- Email: [email protected]
Pseudomonas aeruginosa biofilms are more susceptible to ciprofloxacin than to tobramycin
,\,,“\\,ICI\\I
,,I,“\\,
III
Antimicrobial Agents ELSEVIER
International Journal of Antimicrobial Agents 7 (1996) 251-256
Pseudomonas aeruginosa biofilms are more susceptible to ciprofloxacin
than to tobramycin Carol A.K. Preston”, Antoine E. Khouryb, Gregor Reid”ad,*, Andrew W. Bruce”, J. William Costertone “Centre for Biomaterials Infection Research, Toronto General Division, 200 Elizabeth Street, Toronto, Ontario, Canada MSG 2C4 bDivision of Urology, Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, Canada MSG 1X8 ‘Department of Microbiology and Immunology, The University of Western Ontario, Toronto, Canada dThe Lawson Research Institute, London, Ontario, Canada ‘Montana State University, Butte, Montana, USA
Accepted 30 July 1996
Abstract
The objective of this study was to determine and compare the biofilm elimination concentrations (BEC: the concentration which reduced the viability of biofilm organisms by at least 99.9%) of ciprofloxacin and tobramycin for Pseudomonas aeruginosa, a common cause of nosocomial biomaterial-related infections. Bacterial biofilms were produced in the modified Robbins device using continuous culture flow at 60 ml/h for 40-44 h, and the sessile organisms were then exposed to either ciprofloxacin or tobramycin at a range of concentrations for 12 or 36h. The BEC of ciprofloxacin was 5 pgg/ml for the 12 and 36 h treatments, a value 10 x greater than the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC). In contrast, the BEC of tobramycin was > 100 ,ug/ml after 12h and 75 ug/ml following 36 h of drug exposure, that is 75-100 x the MIC and MBC. The results demonstrated that the BEC is a more suitable indicator of the antibiotic susceptibility of P. aeruginosa biofilms than the MIC and MBC. Ciprofloxacin was significantly more effective than tobramycin in the treatment of P. ueruginosu adherent to biomaterials. With respect to clinical application, if the intention of antibiotic use is to eradicate bacteria adherent to devices, only biofilm-active agents should be used. Copyright 0 1996 Elsevier Science B.V. Keywords:
Biofilms; Pseudomonas;
Ciprofloxacin;
Tobramycin
1. Introduction Infections of prosthetic devices represent an important clinical problem, which is compounded by the fact that the organisms which bind to, and multiply on, the surfaces of the biomaterials, form a thick biofilm which protects them against the host’s immune system and antimicrobial agents [ l-41. Antimicrobial agents such as tobramycin routinely used in hospital settings to * Corresponding author, H414 Lawson Research Institute, St. Joseph’s Health Centre, 268 Grosvenor Street, London, Ontario, N6A 4V2, Canada. Tel.: + 519 646 6100 ext. 5256; fax: + 519 646 61 lo:, e-mail: [email protected] 0924-8579/96/$32.00
treat infections, are generally ineffective clinically in eliminating prosthesis-associated infections 151. At present, the only way to eliminate such infections is to remove and replace the infected device. This process can be traumatic to the patient and expensive to the health care system. Pseudomonas aeruginosa is particularly pathogenic when associated with the nidus of an implanted medical device, and it is often resistant to antimicrobials, especially those used to treat nosocomial infections [6,7]. The study carried out herein was designed to compare the efficacy of ciprofloxacin, a potent fluoroquinolone agent, against tobramycin on the viability of P. aeruginosa in a biofilm.
Copyright 0 1996 Elsevier Science B.V. All rights reserved
PII SO924-8579(96)00330-S
252
C.A.K. Preston et al. 1 international Journal of’ Antimicrobial
Agents 7 (1996) 251-256
2. Materials and methods
2.5. Measurement of bacterial viability
2.1. Bacterial strain
Adherent bacteria were removed from the studs by ultrasonication for 7 min (Branson 1200 ultrasonic cleaner) in 5 ml of PBS. Bacteria released from the stud surface were diluted in PBS and plated with four dilutions per stud, in duplicate, on nutrient agar. Cultures were incubated for 16-24 h at 37°C after which viable counts were made.
