- Email: [email protected]
Combined topical and systemic antibiotic prophylaxis in experimental wound infection
MD, Louisville, Kentucky Thimas#. P&m Y lwammt, m, FRCM, LabvilIe, Kentucky wlkm & Cbdb, MD, Louisvilte, Kentucky Mmm C. Rear, Jr., MO, Louisville, Kentucky
aydemic and topical antiofeurgi4wWWnf~on [I-4]. Both of thefie routes haJsbeen ofad&&t&o~havebeenehowntoreducethe
ehta=d . .
prophylactic
,uacommon place of a combination of both topical and systemic ntary ve efnenta onregnnenaare oftenueedincUnic4practice,althoughtherehasyet to be a full and proper evaluation of their clinical effectin evatuation is highlighted by number ofpatients in whom slop despite appropriafe antibiotic . prop&hsie by a single proven route. ContammaW wounds are at particular risk of subsequent infection [II], and previous clinical and experimentai obeerv&ions have suggeeted that a combir&ionoftopicalandsy&emicprophyluxiemay be more effective then either would be aloe@when bact&al contamination of the wound is great [12-
acitracin, and neomycin sulfate Ma?WllrlaMWe@Wa Adult male albino guinea pige that we&bed approxi-
fascia and were extend
14). The aim of this study was to determhae whether
the addition of a topical antibiotic would be benefirnrrgicsiwoundinftionrateswhen a @e&o, a topical antibiotic alone, a&biotic alone, in the preeence of alowlevelofbacter4uW&nination of the wound. A w&&cribed guinea pig model was Fmm~(hl)Eke#nwt~Ikmrd(k*Rlos~ot~Rgsroh, M~.LG@@&IQJnW9,.
theanimalswerekilledandtbeir~wereopenedand inspeet8d for pus, the obeermv epscifiepllyunaware of the treatment groupto whichtBeenin& belonged. A wound infection was diagnosed only in the presenceof grossly visible PUB. Tbe.St@hylococcua aureu8umdinthbbqerimentwee ieolated fmm a hoepital CkllhlWiOf~glanism
ticaee soy agar (BBL Microbiology @&ems, Cockeysviile,
753
TABLE I
QfouD I
II l
lnodum of Staphyhwcws aureus Producing High and Low Wound lnhcth Rates in the CodrolQnwrp(n=30) Staph. aureus Inboduced’ 1.5 x 10’ 1 x 10’0
wounds Infected (n) 6 13
Infection Rate 20% 43%
Measuredas numberof colonyformingunits.
MD). Before inoculation, the Staph. aureus was grown overnight in static trypticase broth (BBL Microbiology Systems) at 37°C and suspended at the desired concentrations in sterile saline solution. Bacterial concentrations were confirmed by plating and overnight incubation at 37% on trypticase soy agar, and were expressed as the number of colony forming units per milliliter of saline solution. Clindamycin (Cleocin Phosphate@, Upjohn Co., Kalamazoo, MI) was the antibiotic used systemically, being administered intramuscularly in a dose of 5 mg/kg 20 minutes before bacterial inoculation of the wound. A mixture of polymyxin B, bacitracin, and, neomycin sulfate (Neosporina powder, Burroughs Wellcome Co., Triangle Park, NC) was used as the topical antibiotic. Fifteen minutes after the bacterial inoculation, 0.5 mg of powder suspended in 0.1 ml of sterile saline solution was placed in the wound. Each wound therefore received 2.5 units of polymyxin B, 0.2 units of bacitracin, and 2.5 pg of neomycin sulphate. When no antibiotic was given, sterile saline solution was administered as a control (0.5 ml given intramuscularly or 0.1 ml placed in the wound, as appropriate). Sixty wounds were examined in each of four treatment groups, a total of 240 wounds being available for subse-
quent analysis. The treatment groups were as follows: (a) no antibiotic administered, (b) topical antibiotic administered alone, (c) systemic antibiotic administered alone, and (d) combined topical and systemic antibiotics administered in combination. Thirty wounds in each group were inoculated with 1.5 X lo7 colony forming units of Staph. aureus suspended in 0.1 ml of sterile saline solution. The remaining 30 wounds in each group were inoculated with 1 X lOlo colony forming units of Staph. aureus suspended in 0.1 ml of sterile saline solution. The former inocuhrm size was selected to produce a low wound infection rate (20 percent) in the control animals and the latter inoculum size, a high wound infection rate (43 percent) (Table I). Statistical analysis of the wound infection rates was undertaken by use of the Fisher exact test for the comparison of two proportions.
