Antimicrobial cream susceptibility testing

Burns (1985) 12,91-96

Printed in

Antimicrobial

GreatBritain

91

cream susceptibility

J. M. Conly, S. Byrne, J. McLeod,

S. Hoban, G. Robertson

Deoartments of Medicine, Surgery, Medical Microbiology B&n Unit, University of Manitoba, Canada

Summary Utilizing a recently described rapid needle extrusion method for determining the sensitivities of burn wound isolates, we have compiled data on the resistance patterns of over 250 isolates from our burn unit. Major isolate groups were Staphylococcus aureus, l?w_domoand Enterr1a.sauruginma. various Enterobacteriaceae cocci. Excellent correlation was exhibited between the inhibitory zone sizes and the minimal inhibitory concentration for 120 organisms tested. Utilizing the needle extrusion sensitivities to facilitate the selection of the hurn creams, a significant reduction in the microbiologic flora of the burn wound was noted. The utilization of this technique in the selection of burn creams deserves controlled trials to assess whether changes in topical therapy might alter clinical outcome.

INTRODUCTION DCIKINC; the past two decades, topical antimicrobial therapy of burns has become standard practice. In the early 1960s. gentamicin, 0.5 per cent silver nitrate and mafenide acetate were introduced and in lY68 silver sulphadiazine became available (Lindberg et al., 1965; Moyer et al., lY65: Fox, 1968; Stone, 1971). Silver sulphadiazine has been reported to have few side effects and is utilized exclusively in many burn units. Carr et al. (1973) determined the in vitro antibactericidal activity of silver sulphadiazine to burn wound isolates and found all strains tested to be inhibited by levels of the drug achieved topically. However, several reports on the of silver resistant Pseudomonas emergence

Dr J. Conly is supported by a Fellowship from the Medical Research Council of Canada. Presented, in part, at the X&h Annual Meeting of the American Society for Microbiology Meeting, March l%G, St. Louis, Missouri.

testing and A. R. Ronald

and the Health Sciences

(Bridges et al, 197Y) and sulphadiazine resistant Gram-negative bacilli (Bridges and Lowbury. 1977; Gayle et al., 1978; Hendry and Stewart. 1979) have appeared. These reports have prompted the use of in vitro methods to determine topical agent sensitivity. Nathan et al. (1978) noted a reduction or elimination of bacteria in wounds where treatment was based on in vitro sensitivities utilizing an agar cup diffusion method. However, this technique is time consuming and in a busy general microbiology laboratory may not lend itself to being adopted as a routine procedure. Rodeheaver et al. (lY80) described a simple inexpensive needle extrusion method for performing in vitro cream susceptibility testing and found excellent correlation with the agar cup diffusion method. These preliminary reports are encouraging and we report here the results of our evaluation of antimicrobial cream sensitivities. Our study was initiated with the following objectives: (1) to assess the correlation of zone sizes produced by the needle extrusion method with the minimal inhibitory concentration of the antimicrobial agent in the cream; (2) to report the sensitivity patterns of major pathogen groups in our burn unit to specific burn creams; (3) to assess the clinical utilization of the test when introduced as routine and whether a reduction in bacteria occurred in wounds treated on the basis of in vitro cream sensitivity testing. MATERIALS AND METHODS Bacteria tested The correlation of the needle extrusion method with the standard agar dilution method was determined utilizing I17 blood culture isolates. Additional isolates were obtained from patients with burns in our unit.

Burns (1985) Vol. 12/No. 2

92

Antimicrobial

creams

were assessed for sensitivity to the four topical antimicrobial creams, gentamicin, silver sulphadiazine, nitrofurazone and mafenide acetate. The burn wound was cultured twice weekly by rolling a dry swab several times over the area to be sampled and in vitro testing was then performed. Any residual topical cream was removed prior to obtaining the culture. Quantitative cultures from burn wound biopsies were not routinely done in this study. A grading system, based on a standard method of quantitating bacteria in the Health Sciences Centre Clinical Microbiology Laboratory, was instituted l+ (1-12 colonies), 2+ (13-40 colonies). A reduction or increase was defined as a difference of G or 2 2 grades, respectively and no change as a difference of +l grade. Charts of all patients with any burn admitted to our adult burn unit were reviewed. Patients were selected for analysis according to the following criteria: (1) cream sensitivities had been done; (2) appropriate change in cream had been instituted; (3) follow-up cultures from the same area of the bum wound were obtained; (4) the burn wound was intracolonized >2+ ; (5) no broad spectrum venous antimicrobials had been administered for sepsis from any source so that possible confounding would be avoided; and (6) the substituted cream must have been used for at least 3 consecutive days. An appropriate selection of cream was based upon choosing the cream with the greatest or second greatest zone of inhibition on in vitro testing. In our burn unit silver sulphadiazine is utilized as the primary burn cream unless otherwise specified.

