Herruzo-Cabrera R.,* Calle-Puffin E.,* Garcia-Torres V.,** Luengo-Matos W, Lenguas-Portero F,*** Rey-Calero J.*

* DptO Medicina Preventiva y Salud Pùblica, Hospital La Paz, Facultad de Medicina, Universidad Autèrioma de Madrid, Spain
** Dpto Cirurgia Plàstica y Reparadora, Hospital La Paz
*** Unidad de Cuidados Intensivos, Hospital La Paz

SUMMARY. A comparative study was performed to investigate the variations in the septicaemia rate occuring in the Critical Area of the Burn Unit in La Paz Hospital. Three time periods were compared: 1980-84, 1985-86 and 1987-90, coinciding with major modifications in preventive measures in bum patients. The analysis was made by direct standardization and multiple logistic regression. Our results show that during the most recent time period the incidence of septicaemia decreased significantly and that the decrease coincided with the introduction of intensive decontamination treatment.

Burn patient survival has improved over the last 50 years, as a consequence of advances in bum management (1). The reduction in mortality in a given period is due to the changes in treatment during that time (2, 3). Hydric therapy has elin-finated hypovolaemic shock as the major cause of death; the prevention and treatment of bum sepsis, which includes earlier surgical debridement, the use of topical antibiotics and systemic antibiotherapy in patients with systemic or urinary infection, etc., are factors that diminish the risk of microbial invasion and septicaemia (4).
Despite the wide use of antibiotcs, infections caused by multiple agents such as Pseudomonas, Staphylococcus aureus, E. coli, etc. are present in a large number of burn patients, and if septicaemia occurs burn patient mortality is elevated (5). At present, sepsis and its sequelae and respiratory complications are the most important causes of death in serious bum patients.
Environmental control to reduce the number and character of micro-organisms is very important in diminishing bum infections. Although the objective of strict isolation is to eliminate exogenous sources of infection (6), it cannot eliminate endogenous sources, particularly intestinal contamination. Endogenous sources can be direct, from the mouth or anus, and by intestinal translocation, which increases the permeability of the intestinal wall to micro-organisms from the lumen (7, 8, 9).
The prophylactic use of antibiotics to suppress colon flora, in addition to strict isolation techniques (10), has considerably decreased the number of bum infections. This joint effect is the theoretical basis for the establishment of selective intestinal decontan-fination measures exclusively for intestinal pathogens (11), since the indiscriminate use of antibiotics destroys intestinal microflora involved in host resistance to infection (12).
Another classic infection source in Intensive Care Units is the catheter (13).
The present study evaluates all previous therapeutic measures, including a protocolized treatment, called Intensive Burn Decontamination (IBD), which includes local and systemic decontamination, as well as early debridement, avoidance of manipulations, etc.
The incidence of septicaemia morbidity has been studied and compared during three different time periods, in our Burn Unit at La Paz Hospital.

All patients spending two or more days in the Critical Area of the Burn Unit in La Paz Hospital from 1980 to 1990 were studied.
The Bum Unit was a part of the Plastic Surgery, Department until 1985. In 1985, it became a multidisciplinary unit served by the Plastic Surgery Service, the Intensive Medicine Service and the Preventive Medicine Service.
The Critical Bum Unit has twelve single rooms.
There is a written protocol on the management of burn patients that is reviewed every two years and published as a guide.
The major modifications that have prevented infection in bum patients were:

• beginning in 1985, surgical prophylaxis; pharyngeal decontamination with hexetidine- spray; twice weekly microbiological monitorization;
• beginning in 1987, 1BD as follows: intestinal decontamination with three drugs, usually tobramycin 300 mg/day plus cholymycin 400 mg/day plus nystatin 100,000-150,000 U/Kg/day nasal decontamination with hexetidine and topical treatment of the burn itself with clorhexidine 0.5% cream (14);
• beginning in 1988, a topical treatment with clorhexidine 0.5% cream was initiated at the site of catheter insertion.