The clinical isolate of P. aeruginosa was obtained from an infected urine sample from a patient at the Hospital for Sick Children, Toronto, Canada. 2.2. Antimicrobial agents Intravenous tobramycin (tobramycin sulphate, Eli Lilly, Toronto) and ciprofloxacin (lactate salt, kindly provided by Bayer Pharmaceuticals, Toronto) were used. 2.3. Antibiotic susceptibility of planktonic organisms The in vitro minimum inhibitory concentrations (MICs) and minimum bactericidal concentrations (MBCs) of ciprofloxacin and tobramycin were measured with planktonic P. aeruginosa cells using the standard macrobroth dilution method (NCCLS) used in the Hospital for Sick Children. The MIC and MBC values were 0.5 pg/ml and 1.0 pg/ml for ciprofloxacin and tobramycin, respectively. 2.4. Colonization of MRD and drug treatment The modified Robbins device (MRD) has been developed [2,8] as a means of in vitro testing of the efficacy of antimicrobial agents on sessile (biofilm) organisms. In the present study, the MRD was set up largely as described by Nickel et al. [2]. The sampling ports were furnished with metal studs in the lumen which were exposed to an actively growing P. aeruginosa culture. The bacteria were grown in nutrient broth (Difco Laboratories, Detroit, MI) in a 4-l reservoir held in a 37°C water bath and the culture was pumped through the device for 40-44 h at a rate of approximately 60 ml/h. After initial bacterial colonization, two studs were removed and replaced with sterile ones, washed twice with 10 ml PBS to remove non-adherent bacteria, and processed to measure cell viability as described, below. The inflow tubing to the MRD was replaced with a fresh sterile one and the culture flask was exchanged for one containing ciprofloxacin at 2.5 to 10 pg/ml or tobramycin at 50 to 100 ug/ml in nutrient broth. P aeruginosa biofilms were exposed to the antibiotics for 12 or 36 h. Two studs were sampled at the following intervals during the incubation with the drugs: for 12 h exposure, every 4 h; for 36 h exposure, every 12 h. Two concentrations of antibiotic were tested simultaneously with two devices colonized with the same culture.
2.6. Acridine orange staining of A4RD studs The studs were subjected to bacterial colonization and rinsed to remove non-adherent bacteria, as described above. One half of the studs was sonicated as noted above. The other half was rinsed twice more with PBS and were then fixed by exposure to 10% formalin (Fisher Scientific, Toronto) at 37” C for 1 h. Samples were then rinsed thrice with PBS at room temperature and stained with Acridine Orange, prepared as described by Zufferey et al. [9], for 3 min at room temperature. The stained studs were then rinsed thrice with distilled water, air dried and affixed to glass slides with Silastic medical adhesive silicone Type A (Dow Corning Corp., Midland MI). Stained stud surfaces were examined under a fluorescence microscope (Zeiss) at a magnification of x 1000 with oil immersion to assess the presence of bacteria. These steps were carried out in triplicate prior to and after the sonication procedure to verify the efficacy of ultrasonication at removing the adherent bacteria from the surface of the studs. Studs not exposed to bacteria were also stained as a control.
,021. . . 0
’
’
5
.
.
.
.
’
*
.
10
.
’ 15
Period of Exposure to Antibiotic (hours) Fig. 1. Effect of 12 h exposure of tobramycin on P. aeruginosa biofilm bacteria viability. Forty 44-h old biofilms were exposed to 75 ( n ) or 100 (0) mg/ml tobramycin in the MRD for 12 h. At the intervals noted, samples were removed and processed to assess bacterial viability. Each data point represents the weighted mean of 3-4 dilutions for duplicate samples from three experiments.
253
C.A.K. Preston et al. 1 International Journal of’ Antimicrobial Agents 7 (1996) 251-256
._
showed that 99.98% of the viable sessile bacteria were removed from the stud surface by sonication. The studs exposed to bacteria for 40-44 h and sonicated for 7 min were stained with Acridine Orange and assessed microscopically for the presence of bacterial cells. These stud surfaces were compared to those of stained colonized studs which were not subjected to sonication. Results showed that the sonicated studs had only background levels of fluorescence while the surfaces which were not sonicated showed the presence of thick biofilms. These results indicated that sonication is effective, under the conditions utilized, for removing P. aeruginosa in a biofilm from the metal MRD studs. 0
5
10
15
Period of Exposure to Antibiotic (hours) Fig. 2. Effect of 12 h exposure of ciprofloxacin on P. aeruginosa biofilm bacteria viability. Forty 44-h old biofilms were exposed to 2.5 (O), 5.0 (m) or 10 (A) mg/ml ciprofloxacin in the MRD for 12 h. At the intervals noted, samples were removed and processed to assess bacterial viability. Each data point represents the weighted mean of 3-4 dilutions for duplicate samples from 3-4 experiments.
2.7. Quantljication of viable bacteria remaining on studs ufter sonication
Three studs, subjected to bacterial colonization and rinsed to remove non-adherent bacteria, were sonicated as noted above and rinsed twice more with PBS. Flamed forceps were used as the formerly colonized stud surface was streaked across the surface of a nutrient agar plate. Viable counts were made after a 24 h incubation period at 37°C.
3.2. Biojilm elimination concentration at 12 h The effect of tobramycin
and ciproflaxacin
on P.
aeruginosa biofilms was tested by exposing 40-44 h old
biofilms to a range of concentrations of these agents for 12 h. The results are presented in Figs. 1 and 2. The biofilm elimination concentration of the antimicrobial agents was determined from this data as the concentration required to achieve at least a 99.9% reduction in the viability of biofilm bacteria (confidence interval of 95%; one-tailed test). The BEC of ciprofloxacin was found to be 5 pug/ml. This value is 10 x the planktonic MIC and MBC value for this organism. With 100 lug/ml an average reduction of 42% (range O-100%) ensued, The tobramycin BEC was therefore determined to be > 100 pg/ml, which is more than 100 x the MIC and MBC for planktonic cells.
