- Email: [email protected]
Enteral resuscitation and early enteral feeding in children with major burns—Effect on McFarlane response to stress
burns 33 (2007) 464–471
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/burns
Enteral resuscitation and early enteral feeding in children with major burns—Effect on McFarlane response to stress M. Venter *, H. Rode, A. Sive, M. Visser Red Cross War Memorial Children’s Hospital, Paediatric Surgery, Klipfontein Road, Rondebosch, Cape Town 7700, South Africa
article info
abstract
Article history:
Aim: Early enteral feeding has become standard practice for burned patients. The aim of this
Accepted 14 August 2006
study was to determine whether early enteral feeding could be used as an avenue for resuscitation and feeding and the effect it would have on the induction/amelioration of the
Keywords:
hormonal stress response.
Enteral resuscitation
Method: Eighteen children with <20% TBSA were randomly assigned to either early enteral
Early enteral feeding
feeding and resuscitation, or intravenous resuscitation with the induction of enteral feeding
Metabolic response
delayed. The enteral fluid volume was incrementally increased every 3 h with a simulta-
Children
neous equal reduction in the intravenous volume until all the calculated intravenous fluid requirements for resuscitation and maintenance could be administered enterally. In the second group, intravenous resuscitation continued for 48 h when enteral feeding was introduced. Parameters measured were the clinical responses and outcome as well as the concentrations of insulin, insulin-like growth factor 1, glucagon, cortisone and growth hormone. The estimated and calculated energy expenditure was measured calorimetrically and bowel permeability was assessed using a dual sugar absorption test. Results: Three children were excluded from the study because of early death from organ failure or carbon monoxide poisoning. Early enteral resuscitation and feeding (ER/EEF) was initiated within a median of 10.7 h post-burn in nine children and late enteral feeding introduced on an average 54 h post-burn. The ER/EEF group showed an anabolic response with significantly higher insulin concentrations ( p = 0.008) and insulin: glucagon ratios ( p = 0.043). Although blood glucose concentrations were initially slightly elevated (EEF: 10.3 g/l, LEF: 8.1 g/l), they rapidly returned to within the normal range. The cortisol and IGF1 concentrations did not differ significantly between the two treatment groups. Growth hormone concentrations were significantly higher in the late enteral feeding (LEF) group ( p = 0.03). The estimated energy expenditure was not different amongst the groups. Small bowel permeability [lactulose:rhamnose (L:R) ratios] decreased significantly over time ( p = 0.02) in both study groups. No pulmonary aspiration was found. Diarrhoea in the ER/ EEF settled quickly (2–4 days), whereas in the LEF group it persisted for longer than a week. The LEF group lost a median of 7.75% (acceptable range = 5%) of admission body weight, whereas the ER/EEF group lost a median of 3.01%. Patients in the LEF group required antibiotic treatment for a longer period ( p = 0.08) and their hospital stay was longer, though not significant. Conclusions: Enteral resuscitation and early enteral feeding is a safe and effective method and particularly suited for children in developing countries. It resulted in the amelioration of
* Corresponding author. E-mail address: [email protected] (M. Venter). 0305-4179/$32.00 # 2006 Elsevier Ltd and ISBI. All rights reserved. doi:10.1016/j.burns.2006.08.008
burns 33 (2007) 464–471
465
the hormonal stress response and improved outcome. Enteral resuscitation should not be introduced in a patient in shock or with existing gastrointestinal disease. Complications were minimal. # 2006 Elsevier Ltd and ISBI. All rights reserved.
