MORTALITY OF THE PAEDIATRIC BURN POPULATIONTREATED AT THE VIRGEN DEL ROCIO UNIVERSITY HOSPITAL,SEVILLE, SPAIN IN THE PERIOD 1968-1999

Annals of Burns and Fire Disasters - vol. XIII -n. 2 - June 2000

MORTALITY OF THE PAEDIATRIC BURN POPULATION TREATED AT THE VIRGEN DEL ROCIO UNIVERSITY HOSPITAL, SEVILLE, SPAIN IN THE PERIOD 1968-1999

Gornez-Cìa T., Franco A., Mallén J.M., Gimeno M.A., Ferndndez-Mota A., Màrquez T., Portela C., Lòpez I.

Plastic and Reconstructive Service, Burn Unit, Virgen del Rocìo University Hospital, Seville, Spain

SUMMARY. Changes in medical treatment protocols have led to a significant reduction in the mortality of the paediatric population treated at the Burn Unit of the Virgen del Rocfo University Hospital in Seville (Spain) in the last 30 yr. An analysis was made of the probability of death in two groups of patients under 9 yr with burns in:5 45 % body surface area: Group A (1451 admissions between 1968 and 1977) and Group B (312 patients admitted between January 1993 and January 1999). In Group A, approximately one-hundredth of the patients admitted with the above characteristics died, while in Group B the mortality rate was one in three hundred. The findings in these two groups of patients during the two periods are compatible with those published by other authors. Some changes in the clinical protocols might explain the improvements achieved, e.g., resuscitation therapy adapted to each individual case, with early use of colloids; strict hypercaloric and hyperproteic nutrition; early escharectomy and subsequent closing of the wound.

Introduction

Burns in the paediatric population are frequent. According to the Center for Disease Control in Atlanta, USA, every year more than 1000 children under 15 yr of age die in the United States in domestic fires, equivalent to almost two-thirds of deaths in children under 4 yr of age. Approximately 20% of the patients admitted to the Burn Units in our country are under 9 yr of age. In other regions, with a lower social and economic development, this phenomenon is even more frequent since the birth rate is much higher than in our country, the mean age of the population is lower, and the probability of suffering a burns accident burns increases exponentially in relation to inadequate or poorly applied fire prevention regulations.
Certain specific characteristics of children make their response to a thermal accident different from that of an adult. The body proportions differ. Blood volume, for example, after adjustment for body weight, is greater in children. The loss of body fluids through the burned area isproportionally greater in the young, owing to the difference between weight and body surface, with respect to age. These and other circumstances have led many researchers to recommend that the fonnulae for fluid restoration during the shock phase should be adjusted to the paediatric patient's body surface, or that human seroalbuntin should be added early to resuscitation fluids.
According to Wolf et al., lower age, larger burn size, the presence of inhalation injury, delayed intravenous access, sepsis, inotropic requirements, and ventilator dependency during the hospital course are significant predictors - among other variables - of increased mortality.
The high incidence of infantile burns and the severity of sequelae in the child's subsequent development have drawn special attention to this risk group in our Burn Unit, since its inauguration in 1968. The aim of this paper is to analyse the effects that advances in burn patient treatment have had in the increased probability of survival of our paediatric patients.

