Burns (1988) 14, (1). l-6 Printedin Great Britain Fever in burned children Charmaine Childs Regional Paediatric Burns Unit, Booth Hall Children‘s Hospital, Manchester and the North Western Injury Research Centre, University of Manchester, UK Summary The patterns of rectal temperature (r,) and skin sur- face temperatures ( T,h,F:h and T,,,,) of 40 children were studied during the first 48 h after burn injury. The patients’ ages ranged from 5 months to 12 years and all required intravenous resuscitation. On admission T, measured 2-6 h after the burn, ranged from 36.5 to 39+“C and 19 children had temperatures >37,5”C; the upper limit of the normal range for healthy children. Patients S-12 months old had the lowest 7’ , during the early period. During the first 12 h T, rose rapidly and 7.7 oatients (67 oer cent) had a fever (T,>38.5”C). In 10 Children ‘r, kas >40~o”C. There was no relationship between the rise in T, and the sex of the patient, the ambient temperature, the size. site or type of burn. Age was the most important variable. particularly dur- ing the first 12 h. where a positive correlation between r, and age occurred in patients aged 5-23 months (P<O~Ol, r=O.S3). Throughout the first 48h r, re- mained elevated. Trhlfh closely followed the pattern of T,. but T,,,, varied considerably, reflecting cardiovascu- lar as well as thermoregulatory responses. INTRODUCTION 1~ IIAS been observed that many of the children admitted to this Burns Unit develop a fever within a few hours of being burned. This is unexpected so soon after the injury and the mechanism is not yet understood. In contrast, adult patients treated in the Regional Burns Unit do not have such high temperatures during the early hours after the burn. Experimental and clinical studies have shown that injury affects thermoregulation (Stoner, 196X; Little and Stoner. 1981; Little et al.. 1986; Stoner, 1986), but the clinical studies have been confined to adults. Little has been reported about the responses to thermal injury in children, parti- cularly infants and young children, and practical- ly nothing about the changes in their body 0 1988 Butterworth & Co (Publishers) Ltd 0305-4179/x8/010001-06 $03~00 temperature. From the work referred to above it might be expected that body temperature would fall during the first few hours after the burn, then rise above normal after a few days when recovery had begun. This later pyrexia seen in the adult burn patient during what Cuthbertson (1942) has called the ‘flow phase’ is well documented (Wil- more et al., 1975, 1977). The control of body temperature depends on a complex balance of heat production and heat loss. This balance could be disturbed by endoge- nous factors associated with the burn- nociceptive afferent stimulation, peripheral vaso- constriction related to hypovolaemia, loss of skin areas specifically concerned in heat transfer. in- creased evaporative heat loss from the burn sur- face, endocrine changes, production of polypeptides such as interleukin-l--or by exoge- nous factors altering the environment, occlusive dressings and ambient temperature. Until the role of these, or other factors, in the production of this early fever is known, physiologically appropriate treatment cannot be given to these children. As the first step towards this, the char- acteristics of this fever must be determined. as well as identifying groups of children most at risk of developing fever. This paper describes the changes in core and skin temperatures during the first 48h after burn injury in children. A pre- liminary account of this work was presented to the First Congress of the European Burns Asso- ciation (Childs. 1985). PATIENTS AND METHODS With the exception of one patient (male, aged 5 months with a 9 per cent full thickness contact burn) all the patients studied had sustained burns
Transcript
Burns (1988) 14, (1). l-6 Printedin Great Britain
Fever in burned children
Charmaine Childs Regional Paediatric Burns Unit, Booth Hall Children‘s Hospital, Manchester and the North Western Injury Research Centre, University of Manchester, UK
Summary The patterns of rectal temperature (r,) and skin sur- face temperatures ( T,h,F:h and T,,,,) of 40 children were studied during the first 48 h after burn injury. The patients’ ages ranged from 5 months to 12 years and all required intravenous resuscitation. On admission T, measured 2-6 h after the burn, ranged from 36.5 to 39+“C and 19 children had temperatures >37,5”C; the upper limit of the normal range for healthy children. Patients S-12 months old had the lowest 7’, during the early period. During the first 12 h T, rose rapidly and 7.7 oatients (67 oer cent) had a fever (T,>38.5”C). In 10 Children ‘r, kas >40~o”C. There was no relationship between the rise in T, and the sex of the patient, the ambient temperature, the size. site or type of burn. Age was the most important variable. particularly dur- ing the first 12 h. where a positive correlation between r, and age occurred in patients aged 5-23 months (P<O~Ol, r=O.S3). Throughout the first 48h r, re- mained elevated. Trhlfh closely followed the pattern of T,. but T,,,, varied considerably, reflecting cardiovascu- lar as well as thermoregulatory responses.