Numerous fluid resuscitation formulae have been developed to counterbalance thermally induced fluid losses, hypovolaemia, poor tissue perfusion and shock. Collectively they include crystalloid fluids (2–4 ml/kg/%burn), colloids or hypertonic saline solutions, aiming at maintaining haemodynamic stability and an hourly urine output of 0.5–1.5 ml/kg body weight [1–3]. However, during the same period two important additional advances were made: enteral resuscitation and early enteral feeding. Oral rehydration therapy (ORT) which combines the use of a balanced oral solution of electrolytes and carbohydrates, with early refeeding in children with acute diarrhoea was introduced during the early sixties [4,5]. This process has conclusively been shown by numerous studies to satisfy all principles of fluid replacement, restoration of electrolytes and acid based deficiencies and reducing the need for intravenous resuscitation by 70% [6–8]. In addition to the safety and efficacy of ORT, enteral feeding has become the preferred route of nutrient delivery provided the GIT is functioning. There is progressive evidence that early enteral feeding (<12 h post-burn) in burnt children can be administered safely and effectively in increasing volumes, to meet calculated energy requirements within 2–3 days following severe burn [9–12]. This process preserves gut integrity, restores cell mediated immunity, reduces bacterial translocation, decreases the hypermetabolic response, improves nitrogen retention, increases intestinal blood flow, stimulates the secretion of gut trophic hormones and generally decreases morbidity and mortality [9–12]. To date, few studies document the clinical and metabolic responses in children who have received early enteral feeding (EEF) following major burns [12,13]. Moreover, there are no studies documenting the progress of children where enteral feeding was used as an adjunct to intravenous fluids in resuscitating children with burns. This prospective study reports on the clinical course and outcome of children who have received EEF, in comparison to a cohort who were kept nil per mouth for 48 h during the resuscitation phase. In addition, stress hormones, involved in intermediary metabolism, energy expenditure and intestinal permeability were measured.
1.
Materials and methods
Children admitted to the Paediatric Burns Unit of Red Cross War Memorial Children’s Hospital were studied. For inclusion, the children were between 6 months and 13 years; have burns equal to or greater than 20% of total body surface area (TBSA) and have sustained the injury within 24 h of admission. On admission, the children were randomised to receive either conventional IV resuscitation and introduction of enteral feeding after 48 h [late enteral feeding] (LEF), or enteral
resuscitation and early enteral feeding (EEF) within 24 h of having sustained the injury. Enteral feeding was administered via a nasojejunal feeding tube. These were placed either under fluoroscopic guidance [14] or bed-side-technique; the latter’s position confirmed by abdominal radiograph. The radiological service was only available between 8:00 and 23:00 h and therefore for reasons of safety, the last three children admitted outside of these hours were assigned to the LEF group. The total volume of enterally administered fluid consisted of the following: Resuscitation fluid: Day 1, 3.5 ml/kg/% burn; Day 2, 1.5 ml/kg/% burn. Maintenance fluid: 0–1 years, 120 ml/kg/day; 1–2 years, 100 ml/ kg/day; 2–5 years, 80 ml/kg/day; 5–10 years, 60 ml/kg/day; 1–12 years, 50 ml/kg/day. The total volume, i.e. resuscitation and maintenance fluid, was divided by 24 which constituted the hourly administered volume. Resuscitation was initiated intravenously with a balanced crystalloid solution (Plasmalyte B) at the calculated volume with simultaneous commencement of enteral resuscitation (ER) at a volume of 1 ml/kg/h. The enteral feeding consisted of commercial polymeric formula (Abbott Laboratories). The enteral fluid volume was incrementally increased every 3 h by 50% (2 ml/kg), 33% (3 ml/kg), 25% (4 ml/kg), 20% (5 ml/kg), 17% (6 ml/kg), respectively until the calculated hourly volume was reached. At the same time the intravenous volume, resuscitation and maintenance was correspondingly decreased until the total calculated hourly volume was administered enterally. The intravenous line was kept open with 5 ml/h until the end of the 48 h resuscitation period for purposes of analgesia and if additional intravenous therapy should be required. During the study period there was no need to deviate from this regime, none of the children were shocked on admission and all maintained a urine output between 0.5 and 1.5 ml/kg/h. The children were all admitted to a high care area and routinely monitored to evaluate the progress of the resuscitation, i.e. BP, pulse, pulse-oximetry, urine output and routine haematological and biochemical investigations. A nasogastric tube was aspirated every hour. If the residual volume would have exceeded the previous hourly intake by 50%, the ER would be temporarily stopped and intravenous fluid increased to the full hourly volume. However, this scenario remained absent during the study period. Routine burn wound management was administered to all children. The carbohydrate and protein concentrations of the feed were supplemented from Day 3 and Day 5 onwards in the EEF and FEF patients, respectively using glucose polymer and hydrolysed protein (Abbott Laboratories). The final feed