Material and metods

The data regarding children under 9 yr of age admitted to the Burn Unit of the Virgen del Rocìo University Hospital, Seville, Spain between July 1968 and December 1976 (8.5 yr) (Group A) were compared with those of children admitted between January 1993 and January 1999 (6 yr) (Group B).
The data concerning the patients in Group A have been published previously.
The following parameters were prospectively selected from the clinical history of the patients in Group B: age, percentage body surface area (BSA) burned, and survival.
The treatments given to the two groups differed. The resuscitation of children in Group A was performed in relation to body weight, i.e. 1.5 cc Ringer's lactate solution per kg body weight per % burned BSA plus 0.5 cc per kg body weight per % BSA of human plasma, together with the usual daily fluid requirements adjusted in relation to the child's weight.
Half the estimated volume was given in first the 8 h, and half in the subsequent 16 h, with certain limitations: this calculation excluded burns in more than 30% BSA in children under 3 yr, even if the burn was more extensive, and the normal daily fluid requirements were administered as 5% dextrose in water. Oral administration of fluids and intravenous infusion of colloids were initiated in the following 24 h, decreasing to half the requiremerits of electrolytes and fluids calculated for the first day.
In Group B, the resuscitation of the patients with critical burns varied in relation to age: for children of 5 yr the guidelines of the Shriners Burns Institute (Galvestone, Texas) were followed: 5000 cc/m² burned BSA, plus 2000 CC/M2 BSA as normal daily fluid requirements, during the first 24 h. Half this volume was administered during the first 8 h, and the volumes were reduced to 3750 CC/M2 plus 1500 cc/m² , respectively, on the second day. Ringer's lactate solution with alburnin was used to maintain levels of this protein at approximately 2.5 g/dl in plasma. 2,3 Patients with a body weight exceeding 40 kg and those aged 5 yr were reanimated following the B.E.T. guidelines:220 cc/m² BSA of human seroalburnin solution in Ringer's lactate, at a concentration varying in relation to the time period following the thermal injury (10 g/dl between 0 and 8 h, 7.5 g/dl between 8 and 16 h, 5 g/dl until the end of the first 24 h, 2.5 g/dl between 24 and 40 h, and 0 g/dlthereafter).
Escharotomies were performed immediately when necessary.
The topical therapy of burns in both groups consisted of the daily or twice daily application of 1% silver sulphadiazine. In Group B, depending on the aspect of the injury, cerium nitrate-silver sulphadiazine was at times applied (Flammazine and Flammazine cerium, Solvay Pharma Inc.).
While the patients in Group A were given an oral hypercaloric and hyperproteic diet, the severely burned children in Group B received oral nutrition, mixed enteral and oral nutrition, enteral nutrition, or parenteral nutrition alone, depending on the clinical situation. The daily energy requirements of Group B patients were calculated according to the modified Long formula and the estimates were confirmed weekly by indirect calorimetry (Monitor Deltatrac II, Datex Inc., Helsinki, Finland). The protein requirements of Group B patients were obtained from the daily nitrogen balance.
In Group B patients, analgesia was mainly performed by administration of morphine chloride. The levels of analgesia obtained were measured at frequent and regular intervals. Patient-controlled analgesia devices (Pain Management Provider, Abbott Laboratories, North Chicago, Illinois, USA) were frequently used. The most painful procedures and surgery were performed in collaboration with the Department of Anaesthesia.
The surgical treatment of wounds in Group A patients was conservative, pending spontaneous escharectomy of the wounds, at times aided by the use of scissor-tips, and also pending the growth of healthy granulation tissue in the wound to be grafted.
However, in Group B patients with deep burns, tangential or fascial. excisions were made, with temporary or definitive closing of the surgical wound during the first 24-48 h post-burn. For the temporary closing of wounds in Group B children, semi-synthetic biological dressings (Biobrane, Dow Hickan Pharmaceuticals Inc., Texas, USA) or cryopreserved putaneous homografts obtained from multi-organ donors from our hospital were used. Definitive closure of the burn was performed, in both groups, with laminated autografts, meshed skin autografts (Tanner Vanderputt, Zimmer Inc.), or flaps, according to the indications and the availability in each particular case.
In Group A the donor zones in partial-thickness skin grafts were treated with a petrolatum gauze impregnated with an antiseptic ointment, while the donor areas in children in Group B were treated with a hydrocolloid dressing (Varihesive Gel Control, Bristol Myers Squibb Inc.), after haemostasis with the topical application of an adrenaline solution in isotonic saline (1/500,000), which reduced pain and shortened the time needed to treat the donor sites (unpublished data).
The various complications of infectious, respiratory (lesions due to inhalation of toxic substances), traumatological, or other aetiologies caused by the accident or appearing during the cases' clinical course were managed in collaboration with the specialists in our hospital.
The comparison between the two patient groups was made by graphic analysis (level curves) of the patients' probability of death, according to age and percentage BSA burned (published in 1980), for Group A, and according to the incidence observed in Group B.For the purpose of comparison, some of the results published by two different study groups in two different time periods, 1954 and 1991, are represented graphically. No other type of analysis of the differences observed was performed.

Results

Group A consisted of 1490 burned children aged less than 9 yr hospitalized between 1968 and 1976, i.e. over a period of 8.5 yr. Group B consisted of 312 burned patients of similar age hospitalized between 1993 and 1998 inclusive, i.e. a 6-yr period.
The population served by our Burn Unit has been practically constant since its inauguration (about 7,000,000 persons) but an important reduction has been observed in the incidence of annual hospitalizations: from 175 children per yr in Group A to 50 children per yr in Group B. This could be related to the prevention campaigns that have been conducted.
The burned BSA in patients in Group A ranged from I to 90%. All the patients in Group B had burns in 45% BSA (Fig. 1).
In Group A, 1451 children had burns in 45% BSA, with a greater extent in 24 patients. For the sake of adequate comparison, these patients were excluded from the analysis

