Burns 25 (1999) 611±616
www.elsevier.com/locate/burns
Pseudomonas aeruginosa septicaemia in burns Raj Kumar Gang a, Rameshwar L. Bang a, c,*, Suhas C. Sanyal b, Eiman Mokaddas b, Abdul Reda Lari a a
Al Babtain Center for Plastic Surgery and Burns, Ibn Sina Hospital, Kuwait b Department of Microbiology, Ibn Sina Hospital, Kuwait c Department of Surgery, Faculty of Medicine, Kuwait University, P.O. Box 24923, Safat 13110, Kuwait Accepted 15 March 1999
Abstract Out of 1415 patients treated as inpatients at Al-Babtain Center for Burns and Plastic Surgery, Ibn Sina Hospital, Kuwait spanning over a period of 6 years from June 1992 to June 1998, 102 developed clinically and microbiologically proven septicaemia. Only 15 out of them had either single or multiple episodes of septicaemia due to Pseudomonas aeruginosa and were studied during their stay in the hospital. Five of them were males and 10 females, with a mean age of 26 years (range 3±51 years) and mean total body surface area of burns (TBSA) of 66% (range 25±90%). All of them had ¯ame burns and resuscitation was found to be dicult in eight patients either due to delayed hospitalization or accompanied inhalation injury. Seven patients were intubated, four due to inhalation injury and three for septicaemic complications. Among the 15 patients under study, a total of 36 septicaemic episodes were detected of which 21 were due to P. aeruginosa. This organism was found in the ®rst episodes in nine patients, in second episodes in six, in third episodes in three and fourth, ®fth and sixth episodes in one patient, each at a variable postburn day. Ten patients had 38 sessions of excision and skin grafting, six of them survived. Nine of the 15 patients under study died due to septicaemia, but only six of them had P. aeruginosa as the last isolate. Except for one, all patients had >40% TBSA burn, two had dicult resuscitation and four were intubated. The day of death varied between 3 to 52 days postburn (mean 19 days). This study showed that females with ¯ame burns are susceptible to P. aeruginosa septicaemia. Dicult resuscitation and intubation also proved to be important risk factors. Septicaemia could occur quite early in the postburn days and the mortality due to this organism was quite high. Early excision and grafting with other eective management may result in a better outcome. # 1999 Elsevier Science Ltd and ISBI. All rights reserved.
1. Introduction A burn injury causes tissue necrosis and raw area with serous exudation. The devitalized tissues and moist burn wound is favorable for the colonization and proliferation of microorganisms and subsequent infection, therefore the potential risk of burn wound sepsis and septicaemia persists until complete wound healing. The risk of infection is further potentiated due to immune disturbances caused by thermal injury [1± * Corresponding author. Tel.: +965-531-9475; fax: + 965-5319597. E-mail address:
[email protected] (R.L. Bang)
3]. The aim of management of burn wound is early healing without infection. Attempts to achieve this goal have been made by isolation of the patient, early excision and grafting, prompt diagnosis of burn wound sepsis and the institution of appropriate antibiotic therapy. However, this may not always be possible because of the reasons stated above in conjunction with the possibility of gastro-intestinal bacterial translocation and nosocomial infection [4±6]. Though septicaemia is most undesirable in burn patients, once it sets in early, diagnosis and its microbiological con®rmation remains the foremost concern in the management of burn. This prompt eort certainly in¯uences the morbidity and mortality in burn patients.