2.8. Scanning electron microscopy of biojilm bacteria
Studs were subjected to bacterial colonization, with or without subsequent antibiotic exposure of 10 pug/ml ciprofloxacin or 75 pg/ml tobramycin for 24 or 36 h, and rinsed to remove non-adherent bacteria. Bacteria were fixed in 5% glutaraldehyde in cacodylate buffer (0.1 M, pH 7.0) at room temperature. Samples were then exposed to two fresh solutions of cacodylate buffer for 20 min at room temperature and placed in a third buffer ‘wash’ and stored at 40°C prior to dehydration steps (performed as noted by Nickel et al. [2]) and examination in a Hitachi 5540 scanning electron microscope.
-I
\ 8
10’ ! 0
3. Results 3.1. Assessment of efficacy of sonication at removal of sessile bacteria from stud surface The number of viable bacteria remaining on the stud surface after sonication was quantified. The results
10
20
30
40
Period of Exposure to Antibiotic (hours) Fig. 3. Effect of 36 h exposure of tobramycin on P. aeruginosa biofilm bacteria viability. Forty 44-h old biofilms were exposed to 50 (0) or 75 (m) mg/ml tobramycin in the MRD for 36 h. At the intervals noted, samples were removed and processed to assess bacterial viability. Each data point represents the weighted mean of 3-4 dilutions for duplicate samples from four experiments.
254
C.A.K. Preston et al. / International Journal of Antimicrobial Agenrs 7 (1996) 251-256
4. Discussion
Period of Exposure
to Antibiotic (hours)
Fig. 4. Effect of 36 h exposure of ciprofloxacin on P. aeruginosa biofilm bacteria viability. Forty 44-h old biofilms were exposed to 5.0 (0) or 10 (W) mg/ml ciprofloxacin in the MRD for 36 h. At the intervals noted, samples were removed and processed to assess bacterial viability. Each data point represents the weighted mean of 3-4 dilutions for duplicate samples from five experiments.
3.3.
Biojilm elimination concentration at 36 h
To determine whether a longer exposure to the antimicrobial agents could eradicate the biofilm viability or reduce the BEC, P. aeruginosa biofilms were exposed to a range of drug concentrations for 36 h. The results are shown in Figs. 3 and 4. The BEC of ciprofloxacin under these conditions was 5 pg/ml, as was found after the 12 h treatment. Total elimination of viable biofilm cells was achieved in three out of five experiments with the BEC. The BEC of tobramycin was found to be 75 ,ug/ml, a value 75 x the MIC and MBC determined for planktonic cells.
The results showed that a virulent strain of uropathogenie bacteria existing in biofllms is more resistant to antimicrobial agents than planktonic counterparts. More importantly, the latest results showed that in vitro planktonic MIC and MBC levels of antimicrobial agents are inadequate to eliminate biofilm organisms associated with biomaterials. Indeed, the BECs of ciprofloxacin and tobramycin were found to be 10 x or 75-100 x the MIC and MBC values, respectively. The BEC is a more suitable indicator of the antibiotic susceptibility of P. aeruginosa biofilms. Unless provided with the true concentration of drug required to eradicate biofilms, a physician may be unable to adequately treat the patient. In view of previous work showing P. aeruginosa was resistant to 1000 pg/ml tobramycin, the latest findings are not completely unexpected [2]. However, our results are in contrast to those reported by Anwar et al. [lo] where the treatment of P. aeruginosa biofilms with 10 pgg/ml of tobramycin for 3 h was sufficient to reduce bacterial viability by > 99.9%. It is believed that growth rate, which is dependent on the availability of nutrients, can affect the susceptibility of bacteria to antimicrobial agents [ll]. Since Anwar and colleagues
3.4. Electron microscopical examination of metal studs colonized with P. aeruginosa and treated with antimicrobial agents SEM examination demonstrated that after a 44 h exposure of the studs to the P. aeruginosa culture a thick adherent bacterial biofilm had formed (Fig. 5). It should be pointed out that the biofllms did not universally and totally cover the surfaces, thereby explaining why only 100000 or so organisms were found per stud. Following 24 h exposure to 5 pg/ml ciprofloxacin, a marked decrease in the extent of the biofilm was noted, and no adherent organisms were evident with 36 h exposure (Fig. 6). After 24 h treatment with 75 pgg/ml tobramycin much of the biofilm was gone, and after a further 12 h, the biofilm was removed. All SEM results correlated well with the viable plate counts.
Fig. 5. SEM of a stud examined 44 h after colonization with P. The surface is covered by a thick bacterial biofilm. Bar: 5 ,um.
aeruginosa under flow conditions.
C.A.K. Preston et al.
I International Journal of Antimicrobial Agents 7 (1996) 2.51-256
255
The reduction of viability of 99.9% was used to determine the BEC values as this is the accepted killing endpoint in bactericidal tests [14]. Although this endpoint was achievable with 12 h and 36 h exposures to ciprofloxacin at 5 pg/ml (a concentration slightly higher than the attainable serum level of this agent - 3 pg/ml [15]), it was not always possible to completely eradicate viable organisms from the biofilms. It is hoped that the present results will provide a better rationale for future clinical trials aimed at preventing or eradicating highly virulent uropathogens in biofilm mode from patients with infected urinary prosthetic devices.