Group A		Age (yr)
% BSA	0-1	2-3	4-8
0-5	0	0	0
6-10	0	0	0
11-15	0	0	0
16-20	0	0	0
21-25	0.1	0	0
26-30	0.2	0.1	0
31-35	0.4	0.2	0
36-40	0.6	0.3	0.1

Fig.1 - Group B consisted of 312 burned children aged under 9 yr admitted to the Bum Unit of the Virgen del Rocfo University Hospital, Seville between January 1993 and January 1999. All the patients in less than 45% BSA and the majority less than 20%

Table I - Theoretical probability of mortality, for 1,451 children aged up to 8 yr, treated during the 1970s in the Burn Unit of the Virgen del Roefo University Hospital (UVRH) Seville' (1 = 100% mortality; 0 = no mortality). Published with authors' permission.

Fig.1 - Group B consisted of 312 burned children aged under 9 yr admitted to the Burn Unit of the Virgen del Rocfo University Hospital, Seville between January 1993 and January 1999. All the patients in less than 45% BSA and the majority less than 20%.

In Group A, 32 patients died during hospitalization in the Unit, fourteen of whom with burns in 45% BSA, an approximate ratio of one per hundred. Only one patient in Group B died out of the 312 patients hospitalized. This was a one-year-old boy hospitalized after a scald affecting 15% BSA in the face, neck, chest, and abdomen and requiring mechanical ventilation because of laryngeal stridor less than 12 h after admission. The child died 45 days after the accident, in the paediatric ICU of our hospital, where he had been put on mechanical ventilation to treat a distress respiratory syndrome. The cause of death was multi-system organ failure. At the moment of death the epithelialization of the burns was almost complete. Table I shows the theoretical probability of death, published in 1980, of patients in Group A. It can be observed that the probability of death was greater for younger children with more extensive burns.

Group B		Age (yr)
% BSA	0-1	2-3	4-8
0-5	0	0	0
6-10	0	0	0
11-15	0.009	0	0
16-20	0	0	0
21-25	0	0	0
26-30	0	0	0
31-35	0	0	0
36-40	0	0	0

Fig. 2 - Representation on an area plan with equivalent probabilities of death. Twenty years ago (Group A) (see Table I) in our Unit children had a lower probability for survival adjusted according to age and bum extent than at present (Group B) (see Table II) (1 = 100% mortality; 0 = no mortality).

Table II - Incidence of mortality in 312 children aged up to 8 yr of age treated in 1993-98 (1 = 100% mortality; 0 = no mortality).

Fig. 2 - Representation on an area plan with equivalent probabilities of death. Twenty years ago (Group A) (see Table I) in our Unit children had a lower probability for survival adjusted according to age and burn extent than at present (Group B) (see Table II) (1 = 100% mortality; 0 = no mortality).

Table II shows the incidence of death observed in the 312 children in Group B. The probability of death decreased significantly in our Unit from one out of every hundred children in Group A to one out of every three hundred children in Group B.
Fig. 2 compares, by means of level curves, the probability of death calculated for children in Group A versus the incidence of death in Group B. Certain patients in Group A had a high probability of death (up to 0.6) in burns approaching 40%. This was not the case in Group B. Certain patients in Group A had a high probability of death (up to 0.6) in burns approaching 40%. This was not the case in Group B.

	MGH	Seville 1970s	FHS 1996	Seville 1990s
0-10%	0	0	0	0
11-20%	0	0	0.1	0.006
21-30%	0.1	0.066	0.12	0
31-40%	.25	0.266	0.15	0

Table III - Comparative study of the probability of death in different centres and periods for burn patients aged under 9 yr (Massachusetts General Hospital, 1954; Seville, 1970s, Group A~; FSH, 1996` (1 = 100% mortality; 0 = no mortality). The Seville 1990s column is equivalent to Group B and represents the incidence of death measured in the 312 children treated in our Unit in the last 6 yr.

Fig. 3 - The probability of post-burn death in different centres and periods in patients aged under 9 yr, compared with the calculations performed in our Unit. Our results from the 1970s matched those that were acceptable at that period of time and were clearly inferior to results obtained at present (NIGH, 1954;9 Group A, Seville, in the 1970s;' FSH, 1996` (1 = 100% mortality; 0 = no mortality). In the Seville 1970s graph, we represent the incidence of death measured in our Unit for the 312 patients in Group B (see Table III).

Table III and Fig. 3 express the probability of death after burns in different centres and periods for patients aged less than 9 yr. Our probabilities of death, adjusted for age and burn extent in the 1970s, were similar to those published by Bull and Fishr and significantly lower than present-day probabilities.
Mortality in Group B patients was found to be within the ranges of patients currently admitted. The results show an important reduction in the mortality of children hospitalized in our Burn Unit during the 30-yr period analysed.