0305-4179/99/$20.00+0.00 # 1999 Elsevier Science Ltd and ISBI. All rights reserved. PII: S 0 3 0 5 - 4 1 7 9 ( 9 9 ) 0 0 0 4 2 - X
612
R. Kumar Gang et al. / Burns 25 (1999) 611±616
Pseudomonas aeruginosa thrives on the moist burn wound surface [6] and it is highly pathogenic in thermally injured immunosuppressed patients [1±3]. These bacteria usually gain access to burn patients through cross-contamination of burn wounds. Pseudomonas infection is a common complication in burn patients and it contributes to morbidity and mortality [7] amongst them. Septicaemic episodes due to this organism may occur in about 10% of burn patients [8]. However, in an earlier study in this unit, septicaemia due to P. aeruginosa was observed in only 1.6% of thermally injured patients [9]. Despite advances in medical and surgical care, the prognosis remains poor with a mortality rate of about 80% [8] in such patients. This prospective study spanning over a period of six years was undertaken to ®nd out the incidence of clinically and microbiologically proven P. aeruginosa septicaemia in burn patients in our unit, the extent and type of burns susceptible to this infection, time of onset, relationship of resuscitation and intubation in the incidence of septicaemic episodes and ®nally its outcome. 2. Material and methods 2.1. Patients One hundred and two patients had septicaemia amongst 1415 patients treated as inpatients at Al Babtain center for burns and plastic surgery, Ibn Sina Hospital, Kuwait from June 1992 to June 1998. Out of them 15 patients had either single or multiple episodes of septicaemia due to P. aeruginosa and they were studied during their stay in the hospital. 2.2. Assessment of burn, resuscitation and nutrition The percentage of burns was assessed using the Lund and Browders' chart, and the depth was determined by clinical observations. The clinically suspected inhalation injury was con®rmed by blood gas analysis, and radiological and bronchoscopical examinations. The patients with positive signs of inhalation injury were intubated on admission and mechanically ventilated throughout the period of treatment. All the patients were resuscitated based on the Parkland formula guidelines using crystalloids and were modi®ed according to the patients' response. Generally the colloid solutions were given 24 h postburn, but in patients with dicult resuscitation colloid was instituted early. Nutritional support was initiated from 48 h postburn with 1 cal/ml mixture and increased gradually to the required full amount by the ®fth to seventh day, until complete healing.
2.3. Antibiotic prophylaxsis Piperacillin and amikacin were given as prophylactic antibiotics for 5 days from the time of admission in adults with e40% TBSA burns and in children with e20% TBSA burns. However, this practice was stopped from January 1997 and no prophylactic antibiotic is administered to burn patients in our unit since then [10]. 2.4. Microbiology The culture swabs from the burn wounds, nose, throat and perineum were taken at the time of admission. Thereafter only burn wound swabs from dierent sites were cultured twice a week as per the microbiological surveillance protocol for all patients. Sets of two± three blood cultures were collected at an interval of at least 1 h from patients with clinical signs of septicaemia prior to institution of any antibiotic therapy. The blood specimens were processed in a Bactec 9240 (Becton Dickinson, MD, USA) machine which is known for quick detection of bacterial growth leading to an early diagnosis. Repeat specimens were collected if cultures were negative after 24 h. 2.5. Dressings of the wounds and surgery Silver sulphadiazine 1% (¯amazine) was used topically in all patients. The dressings were changed daily and the wounds were inspected. Primary excision and skin grafting with either autografts or auto and homografts were done in some of the second degree deep, and all the third degree burn patients within 3±5 days of admission and about 15% of burn areas were dealt with at each session. 2.6. Signs of septicaemia and empirical antibiotic therapy Septicaemia was suspected when a patient showed signs of disorientation, hyperpyrexia or hypothermia, circulatory embarrassment, petechial hemorrhages, bleeding in subcutaneous tissues, black or dark hemorrhagic discoloration in a previously normal appearing burn wound, increased oedema in unburned areas, early and rapid eschar separation, leucocytosis and thrombocytopenia. Immediate institution of empirical antibiotic therapy was done based on previous burn wound swab cultures. In patients with positive culture the antibiotic therapy was modi®ed if necessary. 2.7. Analysis of data The data of P. aeruginosa septicaemia patients in relation to percentage and causes of burns, response to
R. Kumar Gang et al. / Burns 25 (1999) 611±616
613
Table 1 Clinical details with number of episodes and the outcome No.
Age (yr)
Sex
% Burn
Number of Episodes (postburn day/microorganisma)
Outcome
1
2
3
4
5
6
7
12/Ps
18/Ps
23/Ps
25/ Strep
30/MRSA Ps
40/Kl
1
4
F
82
9/Ps
2 3
11 40
M F
42 61
4
35
F
90
23/Ps 36/Ac. MRSA 15/MRSA
5 6 7 8 9 10 11 12 13
47 50 3 30 4 18 35 29 27
M M F F F M F M M
90 25 75 70 50 50 45 90 85
14
51
F
67
15
12
F
70
2/Ps 10/Ps 15/Ps, Kl 2/Ps 14/Ps, Sa 45/Ps 11/Ps 4/MRSA 7/Entero Ac. 6/ Ac. MRSA 5/MRSA
65/Ps 21/Ps
30/Ps
45/MRSA Ac.
died survived died died died died survived survived Died
46/Ps 58/MRSA 14/Ps
10/MRSA
14/MRSA
survived survived died
40/ MRSA
21/Ps 8/Entero Ac. 19/Ps
died
survived 17/MRSA
23/Ps
died
a
Ps=P. aeruginosa, MRSA=methicillin resistant Staphylococcus aureus, Sa=Staphylococcus aureus, Strep=Streptococcus, Ac=Acinetobacter, Kl=Klebsiella, Entero=Enterobacter..
resuscitation, type of burns and their relation to septicaemic episodes, number of episodes and their occurrence on postburn day, antibiotic therapy and the outcome of patients were analyzed. The data of these patients were compared with all the septicaemic patients during the period of study. 3. Results Amongst the 102 patients who had septicaemia during the six years of study period from June 1992 to June 1998, P. aeruginosa septicaemia was diagnosed in 15 patients. Five of them were males and 10 females with a mean age of 26 years (range 3±51 years) (Table
Fig. 1. Distribution of patients with Pseudomonas septicaemia and due to other organisms in relation to causes of burn.