Acknowledgements
This work was supported by a grant from Bayer Pharmaceutical Canada and by NSERC. Statistical help was provided by Mr. James MacMillan. The advice given by Dr. David Vaughan and Dr. Martin Preston was greatly appreciated. We thank Mr. Kan Lam for equipment coordination and Mr. Ushi for fixation and examination of studs for SEM.
References Fig. 6. SEM of stud examined 43 h after colonization with P. aeruginosa and 36 h exposure to 5 mg/ml ciprofloxacin. No adherent bacteria are evident. Bar: 5 pm.
performed their experiments under conditions of slow growth rate and iron restriction in a chemostat, it is likely that the high level of susceptibility which they noted was attributed to these experimental conditions. Ciprofloxacin at low (sub-MIC) and high ( > 100 pg/ml) concentrations has been shown to kill young ( < 24 h) biofilms on device and cell surfaces [ 12,131. The present data demonstrated 59-99.95% reduction in viability after only 4 h exposure to 5 pg/ml ciprofloxacin. While individual uropathogenic isolates will differ in their susceptibility to antimicrobial agents, it is essential that the bacterial retention conditions existing in the host and on the device are appreciated, and that those are accounted for when administering antimicrobial treatment in vivo. The fact that low concentrations of ciprofloxacin were required to effect a 99.9% reduction in the viability of P. aeruginosa biofilms, indicates the potential for lower dosage regimens to be implemented clinically. This is particularly so for antibiotics, such as ciprofloxacin, which is cleared by the kidneys and maintains high concentrations in the urine. Another advantage for the fluoroquinolone, compared to tobramycin, is its safety in neutropenic patients.
111Costerton
JW. Irvin RT. Cheng KJ. The bacterial glycocalyx in nature and disease. Annu. Rev. Microbial. 1981;35:2999324. 121Nickel JC, Ruseska I, Wright JB, Costerton JW. Tobramycin resistance of Pseudomonas aeruginosa cells growing as a biofilm on urinary catheter material. Antimicrob. Agents Chemother. 1985;27:619-624. JW, Cheng KJ, Geesey GG, Ladd Tl, Nickel JC. [31 Costerton Dasgupta M, Marrie TJ. Bacterial biofilms in nature and disease. Annu. Rev. Microbial. 1987;41:435-464. urinary tract [41 Hustinx WNM, Verbrugh HA. Catheter-associated infections: epidemiological, preventive and therapeutic considerations. Int. J. Antimicrob. Agents 1994;4:117- 123. [51 Warren JW. Muncie HL, Berquist, EJ, Hoopes JM. Sequelae and management of urinary infections in the patient requiring chronic catheterization. J. Ural. 1981;125:1-8. T. Uete T, Kawada T. Gkuma A. Resistance to 161 Furusawa cefsulodin and gentamycin in five areas of Japan between 1980 and 1983. J. Antimicrob. Chemother. 1986;17:7555762. of antimicrobial resis[71 Murray BE. Problems and mechanisms tance. Infect. Dis. Clin North Am. 1989:3:423-439. JW. Fouling biofilm development in PI McCoy WF, Costerton tubular flow systems. Dev. Ind. Microbial. 1982;23:551-558. [91 Zufferey J, Rime B, Francioli P, Bille J. Simple method for rapid diagnosis of catheter-associated infection by direct acridine orange staining of catheter tips. J. Clin. Microbial. 1988;26:175177. UOI Anwar H. van Biesen T, Dasgupta M, Lam K, Costerton JW. Interaction of biofilm bacteria with antibiotics in a novel in vitro Chemother. chemostat system. Antimicrob. Agents 1989;33:1824-1826. s 11 Gilbert P, Collier PJ, Brown MRW. Influence of growth rate on susceptibility to antimicrobial agents: biofilms, cell cycle, dormancy and stringent response. Antimicrob. Agents Chemother. 1990;34:186551868. u21 Evans DJ, Allison DG. Brown MRW, Gilbert P. Susceptibility
256
C.A.K. Preston et al. I International Journal of Antimicrobial Agents 7 (1996) 251-2.56
of Pseudomonas aeruginosa and Escherichia coli biofilms towards ciprofloxacin: effect of specific growth rate. J. Antimicrab. Chemother. 1991;27:177-184. [13] Reid G, Dafoe L, Delaney G, Lacerte M, Valvano M, Hayes KC. Use of adhesion counts to help predict symptomatic infection and the ability of fluoroquinolones to penetrate bacterial biofilms on the bladder cells of spinal cord injured
patients. Paraplegia 1994;32:468-472. [14] Pearson RD, Steigbigel RT, Davis HT, Chapman SW. Method for reliable determination of minimal lethal antibiotic concentration. Antimicrob. Agents Chemother. 1980;18:699-708. [15] Andriole VT. Quinolones. In G.L. Mandell, R.G. Douglas. Jr., and J.E. Bennet (ed.), Principles and practices of infectious disease. Churchill Livingstone, New York, pp. 3344345, 1990.