Each of the topical antimicrobial creams accepted for use in our hospital for the treatment

of burns was included in our test system. The gentamicin sulphate (0.1 per cent), silver sulphadiazine (1.0 per cent), mafenide acetate (8-S per cent) and nitrofurazone (0.2 per cent) were all commercially available. Cream sensitivity testing Bacterial isolates were tested for sensitivity to topical antimicrobial creams using the needle extrusion method described by Rodeheaver et al. (1980). Briefly, Meuller Hinton agar plates were inoculated by swabbing the surface from a suspension of the test organism (10’ CFUlml). Using a 19 gauge needle, each of the four topical creams was laid onto the surface of the agar as a thin line approximately 25 mm in length. After 18 h of incubation at 37°C the total zone of growth inhibition at the midpoint and perpendicular to the line of cream was measured. If the zone was not clearly delineated the inhibitory zone was taken as that which had approximately 80 per cent inhibition of growth. Minimum inhibitory concentrations (MIC) Bacterial isolates were grown overnight at 37” C on blood agar. Each isolate was then inoculated into 4 ml of Mueller Hinton broth and incubated at 37” C for 4 h. The turbidity of the broth was then adjusted to approximately 10sCFU and further diluted 100 fold. The MIC of mafenide acetate, nitrofurazone, gentamicin and silver sulphadiazine was determined on 117 blood isolates using standard agar dilution methods (Lenette, 1980).

RESULTS Correlation of inhibitory zone size to MIC A total of 117 bacterial blood isolates were tested for sensitivity to the topical antimicrobial creams

Clinical analysis During a la-month period, bacteria isolated from adult patients (a18 years of age) in our burn unit

Table 1. Minimum isolates

concentration

of antimicrobials

Gentamicin (@ml) Organism

(nl

P. aeruginosa (24) S. aureus (24) S. faecalis (24) Enterobacter spp. (13) E. co/i (12)

M&I 2 0.125 4 0.25 0.5

Ml& 2 0.25 8 4 4

required to inhibit growth of blood culture bacterial

Silver sulphadiazine Wmll

Mafenide acetate h@nll

Ml’%,

MlCw

MlC,

MICw

64 64 64 64 64

64 64 64 64 64

1024 512 256 4096 4096

2048 512 512 4096 4096

Nitrofurazone hg/mll Ml& 128 4 16 32 8

M&O 128 8 64 64 16

Conly et

al.: Burn cream

susceptibility testing

93

Fig. 2. Relationship between MIC of silver sulphadiazine and inhibitory zone size (mean fl s.d.).

Fig. 3. Relationship between MIC of mafenide acetate and inhibitory zone size (mean fl s.d.).

Fig, 4. Relationship between MIC of nitrofurazone and inhibitory zone size (mean 2~1s.d.).

utilizing the needle extrusion method and the standard agar dilution method. The sensitivity to the antimicrobial agents in burn creams of the major bacterial groups is presented in Table 1. Gentamicin was most effective, conversely the MIC of mafenide acetate was remarkably high against all bacteria tested. The effectiveness of nitrofurazone was greatest against S. aureu.~ and least against P. aeruginosa. Silver sulphadiazine had consistent MI&, of 64 pg/l against most major isolates. The relationship of the MIC to the inhibitory zone size for each of the antibiotics is shown in Figs. l-4. Significant correlations were observed between these parameters for all the creams tested (P
silver sulphadiazine inhibitory zone sizes with MIC was lower and reflects the large number of isolates which had MICs of 64 pg/ml. Cream sensitivity profiles of bacterial burn wound isolates Data on the mean inhibitory zone sizes of the four major bacterial groups was compiled and is presented in Table II. On the basis of these measurements, resistance to topical creams is evident in P. aeruginosu to nitrofurazone, Enterobacteriaceae to mafenide acetate and S. faecalis to gentamicin. Due to the apparent poor diffusibility of silver sulphadiazine through agar, inhibitory zone sizes

Bums (1985) Vol. 12/No. 2

94

Table II. Mean inhibitory zone size (mm) around topical burn creams against major burn wound isolates

Organism

Gentamicin

In)

P. aeruginosa (26) S. aureus (157) Enterobacter spp. (67) S. faecalis (34)

14 19 20 5

Table 1/L Major burn wound organisms in patients with evaluable antimicrobial cream switches 2-8 weeks*

>8 weeks

No. of burn wounds

9

15

Organism P. aeruginosa S. aureus S. epidermidis E. coli E. cloace

6 4 1 1 1

8 7 -

Length of stay

??