Discussion

As long ago as 1980 we stated that the treatment of burn patients had made important progress in our country because such treatment was now being given in the specially designed centres to which patients were normally referred. There were, however, certain problems: children with burns frequently arrived late at the Unit, having been sent on from other medical or surgical units that were not specialized in burns management and having received inadequate therapy. Problems were thus encountered in relation to the primary attendance and evacuation of the critically ill burn patient. These problems have now been solved by the creation in our country of a National Healthcare Service that co-ordinates specialized care and of Emergency Healthcare services that facilitate a more satisfactory emergency treatment at the location of accidents as well as the evacuation of patients with multiple trauma, in particular those suffering from burns. Mortality related to burn injuries in Spain is thus improving.
Medical care protocols for burn patients, particularly in relation to the paediatric population, have also shown an outstanding improvement since the creation of this Unit in 1968. Essentially, the difference lies in the more active treatment of the patients with the intention of preventing complications and treating them early as soon as they occur: resuscitation adapted to the individual necessities of each patient, early hypercaloric and hyperproteic nutrition, excision of the burn wound, immediate temporary or definitive closure, correct application of medication, the use of a burn unit antimicrobial policy adapted to suit the particular context of the Hospital, etc.
It is our logical wish to analyse whether the application of updated protocols for the assistance of burn patients has led to an improvement in our clinical results. Of the many possible methods of conducting such a study, we chose that of comparing the probability of survival, adjusted in relation to age and burn extent, in two different periods in our Burn Unit, which has been open and operative for over 30 yr - since 1968. The option for this type of analysis was partially motivated by the fact that the publication of Franco et al. in 1980 was the only study, among those performed in our Unit, that was old enough - with a 20-yr difference - to permit a reliable comparative analysis with recent data. No other similar studies exist that were, performed with other groups of patients treated at this Unit.
The percentage of burned BSA and age are determinant factors in burn severity, and these two parameters are therefore frequently considered in the prognosis of patients admitted to burn units. It has also been demonstrated that besides age and burned BSA there are other decisive risk factors regarding burn severity. However, predictions based on age and the percentage of burned BSA alone have sufficient specificity to fulfil clinical and scientific goals, although the low sensitivity decreases the precision of the prediction at the moment of hospitalization if additional factors are not considered.
We therefore decided that the survival curves of the paediatric population attending our Unit, in the two periods during which the care protocols were clearly different,could be used to evaluate the effectiveness of the medical treatment given. We selected the paediatric population as being a representative sample because of the high proportion of hospitalizations in relation to the total number of admissions to our Centre, and to the existence of an extensive historical control group. Unfortunately, no comparable study has been performed using the total number of patients, both paediatric and adult, seen in our Unit during the period of time in question.
The present results, at least in our view, would appear to confirm the improvement in the clinical care provided in our Unit for burned paediatric patients since its creation 30 yr ago.
However, these results alone do not confirm the validity, of our clinical activity and we therefore compared our results with those published by other researchers. We selected a classic reference published in 1954 by Bull and Fisher and - as a representation of burn units with currently confirmed results - findings from the Burn Unit of the United States Army Institute of Surgical Research, among the many possible units. It is evident, when comparing data of the early 1950s with those of the early 1990s, that the improvement in the clinical protocols of the burn patient care has significantly increased postburn survival, including that of paediatric patients (Fig. 3).
An in-depth statistical analysis has not been possible because of the lack of comparable data between the groups analysed.
We conclude that the application of updated clinical protocols to the treatment of burned children permits the achievement of results that can be homologized, with a significant decrease of mortality in our Unit compared with historical data over the thirty years studied. In our opinion, mortality curves related to age and percentage burned BSA are a simple and effective tool for the comparison of clinical results from different burn units, as long as the population analysed is as large and as homogeneous as possible. The time-course of this type of curve can be useful for the assessment of the clinical efficacy of a particular burn unit and the impact that changes in strategy have on the final outcome.