,,I,“\\,
III
Antimicrobial Agents ELSEVIER
International Journal of Antimicrobial Agents 7 (1996) 251-256
Pseudomonas aeruginosa biofilms are more susceptible to ciprofloxacin
than to tobramycin Carol A.K. Preston”, Antoine E. Khouryb, Gregor Reid”ad,*, Andrew W. Bruce”, J. William Costertone “Centre for Biomaterials Infection Research, Toronto General Division, 200 Elizabeth Street, Toronto, Ontario, Canada MSG 2C4 bDivision of Urology, Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, Canada MSG 1X8 ‘Department of Microbiology and Immunology, The University of Western Ontario, Toronto, Canada dThe Lawson Research Institute, London, Ontario, Canada ‘Montana State University, Butte, Montana, USA
Accepted 30 July 1996
Abstract
The objective of this study was to determine and compare the biofilm elimination concentrations (BEC: the concentration which reduced the viability of biofilm organisms by at least 99.9%) of ciprofloxacin and tobramycin for Pseudomonas aeruginosa, a common cause of nosocomial biomaterial-related infections. Bacterial biofilms were produced in the modified Robbins device using continuous culture flow at 60 ml/h for 40-44 h, and the sessile organisms were then exposed to either ciprofloxacin or tobramycin at a range of concentrations for 12 or 36h. The BEC of ciprofloxacin was 5 pgg/ml for the 12 and 36 h treatments, a value 10 x greater than the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC). In contrast, the BEC of tobramycin was > 100 ,ug/ml after 12h and 75 ug/ml following 36 h of drug exposure, that is 75-100 x the MIC and MBC. The results demonstrated that the BEC is a more suitable indicator of the antibiotic susceptibility of P. aeruginosa biofilms than the MIC and MBC. Ciprofloxacin was significantly more effective than tobramycin in the treatment of P. ueruginosu adherent to biomaterials. With respect to clinical application, if the intention of antibiotic use is to eradicate bacteria adherent to devices, only biofilm-active agents should be used. Copyright 0 1996 Elsevier Science B.V. Keywords:
Biofilms; Pseudomonas;
Ciprofloxacin;
Tobramycin
1. Introduction Infections of prosthetic devices represent an important clinical problem, which is compounded by the fact that the organisms which bind to, and multiply on, the surfaces of the biomaterials, form a thick biofilm which protects them against the host’s immune system and antimicrobial agents [ l-41. Antimicrobial agents such as tobramycin routinely used in hospital settings to * Corresponding author, H414 Lawson Research Institute, St. Joseph’s Health Centre, 268 Grosvenor Street, London, Ontario, N6A 4V2, Canada. Tel.: + 519 646 6100 ext. 5256; fax: + 519 646 61 lo:, e-mail: [email protected] 0924-8579/96/$32.00
treat infections, are generally ineffective clinically in eliminating prosthesis-associated infections 151. At present, the only way to eliminate such infections is to remove and replace the infected device. This process can be traumatic to the patient and expensive to the health care system. Pseudomonas aeruginosa is particularly pathogenic when associated with the nidus of an implanted medical device, and it is often resistant to antimicrobials, especially those used to treat nosocomial infections [6,7]. The study carried out herein was designed to compare the efficacy of ciprofloxacin, a potent fluoroquinolone agent, against tobramycin on the viability of P. aeruginosa in a biofilm.
Copyright 0 1996 Elsevier Science B.V. All rights reserved
PII SO924-8579(96)00330-S
252
C.A.K. Preston et al. 1 international Journal of’ Antimicrobial
Agents 7 (1996) 251-256
2. Materials and methods
2.5. Measurement of bacterial viability
2.1. Bacterial strain
Adherent bacteria were removed from the studs by ultrasonication for 7 min (Branson 1200 ultrasonic cleaner) in 5 ml of PBS. Bacteria released from the stud surface were diluted in PBS and plated with four dilutions per stud, in duplicate, on nutrient agar. Cultures were incubated for 16-24 h at 37°C after which viable counts were made.
