Ramos G.E.¹, Resta M.², Durlach R.², Patiño O.¹, Bolgiani A.¹, Prezzavento G.,Fernandez Canigia L.³, Benaim F.¹

German Hospital, Buenos Aires, Argentina
¹ CEPAQ (Burns Unit), Benaim Foundation
² Infectology Department
³ Microbiology Laboratory

SUMMARY. Bacteraemias during burn wound manipulation are frequent, especially following burn wound excision. However, these bacteraemias seem not to have any clinical consequences, and their treatment is therefore controversial. Over a 20-month period 35 surgical debridement procedures were recorded prospectively in 18 burn patients. Blood culture samples were drawn before, during, and after surgical excision. Bacteraemias were found in ten out of the 35 patients (28%), and 16 of the 105 blood samples (15%) were positive. All three blood samples were positive in one case (“primary bacteraemia”), while others were “transient bacteraemia”. Six positive blood cultures were considered to be “bacteraemias induced by wound manipulation” and seven “bacteraemias of unknown source”. Bacteraemias of unknown source were not recorded at any time while “bacteraemias induced by wound manipulation” were recorded after day 5 post-burn. Patients with more than 40% TBSA had 4.3 times more bacteraemic risk than patients with less extensive TBSA. Blood pressure and white blood cell variations were observed in bacteraemic patients but without any clinical relevance. We conclude that bacteraemic rates were high and that there were two different patterns of bacteraemia - both transient and with no clinical relevance.

Introduction

The presence of viable micro-organisms (MOs) in the blood usually means a systemic infection with high morbidity and mortality. It may be associated either with the failure of host defences to contain the infection of a remote source or with the failure of treatment to eradicate the infectious source.1 However, positive blood cultures may also be due to sample contamination or transient and self-limited bacteraemia, which has a better outcome.2 Transient bacteraemia after medical intervention has been described3 - it can be dangerous when patients have previous heart valve damage. It has been demonstrated during odontological procedures,4 urological instrumentation,5 sigmoidoscopy,6 minor cutaneous surgery,7 and intravascular catheter placement.

In burn patients, bacteraemias produced by cutaneous surgery are ten times more frequent than in immune-competent patients.9,10 This difference has also been attributed not only to the immune system dysfunction that follows a thermal insult but also to the degree of wound contamination.

In spite of experimental descriptions in which endotoxaemia triggered haemodynamic changes indistinguishable from sepsis,11,12 there has been insufficient clinical evidence to support the hypothesis that bacteraemia induced by wound manipulation would have immediate clinical consequences. Prophylaxis with antibiotics would not therefore change the natural history of these bacteraemias.

On the other hand, antibiotics can be associated with side effects such as allergic reactions, organ damage, high cost, and the generation of MOs with antibiotic resistance.15 A study conducted in England16 showed that, in half of the hospitals considered, prophylactic antibiotics were administered for some specific surgery. Furthermore, in three out of the 39 hospitals, antibiotics were used in all surgical manoeuvres, even simple dressing changes. Mozingo et al.17 suggested the use of prophylactic antibiotics in patients with a high risk of bacteraemia, such as those with burns in more than 40% TBSA and more than 10 days post-burn.

The low availability of allografts and the high cost of biosynthetic skin substitutes prolong open wound duration in our unit, which can increase bacteraemic risk.

The objective of this prospective trial was to determine the incidence, aetiology, and complications of peri-operative bacteraemia during escharectomy in burn patients.

Materials and methods

Design: prospective observational study.

Period: from May 1999 to December 2000 (20 months).

Study group: 35 surgical procedures (SPs) with general mixed anaesthesia were studied in 18 burn patients.

Inclusion criterion: escharectomy of acute burn wound.

Exclusion criteria: antibiotics started within 72 h, absence of undamaged site from which to obtain a contamination-free blood sample.

Cultures: surface wound skin cultures were taken after cleaning with crystalloid solution.

Blood culture (BC): an undamaged skin preparation was made with isopropanol 70% and iodopovidone. BC1: the first sample was taken during anaesthesia induction. BC2: the second sample was taken while escharectomy was being performed. BC3: the third sample was taken before the patient awoke after anaesthesia.

Antibiotics: eight patients were receiving antibiotics as treatment (four burn wound infections, two pneumonias, two central venous catheter-related infections) during 14 SPs, but none as prophylaxis.

Severity scale: APACHE II19 scores were recorded on day 1 and multi-organ dysfunction (MOD) from Marshall20 was recorded on the day of surgery.