Results Figure 1 shows the wound infection rates in each group in the presence of each of the two different concentrations of organisms. A total of 240 wounds were available for analysis. With the lower dose of bacterial inoculum (Group I), both systemic antibi-
754
otic used alone and combined topical and systemic antibiotics were equally effective in the reduction of wound infection rates when compared with the control groups (p <0.05). The topical antibiotic, when used alone, also produced a reduction in the wound infection rate that was not significant when compared with the control group. With a higher dose of bacterial inoculum (Group II), topical antibiotic administered alone did not reduce the infection rate when compared with the control group. Systemic antibiotic administered alone did reduce the infection rate, but not as effectively as it did in Group I. Combined topical and systemic antibiotics produced the most effective reduction in the infection rate, performing significantly better than in either the control group (p
The American JournalOl %irgOIY
combined regimen are confirmed by our finding that the combined regimen provided the most eff8ctive prophylaxis in the presence of dense bacterial inocuhun. When the doee of bacterial inoculum was low, the combiition regimen was no more effective than the sy&xnic antibiotic used alone. Such an apparent contradiction may be related to the low infection rates in these two groups (3 percent). At this lower level of wound contamination, the demonstration of any difference between the two groups would require a much larger study population. In a pmspe&ve clinical trial that involved patients whose wounds were contaminated at emergency laparotomy, Stone and Hester [12] demonstrated that prophylaxis with a combination of systemic and topical antibiotics produced a lower wound infection rate than systemic antibiotic alone. Unfortunately, there WBBno topical antibiotic alone group, and cephalothin has subsequently been discredited as a prophylacticagent because of its rapid disappearance from the surgicalincision 1241.Pitt et al [ 71compared the route of administration of prophylactic antibiotics upon wound infection rates in the cleen wound8 of patients und8rgoing vascular surgery.They found no dBer8nce in the effectiveness of topical antibiotic admimst8md alone, systemic antibiotic administered alone, or when a combination of the two was admin-
vm
wr, Jmo f8u
istered. The .resultaof these two clinical studies appear to reflect our experimental f&rdingsthat the combination regimen is more effe&v8 than systemic prophylaxis alone only in heavily contaminated wounds. This principle does not appear to be confined to the particular combination of antimicrobial agenta and bacteria uaed in this study. Galland et al [13] have shown experimentally that a similar pattern is seen when the infecting organism is Escherichia coli, and the topical antiseptic povidone iodine is used with the systemic antibiotic cephaloridine. The effectiveness of a combined topical and systemic antibiotic regimen was ahdied in an animal model previously shown to simulate chnical surgical wound infection. At a high level of bacterial contamination, the combination regimen produced a lower infection rate than either a placebo (p
755
otica is the more effective regimen. Where wound contamination is less severe, systemic antibiotic prophylaxis is all that is required;no further benefit is obtained by the additional administration of topical antibiotics. Clinical trials appear justified to confirm or refute this hypothesis. References 1. Bernard w1, Cok WR. The prophylaxis of surgical infection: the effect of prophykctk antimkrobial drugs on the incidence of infectkn tolkwing potentially contaminated operatkns. sugary 1984$8:151-7. 2. Evans C, Potkck AV. The re&cfkm of sqkal wo& infections by prophykctk parenkml cephalorkine. Br J Surg 1973; 60:434-7. 3. Polk HC Jr, LopeMayer JF. Postoperative wound infection. A proepective study of determinant factors and prevention. Sugery 19fI9$8*7-103. 4. yz r?ts TR. Topkal antibiotic and delayed primary maqx~~M of contaminated surgical incisions. J Sq Res 1972;12:70-8. 5. Erkson C, Lk@en L, LiMberg L. Ckxacillin In the prophylaxis $7~1$f=t=tons of the hip. J Joint -ci
:
8. Mkn w
&D, Hunt J, A&m&r WA. Prophylactk antibiotics in wotmds of the abdomen. J Trauma 1972;12:
7. Pftt HA~F&stkr RG McGowan WAL, et al. Prophylactic antibktks In vascular surgsry: topical. systemk or both? Ann surg 1980;192:358-84. 8. Stone F&f,Hooper CA, Kofb LD, et al. Antibiotic prophykxis in gastric, biltary and coknk surgery. Ann Surg 1978;184: 443-50. 9. Baum ML, Amish DB. Chalmers TC, et al. A survey of clinical trials of antibiotic prophylaxis In cokn surgery: evidence against furthar use of nc&eatment controfs. N Engl J Med 1981;305z795-9. 10. Pofk HC Jr, Ausobsky JR. The role of antibktics in surgical inhsugery.Chkago: fectkns.frxf&&smlLD,ed.,Adwmces
YearBook Medical 1983:225-75. 11. DavksonAlG,ClerlcC,SmlthG postoperathfewolJlldMctkn: a computer analysts. Br J Surg 1971;58:333-7. 12. Stone HH, Hester TFt. Incisional and peritoneal infection after emergency celktomy. Ann Surg 1973;177:889-78. 13. Galland RB, Heine KJ, Trachtenberg LS, Pofk HC. Reductkn of surgical wound infection rates in contaminated wounds treated with antiseptics combined with systemk anubktksz an experimental study. Surgery 1982;91:329-32. 14. GingrassRP, Ckse AS, EMsonEH. The effect of vt%kus topkal and parenteral agents on the preventkn of infection In experimental contaminated wounds. J Trauma 1984;4:78383. 15. Edlich RF, Tsung MS, Rogers W, et al. Studies in the management of the contaminated wound. J Surg Res 1988;8: 585-92. 18. Bray FJ, KtddEE.Topical dwwnotherewk~of~ infection. Surgery 1963;54:891-9. 17. McGeehan D, Hunt D, Chaudhuri A, Rutter P. An experimental study of the relationship between synergistic wound sepsis and suture materkla. Br J Surg 1980;87:838-8. 18. Foster GE, Bourke JB, Bolwell J, et al. Clkkal and economic consequences of wound sepsis after appendiceotomy and their modification by metrontdazole or povkone kdlne. Lancet 1981:1:789-71. 19. Flint Lhf, Fry DE,eds. &q$cal tnfectkns: discu&ms in sugkal managemant. Garden City, NY: Medkal Examkatkn Publishing, 1982. 20. Altemeier WA, Cukertson WR, Hummel RP. Sugkal conskerations of endogenous infectkns--sauces, types and methods of control. Surg Clin North Am l-48:227-40. 21. Burke JF. The effective perkd of preventtve anttbktk actkn in experimental inclskns and dermai kskns. Surgery 1981;50:181-8. 22. Matsumoto T, Dobek AS, Kovark JJ, Ham5 HF. Topical spray of antibiotics in simulated combat wounds. Arch Surg 1988;97:81-7. 23. Gilmore DJA, Martin TDM. Aetiology and preventkn of wound infection in appendkectomy. Br J Surg t974;81:281-7. 24. Polk HC Jr, Trachtenberg LS, Finn MP. Antlbktk actfvlty In surgical incisions. The basis for prophyfaxis in r&kted operations. JAMA 198Q244zl353-4.
See page 838 for a related editorial comment.
756
aydemic and topical antiofeurgi4wWWnf~on [I-4]. Both of thefie routes haJsbeen ofad&&t&o~havebeenehowntoreducethe
ehta=d . .
prophylactic
,uacommon place of a combination of both topical and systemic ntary ve efnenta onregnnenaare oftenueedincUnic4practice,althoughtherehasyet to be a full and proper evaluation of their clinical effectin evatuation is highlighted by number ofpatients in whom slop despite appropriafe antibiotic . prop&hsie by a single proven route. ContammaW wounds are at particular risk of subsequent infection [II], and previous clinical and experimentai obeerv&ions have suggeeted that a combir&ionoftopicalandsy&emicprophyluxiemay be more effective then either would be aloe@when bact&al contamination of the wound is great [12-
acitracin, and neomycin sulfate Ma?WllrlaMWe@Wa Adult male albino guinea pige that we&bed approxi-
fascia and were extend
14). The aim of this study was to determhae whether
the addition of a topical antibiotic would be benefirnrrgicsiwoundinftionrateswhen a @e&o, a topical antibiotic alone, a&biotic alone, in the preeence of alowlevelofbacter4uW&nination of the wound. A w&&cribed guinea pig model was Fmm~(hl)Eke#nwt~Ikmrd(k*Rlos~ot~Rgsroh, M~.LG@@&IQJnW9,.
theanimalswerekilledandtbeir~wereopenedand inspeet8d for pus, the obeermv epscifiepllyunaware of the treatment groupto whichtBeenin& belonged. A wound infection was diagnosed only in the presenceof grossly visible PUB. Tbe.St@hylococcua aureu8umdinthbbqerimentwee ieolated fmm a hoepital CkllhlWiOf~glanism
ticaee soy agar (BBL Microbiology @&ems, Cockeysviile,
753
TABLE I
QfouD I
II l
lnodum of Staphyhwcws aureus Producing High and Low Wound lnhcth Rates in the CodrolQnwrp(n=30) Staph. aureus Inboduced’ 1.5 x 10’ 1 x 10’0
wounds Infected (n) 6 13
Infection Rate 20% 43%
Measuredas numberof colonyformingunits.