Between 1980 and 1984, patients were followed in a record kept by the plastic surgeons. Since 1985 this information has been recorded by the Preventive Medicine Service, using the same protocol during the entire period that the patient is in the Critical Burn Unit. The records contain information on the variables sex, age, percentage of burned body surface, previous disease, type of burn, etc., as well as data on manipulation, surgical treatment of the burns, treatment, burn infections and aetiology of the infection.
Septicaemia was diagnosed following CDC criteria and measured by:
Accumulated Incidence (A.I.): number of infected patients x 100 divided by the total number of patients spending two or more days in the Unit. The association between infection and qualitative variables was calculated with the Chi square test, and the association between infection and numerical variables was calculated with analysis of variance or Student's T test.
Multivariant analysis was performed with multiple logistic regression to control confounding factors. All variables that were significant in bivariant analysis were included in the multiple logistic regression analysis as independent.
Standardization was made in two different ways:

• by the direct method (15), which compared the incidence of septicaemia in the three periods with that of the first period;
• by individual probability estimation (16), which included each patient's value in the logistic regression equation. The probability of septicaemia during the first period was taken as the referent level.

The characteristics of the population are described in Table 1, for the different time period studied.
During the second and third periods the burns were significantly larger. The number of patients with more than 30% BSA affected was high (over 30% of the total patient number). During the period 1985-86 an increase in bum depth was observed, the number of subdermal bums being higher than in the period 1987-89. These differences were not however statistically significant.
Patients in the most recent period were older, they presented more severe bums and they had a high percentage of previous diseases. the frequency of the different aetiological agents. Septicaemia became less frequent during the study period, excluding fungal infections, which did not occur during the first time period at all, and decreased in the last period compared to the second period. The significant increase of Gram+ cocci different from S. aureus coincided with the decrease in the presence of S. aureus.
Figs. 1 and 2 show septicaemia frequency, and Tables III and IV refer to the A.I. of septicaemia, i.e. the real incidence, the standardized incidence calculated by the percentage of burned body surface, and the accumulated incidence of septicaemia calculated by multiple logistic regression, taking the first period as the reference. The incidence of septicaemia decreased during the second (30%) and the third period (60%).
With regards to septicaemia aetiology, differences between real incidence and estimated incidence were significantly reduced in the second and third periods (by 50% and 85% respectively), excepting septicaen-iias caused by fungi and Staphylococcus different from S. aureus.
In general, the septicaemia percentage decreased during 1987-90 compared to 1985-86.

Fig. 1* = statistically significant difference p<0.05 aetiology &= p<0,1 Fig. 2: Septicaemia aetiology in burn patients over 10 years (3 periods)

Discussion

The incidence of septicaemia decreased during the study period by about 51-60%, depending on the method of measurement and the aetiological agents. The decrease was associated to the treatment used in each time period. The reduction in the A.I. of sepsis for some aetiological agents, such as S. aureus, P. aerugi . nosa and Enterobacteriaceae, which was 58% lower in the less favourable situation, was marked. As we demonstrated in a previous paper (14), and in agreement with the Manson study (11), intensive treatment decreases nasal and intestinal colonization, which is the most important source of infection in the bum patient.
The number of non S. aureus isolated Gram+ cocci increased during 1985-86, but this percentage decreased when clorhexidine 0.5% replaced argentic sulphadiazine in topical bums therapy and the use of clorhexidine 0.5% cream was begun at the sites of catheter insertion. However, we think that the increase in the number of Gram+ cocci infections ' compared to 1980-84, was due to intensive therapy with parenteral feeding via the central venous system.
The appearance of fungal septicaemia in the second time period, but not the first, was probably due to the intensive therapy, which increased burn patient survival although selecting some fungi (17). Fungal selection explains the large increase in fungal septicaemia during the second study period and the decrease in its incidence, to one quarter of the second study period value, in the third period, coinciding with 1BD.
Local and systemic decontamination, plus other treatment measures, such as surgical debridement, frequent surgical interventions and prophylaxis at catheter insertion sites, have significantly decreased the number of septicaemias in the Critical Bum Unit at our Hospital.