Complications: intra-operative complications, hypotension, cardiac arrhythmia.

Post-operative complications: fever (> 38 ºC),hypotension (< 90 mm Hg systolic arterial pressure or decrease of 40 mm Hg from normal), oliguria (< 0.5 ml/kg urine output in 3 h).

Definitions

Bacteraemia: viable MOs in the blood.

Primary bacteraemia: the same bacteria in two or more consecutive blood cultures.

Transient bacteraemia: bacteria were isolated in only one blood sample.

Bacteraemia related to escharectomy: bacteria isolated in blood sample while escharectomies were performed.

Wound colonization: MOs isolated from burn wound without any evidence of clinical infection.

Source of bacteraemia: the eschar was considered the source of bacteraemia when the bacteria isolated in the wound and blood shared the same phenotypic features.

Statistics: chi square was used to assess differences between categorical variables, and the Student test was used for differences between averages.

Results

Thirty-five surgical procedures in 18 patients with acute burns caused by fire, hot liquids, and high-energy electricity were evaluated. The average TBSA was 34% (range, 6-80%), APACHE II 8.5 (range, 2-32), age 36.8 yr (range, 6-80), and SPs 1.9 (range, 1-5) (Table I).

In 16 out of the 105 blood samples (15%), bacteraemias were confirmed. This also occurred in 10 out of the 35 surgical procedures (28%) studied. TBSA was statistically different in procedures where positive blood cultures were observed (31.5% vs. 54%, p < 0.01). No significant statistical differences were found as regards evolution time, APACHE II, MOD, percentage of excised wound by procedure, and antibiotics (Table II).

Bacteraemias were found in 10 SPs. Only one patient had the three blood cultures (BC1, 2, 3) positive with the same MO (Streptococcus pyogenes). This was therefore primary bacteraemia. The other bacteraemias were transient, with a positive BC1 in four out of 34 (11%), a positive BC2 in six out of 34 (17%), and a positive BC3 in three out of 34 (8%) (Fig. 1).

The MOs isolated are shown in Table III.

One bacterium (Pseudomonas stutzeri) from BC1 was also isolated from the surface wound culture. None of the transient bacteraemias from BC3 were isolated from surface wound cultures. All six transient bacteraemias from BC2 were also isolated from surface wound cultures.

MOs isolated from central venous catheters were different from those isolated in blood.

Concordance between MOs isolated from blood (excluding primary bacteraemia) and wound colonization is shown in Fig. 3. All MOs from transient bacteraemia during surgical excision (BC2) were compatible with wound MOs, while just one out of seven MOs isolated before and after the excision (BC1 and 3) was compatible with wound colonization (p = 0.01) (Fig. 2).

All bacteraemias during BC2 occurred when the procedures were performed within five days of the burn, whereas the other bacteraemias during BC1 and BC3 occurred at any time (p = 0.02) (Fig. 3).

TBSA greater than 40% increased bacteraemic risk. Antibiotic administration and the number of post-burn days were not related to bacteraemic risk (Table IV).

Arterial hypotension was observed in one non-bacteraemic procedure during surgical procedures and in two cases of bacteraemic post-operative surgery. Post-operative fever was observed in six bacteraemic procedures (60%) and 12 non-bacteraemic procedures (48%) (p non-significant). Post-operative oliguria was observed in one bacteraemic procedure (10%) and one non-bacteraemic procedure (4%) (p non-significant).

The reported incidence of peri-operative bacteraemia is variable.22,23 The reason for this may reflect differences in patient severity and treatment. Several works relate the risk of bacteraemia to burn extent.17,24-27 The increased risk of bacteraemia in patients with extensive burns may be related either to a greater bacterial load or to a greater derangement of the immune system.

In the present study the bacteraemic risk was also related to burn extent. Nevertheless, it is possible that some of the bacteraemias were not real bacteraemias, since it has been reported that the likelihood of finding false positive blood cultures increases with burn severity.28 Weinstein et al.2 reported that nearly half the number of positive blood cultures were considered contaminated or of indeterminate meaning, and argued that many of these blood cultures could have been drawn through central venous catheters or by inadequate antisepsis technique. In the present study all the samples were drawn by direct venous puncture, using the antisepsis protocol described. Notwithstanding, falsely positive blood cultures cannot be ruled out.