466
burns 33 (2007) 464–471
composition provided 17–20% protein, 22–28% fat and 55–58% carbohydrate. In both groups, once full enteral nutrition was established, solid meals were offered as an adjunct to the enteral feed. In addition to blood drawn for immediate clinical management, venous blood was sampled for purposes of stress hormone bends determinations on admission, daily for 5 days, and then weekly on days 12, 19 and 26 after admission. Serum and plasma were separated and aliquots stored at 20 8C until hormone concentrations were measured in batches. Routine chemistry was performed immediately. Serum IGF1 was measured using the Medgnix ‘‘IGF-D-RIA’’ kit assay [15]; plasma glucagons [16], serum insulin [17] and serum growth hormone [18] by radioimmunoassay (RIA). Serum cortisol by a ‘heterogenous competitive magnetic separation assay (MSA) [19]. The intra-assay coefficients of variation (CV) for the concentrations of cortisol, glucagon, insulin, IGF1 and GH were 3.1, 3.9, 4.9, 3.9 and 3.5%, respectively. In each patient, resting energy expenditure (REE) was measured daily using a Paediatric Calorimeter (Delta Trac). The machine was calibrated before use with 100% alcohol and a combination of 95% oxygen and 5% carbon dioxide. Daily barometric pressure was obtained from the meteorological office. Full calibration was repeated before daily measurements. The estimate energy requirements (EEE) were extrapolated from these measurements by adding 30% for daily activity [20] and compared to the Recommended Daily Allowances (RDA) [21] for energy and to energy requirements derived by Solomons [22] and Galveston [23] formulae, respectively for children with burns.
Testing for permeability of the small intestine was undertaken in 15 children using a dual sugar absorption test (SAT). Each child was tested on two occasions, once on Day 3 after admission and again 48 h later. The children were fasted for 2 h before and isotonic solution, of lactulose 5 g and L-rhamnose 1 g in 100 ml of water, was administered via feeding tube. The nasojejunal tube was clamped during the procedure. Fasting was maintained for a further 5 h during which time all urine was collected. Hydration and plasma glucose was maintained using the intravenous route. Feeds withheld during this period were subsequently made up by increasing the rate of the nasojejunal feeds. Urinary sugar excretion was measured using thin layer chromatography [24,25] and expressed as a ration of the disaccharide to the monosaccharide. Descriptive statistics for the groups are provided. Where appropriate, normally distributed inter-group data are compared using the Student’s t-test. MANOVA was used for the analysis of repeated measures. Log-transformation was applied to the growth hormone, which were not normally distributed, before comparison. The data were analysed using Statistica (Version Z.34, 1998) and Microsoft Excell packages, 2002. Significance was accepted at the 5% level. The Ethics Research Committee of the University Of Cape Town approved the study. Each child’s legal caregiver signed written consent.
2.
Results
Nine patients in each group completed the study. Twenty-one children were originally recruited for the study, 11 in the EEF
Table 1 – Clinical characteristics of the patients in the study Subject
Age (yr, years, m, months)
TBSAB (%)
Admission weight (kg)
Cause (HWB, hot water burn; FB, fire burn)
Inhalation injury (U) ventilated (V)
EEF group (n = 11) 1 2 3 4 5 6 7 8 9 *10 *11 Medians
8m 2 yr 2 yr, 9 m 4 yr, 7 m 5 yr, 4 m 6 yr, 4 m 7 yr 7 yr 11 yr, 3 m 1 yr, 10 m 1 yr, 4 m 4.54
30 23 30 31 23 35 20 33 30 75 30 29.5
8.72 12.3 13 15.5 18 22 24.5 24.5 30 13.8 11 14.38
HWB HWB HWB HWB FB FB FB FB FB FB HWB
U (V) U (V) U (V)
LEF group (n = 10) 12 13 14 15 16 17 18 19 20 *21 Medians
8 yr, 11 m 3 yr, 8 m 1 yr 1 yr, 1 m 9 yr, 1 m 3 yr, 5 m 1 yr, 9 m 4 yr, 6 m 1 yr *9 yr 4.45
20 25 25 20 28 32 29 30 62 32 30.00
25 13.25 10.5 10.95 15 13 11 19 8 28 14.25
HWB FB HWB HWB HWB FB FB FB FB FB
U (V) U (V) U (V)
burns 33 (2007) 464–471
Fig. 1 – Median serum cortisol concentrations (nmol/l) over time in the EEF group and in the LEF group (normal range: 5.52–800 nmol/l).