INTRODUCTION 1~ IIAS been observed that many of the children admitted to this Burns Unit develop a fever within a few hours of being burned. This is unexpected so soon after the injury and the mechanism is not yet understood. In contrast, adult patients treated in the Regional Burns Unit do not have such high temperatures during the early hours after the burn.
Experimental and clinical studies have shown that injury affects thermoregulation (Stoner, 196X; Little and Stoner. 1981; Little et al.. 1986; Stoner, 1986), but the clinical studies have been confined to adults. Little has been reported about the responses to thermal injury in children, parti- cularly infants and young children, and practical- ly nothing about the changes in their body
0 1988 Butterworth & Co (Publishers) Ltd 0305-4179/x8/010001-06 $03~00
temperature. From the work referred to above it might be expected that body temperature would fall during the first few hours after the burn, then rise above normal after a few days when recovery had begun. This later pyrexia seen in the adult burn patient during what Cuthbertson (1942) has called the ‘flow phase’ is well documented (Wil- more et al., 1975, 1977).
The control of body temperature depends on a complex balance of heat production and heat loss. This balance could be disturbed by endoge- nous factors associated with the burn- nociceptive afferent stimulation, peripheral vaso- constriction related to hypovolaemia, loss of skin areas specifically concerned in heat transfer. in- creased evaporative heat loss from the burn sur- face, endocrine changes, production of polypeptides such as interleukin-l--or by exoge- nous factors altering the environment, occlusive dressings and ambient temperature. Until the role of these, or other factors, in the production of this early fever is known, physiologically appropriate treatment cannot be given to these children. As the first step towards this, the char- acteristics of this fever must be determined. as well as identifying groups of children most at risk of developing fever. This paper describes the changes in core and skin temperatures during the first 48h after burn injury in children. A pre- liminary account of this work was presented to the First Congress of the European Burns Asso- ciation (Childs. 1985).
PATIENTS AND METHODS With the exception of one patient (male, aged 5 months with a 9 per cent full thickness contact burn) all the patients studied had sustained burns
2 Burns (1988) Vol. 14/No. 1
to 10 per cent or more of the body surface and were admitted to the Paediatric Burns Unit, Booth Hall Hospital, Manchester within 6h of injury. They came either direct from the site of the accident or, more usually, from accident and emergency departments in the hospitals of the city and North Western Health Region.
On admission all patients were resuscitated with crystalloid (0.45 per cent NaCl and 5 per cent dextrose) and colloid (freeze-dried plasma,
later replaced by plasma protein fraction and fresh frozen plasma) guided by the Mount Ver- non Formula (Muir and Barclay, 1962). The ini- tial cleaning and dibridement of the burns was carried out within approximately 4 h of arrival. Wet soaks of either 0.05 per cent chlorhexidine gluconate (Hibidil), povidone iodine (Betadine) or silver sulphadiazine (Flamazine) were applied and covered with gauze and crepe dressings until tangential excision of the wound was performed,
Table 1. Details of patients studied; all survived except patient number 31 who died on day 11
Contact burn Fa, T, L, H Scald: coffee Fa, T Scald: kettle T, L, H Scald: kettle He, T, L, F Scald: kettle T. L Scald: kettle L Scald: kettle Fa, T, L, H Scald: kettle T. L, F Scald: flask Fa, T, L Scald: kettle Fa, T, L Scald: coffee Fa, T, L Scald: tea Fa, T, L Scald: tea Fa, T, L Scald: tea Fa, T, L Burn: flame T, L Scald: kettle Fa, He, T, L Scald: tea Fa, T, L Scald: tea Fa, T, L Scald: teapot Fa, T, L Scald: kettle Fa, T, L, F Scald: teapot Fa, T, L Scald: tea Fa, T, L Scald: teapot T. L, H Scald: teapot Fa, T. L, F Scald: bath P. L, F Scald: bath T, L Scald: kettle T, L Scald: teapot Fa, T, L Scald: kettle Fa, T, L, F
usually on day 3 post-burn. On arrival at the Burns Unit the patients were immediately trans- ferred to a cubicle where they were nursed in a warm environment for a minimum period of 36 48 h. Temperature monitoring was begun 1-2 h after patients were introduced to the warm environment and usually before dressings were applied. Occasionally, however, temperature monitoring was delayed until after the burn was bandaged. All patients were given morphine sulphate (0.2mg/kg body wt) on admission and at regular intervals throughout the resuscitation period. The antipyretic. acetaminophen (para- cetamol) I2 mdkg body wt was administered on occasions, usually when T, was >4OVC.