MD). Before inoculation, the Staph. aureus was grown overnight in static trypticase broth (BBL Microbiology Systems) at 37°C and suspended at the desired concentrations in sterile saline solution. Bacterial concentrations were confirmed by plating and overnight incubation at 37% on trypticase soy agar, and were expressed as the number of colony forming units per milliliter of saline solution. Clindamycin (Cleocin Phosphate@, Upjohn Co., Kalamazoo, MI) was the antibiotic used systemically, being administered intramuscularly in a dose of 5 mg/kg 20 minutes before bacterial inoculation of the wound. A mixture of polymyxin B, bacitracin, and, neomycin sulfate (Neosporina powder, Burroughs Wellcome Co., Triangle Park, NC) was used as the topical antibiotic. Fifteen minutes after the bacterial inoculation, 0.5 mg of powder suspended in 0.1 ml of sterile saline solution was placed in the wound. Each wound therefore received 2.5 units of polymyxin B, 0.2 units of bacitracin, and 2.5 pg of neomycin sulphate. When no antibiotic was given, sterile saline solution was administered as a control (0.5 ml given intramuscularly or 0.1 ml placed in the wound, as appropriate). Sixty wounds were examined in each of four treatment groups, a total of 240 wounds being available for subse-
quent analysis. The treatment groups were as follows: (a) no antibiotic administered, (b) topical antibiotic administered alone, (c) systemic antibiotic administered alone, and (d) combined topical and systemic antibiotics administered in combination. Thirty wounds in each group were inoculated with 1.5 X lo7 colony forming units of Staph. aureus suspended in 0.1 ml of sterile saline solution. The remaining 30 wounds in each group were inoculated with 1 X lOlo colony forming units of Staph. aureus suspended in 0.1 ml of sterile saline solution. The former inocuhrm size was selected to produce a low wound infection rate (20 percent) in the control animals and the latter inoculum size, a high wound infection rate (43 percent) (Table I). Statistical analysis of the wound infection rates was undertaken by use of the Fisher exact test for the comparison of two proportions.
Results Figure 1 shows the wound infection rates in each group in the presence of each of the two different concentrations of organisms. A total of 240 wounds were available for analysis. With the lower dose of bacterial inoculum (Group I), both systemic antibi-
754
otic used alone and combined topical and systemic antibiotics were equally effective in the reduction of wound infection rates when compared with the control groups (p <0.05). The topical antibiotic, when used alone, also produced a reduction in the wound infection rate that was not significant when compared with the control group. With a higher dose of bacterial inoculum (Group II), topical antibiotic administered alone did not reduce the infection rate when compared with the control group. Systemic antibiotic administered alone did reduce the infection rate, but not as effectively as it did in Group I. Combined topical and systemic antibiotics produced the most effective reduction in the infection rate, performing significantly better than in either the control group (p
The American JournalOl %irgOIY
combined regimen are confirmed by our finding that the combined regimen provided the most eff8ctive prophylaxis in the presence of dense bacterial inocuhun. When the doee of bacterial inoculum was low, the combiition regimen was no more effective than the sy&xnic antibiotic used alone. Such an apparent contradiction may be related to the low infection rates in these two groups (3 percent). At this lower level of wound contamination, the demonstration of any difference between the two groups would require a much larger study population. In a pmspe&ve clinical trial that involved patients whose wounds were contaminated at emergency laparotomy, Stone and Hester [12] demonstrated that prophylaxis with a combination of systemic and topical antibiotics produced a lower wound infection rate than systemic antibiotic alone. Unfortunately, there WBBno topical antibiotic alone group, and cephalothin has subsequently been discredited as a prophylacticagent because of its rapid disappearance from the surgicalincision 1241.Pitt et al [ 71compared the route of administration of prophylactic antibiotics upon wound infection rates in the cleen wound8 of patients und8rgoing vascular surgery.They found no dBer8nce in the effectiveness of topical antibiotic admimst8md alone, systemic antibiotic administered alone, or when a combination of the two was admin-