Mafenide acetate 19 13 3 11

Nitrofurazone

Silver sulphadiazine

1 21 10 9

8 7 5 5

Tab/e IV. Microbial response following topical antimicrobial Microbial

changes in

response*

Reduction No change Increase No changes Changes instituted

4 15

9 7

12 0

* P
More than one organism present per patient.

of most isolates were small. However, within the four major groups, the proportion of isolates growing within 4mm of the cream was significantly different. Only 4 per cent of S. aureus and P. aeruginosa isolates grew within this zone compared to 34 per cent of the Enterobacteriaceae and 53 per cent of S. fuecalis isolates.

Quantitative changes in colonization after switching burn creams A total of 92 charts were reviewed during 1983, representing all the patients admitted to the Burn Unit during this period. A total of 22 evaluable switches in topical therapy were made from a total of 65 switches. Lack of follow-up culture information or concomitant use of broad spectrum parenteral antibiotics accounted for the vast majority of inevaluable switches. The bacteriology of the wounds for those switches which were deemed evaluable in each group is illustrated in Table III. Mixed cultures were present in several cases. The majority of the isolates were P. aeruginosa and S. aureus. Utilizing similar criteria as outlined for assessing an evaluable switch (with the exception of an appropriate choice of cream based on the largest zone size), the burn wounds on which silver sulphadiazine was continued despite having a

zone size 6 10 mm were evaluated and compared to follow-up cultures which had appropriate switches. After instituting changes in topical antimicrobial therapy on the basis of in vitro tests (Table IV), a significant number of burns (68 per cent) had a demonstrable decrease in the number of colonizing organisms compared to patients who did not have creams changed (16 per cent).

DISCUSSION In this study, we have used a simple method to determine the sensitivity of burn wound bacterial isolates to topical antimicrobial therapy. This method, as described by Rodeheaver et al. (1980), requires no special apparatus and requires approximately the same time as a standard Kirby Bauer disc susceptibility assay. The inhibitory zone size of the antimicrobial cream is determined 18 h after planting, making the total time required to send out a final report after receiving the swab under 48 h. In the present study we have found significant correlation between the MIC and inhibitory zones as measured with needle extrusion. This relationship provides evidence that the zone size for nitrofurazone, gentamicin and sulfamylon (mafenide acetate) and probably silver sulphadiazine truly reflect sensi-

95

Conly et al.: Burn cream susceptibility testing tivity to the drug and is not a reflection of possible antimicrobial effects of the carrier. A paucity of information exists regarding the diffusion of antimicrobials through the burn data presented by Barry wound. However. (1983) suggest the dermis, which may be exposed after a severe burn. favours diffusion of aqueous solvents. It therefore seems likely that diffusion of antibiotics in agar is similar to diffusion into a fresh partial thickness skin loss burn. The sensitivity patterns as indicated by the inhibitory zone size around the topical creams illustrates significant differences between the four major groups of bacteria. On the basis of zone size, nitrofurazone generally appears to be most active against S. uuwus yet has minimal activity against P. rreruginosa. Mafenide acetate produces the largest mean zone of inhibition against P. ueruginosa and a much smaller mean zone size against Enterobacteriaceae. Most S. faecalis isolates appear to be resistant to gentamicin and silver sulphadiazine. The sensitivity of this group to nitrofurazone and mafenide acetate is extremely variable. Such variability could indicate the necessity for topical antimicrobial sensitivity testing to ensure appropriate selection of creams. Of the creams tested, the Enterobacteriaceae were most sensitive to gentamicin cream and nitrofurazone. A wide range of zone sizes around nitrofurazone also illustrates the necessity for cream sensitivity testing. Since the introduction of cream sensitivity testing in our burn unit the information provided has been utilized extensively by the clinicians. The utilization of this procedure is particularly evident in those patients with larger burns who had prolonged hospital stays (X3 weeks) where approximately three switches were made per patient. A reduction in burn wound colonization in 68 per cent of cases and the absence of an increase in colonization following appropriate switches is encouraging. In comparison, in the instances where silver sulphadiazine was continued despite in vitro evidence indicating it to be less effective, a reduction in bacterial count was noted in 16 per cent and an increase in 48 per cent. However, no controlled studies are available to indicate whether changes in topical agents based on in vitro testing would improve clinical outcome in terms of burn wound sepsis, length of hospital stay, percentage of graft take, or mortality. Gayle et al. (1978) noted that changing to mafenide acetate from silver sulphadiazine when the patient was already ill had no effect on clinical outcome. However, with proper selection of antimicrobial creams, it is possible less burn