The clinical isolate of P. aeruginosa was obtained from an infected urine sample from a patient at the Hospital for Sick Children, Toronto, Canada. 2.2. Antimicrobial agents Intravenous tobramycin (tobramycin sulphate, Eli Lilly, Toronto) and ciprofloxacin (lactate salt, kindly provided by Bayer Pharmaceuticals, Toronto) were used. 2.3. Antibiotic susceptibility of planktonic organisms The in vitro minimum inhibitory concentrations (MICs) and minimum bactericidal concentrations (MBCs) of ciprofloxacin and tobramycin were measured with planktonic P. aeruginosa cells using the standard macrobroth dilution method (NCCLS) used in the Hospital for Sick Children. The MIC and MBC values were 0.5 pg/ml and 1.0 pg/ml for ciprofloxacin and tobramycin, respectively. 2.4. Colonization of MRD and drug treatment The modified Robbins device (MRD) has been developed [2,8] as a means of in vitro testing of the efficacy of antimicrobial agents on sessile (biofilm) organisms. In the present study, the MRD was set up largely as described by Nickel et al. [2]. The sampling ports were furnished with metal studs in the lumen which were exposed to an actively growing P. aeruginosa culture. The bacteria were grown in nutrient broth (Difco Laboratories, Detroit, MI) in a 4-l reservoir held in a 37°C water bath and the culture was pumped through the device for 40-44 h at a rate of approximately 60 ml/h. After initial bacterial colonization, two studs were removed and replaced with sterile ones, washed twice with 10 ml PBS to remove non-adherent bacteria, and processed to measure cell viability as described, below. The inflow tubing to the MRD was replaced with a fresh sterile one and the culture flask was exchanged for one containing ciprofloxacin at 2.5 to 10 pg/ml or tobramycin at 50 to 100 ug/ml in nutrient broth. P aeruginosa biofilms were exposed to the antibiotics for 12 or 36 h. Two studs were sampled at the following intervals during the incubation with the drugs: for 12 h exposure, every 4 h; for 36 h exposure, every 12 h. Two concentrations of antibiotic were tested simultaneously with two devices colonized with the same culture.
2.6. Acridine orange staining of A4RD studs The studs were subjected to bacterial colonization and rinsed to remove non-adherent bacteria, as described above. One half of the studs was sonicated as noted above. The other half was rinsed twice more with PBS and were then fixed by exposure to 10% formalin (Fisher Scientific, Toronto) at 37” C for 1 h. Samples were then rinsed thrice with PBS at room temperature and stained with Acridine Orange, prepared as described by Zufferey et al. [9], for 3 min at room temperature. The stained studs were then rinsed thrice with distilled water, air dried and affixed to glass slides with Silastic medical adhesive silicone Type A (Dow Corning Corp., Midland MI). Stained stud surfaces were examined under a fluorescence microscope (Zeiss) at a magnification of x 1000 with oil immersion to assess the presence of bacteria. These steps were carried out in triplicate prior to and after the sonication procedure to verify the efficacy of ultrasonication at removing the adherent bacteria from the surface of the studs. Studs not exposed to bacteria were also stained as a control.
,021. . . 0
’
’
5
.
.
.
.
’
*
.
10
.
’ 15
Period of Exposure to Antibiotic (hours) Fig. 1. Effect of 12 h exposure of tobramycin on P. aeruginosa biofilm bacteria viability. Forty 44-h old biofilms were exposed to 75 ( n ) or 100 (0) mg/ml tobramycin in the MRD for 12 h. At the intervals noted, samples were removed and processed to assess bacterial viability. Each data point represents the weighted mean of 3-4 dilutions for duplicate samples from three experiments.
253
C.A.K. Preston et al. 1 International Journal of’ Antimicrobial Agents 7 (1996) 251-256
._
showed that 99.98% of the viable sessile bacteria were removed from the stud surface by sonication. The studs exposed to bacteria for 40-44 h and sonicated for 7 min were stained with Acridine Orange and assessed microscopically for the presence of bacterial cells. These stud surfaces were compared to those of stained colonized studs which were not subjected to sonication. Results showed that the sonicated studs had only background levels of fluorescence while the surfaces which were not sonicated showed the presence of thick biofilms. These results indicated that sonication is effective, under the conditions utilized, for removing P. aeruginosa in a biofilm from the metal MRD studs. 0
5
10
15
Period of Exposure to Antibiotic (hours) Fig. 2. Effect of 12 h exposure of ciprofloxacin on P. aeruginosa biofilm bacteria viability. Forty 44-h old biofilms were exposed to 2.5 (O), 5.0 (m) or 10 (A) mg/ml ciprofloxacin in the MRD for 12 h. At the intervals noted, samples were removed and processed to assess bacterial viability. Each data point represents the weighted mean of 3-4 dilutions for duplicate samples from 3-4 experiments.
2.7. Quantljication of viable bacteria remaining on studs ufter sonication
Three studs, subjected to bacterial colonization and rinsed to remove non-adherent bacteria, were sonicated as noted above and rinsed twice more with PBS. Flamed forceps were used as the formerly colonized stud surface was streaked across the surface of a nutrient agar plate. Viable counts were made after a 24 h incubation period at 37°C.
3.2. Biojilm elimination concentration at 12 h The effect of tobramycin
and ciproflaxacin
on P.
aeruginosa biofilms was tested by exposing 40-44 h old
biofilms to a range of concentrations of these agents for 12 h. The results are presented in Figs. 1 and 2. The biofilm elimination concentration of the antimicrobial agents was determined from this data as the concentration required to achieve at least a 99.9% reduction in the viability of biofilm bacteria (confidence interval of 95%; one-tailed test). The BEC of ciprofloxacin was found to be 5 pug/ml. This value is 10 x the planktonic MIC and MBC value for this organism. With 100 lug/ml an average reduction of 42% (range O-100%) ensued, The tobramycin BEC was therefore determined to be > 100 pg/ml, which is more than 100 x the MIC and MBC for planktonic cells.