suffered these iniuries. Face, trunk and limbs were the most commonly affected areas in both burns and scalds. The patients were treated in a 27.7-32.5”C environment, r.h. 15-60 per cent (median 30 per cent). Thirty-nine patients sur- vived. Patient 31 (male, aged 4 years, 65 per cent bath scald) died on day 11.
Temperatures were measured with an electric thermometer (Thermocouple Instruments Ltd, Cardiff). Core temperature was measured in the rectum, 4-6cm from the anus, and skin tempera- tures on undamaged skin over the medial surface of the thigh and the pad of the great toe. An air-temperature probe was placed adjacent to the patient’s bed. Relative humidity (r.h.) was mea- sured with a hair thermohygrometer. The Man- chester North District Ethical Committee gave their approval for this study.
Core temperature Rectal temperature on admission Rectal temperature (T,) was first measured 2-6 h after the accident. The majority of children had been exposed to a cool environment when in transit to the Unit, some with wet soaks applied to the wound as ‘first aid’. At the time of the first measurement. l-2 h after admission to the Burns Unit, T, ranged from 36.5 to 39.o”C (median 37.4”(Z). Nineteen children had a rectal tempera- ture (T,) >37.5”C. Infants (5-12 months) had significantly lower temperatures on admission than the other groups of children (Table If).
RESULTS Forty patients. 28 male (Table I), were studied during the first 48 h post-burn. Their ages ranged from 5 months to 12 years (median 18 months). The total burn size ranged from 9 to 65 per cent body surface area; areas of partial/full thickness skin loss are shown also in Table 1. The majority of the injuries (32; 80 per cent) were scalds. 25 of which occurred during the process of making hot beverages. A small proportion of injuries (8; 20 per cent) resulted from flame burns and generally it was the older children who
Throughout the first 48 h after injury, during which T, was recorded hourly, all the patients showed an increase in core temperature. Typical patterns are illustrated in Fig. 1 (antipyretics were not administered to either of the patients). Fiy. la shows the pattern of T, during the first 48 h after the burn in Patient 13. a lCmonth-old male with I5 per cent scalds. T, reached a peak at around hour 12 and although elevated throughout most of the first 24 h there were considerable fluctuations in the pattern of rectal temperature 3648 h after the burn. Fig. lb shows a different pattern of T, in an older patient, a 7-year-old male with a 50 per cent flame burn and electrical injury. Although T, is elevated throughout the 48 h period, the pattern is rather more constant than in the younger child. During the first 12 h a fever (T,>38.S°C) developed in 27 patients (67
Table Il. Rectal temperature on admission and at 12 h post-burn
Patients
Infants 5-12 months (n=lO)
Toddlers 13-23 months (n=14)
Children 26 months-12 years (r-16)
Rectal temperature (mean5s.e.) On admission 12 h post-burn
37.2+0.15* 38.6*0.23**
37.7f0.19 39.5*0.24**
37.7f0.16 38.9?0.23**
*Significantly different from initial T, in other two groups Pc0.05 (unpaired t test). **Significantly different, f<O~OO1, from initial T, (paired r test).