RESUME. Les modifications des protocoles du traitement médical ont porté a une réduction significative de la mortalité de la population pédiatrique traitée dans l'Unité des Brûlures de l'Hôpital Universitaire Virgen del Rocìo A Séville (Espagne) pendant les derniers 30 ans. Les Auteurs analysent la probabilité de la mort dans deux groupes de patients âgés de moins de 9 ans atteints de brûlures dans £ 45% de la surface corporelle. Le groupe A était composd de 1451 enfants hospitalis6s entre janvier 1968 et 1977 et le groupe B de 312 enfants hospitalisés entre janvier 1993 et janvier 1999. Dans le groupe A, approximativement un patient sur 100 atteints de brûlures de l'extension en question est mort, tandis que le taux de mortalité dans le groupe B était un sur 300. Les résultats des Auteurs pendant les deux périodes analysées sont compatibles avec les résultats publiés par d'autres Auteurs.Certaines modifications dans les protocoles cliniques pourraient etrê responsables des améliorations, c'est-a-dire une thérapie réanimatoire adaptée selon les cas, avec l'emploi précoce de colloïdes, une nutrition stricte hypercalorique et hyperprotéique,l'escharectomie précoce, et la fermeture de la lésion.

BIBLIOGRAPHY

Centre for Disease Control. 1986 Mortality Data. Atlanta, Georgia. CDC, 1987.

Rutan R., Benjamin D.: Special considerations of age: The paediatric burned patient. In: "Total Burn Care", Herndon D. (ed.), W.B. Saunders, London, 350, 1996.

Carvajal H.: A physiological approach to fluid therapy, in severely burned children. Surg. Gynecol. Obstet., 150: 379-84, 1980.

Wolf S., Rose L, Desai M., Mileski L, Barrow R., Herndon D.: Mortality determinants in massive paediatric burns. An analysis of 103 children with ³ 80% TBSA burns (³ 70% full-thickness). Aim. Surg., 225: 554, 1997.

Franco A., Rus L, Gimeno M., Ferndndez-Mota A., Ambel R, 17Alonso L: Estudio de las causas de muerte en niflos quernados.Cirugfa PIdstica Ibero-Latinoamer, 6: 133-44, 1980.

Gòmez-Cìa T., Roa L.: A burn patient resuscitation therapy 18 designed by computer simulation (B.E.T). Part 2: Initial clinicalvalidation. Burns, 19: 332, 1993.

Long C.: Energy expenditure of major burns. J. Trauma, 19: 904,1979.

Lochbultler H., Mettli M.: Current concepts in paediatric burn care: Surgery of severe burns. Eur. J. Paediatr. Surg., 2: 201-4, 1992.

Bull J.P., Fisher A.J.: A study of burns at the Massachusetts General Hospital, 1939-1954. Arm. Surg., 145: 210, 1954.

Pruitt B.A., Mason A.: Epidemiological, demographic and outcome characteristics of burn injury. In: "Total Burn Care", Herndon D. (ed.), W.B. Saunders, London, 13, 1996.

Benito-Ruiz L., Navarro-Monzonis A.,Baena-Montilla P., MirabetHipolito V.: An analysis of burn mortality: A report from a Spanish regional burn centre. Burns, 17: 201, 1991.

Reig A., Tejerina C., Baena P., Mirabet V.: Massive burns: A study of epidemiology and mortality. Burns, 20: 51, 1994.

Herruzo-Cabrera R., Femandez-Arjona M., Garcia-Torres V. , Martinez-Ratero S., Lenguas-Portero R, Rey-Calero L: Mortality evolution study of burns patients in a critical care burn unit between 1971 and 1991. Burns, 21: 106, 1995.

Barret P., Gomez P., Solano L, Gonzalez-Dorrego M., Crisol F.: Epidemiology and mortality of adult burns in Catalonia. Burns, 25: 325, 1999.

Tobiasen L, Hiebert L, Edlich R.: A practical burn severity index. J. Burn Care, 3: 229, 1982.

Germann G., Barthold U., Lefering R., Raff T., Hartmann B.: The impact of risk factors and pre-existing conditions on the mortality of burn patients and the precision of predictive admission-scoring systems. Burns, 23: 195-203, 1997.

Gòmez-Cía T., Mallén J, Márquez T., Portela C., Lòpez I.: Mortality according to age and burned body surface in the Virgen del Roefo University Hospital. Burns, 25: 317, 1999.

Smith D., Caims B., Rarnadan R, Dalston L, Falchry S., Rutledge R., Meyer A., Peterson H.D.: Effect of inhalation injury, burn size, and age on mortality: A study of 1447 consecutive burn patients. J. Trauma, 37: 655, 1994.

Ryan C., Shoenfeld D., Thorpe W., Sheridan R., Cassent E., Tompkins R.: Objective estimates of the probability of death from burn injuries. N. Engl. J. Med.: 338: 362, 1998.

This paper was received on 11 February 2000

Address correspondence to: Dr Tornds Gòmez-Cìa, Ayda. Jerez 8, Bajo A, 41012 Seville, Spain. E-mail: tgcia@hotinail.com