2.8. Scanning electron microscopy of biojilm bacteria
Studs were subjected to bacterial colonization, with or without subsequent antibiotic exposure of 10 pug/ml ciprofloxacin or 75 pg/ml tobramycin for 24 or 36 h, and rinsed to remove non-adherent bacteria. Bacteria were fixed in 5% glutaraldehyde in cacodylate buffer (0.1 M, pH 7.0) at room temperature. Samples were then exposed to two fresh solutions of cacodylate buffer for 20 min at room temperature and placed in a third buffer ‘wash’ and stored at 40°C prior to dehydration steps (performed as noted by Nickel et al. [2]) and examination in a Hitachi 5540 scanning electron microscope.
-I
\ 8
10’ ! 0
3. Results 3.1. Assessment of efficacy of sonication at removal of sessile bacteria from stud surface The number of viable bacteria remaining on the stud surface after sonication was quantified. The results
10
20
30
40
Period of Exposure to Antibiotic (hours) Fig. 3. Effect of 36 h exposure of tobramycin on P. aeruginosa biofilm bacteria viability. Forty 44-h old biofilms were exposed to 50 (0) or 75 (m) mg/ml tobramycin in the MRD for 36 h. At the intervals noted, samples were removed and processed to assess bacterial viability. Each data point represents the weighted mean of 3-4 dilutions for duplicate samples from four experiments.
254
C.A.K. Preston et al. / International Journal of Antimicrobial Agenrs 7 (1996) 251-256
4. Discussion
Period of Exposure
to Antibiotic (hours)
Fig. 4. Effect of 36 h exposure of ciprofloxacin on P. aeruginosa biofilm bacteria viability. Forty 44-h old biofilms were exposed to 5.0 (0) or 10 (W) mg/ml ciprofloxacin in the MRD for 36 h. At the intervals noted, samples were removed and processed to assess bacterial viability. Each data point represents the weighted mean of 3-4 dilutions for duplicate samples from five experiments.
3.3.
Biojilm elimination concentration at 36 h
To determine whether a longer exposure to the antimicrobial agents could eradicate the biofilm viability or reduce the BEC, P. aeruginosa biofilms were exposed to a range of drug concentrations for 36 h. The results are shown in Figs. 3 and 4. The BEC of ciprofloxacin under these conditions was 5 pg/ml, as was found after the 12 h treatment. Total elimination of viable biofilm cells was achieved in three out of five experiments with the BEC. The BEC of tobramycin was found to be 75 ,ug/ml, a value 75 x the MIC and MBC determined for planktonic cells.
The results showed that a virulent strain of uropathogenie bacteria existing in biofllms is more resistant to antimicrobial agents than planktonic counterparts. More importantly, the latest results showed that in vitro planktonic MIC and MBC levels of antimicrobial agents are inadequate to eliminate biofilm organisms associated with biomaterials. Indeed, the BECs of ciprofloxacin and tobramycin were found to be 10 x or 75-100 x the MIC and MBC values, respectively. The BEC is a more suitable indicator of the antibiotic susceptibility of P. aeruginosa biofilms. Unless provided with the true concentration of drug required to eradicate biofilms, a physician may be unable to adequately treat the patient. In view of previous work showing P. aeruginosa was resistant to 1000 pg/ml tobramycin, the latest findings are not completely unexpected [2]. However, our results are in contrast to those reported by Anwar et al. [lo] where the treatment of P. aeruginosa biofilms with 10 pgg/ml of tobramycin for 3 h was sufficient to reduce bacterial viability by > 99.9%. It is believed that growth rate, which is dependent on the availability of nutrients, can affect the susceptibility of bacteria to antimicrobial agents [ll]. Since Anwar and colleagues
3.4. Electron microscopical examination of metal studs colonized with P. aeruginosa and treated with antimicrobial agents SEM examination demonstrated that after a 44 h exposure of the studs to the P. aeruginosa culture a thick adherent bacterial biofilm had formed (Fig. 5). It should be pointed out that the biofllms did not universally and totally cover the surfaces, thereby explaining why only 100000 or so organisms were found per stud. Following 24 h exposure to 5 pg/ml ciprofloxacin, a marked decrease in the extent of the biofilm was noted, and no adherent organisms were evident with 36 h exposure (Fig. 6). After 24 h treatment with 75 pgg/ml tobramycin much of the biofilm was gone, and after a further 12 h, the biofilm was removed. All SEM results correlated well with the viable plate counts.
Fig. 5. SEM of a stud examined 44 h after colonization with P. The surface is covered by a thick bacterial biofilm. Bar: 5 ,um.
aeruginosa under flow conditions.
C.A.K. Preston et al.
I International Journal of Antimicrobial Agents 7 (1996) 2.51-256
255
The reduction of viability of 99.9% was used to determine the BEC values as this is the accepted killing endpoint in bactericidal tests [14]. Although this endpoint was achievable with 12 h and 36 h exposures to ciprofloxacin at 5 pg/ml (a concentration slightly higher than the attainable serum level of this agent - 3 pg/ml [15]), it was not always possible to completely eradicate viable organisms from the biofilms. It is hoped that the present results will provide a better rationale for future clinical trials aimed at preventing or eradicating highly virulent uropathogens in biofilm mode from patients with infected urinary prosthetic devices.