and 10 in the LEF group. Two patients in the EEF group died from organ failure on day 3 and 19, respectively, and one patient in the LEF group died from monoxide poisoning and brain ischaemia on day 5 post-burn. The groups were similar in terms of age, body weight and total body surface area of the burn (Table 1). There were no significant difference in the groups with respect to the initial time to debridement, skin grafts or the total number of grafts performed. Feeding in the EEF group was started at a median of 10.7 h (range: 5.5–23 h) after sustaining the burn and at 16 h (range: 13–24 h) all were on full enteral feeds. In the LEF group, enteral nutrition was started at a median of 54 h (range: 51–58 h) post-burn and full feeds were achieved at a median of 10 h (range: 7–10 h) later. There is a shorter time range in the LEF group, since intravenous resuscitation was completed. Although the LEF group consumed fewer calories over the entire period of 26 days; (excluding the first 2 days) there was no statistical significance ( p = 0.7). Serum glucose was equivalently elevated in both groups on admission (median: 10.3 mmol/l in EEF and median: 8.1 mmol/l) in LEF). For the remainder of the study, the glucose concentrations for both groups were within the normal range.
467
Fig. 3 – Median serum insulin concentrations (mU/l) over time in the EEF group and the LEF group (normal range: 0– 30 mU/l).
Fig. 4 – Median plasma glucagon concentrations (pg/ml) over time in the EEF group and the LEF group (normal range: 25–250 pg/ml).
Fig. 5 – Medians for the insulin: glucagon ratios over time in the EEF group and the LEF group.
Fig. 2 – Median serum insulin-like growth factor 1 concentrations over time in the EEF group and LEF group (normal range: 36–350 ng/ml).
There were no significant differences in the serum cortisol or IGF levels between the groups either on admission or subsequently (Figs. 1 and 2, respectively). The range of values for patients in the LEF group appeared to be higher than those in the EEF group, but was not significant. Insulin concentrations were significantly higher ( p = 0.008) in the EEF for the entire study period, as shown in (Fig. 3). Although the plasma glucagon concentrations (Fig. 4) appeared lower in the EEF
468
burns 33 (2007) 464–471
Fig. 6 – The log transformation of serum growth hormone concentrations (mU/l) over time in the EEF group and the LEF group (normal range: 0.5–150 mU/l).
Fig. 7 – Medians for the estimated energy expenditure (REE T 1.3), as measured by indirect calorimetry in the EEF vs. LEF groups.
group on days 3 and 4, the difference failed to reach significance. However, insulin:glucagon (IG) ratio (Fig. 5) were significantly ( p = 0.04) higher in the EEF group till day 4. From admission to day 12, the LEF group had significantly ( p = 0.03) higher hGH concentrations (Fig. 6). The estimated energy expenditure (EEE) for the groups is shown in Fig. 7. Although the median EEE was higher in the EEF
Fig. 8 – The median values of the lactulose:rhamnose ratios for the EEF and LEF groups on day 3 and day 5 post-burn.