1). Female preponderance was observed in the incidence of P. aeuruginosa septicaemia. All the P. aeruginosa septicaemia patients sustained ¯ame burns in contrast to the other 87 septicaemic patients whose burns were due not only to ¯ame but to other causes as well (Fig. 1). The mean percentage of burns in P. aeruginosa septicaemia patients was 66% (range 25±90%) (Table 1) compared to the total patients whose mean was only 41% (range 1±90%) (Fig. 2). They all had more than 40% burn areas except one 50-year-old male who had 25% TBSA burns (Table 1). The resuscitation was dicult in eight patients and six of them had more than 60% burn areas (Table 2). The dicult resuscitation in ®ve patients was mainly due to delay in transportation of the patients to the burns unit and thus late institution
Fig. 2. Patients distribution according to percentage TBSA burns.
614
R. Kumar Gang et al. / Burns 25 (1999) 611±616
Table 2 Resuscitation, inhalation and last isolate/s amongst P. aeruginosa septicaemia patients (N = 15). D means dicult and S satisfactory No.
% Burn
Resuscitation
Intubation
Last isolate(s)
Outcome
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
82 42 61 90 90 25 75 70 50 50 45 90 85 67 70
D S D D S S D D D S D S S S D
ARDS not intubated not intubated inhalation inhalation not intubated inhalation inhalation ARDS not intubated not intubated not intubated ARDS not intubated not intubated
Klebsiella P. aeruginosa P. aeruginosa MRSA, Acenetobacter P. aeruginosa P. aeruginosa MRSA P. aeruginosa P. aeruginosa, S, aureus P. aeruginosa P. aeruginosa MRSA P. aeruginosa P. aeruginosa P. aeruginosa
died survived survived died died survived died died died died survived survived died survived died
of resuscitation, in three other patients associated inhalation injury was probably responsible. However, these patients responded to aggressive management with the needed amount of ¯uid administration, early colloid infusion, correction of acidosis and ventilatory support. Four patients were intubated on admission because of inhalation injury and three others at a later postburn day because of septicaemic complications (Table 2). They were intubated on the development of dysponea, decreased oxygen saturation, low arterial oxygen tension and chest X-ray suggestive of ARDS due to septicaemia. All seven of these intubated patients died and four of them had P. aeruginosa as their blood isolate prior to death, while three had infections with Klebsiella, Acinetobacter and MRSA, S. aureus and Pseudomonas as their last isolates prior to fatal outcome (Table 2). The isolates of 147 septicaemic episodes amongst 102 patients who had septicaemia during the study Table 3 Microorganism isolated amongst 102 patients with 147 septicaemic episodes Organisms isolated
No. of episodes
Percentage
MRSA MRSE Pseudomonas Acinetobacter Klebsiella Streptococcus Enterococcus Serratia Pasteurella hemolytica Staphylococcus aureus Mixed growth Total
64 17 18 12 2 6 4 1 1 4 18 147
43.5 11.6 12.2 8.2 1.4 4.1 2.7 0.7 0.7 2.7 12.2 100
period are depicted in Table 3. A total of 21 P. aeruginosa strains were isolated from 15 patients. These patients had a total of 36 septicaemic episodes during their stay in the hospital (Table 1). P. aeruginosa alone was isolated in 18 septicaemic episodes, while in the other three S. aureus, MRSA and Klebseilla were found in one episode each along with P. aeruginosa. The incidence of septicaemic episodes in relation to postburn day is depicted in Table 1. A 4-year-old female child with 82 % burn and dicult resuscitation had seven septicaemic episodes and ®ve of these were due to P. aeruginosa on 9, 12, 18, 23 and 30 days postburn, but her seventh episode was due to Klebseilla, which was associated with fatality. P. aeruginosa was found in the ®rst episodes in nine patients, in second episodes in six, in third episodes in three, and fourth ®fth and sixth episodes in one patient, each at a variable postburn day. Five episodes due to this organism were observed in one patient, two episodes each in two patients and one episode each in 12 patients thus accounting for 21 episodes in 15 patients. Septicaemic
Table 4 Surgical treatment and outcome Age Sex
% Burn No. of episodes No. of operations Outcome
4 11 40 35 50 3 35 29 27 51
82 42 61 90 25 75 45 90 85 67