DESCRIPTION OF POPULATION STUDIED TIME PERIOD   80-84 85-86 87-90 N' of Patients 309 299 457 Type of bum no % no % no % Flame 128 (41.8) 122 (41.1) 240 (52.5) Scald 89 (28.8) 48 (16.2)* 55 (12.0) Electrical 32 (10.3) 36 (12.2) 52 (11.4) Chemical 5 (1.6) 9 (3.0) 16 (3.5) Explosion 35 (11.3) 19 (6.3) 45 (10.5) Other 20 (6.5) 63 (21.2)** 46 (10.1)**         Previous Disease       Epilepsy 9 (2.9) 11 (3.7) 9 (2.0) Neurological 19 (9.2) 14 (4.7) 25 (4.9) Heart 8 (2.5) 2 (0.7) 8 (1.7) Alcoholism 5 (1.6) 3 (1.0) 22 (4.8)* Drug abuse - - 2 (0.7) 10 (2.0) Others 83 (26.9) 45 (15.1)* 129 (28.3)** No known prev. dis. 185 (60.6) 220 (74.1) 254 (55.6)*         Bum depth       Superficial 128 (41.9) 88 (29.6)* 100 (21.9) Deep 103 (33.8) 157 (52.9)* 195 (42.7) Subdermic 74 (24.3) 52 (17.5) 161 (35.0)**         % Burned Body surface 13.5+ (0.9) 21.9+ (1.3)** 22.9+ (1. 1) <10% 171 (55.3) 115 (39.9)** 172 (39.3) 10-30% 103 (33.3) 85 (29.5) 128 (29.2) >30% 35 (11.3) 88 (30.6)** 138 (31.5)         Mean Age 37.2 1.1 38.1 1.1 42.9 1 1 st Debrid. (day) 6.7 0,4 10.3 0.7 6.5 0.6** last debrid. (day) 16.6 0.9 14.6 0.9 15.7 1.5 Operations per patient 2.1 0.1 1.7 1.1 2.1 0.1 Length of stay (days) 23 1.2 13.5 0.8** 16.7 0.8** Parenteral feeding (%) 41 (13.7%) 83 (28%)* 102 (22.3%) Accumulated incidence of septicaernias 34 (11.0%) 36 (12.2%) 32 (7.4%)* Table 1

 

STUDIED PERIODS 80-84 85-86 87-90 Real A.I. 11% 12% 7.4% A.I. by direct standardization 11% 17.8% 18.6% Estimated A.I. by MLR standardization 11% 14.8% 15.1% Changes in real A.I. with regard to 1 st period* 0% +10.9% -32.7% Changes in estimated A.I. by MLR standardization* 0% -17.6% -51% Table II

 

STUDIED PERIODS   80-84 85-86 87-90   percentage real incidence/estimated incidence S. Aureus 2.1 13/3.4 L5/3.6 Gram+ Cocci (no S. aureus) 0.6 7/0.97 2.6/1 P. Aeruginosa 5.8 2.7/9.4 2.2/9.8 Enterobacterial 2.4 13/3.9 0.6/4.1 Fungi* 0* 1,10.2 0.4,10.2 Others 0.7 0.3,11.1 0.4/1.2 Table III

 

STUDIED PERIODS   85-86 87-90 S. Aureus -61.8% -58.3% Gram+ Cocci (no S. aureus) +521% +260% P. Aeruginosa -71.3% -77.4% Enterobacterial -66.7% -85.4% Fungi +400% +100% Others -72.3% -66.7% Table IV