4 Burns (1988) Vol. WNO. 1
a, 39- ; z $ 38-
Er
; 37-
z
n” 367
I I I I I
t 12 24 36 48
Burn Time After Burn (hours >
a
4l-
36L
I
t Burn
12 24 36
Time After Burn (hours)
b
I 48
Fig. I. Typical patterns of rectal temperature changes during the first 48 h post-burn. a, Patient 13: a 14-month- old male with a 15 per cent scald. b, Patient 37: a 7-year-old male with 50 per cent burn and electrical injury. Neither of the patients received antipyretic drugs.
per cent), and in 10 T, rose above 40,O”C; the highest recorded value, 41.O”C, occurred in a 15month-old male with a 22 per cent kettle scald who made an uneventful recovery. Twelve hours after the burn 7’, was significantly higher than on admission (Table II). The mean T, measured during the first 12 h showed a positive linear correlation between T, and age in patients aged 5-23 months (PtO.O1, r=0.53). This correlation was lost in the older group of patients.
Fig. 2 shows the early rise in 7’, during the first
12 h after the burn and the mean rectal tempera- tures 1324,25-36 and 3748 h after injury. Dur- ing this 48 h period no relationship could be demonstrated between the rise in rectal tempera- ture and the sex of the patient, the ambient temperature in which the child was nursed, the size, site or type of burn.
Despite the high rectal temperatures which developed in these patients. only one had a con- vulsion. This occurred in Patient 25, 12 h after the burn when his rectal temperature was 38.5”C.
Childs: Fever in burned children
40,
38-
37 _
38,
T 1
Burn Time After Burn (hours)
Fig. 2. Changes in rectal temperature (meanfs.e.) during the first 4X h post-burn in children aged 5-12 months (m), 13-23 months (0) and 2-12 years (0). o/a=on addition.
5
and was attributed to hyponatraemia (serum Na 124 mmol/l) rather than pyrexia.
Skin temperature The temperature of the skin over the thigh close- ly followed T,. albeit at a lower level, but the temperature of the pad of the great toe behaved differently (Fig. 3). The toe temperature varied by as much as 8-10°C. Most of the time this temperature remained low but rose rapidly at around the peak elevations in T,.
DISCUSSION In adult patients during the later phase of the response to burn injury. around the second week when the metabolic rate is maximal, mean body temperature rises 1.7-2°C above normal values (Wilmore et al., 1975.1977). The resultsin this paper show that a moderate to severe burn in a child has a much earlier effect on body temperature, causing a rise of l-3YC and starting as early as 2 h after the burn and continuing throughout the first 48 h. This rise, in young children, contrasts with the early fall seen after severe non-thermal injury in adults (Little and Stoner, 1981) and after thermal and other forms of injury in labora- tory animals (Stoner. 1968. 1981) provided that
41-
39-
37-
G
%35-
L??
2 33- &
i31- +
29-
2?_
I L t 4 8 12 16 20 24 28 32 36 40 44 48
Time After Burn (hours)
Fig. 3. Rectal (O-0-0). thigh (m-m-w) and toe temperatures (*-O-O) of patient 13. a 14-month-old male with 15 per cent scalds, during the first 48 h post-burn. Mean air temperature during the first 12 h, 29.5”C; 1%24h. 31.I”C; 25-36h. 29.7”C; 37-48 h. 27.9”C.
6
the ambient temperature is below the thermo- neutral zone. The limits of this zone are about 2830°C for adults and significantly higher, 33- 34°C for the healthy naked newborn infant in draught-free surroundings (Hey, 1972).
En route to the burns unit the environment in the home, ambulance and accident and emergen- cy department would always have been below thermoneutrality. Therefore it might be expected that the first deep body temperature measure- ment would be normal or slightly below normal, rather than high. Although the first measurement of If, was not taken immediately on arrival (there was usually a l-2 h delay) it is clear that once in the burns unit the core temperature of most of the patients rose rapidly and remained high for at least 48 h. Further work is needed to determine body temperature immediately on arrival at the unit as well as the effect of the environment on deep body temperature.