All three blood cultures were positive in one patient without any evident source of infection; this is defined as “primary bacteraemia”. In the other surgical procedures, MOs were isolated in only one blood culture, and this is defined as “transient bacteraemia”. Some of the transient bacteraemias were found during eschar excision and all the MOs matched those found in the wound burn, and these are defined as “bacteraemia induced by wound excision”. Other transient bacteraemias were not related to the surgical excision, and the MOs in 6 out of 7 procedures (85%) did not match those from wounds; these are defined as “bacteraemia of unknown source”. “Bacteraemias induced by wound excision” were found 5 days post-burn, while “unknown source bacteraemias” occurred any day post-burn.

Bacteraemias before wound manipulation have already been reported.17,24,25,27 Hypotheses regarding bacteraemias of hidden source include bacterial translocation,29 intravascular device mobilization,30 contamination of instilled solutions, distant infectious source, involuntary wound manipulation during patient transfer, laryngoscopy, orotracheal intubation,31 and simple contamination of the sample with resident skin MOs. In the present study MOs isolated from central venous lines did not match the bacteraemias.

The intensity of the procedure may be related to bacteraemic risk.24 In this study, only bacteraemias during surgical excision were investigated; however, less aggressive procedures in burn patients can also produce bacteraemia.

The number of post-burn days has already been reported as being an important risk factor for bacteraemia that may be related to the degree of colonization or infection of wounds.17 This finding is not however universally accepted.32 In the present study post-burn time was important only for bacteraemias induced by excision but not for those of unknown source. Mozingo et al.17 suggested that burns more than ten days old had a high bacteraemic risk that justified antibiotic prophylaxis in extensive burns. However, Beard et al.26 reported that bacteraemias occurring ten days post-burn were eliminated quickly by guest defences. This is supported by the fact that, at this stage, guest immunity is fully recovered - antibiotics should therefore not be necessary for the treatment of these transient bacteraemias.

In our study antibiotic prophylaxis was not used. Nevertheless, some patients were receiving antibiotics when the study was being performed only for therapeutic reasons. These patients presented a lower percentage of bacteraemias, although the differences were not statistically significant. Some studies33,34 have not demonstrated any benefits with prophylactic antibiotics in burn patients when they have undergone some surgical procedure. On the other hand, the use of antibiotics use is not free of complications. Vancomycin has been associated with skin rash and severe hypotension, independently of the speed of infusion during prophylactic use.

Widespread antibiotic use in burn units has been associated with the appearance of multi-resistant MOs,37 which has been related to a worse outcome.38 Despite the absence of any scientific evidence that supports the use of prophylactic antibiotics during surgical procedures, it has been a common practice during wound manipulation and grafts.

Some researchers have found an association between transient bacteraemia and low blood pressure, shock, and shivering,40 while others have not seen any clinical association.26 Surgery-induced sepsis in burn patients is a dreadful but infrequent complication,13,23 despite the high incidence of bacteraemias reported.17,24,25 In our study post-surgery sepsis was not observed in any procedure and other complications were not different in either group. Some distant infections due to haematogenous spread, such as arthritis, osteomyelitis, and endocarditis,40-42 have been associated with bacteraemias, but in the present study neither acute endocarditis nor any other suppurative complications attributable to bacteraemias were found.

Conclusion

We concluded that the incidence of peri-operative bacteraemia in burn patients was high and that it was related to burn extent. Some transient bacteraemias may have been induced by surgical excision, while others had no evident source. None of the transient bacteraemias had any clinical relevance. We therefore found no evidence to support the use of prophylactic antibiotics in burn patients when they undergo this type of surgical procedure.

RESUME. Les bactériémies au cours de la manipulation des lésions causées par les brûlures sont fréquentes, particulièrement à la suite de l’excision de la lésion. Cependant, ces bactériémies ne semblent avoir aucune conséquence clinique et pour cette raison leur traitement est controverse. Pendant une période de 20 mois les Auteurs ont enregistré prospectivement 35 procédures de débridement chirurgical chez 18 patients brûlés. Des échantillons de culture sanguine ont été prélevés avant, pendant et durant l’excision chirurgicale. Des bactériémies ont été observées dans 10 sur les 35 patients (28%), et 16 des 105 échantillons (15%) étaient positifs. Tous les trois échantillons étaient positifs dans un cas de “bactériémie primaire”, tandis que d’autres échantillons étaient “bactériémie transitoire”. Six cultures positives ont été définies “bactériémies induites par manipulation des lésions” et sept “bactériémies de source inconnue”. La bactériémie de source inconnue n’a jamais été jamais enregistrée, tandis que les “bactériémies induites par la manipulation des lésions” ont été enregistrées à la suite du cinquième jour après la brûlure. Les patients ayant des brûlures en plus de 40% de la surface corporelle totale étaient exposés à un risque majeur 4,3 fois de plus par rapport aux patients atteints d’une extension brûlée mineure. Les Auteurs ont observé des variations dans la pression sanguine et les cellules blanches du sang dans les patients bactériémiques, mais ce résultat n’avait aucune signifiance clinique. Les Auteurs concluent que les taux bactériens étaient élevés et qu’il y avait deux modèles différents de bactériémie - tous les deux transitoires et sans importance clinique.