vm
wr, Jmo f8u
istered. The .resultaof these two clinical studies appear to reflect our experimental f&rdingsthat the combination regimen is more effe&v8 than systemic prophylaxis alone only in heavily contaminated wounds. This principle does not appear to be confined to the particular combination of antimicrobial agenta and bacteria uaed in this study. Galland et al [13] have shown experimentally that a similar pattern is seen when the infecting organism is Escherichia coli, and the topical antiseptic povidone iodine is used with the systemic antibiotic cephaloridine. The effectiveness of a combined topical and systemic antibiotic regimen was ahdied in an animal model previously shown to simulate chnical surgical wound infection. At a high level of bacterial contamination, the combination regimen produced a lower infection rate than either a placebo (p
755
otica is the more effective regimen. Where wound contamination is less severe, systemic antibiotic prophylaxis is all that is required;no further benefit is obtained by the additional administration of topical antibiotics. Clinical trials appear justified to confirm or refute this hypothesis. References 1. Bernard w1, Cok WR. The prophylaxis of surgical infection: the effect of prophykctk antimkrobial drugs on the incidence of infectkn tolkwing potentially contaminated operatkns. sugary 1984$8:151-7. 2. Evans C, Potkck AV. The re&cfkm of sqkal wo& infections by prophykctk parenkml cephalorkine. Br J Surg 1973; 60:434-7. 3. Polk HC Jr, LopeMayer JF. Postoperative wound infection. A proepective study of determinant factors and prevention. Sugery 19fI9$8*7-103. 4. yz r?ts TR. Topkal antibiotic and delayed primary maqx~~M of contaminated surgical incisions. J Sq Res 1972;12:70-8. 5. Erkson C, Lk@en L, LiMberg L. Ckxacillin In the prophylaxis $7~1$f=t=tons of the hip. J Joint -ci
:
8. Mkn w
&D, Hunt J, A&m&r WA. Prophylactk antibiotics in wotmds of the abdomen. J Trauma 1972;12:
7. Pftt HA~F&stkr RG McGowan WAL, et al. Prophylactic antibktks In vascular surgsry: topical. systemk or both? Ann surg 1980;192:358-84. 8. Stone F&f,Hooper CA, Kofb LD, et al. Antibiotic prophykxis in gastric, biltary and coknk surgery. Ann Surg 1978;184: 443-50. 9. Baum ML, Amish DB. Chalmers TC, et al. A survey of clinical trials of antibiotic prophylaxis In cokn surgery: evidence against furthar use of nc&eatment controfs. N Engl J Med 1981;305z795-9. 10. Pofk HC Jr, Ausobsky JR. The role of antibktics in surgical inhsugery.Chkago: fectkns.frxf&&smlLD,ed.,Adwmces
YearBook Medical 1983:225-75. 11. DavksonAlG,ClerlcC,SmlthG postoperathfewolJlldMctkn: a computer analysts. Br J Surg 1971;58:333-7. 12. Stone HH, Hester TFt. Incisional and peritoneal infection after emergency celktomy. Ann Surg 1973;177:889-78. 13. Galland RB, Heine KJ, Trachtenberg LS, Pofk HC. Reductkn of surgical wound infection rates in contaminated wounds treated with antiseptics combined with systemk anubktksz an experimental study. Surgery 1982;91:329-32. 14. GingrassRP, Ckse AS, EMsonEH. The effect of vt%kus topkal and parenteral agents on the preventkn of infection In experimental contaminated wounds. J Trauma 1984;4:78383. 15. Edlich RF, Tsung MS, Rogers W, et al. Studies in the management of the contaminated wound. J Surg Res 1988;8: 585-92. 18. Bray FJ, KtddEE.Topical dwwnotherewk~of~ infection. Surgery 1963;54:891-9. 17. McGeehan D, Hunt D, Chaudhuri A, Rutter P. An experimental study of the relationship between synergistic wound sepsis and suture materkla. Br J Surg 1980;87:838-8. 18. Foster GE, Bourke JB, Bolwell J, et al. Clkkal and economic consequences of wound sepsis after appendiceotomy and their modification by metrontdazole or povkone kdlne. Lancet 1981:1:789-71. 19. Flint Lhf, Fry DE,eds. &q$cal tnfectkns: discu&ms in sugkal managemant. Garden City, NY: Medkal Examkatkn Publishing, 1982. 20. Altemeier WA, Cukertson WR, Hummel RP. Sugkal conskerations of endogenous infectkns--sauces, types and methods of control. Surg Clin North Am l-48:227-40. 21. Burke JF. The effective perkd of preventtve anttbktk actkn in experimental inclskns and dermai kskns. Surgery 1981;50:181-8. 22. Matsumoto T, Dobek AS, Kovark JJ, Ham5 HF. Topical spray of antibiotics in simulated combat wounds. Arch Surg 1988;97:81-7. 23. Gilmore DJA, Martin TDM. Aetiology and preventkn of wound infection in appendkectomy. Br J Surg t974;81:281-7. 24. Polk HC Jr, Trachtenberg LS, Finn MP. Antlbktk actfvlty In surgical incisions. The basis for prophyfaxis in r&kted operations. JAMA 198Q244zl353-4.
See page 838 for a related editorial comment.
756