The mechanism of the pyrexia is not under- stood. Provided the burn was >9 per cent body surface area the response was independent of burn size, depth or situation. (The possibility of a stimulus-response relationship for smaller burns has not been investigated). Although the pyrexia was seen at all the ages studied, the response was greatest in the 13-23-month-old age-group. Antecedent infection was not present in the patients and infection seems unlikely as a cause of pyrexia at this stage (Parish et al., 1987). How- ever the possible role of endogenous factors such as interleukin-1 remains to be investigated, partic- ularly as the fever can be reduced, at least for a short time in most patients, by paracetamol (Childs, 1987). The behaviour of the toe temper- ature in relation to that of the core and the aggravation of the fever by surface cooling (Addy, 1983; author’s unpublished observations) suggests that the thermoregulatory system is ac- tive, albeit with a higher set-point. The rise in toe temperature at peak elevations in T, may be due
to a reduction in sympathetic vasoconstrictor tone as part of the normal thermoregulatory re- sponse to increase heat loss. Work is in progress on the heat balance of these burned children in an attempt to elucidate the mechanisms involved and this work will take into account iatrogenic factors such as drugs, occlusive dressings and the high environmental temperatures used in the management of these patients. With body temperature rising as high as 41°C a therapeutic response is called for, but until the mechanism
Burns (1988) Vol. 14/No. 1
for the pyrexia is understood a rational approach to management cannot be offered.
Acknowledgements My thanks are due to Mr P. J. Davenport FRCS, and Mr J. Lendrum FRCS for permission to study the patients under their care, to Professor H. B. Stoner, Dr R. A. Little and Dr I. A. MacDonald for their advice and to the sisters and nurses of the Paediatric Burns Unit for their help. Some of these results have been used in a thesis for an MPhil degree of the University of Nottingham.
REFERENCES Addy D. J. (1983) Cold comfort for hot children. Br.
Med. J. 286, 1163. Childs C. (1985) Proceedings of the First Congress of
the European Burns Association, European Burns Association, Groningen, The Netherlands.
Childs C. (1987) Mechanisms for the development of fever in young children after burn injury. Circ. Shock 21, 369.
Cuthbertson D. P. (1942) Post-shock metabolic re- sponse. Lancet i, 433.
Hey E. N. (1972) Thermal regulation in the newborn. Br. .I. Hosp. Med. 8, 51.
Little R. A. and Stoner H. B. (1981) Body temperature after accidental injury. Br. J. S&g. 68; 221..
Little R. A.. Stoner H. B.. Randall P. et al. (1986) An \ , effect of injury on thermoregulation in man. Quart. J. Exp. Physiol. 71, 295.
Muir I. F. K. and Barclay T. L. (1962) Burns and Their Treatment. London: Lloyd-Luke, p. 13.
Parish R. A., Novack A. H., Heimbach D. M. et al. (1987) Fever as a predictor of infection in burned children. J. Trauma 27, 69.
Stoner H. B. (1968) Mechanism of body temperature changes after burns and other injuries. Ann. N. Y. Acad. Sci. 150, 722.
Stoner H. B. (1981) Thermorenulation after trauma. Adv. Physioi Sci.‘26, 25. _
Stoner H. B. (1986) A role for the central nervous system in the responses to trauma. In: Little R. A. and Frayn K. N. (eds), The Scientific Basis for the Care Of The Critically Ill. Manchester: Manchester University Press, p.215.
Wilmore D. W., Mason A. D., Johnson D. W. et al. (1975) Effect of ambient temperature on heat pro- duction and heat loss in burn patients. J. Appl. Physiol. 38, 593.
Wilmore D. W., Aulick L. H., Mason A. D. et al. (1977) Influence of the burn wound on local and systemic responses to injury. Ann. Surg. 186. 444.
Paper accepted 19 September 1987.
Cormspondence .shordd he uddrevsed 10: MS C. Childa, Regional Pacdiatric Burns Unit, Booth Hall Children’s Hospital, Charlcstown Road. Blacklcy, Manchcstcr MY ?AA. UK.