Bibliography

1. Raymond D., Pelletier S., Crabtree T., Gleason T., Pruett T., Sawyer R.: Impact of bloodstream infection on outcome among infected surgical inpatients. Ann. Surg., 233: 549-55, 2001.
2. Weinstein M., Towns M., Quartey S. et al.: The clinical significance of positive blood cultures in the 1990s: A prospective comprehensive evaluation of the microbiology, epidemiology, and outcome of bacteraemia and fungaemia in adults. Clin. Inf. Dis., 24: 584-602, 1997.
3. Kreger B.E., Craven D.E., Carling P.C., McCabe W.R.: Gram-negative bacteraemia. III. Reassessment of aetiology, epidemiology, and ecology in 612 patients. Am. J. Med., 68: 332-43, 1980.
4. Felix J.S., Rosen S., App G.: Detection of bacteraemia after the use of an oral irrigation device in subjects with peridontitis. J. Peridontol., 42: 785-87, 1971.
5. Slade N.: Bacteraemia and septicaemia after urological operations. Proc. R. Soc. Med., 51: 331-4, 1958.
6. Le Frock J., Ellis C., Turchick J., Weinstein L.: Transient bacteraemia associated with sigmoidoscopy. N. Engl. J. Med., 289: 467-9, 1973.
7. Sabetta J., Zitelli J.: The incidence of bacteraemia during skin surgery. Arch. Dermatol., 123: 213-5, 1987.
8. Griffin M., Wilson R., Edwards W.: Infective endocarditis, Olmstead County, Minnesota, 1950 through 1981. JAMA, 254: 1199-202, 1985.
9. Carmichael A., Flanagan P., Holt P., Duerden B.: The occurrence of bacteraemia with skin surgery. Br. J. Dermatol., 134: 120-2, 1996.
10. Baskin J., Rosenthal A., Pruitt B.: Acute bacterial endocarditis: A silent source of sepsis in the burns patient. Ann. Surg., 184: 618-21, 1976.
11. Seyfer A., Zajtchuk R., Hazlett D., Mologne L.: Systemic vascular performance in endotoxic shock. Surg. Gynecol. Obstet., 145: 401-7, 1977.
12. Hopkins R.W., Damewood C.A.: Septic shock: Hemodynamics of endotoxin and inflammation. Am. J. Surg., 127: 476-83, 1974.
13. Kagan R., Hanumadass M., Matsuda T.: Burn wound excision: Predicting intra-operative bacteraemia and immediate post-operative sepsis. Proc. American Burn Association Twenty-Second Annual Meeting, p. 121, 1990.
14. Papini R.P., Wilson A.P., Steer J.A., Hill G., McGrouther D.A., Parkhouse N.: Plasma concentrations of tumour necrosis factor-alpha and interleukin-6 during burn wound surgery or dressing. Br. J. Plast. Surg., 50: 354-61, 1997.
15. Masur H.: Antimicrobials. In: Chernow B. (ed.), “The Pharmacologic Approach to Critically Ill Patients” (3rd edition), 685-99, Williams & Wilkins, 1994.
16. Papini R., Wilson A., Steer J., McGrouther D., Parkhouse N.: Wound management in burn centres in the United Kingdom. Br. J. Surg., 82: 505-9, 1995.
17. Mozingo D., McManus A., Kim S., Pruitt S., jr: Incidence of bacteraemia after wound manipulation in the early post-burn period. J. Trauma, 42: 1006-11, 1997.
18. Burke J., Bondoc C., Quinby W.: Primary burn excision and immediate grafting: A method shortening illness. J. Trauma, 14: 383, 1974.
19. Knaus W.A., Draper E.A., Wagner D.P.: Apache II: A severity of disease classification system. Crit. Care Med., 13: 818-24, 1985.
20. Marshall J.C., Cook D., Christou N.V.: MOD syndrome: A reliable descriptor at a complex clinical outcome. Crit. Care Med., 23: 1638-652, 1995.
21. Peck M.D., Weber J., McManus A.: Surveillance of burn wound infections: A proposal for definitions. J. Burn Care Rehabil., 19: 386-9, 1998.
22. Petersen S., Umphred E., Warden G.: The incidence of bacteraemia following burn wound excision. J. Trauma, 22: 274-9, 1982.
