. Geneva: World Health Organization; 1989. 14. Miller BM, Gorschel DHM, Richardson JH, et al. Laboratory safety, principles and practices. Washington, DC: American Society for Microbiology; 1986. p. 56. 15. Sagripanti JL, Routson LB, Lytle CD. Virus inactivation by copper or iron ions alone and in the presence of peroxide. Appl Environ Microbiol 1993;59:4374-6. 16. Sagripanti JL, Bonifacino A. Comparative sporicidal effects of liquid chemical agents. Appl Environ Microbiol 1996; 62:545-51. 17. SAS Institute. The NPARlWAY Procedure. In: SAS/STAT user's guide. Release 6.03 edition. Cary (NC): SAS Institute; 1988. Chap. 24. 18. Conover WJ. Practical nonparametric statistics. 2nd ed. New York: John Wiley & Sons. 19. Favero MS, Bond WW. Chemical disinfection of medical and surgical materials. In: Block SS, editor. Disinfection, sterilization, and preservation. 4th ed. Philadelphia: Lea & Febiger; 1991. p. 617-41

W H A T ' S N E X T F O R AJIC? Issue

Topic

Guest editor(s)

October December

Organized systems of care Research

Jerri Bryant Helen Rosen Kotilainen