group on certain days, the difference for the entire study period was not significantly different ( p = 0.2). The EEE derived from indirect calorimetry in this study was similar to that calculated using the Galveston equation. Small bowel permeability was abnormal in both groups on Day 3 with median lactose:rhamnose ratios (L:R) of 24.3 and 20.9 for the EEF and LEF groups, respectively. Fourty-eight h later, the mean L:R had returned to the normal range in the LEF group but remained elevated in the EEF group (Fig. 8). With respect to the clinical course of the patients, two children in the EEF and 1 child in the LEF group died on days 3, 19 and 5. The EEF patients spent a total of 222 days in hospital compared to a total of 298 days in the LEF group. The difference was not significant ( p = 0.9). The LEF group lost 7.75% of body weight between admission and discharge compared to 3.0% in the EEF group ( p = 0.1). Three patients in the EEF group but none in the LEF group experienced one to two vomiting episodes each during the period of PER. The vomiting did not have adverse consequences. Five of the children in the EEF group had loose stools for between 2 and 4 days after starting the enteral feed as did three in the LEF group. In LEF children, the loose stools lasted for longer than a week. Antibiotic treatment was only given after a positive blood culture, or positive surface swab culture. All 18 children received antibiotics during their hospital treatment, however, the EEF group required a median of 11 days per patient compared to 14 days in the LEF group ( p = 0.08). None of the 18 children who completed the study had blood-culture proven septicaemia.
3.
Discussion
The small bowel has become a central organ for resuscitation and early enteral feeding which came about because of a better understanding of the pathophysiological function of the gut during the early phases of acute illness and trauma. The discovery of the glucose/sodium co-transport system in the 1960s has provided the basis for the development of oral resuscitation methods [4,5]. Essentially sodium is absorbed from the lumen of the small intestine by three basic mechanisms. Sodium moves down an electrochemical gradient, and is exchanged for hydrogen. In addition chloride is exchanged for bicarbonate in an active coupled mechanism. Sodium absorption is also coupled to glucose and other small ions (e.g. amino acids). The co-transport mechanism remains intact during the early phases of hypotension and mucosal injury resulting from enteric infections. Intestinal sodium absorption is critical for fluid homeostasis, since water transfer is always passive and occurs secondarily to the osmotic gradient created by the movement of sodium ions during this process. This process has made possible the effective introduction of an oral rehydration method using a balanced sodium/glucose solution [6–8]. Adult literature has shown several advantages of EEF, and also provided a basis to point out some differences in the paediatric thermal patient’s response to EEF. Adults show a decreased catabolic response as shown by catecholamine concentrations which stayed within normal ranges [12], elevated cortisol concentrations which returned within
burns 33 (2007) 464–471
normal ranges much faster [23,24] and glucagon concentrations which were significantly lower in the EEF group [12]. Anabolism was shown in significantly higher insulin concentrations on days 4 and 8 post-burn [12]. In addition EEF prevents mucosal breakdown, bacterial translocation and decreases small bowel permeability [25,26], maintains (within a 5% range) admission weight [27], with a shorter wound healing period and hospital stay [28], as well as decreased wasting and mortality [23,29]. This study showed that children who have suffered major burns can be resuscitated and nourished via the enteral route from the time of admission to hospital. None of the children who completed this study had any major complications as a result of early enteral feeding. A striking difference between the two groups is the significantly higher insulin response in the EEF group. In adults, delayed feeding was associated with low insulin concentrations, Wilmore et al. [30] Chiarelli et al. [12] found significantly higher insulin concentrations in adults receiving EEF on days 4 and 8 after burn. Insulin is an anabolic hormone and it is feasible that the higher concentrations were involved in earlier tissue repair and were partly responsible for the smaller decline in body weight in the EEF group. The higher I:G ratio in the EEF group resulted not only from higher insulin but also from lower glucagon in the EEF patients. Although the EEF group had lower glucagon concentrations, most patients in both groups had elevated levels for the entire study period. A similar glucacon pattern was documented by Jenkins et al. [31]. It was shown that elevated hGH concentrations during periods of stress, trauma and starvation, enhances gluconeogenesis and lipolysis [32–35]. Significantly higher hGH concentrations in the LEF group confirmed this. The catabolic properties of hGH during times of severe stress have been documented as the cause of increased protein breakdown, increased free fatty acid levels, increased catecholamine levels, and the induction of insulin release resulting in hyperglycaemia [36–38]. The higher hGH concentrations in the LEF group may also have