F M F F M F F M M F
7 1 2 4 1 2 1 3 3 2
1 1 3 1 2 6 6 11 2 5
died survived survived died survived died survived survived died survived
R. Kumar Gang et al. / Burns 25 (1999) 611±616
episode due to P. aeruginosa occurred as early as the second day postburn and as late as the 65th day postburn. Over the 6-year period of the study several antipseudomonal drugs, such as ceftazidime, piperacillin, aztreonam, cipro¯oxacin, piperacillin/tazobactam, imipenem and meropenem along with aminoglycosides, such as amikacin were used depending on the culture and sensitivity results. Out of 15 patients 10 had 38 sessions of excision and skin grafting during the course of their treatment (Table 4). One patient with 90% burns had 11 sessions of skin grafting; he survived and was discharged on the 149th day postburn. Six out of 10 patients who were operated on survived. Five patients who were not ®t to be operated on died, two of them had septicaemia on the 2nd day postburn (Table 1) and one, who arrived from another country in critical condition on the 45th postburn day, died the day after. Twenty-six deaths amongst 102 total septicaemic patients during this period represent a mortality rate of 25.5%; in contrast, the mortality amongst P. aeruginosa septicaemia patients was higher. Nine patients died amongst 15 who had septicaemia due to this organism, representing a mortality rate of 60%, but only six patients (40%) (Table 5) had P. aeruginosa, either alone or in combination with others as isolates during their last septicaemic episodes prior to death (Table 1). Therefore, the mortality due to P. aeruginosa septicaemia is 40% because other three patients recovered following earlier P. aeruginosa septicaemic episodes (Table 1). However, out of the nine patients who died, three had organisms other than P. aeruginosa as their last blood isolate prior to death (Table 1). The mean percentage of burn in those who died was 73.5% (S.D.2 15.3%) compared to those who survived with 55.0% (S.D. 222.7%). Amongst nine patients who died, six had dicult resuscitation and seven were intubated (Table 2). The day of death was between 3 to 52 days postburn (mean 19 days) (Table 5) and all died due to septicaemia and its complications.
615
4. Discussion P. aeruginosa is a highly evolved nosocomial pathogen that is prevalent in the hospital environment [11] and is highly pathogenic in burn patients, contributing considerably to morbidity and mortality. Of the burned patients 25±29% became colonized with this organism during the course of their stay in hospital and a quarter of the colonized patients developed invasive infection [7]. The incidence of P. aeruginosa septicaemia in 15 (14.7%) out of 102 total septicaemic patients is low as compared to 18% reported by McManus et al. [7] and 26% observed by Lesseva and Hadjiiski [12]. The low incidence of P. aeruginosa septicaemia in this unit might have been due to better isolation facilities [13], dressing of the burn wound with ¯amazine [14], care for the nutrition and the early excision and grafting of the burn wounds. As noted in this study, the male to female ratio was 1:1.5, which is in contrast to the total incidence of burns as observed earlier in this unit where this ratio was 1.7:1 [15]. This was probably due to the fact that the majority of severe ¯ame burns occurred in females at home, either due to cooking gas accidents or due to clothes catching ®re. The loose clothes worn due to climatic and traditional reasons are more prone to catch ®re. The patients with ¯ame burns were more susceptible to P. aeruginosa septicaemia as observed in this study in contrast to the septicaemic episodes due to other organisms which was seen in all types of burns. As discussed earlier, ¯ame burns tend to be deeper with considerable tissue destruction which is ideal for P. aeruginosa colonization and proliferation. Moreover, extensive ¯ame burns cause considerable immunosuppression and this in turn makes the patient susceptible to sepsis. The burns involving >40% TBSA were more prone to P. aeruginosa septicaemia, while septicaemia due to other organisms may not be dependent on the percentage of burns. As observed in this study, resuscitation and intuba-
Table 5 Mortality amongst Pseudomonas aeruginosa septicaemia patients (N = 15) No.