23. Levine B., Sirinek K., Pruitt B.: Wound excision to fascia in burn patients. Arch. Surg., 113: 403-7, 1978.
24. Sasaki T., Welch G., Herndon D., Kaplan J., Lindberg R., Pruitt B.: Burn wound manipulation-induced bacteraemia. J. Trauma, 19: 46-8, 1979.
25. Vindenes H., Bjerknes R.: The frequency of bacteraemia and fungaemia following wound cleaning and excision in patients with large burns. J. Trauma, 35: 742-9, 1993.
26. Beard C., Ribeiro C., Jones D.: The bacteraemia associated with burns surgery. Br. J. Surg., 62: 638-41, 1975.
27. Piel P., Scarnati S., Goldfarb W., Slater H.: Antibiotic prophylaxis in patients undergoing burn wound excision. J. Burn Care Rehabil., 6: 422-4, 1985.
28. Keen A., Knoblock L., Edelman L., Saffle J.: Effective limitation of blood culture use in the burn unit. J. Burn Care Rehabil., 23: 183-9, 2002.
29. Herndon D.: Is bacterial translocation a clinically relevant phenomenon in burns? Crit. Care Med., 28: 1682-3, 2000.
30. Renaud B., Brun-Buisson C.: Outcome of primary and catheter-related bacteraemia. A cohort and case-control study in critically ill patients. Am. J. Resp. Crit. Care Med., 163: 1584-90, 2001.
31. Goldstein S., Wolf G., Kim S., Sierra M., Whitmire C., Tolentino E.: Bacteraemia during direct laryngoscopy and endotracheal intubation: A study using a multiple-culture, large-volume technique. Anaesth. Intensive Care, 25: 239-44, 1997.
32. Steer J., Papini R., Wilson A., McGrouther T., Parkhouse N.: Quantitative microbiology in the management of burn patients. II. Relationship between bacterial counts obtained by burn wound biopsy culture and surface alginate swab culture, with clinical outcome following burn surgery and change of dressing. Burns, 22: 177-81, 1996.
33. Rodgers G., Fisher M., Lo A., Cresswell A., Long S.: Study of antibiotic prophylaxis during burn wound debridement in children. J. Burn Care Rehabil., 18: 342-6, 1997.
34. Steer J., Papini R., Wilson A., McGrouther D., Nakhla L., Parkhouse N.: Randomized placebo-controlled trial of teicoplanin in the antibiotic prophylaxis of infection following manipulation of burn wounds. Br. J. Surg., 84: 848-53, 1997.
35. Valero R., Gomar C., Fita G., Gonzalez M., Pacheco M., Molet J., Velda M.: Adverse reactions to vancomycin prophylaxis in cardiac surgery. J. Cardiothorac. Vasc. Anesth., 5: 574-6, 1991.
36. Ngeow W.C., Chai W.L., Moody A.B.: Red man syndrome during administration of prophylactic antibiotic against infective endocarditis. J. Ir. Dent. Assoc., 46: 92-4, 2000.
37. Kollef M., Fraser V.: Antibiotic resistance in the intensive care unit. Ann. Intern. Med., 134: 298-314, 2001.
38. Goldman D., Weinstein R., Wenzel R., Tablan O., Duma R., Gaynes R.: Strategies to prevent and control the emergence and spread of antimicrobial resistant in hospitals. JAMA, 275: 234-40, 1996.
39. Yowler C.: Recent advances in burn care. Current Opinion in Anaesthesiology, 14: 251-5, 2001.
40. Sullivan N., Sutter V., Mims M., Marsh V., Finegold S.: Clinical aspects of bacteraemia after manipulation of the genitourinary tract. J. Infect. Dis., 127: 49-54, 1973.
41. Munster A.M., Di Vincenti F.C., Foley F.D., Pruitt B.A., jr: Cardiac infections in burns. Am. J. Surg., 122: 524-7, 1971.
42. Cartotto R., McDonald D., Wasan S.: Acute bacterial endocarditis following burn: Case report and review. Burns, 24: 369-73, 1998.
43. Paterson P., Dunn K.: Bacterial endocarditis following a minor burn injury. Case report and review. Burns, 25: 515-7, 1999.