wound sepsis would occur. Both Nathan et al. (1978) and Rodeheaver et al. (1980) indicate a reduction or elimination of bacteria in wounds where the creams were selected based on in vitro sensitivity testing. This simple method could also facilitate the recognition and elimination of silver sulphadiazine resistant organisms in a burn unit. The potential for the routine monitoring of the susceptibility of the bacterial flora of the burn wound utilizing this method is substantial. It is simple, inexpensive, rapid, and correlates extremely well with the more standard methods. Further controlled clinical trials utilizing this method to facilitate selection of antimicrobial creams are warranted to assess whether clinical outcome may be improved.

REFERENCES Barry B. W. (1983) Dermatological formulations, pcrcutaneous absorption. Drugs and Pharmaceutical Services, Vol. 18. Bridges K., Kidson A. and Lowbury E. J. L. (1979) Gentamicin and silver resistant Pseudomonas in a burn unit. Br. Med. J. 1, 446. Bridees K. and Lowburv E. J. (1977), Drue- resistance in &ion to use of silver sulfadiazine cream in a burn unit. J. C/in. Pathol. 30, 160. Carr H. S., Wlodkowski T. J. and Rosenkranz H. S. (1973) S’ Ilver sulfadiazine: in vitro antibacterial activity. Antimicrob. Agent Chemother. 4, 585. Fox C. L. Jr. (1968) Silver sulfadiazine-a new topical therapy for pseudomonas in burns. Arch. Surg. 96, 184.

Gayle W. E., Mayhall C. G., Lamb V. A., et al. (1978) Resistant Enterobacter cloacae in a burn center: the ineffectiveness of silver sulfadiazine. J. Trauma 18, 317.

Hendry A. T. and Stewart I. 0. (1979) Silver resistant Enterobacteriaceae from hospital . .patients. Can. J. Microbial. 25. 915.

Lenette E. H. (ed.) (1980) Manual of Clinical Microbiology, 3rd kd. ‘W’ashington, American Society for Microbiology, p. 446. Lindberg R. B., Moncrief J. A., Switzer W. E. et al. (1965) The successful control of burn wound sepsis. J. Trauma 5, 601.

Moyer C. A., Brentano L., Gravens D. L. et al. (1965) Treatment of large human burns with 0.5% silver nitrate solution. Arch. Surg. 90, 812. Nathan P., Law E. J., Murphy D. F. et al. (1978) A laboratory method for selection of topical antimicrobial agents to treat infected burn wounds. Burns 4. 177.

Burns (1985) Vol. 12/No. 2

96

Rodeheaver G. T., Gentry S., Saffer L. et al. (1980) Topical antimicrobial cream sensitivity testing. Surg. Gynecol. Obstet. 151, 747. Stone H. H. (1971) Wound care with topical gentamitin. In: Polk H. C. Jr. and Stone H. H. (eds.)

Correspondence Manitoba.

should be addrewed IO: Dr _I. M. Conly,

Canada

R3E

Room

Confemporary Burn Management. Brown, p. 203.

Paper

accepted

530, Basic Medical

18 June

Sciences

Boston:

Little and

1985.

Bldg.,

730 William

Avenue,

0W3.

BRITISH BURN ASSOCIATION The 1986 Annual Meeting will be held at Birmingham University on Thursday, 10 April and Friday, 11 April. On Wednesday, 9 April, there will be a Teaching Symposium on Burn Care. This symposium is designed for casualty medical and nursing staff; the content includes burn epidemiology, pathophysiology, smoke and noxious gas inhalation, together with treatment of major and minor burns including their nursing care. Both meetings are open to non-members. Anticipated costs are &12 for the symposium and f55 (non-members) for the annual meeting. Accommodation is available (f12.50 per night). Full details can be obtained from: Dr J. C. Lawrence, MRC Burns Research Group, Birmingham Accident Hospital, Bath Row, Birmingham B1.5 1NA.

Winnipeg,