Age (yr)
Sex
% Burn
Resuscitation
Intubation
No. of episodes
Death days postburn
1 2 3 4 5 6
47 30 4 18 27 12
M F F M M F
90 70 50 50 85 70
satisfactory dicult dicult satisfactory satisfactory dicult
inhalation inhalation ARDS no ARDS no
1 1 1 2 3 5
3 6 15 52 18 23
616
R. Kumar Gang et al. / Burns 25 (1999) 611±616
tion in¯uenced the occurrence of septicaemia in burn patients and ®nally their outcome. The Parkland formula proved to be quite adequate, however, early institution of ¯uid therapy also showed improved results. The delayed and dicult resuscitation proved to be an important risk factor for the acquisition of P. aeruginosa and resulted in a worse outcome. Both groups of patients either intubated prior to sepsis or following sepsis, were at grave risk as observed in this study as all the intubated patients had a fatal outcome. Routine microbiological surveillance of burn wounds was of great help in presumptive diagnosis of P. aeruginosa septicaemia and institution of early antibiotic therapy. As this study indicated early excision and skin grafting in¯uenced the outcome in P. aeruginosa septicaemia. Similar observations were also made by Sorensen and Thomsen [16] who observed a decline in severe P. aeruginosa infection following surgical removal of all necrotic tissues and immediate grafting. In spite of the decline in incidence, the mortality due to P. aeruginosa septicaemia, as observed in this study was high, and similar observations were also made by other workers [17]. The study thus showed that females with ¯ame burns of >40% TBSA coupled with inhalation injuries are more susceptible to P. aeruginosa septicaemia. Though the relative incidence of septicaemia due to this organism was low, mortality was quite high. Further septicaemia could occur quite early in postburn days. Delayed and dicult resuscitation and intubation proved to be signi®cant risk factors for acquisition of P. aeruginosa septicaemia and its fatal outcome. The early excision and grafting with other eective management may result in better outcome. References [1] Ninneman JL. Immunological defenses against infection: altera-
[2] [3] [4]
[5]
[6]
[7]
[8]
[9] [10]
[11]
[12] [13] [14]
[15] [16] [17]
tions following thermal injuries. J Burn Care Rehabil 1987;3:355±66. Winkelstein AU. What are the immunological alterations induced by burn injury? J Trauma 1984;24:S72±S83. Pruit Jr BA, MacManus AT. The changing epidemiology of infection in burn patients. World J Surg 1992;16:57±67. Deitch EA, Berg RD. Bacterial translocation from the gastrointestinal tract: a mechanism of infection. J Burn Care Rehabil 1987;8:475±82. Jones WG, Minei JP, Barber AE, et al. Bacterial translocation and intestinal atrophy after thermal injury and burn wound sepsis. Ann Surg 1990;211:399±405. Tredget EE, Shankowsky HA, Mark Joe A, et al. Epidemiology of infection with Pseudomonas aeruginosa in burn patients: the role of hydrotherapy. Clin Infect Dis 1992;15:941± 9. McManus AT, Mason Jr AD, McManus WF, Pruitt Jr BA. Twenty-®ve years review of Pseudomonas aeruginosa bacteremia in burn center. Eur J Clin Microbiol 1985;4:219±23. Richard P, Le Floch R, Chamoux C, et al. Pseudomonas aeruginosa outbreak in a burn unit: role of antimicrobials in the emergence of multiple resistant strains. J Infect Dis 1994;170:377±83. Bang RL, Gang RK, Sanyal SC, et al. Burn septicaemia: an analysis of 79 patients. Burns 1998;24:354±61. Bang RL, Gang RK, Sanyal SC, Mokaddas E, Lari ARA. Beta-haemolytic Streptococcus infection in burn patients. Burns 1999;25:242±6. Cross A, Allen JR, Burke J, et al. Nosocomial infection due to Pseudomonas aeruginosa: review of recent trends. Rev Infect Dis 1983;5(Suppl. 5):S837±S845. Lesseva MI, Hadjiiski OG. Staphylococcal infections in the So®a Burn Centre, Bulgaria. Burns 1996;22:279±82. McManus AT. Pseudomonas aeruginosa: a controlled burn pathogen? Antibiot Chemother 1989;42:103±8. Brouses BH, Coldwell FT, Conc JB, et al. A prospective analysis of silver sulphadiazine with and without cerium nitrate as a topical agent in the treatment of severely burned children. J Trauma 1981;21:558±63. Bang RL, Mosbah KM. Epidemiology of burns in Kuwait. Burns 1988;14:194±200. Sorensen B, Thomsen M. Clinical features and pathology of burns. Antibiot Chemother 1989;42:97±102. Kolmos HJ, Thuesen B, Neilsen SV, et al. Outbreak of infection in a burns unit due to Pseudomonas aeruginosa originating from contaminated tubing used for irrigation of patient. J Hosp Infect 1993;24:11±21.