RESUME. Les auteurs ont effectué une étude comparative pour examiner les variations de la fréquence de la septicémie qui s'est produite âa le secris cteur critique de l'Unité de Brûlures à l'Hôpital La Paz. Ils ont confronté trois différentes périodes: 1980-84, 1985-86 et 1989-90, qui coïncidaient avec des modifications importantes dans les mesures préventives chez les patients brûlés. L'analyse a été effectuée par la standardisation directe et la régression logistique multiple. Les résultats démontrent que pendant la période la plus récente la fréquence de la septicémie a diminué en % en manière significative et que cette diminution a coïncidé avec l'introduction d'un traitement intensif de décontamination.

---

BIBLIOGRAPHY

1. Thompson P.B., Herndon D.N. Traber D.L. et al.: Effect on mortality of inhalation injury. J. Trauma, 26: 63-9, 1986.
2. Tompkins R.G., Burke J.F., Schoenfeld D.A. et al:. Prompt scar excision: a treatment system contributing to reduced bum mortality. Ann. Surg., 128: 272-81, 1986.
3. Dobke M.K., Simoni J., Ninneman J.L., Garret J., Dressel L.: Postburn endotoxemia: effect of early excision on circulating endotoxin levels. Paper presented at Annual Meeting, American Bum Association, Seattle, 1988.
4. Herruzo-Cabrera R., Vizcaino-Alcaide M.J., Mayer R.F., Rey-Calero J.: A new in vitro model to test the effectiveness of topical antimicrobial agents. Use of an artificial scar. Bums, 18: 35-8, 1992.
5. Bang R.L., Saif J.K.: Mortality from bums in Kuwait. Bums, 15: 315-21, 1989.
6. Settle J.A.D.: Infections in burns. J. Hosp. Infect, 6: 19, 1985.
7. Carter E.A., Gonnella A., Tompkins R.G.: Increased transcellular permeability of rat small intestine after thermal injury. Bums, 18: 117-20, 1992.
8. Ziegler T.R., Smith R.J., O'Dwyer S.T. et al.: Increased intestinal permeability associated with infection in burn patients. Arch. Surg., 123: 1313-7, 1988.
9. Border J.R., Hasset J., La Duca J. et al.: The gut origin septic states in blunt multiple trauma (ISS=40) in the ICU. Ann. Surg., 206: 427-48, 1987.
10. Jarret F., Balish E., Moylan J.A. et al.: Clinical experience with prophylactic antibiotic bowel suppression in burn patients. Surgery, 83: 523, 1978.
11. Manson W.L., Klasen H.J., Sauer E.W., Olieman A.: Selective intestinal decontamination for prevention of wound colonization in severely burned patients: a retrospective analysis. Bums, 18: 98-102,1992.
12. Van der Waay D., Manson W.L., Arends J.A. et al.: Clinical use of selective intestinal decontamination: the concept. Intensive Care Med., 16: 5212, 1990.
13. McMillan B.G., Holder I.A., Alexander J.V.: Infections in burn wounds. In: "Hospital infections", Bennet J.V., Brachman P.S. (Eds), Little Brown, Boston, 1986.
14. Herruzo-Cabrera R., Garcfa-Torres V.; Rey-Calero J., Vizcaino-Alcaide M.J.: Evaluation of the penetration, strength, bactericidal efficacy and spectrum of action of several antimicrobial creams against isolated microorganisms in a burn center. Bums, 18: 39-44, 1992.
15. Rotman K.J.: "Epidemiologia Moderna". Dfaz de Santos,Madrid, 1987.
16. Kleinbaurn D.G., Kupper L.L., Morgenstern H.: "Epidemiologic Research", Van Nostram Reinhold, New York, 1983.
17. Inoue S., Peck M.D., Alexander W.: Fungal translocation is associated with increased mortality after thermal injury in guinea pigs. J. Bum Care Rehabil, 12: 19-22, 1991.

---