reflected the need for increased glucose production. Taken together, the findings suggest that early feeds provide sufficient energy and that the patients do not have to rely on glycogenolyis or gluconeogenesis for their glucose requirements. Serum cortisol concentrations were moderately elevated in both groups on admission but settled rapidly. The fall in cortisol concentration may reflect appropriate pain control and fluid replacement in the children. IGF-I concentrations were similar in the two groups, also reflecting that early enteral feeding did not disadvantage the children. Unfortunately we were unable to measure nitrogen balance during the study to document an anabolic effect in the EEF patients. The estimated energy expenditure (EEE), measured by indirect calorimetry (EEE 1.3), is in keeping with that derived from the Galveston equation. Several authors found Galveston Equation to be the most reliable [39,40]. The EEE was also similar to that predicted using the RDA. It has been proposed (Gottschlich) [41] that the reduced physical activity in the patients offset the increased energy requirements imposed by the injury and the findings in this study accord with that supposition. Although the median EEE was higher in the EEF
469
group on days 5–7,9,10, 12 and 26, the difference for the entire study period was not significantly different ( p = 0.2). The higher energy expenditure initially did not result in greater weight loss and may have been offset by a greater but not significant energy intake in the EEF group, or possible energy spent on increased wound healing. Although not significant, the EEF group spent less days/patient in hospital than LEF group. Despite most of the children coming from poor socioeconomic backgrounds, none of the children were wasted or stunted on admission. The weight loss of 3% in the EEF group is in the acceptable limit (<5%) for children who have suffered major burns whereas the 7.7% experienced by the children in the LEF group is outside this limit [27]. Loss of the intestinal mucosal barrier results in increased permeability of the small bowel and increases the risk of bacterial translocation and infection [25,26]. The sugar absorption tests confirmed an abnormal small bowel permeability which was more than double normal using a L:R ratio of 10 as the upper limit on admission [45]. The L:R ratio returned to normal in the LEF group and improved in the EEF group. We were surprised that EEF seemed to delay the recovery of small bowel integrity because in other situations e.g. after intestinal surgery, early feeding enhances the recovery of intestinal integrity [46]. However, in spite of the abnormal small bowel permeability, an increased propensity to septicaemia, infection or diarrhoea was not seen. Antibiotics may have prevented infection but would have been more likely to be associated with prolonged diarrhoea, a complication not encountered in the EEF group of patients. Whether the return to normal ranges in intestinal permeability is purely a function of time or can be accelerated by altering the feed, for example, by adding multi-fibre [42–44], requires further study in paediatric patients. From a clinical standpoint, the children who received EEF as PER were not disadvantaged. There was concern about pulmonary aspiration but the recorded vomits were small and infrequent and resolved following revised placement of the nasojejunal tubes. The correct siting of the feeding tube is fundamental to this form of resuscitation and feeding. None of the children In the EEF group experienced ileus, intestinal obstruction or intestinal blood loss and absorption was not impaired as reflected by maintenance of weight. None of the children who completed the study had proven septicaemia. Antibiotic use and total period of hospitalisation was lower in the EEF group, but given the wide range and the small numbers in the study, failed to reach statistical significance. However, even in this study, the total of 76 days shorter hospitalisation and the lower antibiotic use, had cost-saving implications. A study designed to compare the costs of EEF with LEF in terms of hospitalisation, pharmaceuticals, nutrition and human resource utilisation is necessary to determine whether EEF is financially efficient in the long term. Contraindications for enteral feeding in thermal patients include hypovolaemic shock, instability and hypoperfusion, intolerance to oral glucose, pre-existing bowel disease, visible abdominal distention and established diarrhoea. These patients may experience difficulties in maintaining a positive fluid balance. In the presence of severe postburn dehydration or shock the patient should first be resuscitated parenterally
470
burns 33 (2007) 464–471
until circulatory parameters have returned to normal. A relative limitation may be persistent vomiting. The position of the NJT should be corrected immediately. We believe that this study demonstrated that enteral resuscitation and early enteral feeding in burns is safe, effective and not complicated by untoward effects. This approach is particularly suited to the management of children in developing countries, provided that nasojejunal tubes can be positioned correctly. In the paediatric burned population studied, EEF enhanced insulin secretion/anabolism, facilitated maintenance of admission weight, decreased antibiotic treatment days and shortened hospital stay.