Acknowledgements
This work was supported by a grant from Bayer Pharmaceutical Canada and by NSERC. Statistical help was provided by Mr. James MacMillan. The advice given by Dr. David Vaughan and Dr. Martin Preston was greatly appreciated. We thank Mr. Kan Lam for equipment coordination and Mr. Ushi for fixation and examination of studs for SEM.
References Fig. 6. SEM of stud examined 43 h after colonization with P. aeruginosa and 36 h exposure to 5 mg/ml ciprofloxacin. No adherent bacteria are evident. Bar: 5 pm.
performed their experiments under conditions of slow growth rate and iron restriction in a chemostat, it is likely that the high level of susceptibility which they noted was attributed to these experimental conditions. Ciprofloxacin at low (sub-MIC) and high ( > 100 pg/ml) concentrations has been shown to kill young ( < 24 h) biofilms on device and cell surfaces [ 12,131. The present data demonstrated 59-99.95% reduction in viability after only 4 h exposure to 5 pg/ml ciprofloxacin. While individual uropathogenic isolates will differ in their susceptibility to antimicrobial agents, it is essential that the bacterial retention conditions existing in the host and on the device are appreciated, and that those are accounted for when administering antimicrobial treatment in vivo. The fact that low concentrations of ciprofloxacin were required to effect a 99.9% reduction in the viability of P. aeruginosa biofilms, indicates the potential for lower dosage regimens to be implemented clinically. This is particularly so for antibiotics, such as ciprofloxacin, which is cleared by the kidneys and maintains high concentrations in the urine. Another advantage for the fluoroquinolone, compared to tobramycin, is its safety in neutropenic patients.
111Costerton
JW. Irvin RT. Cheng KJ. The bacterial glycocalyx in nature and disease. Annu. Rev. Microbial. 1981;35:2999324. 121Nickel JC, Ruseska I, Wright JB, Costerton JW. Tobramycin resistance of Pseudomonas aeruginosa cells growing as a biofilm on urinary catheter material. Antimicrob. Agents Chemother. 1985;27:619-624. JW, Cheng KJ, Geesey GG, Ladd Tl, Nickel JC. [31 Costerton Dasgupta M, Marrie TJ. Bacterial biofilms in nature and disease. Annu. Rev. Microbial. 1987;41:435-464. urinary tract [41 Hustinx WNM, Verbrugh HA. Catheter-associated infections: epidemiological, preventive and therapeutic considerations. Int. J. Antimicrob. Agents 1994;4:117- 123. [51 Warren JW. Muncie HL, Berquist, EJ, Hoopes JM. Sequelae and management of urinary infections in the patient requiring chronic catheterization. J. Ural. 1981;125:1-8. T. Uete T, Kawada T. Gkuma A. Resistance to 161 Furusawa cefsulodin and gentamycin in five areas of Japan between 1980 and 1983. J. Antimicrob. Chemother. 1986;17:7555762. of antimicrobial resis[71 Murray BE. Problems and mechanisms tance. Infect. Dis. Clin North Am. 1989:3:423-439. JW. Fouling biofilm development in PI McCoy WF, Costerton tubular flow systems. Dev. Ind. Microbial. 1982;23:551-558. [91 Zufferey J, Rime B, Francioli P, Bille J. Simple method for rapid diagnosis of catheter-associated infection by direct acridine orange staining of catheter tips. J. Clin. Microbial. 1988;26:175177. UOI Anwar H. van Biesen T, Dasgupta M, Lam K, Costerton JW. Interaction of biofilm bacteria with antibiotics in a novel in vitro Chemother. chemostat system. Antimicrob. Agents 1989;33:1824-1826. s 11 Gilbert P, Collier PJ, Brown MRW. Influence of growth rate on susceptibility to antimicrobial agents: biofilms, cell cycle, dormancy and stringent response. Antimicrob. Agents Chemother. 1990;34:186551868. u21 Evans DJ, Allison DG. Brown MRW, Gilbert P. Susceptibility
256
C.A.K. Preston et al. I International Journal of Antimicrobial Agents 7 (1996) 251-2.56
of Pseudomonas aeruginosa and Escherichia coli biofilms towards ciprofloxacin: effect of specific growth rate. J. Antimicrab. Chemother. 1991;27:177-184. [13] Reid G, Dafoe L, Delaney G, Lacerte M, Valvano M, Hayes KC. Use of adhesion counts to help predict symptomatic infection and the ability of fluoroquinolones to penetrate bacterial biofilms on the bladder cells of spinal cord injured
patients. Paraplegia 1994;32:468-472. [14] Pearson RD, Steigbigel RT, Davis HT, Chapman SW. Method for reliable determination of minimal lethal antibiotic concentration. Antimicrob. Agents Chemother. 1980;18:699-708. [15] Andriole VT. Quinolones. In G.L. Mandell, R.G. Douglas. Jr., and J.E. Bennet (ed.), Principles and practices of infectious disease. Churchill Livingstone, New York, pp. 3344345, 1990.