references
[1] Muller MJ, Herndon DN. The challenge of burns. Lancet 1994;343:216–20. [2] Graves TA, Cioffi WG, McManus WF, Pruitt BA, et al. Fluid resuscitation of infants and children with massive thermal injury. J Trauma 1988;28:1656–9. [3] Merrel SW, Saffle JR, Warden GD, et al. Fluid resuscitation in thermally injured children. Am J Surg 1986;152:664–9. [4] Fischer RB, Parsons DS. Glucose movements across the wall of the rat small intestinal. J Physiol 1953;119:210–23. [5] Hirschborn N, Kinzie JL, Sachar DB, et al. Decrease in net stool output in cholera during intestinal perfusion with glucose-containing solutions. N Engl J Med 1968;297:176–81. [6] Schedl HP. Scientific rationale for oral rehydration therapy. Clin Ther 1990;12:14–21. [7] Santosham M, Greenough WB. Oral rehydration therapy. A global perspective. J Paediatr 1991;118:S44–51. [8] Farthing MJG. Oral rehydration therapy—past, present and future. Trans Coll Med S Afr 1994;82–9. [9] Scott McDonald W, Sharp CW, Deitch EA. Immediate enteral feeding in burn patients is safe and effective. Ann Surg 1991;213:177–83. [10] Moore FA, Moore EE. The benefits of enteric feeding. Adv Surg 1997;30:141–54. [11] Engelbrecht VJ, Clarke SM. Early enteral feeding of a severely burned pediatric patient. J Burn Care Rehabil 1994;15:293–7. [12] Chiarelli A, Enzi G, Casadei A, et al. Very early nutrition supplementation in burned patients. Am J Clin Nutr 1990;51:1035–9. [13] Klasen HJ, ten Duis HJ. Early oral feeding op patients with extensive burns. Burns 1987;13(1):49–52. [14] Chellis M, Sanders SV, Webster H, et al. Bedside transpyloric tube placement in the paediatric intensive care unit. J Paren Enter Nutr 1996;20(1):88–90. [15] Medgenix IGF-1-D-RIA-CT Kit, Catalog Number: 30 158 60, BioSource Europe S.A., Nivelles Belgium. [16] Glucagon-RIA Kit, Catalog Number: 07-152101, Separations, ICN Pharmaceuticals Inc. Diagnostic Division, 3300 Hyland Avenue, Costa Mesa, California. [17] Coat-A-Count Insulin, Catalog Number: 2 D01-20, Abbott Laboratories (Pty), 149 Samuel Evans Road, Aeroton, Johannesburg S.A. [18] Pharmacia hGH RIA, Pharmacia and Upjohn Diagnostics AB, Uppsala Sweden. [19] Cortisol (Cort), Technicon Immune 1 Septems Bayer Corporation, Business Group Diagnostics, Tarrytown, NY 1059-5097 USA. [20] Gore DC, Rutan RL, Hildreth M, et al. Comparison of resting energy expenditure and caloric intake in children with severe burns. J Burn Care Rehabil 1990;11:400–4.
[21] Food and Nutrition Board. National Research Council Recommended Dietary Allowances. 9th ed. Washington DC: National Academy of Science, 1989. [22] Solomon JR. Nutrition in the severely burned child. Progr Pediatr Surg 1981;14:653–79. [23] Mochizuki H, Trocki O, Dominioni L, et al. Mechanism of prevention of postburn hypermetabolism and catabolism by early enteral feeding. Ann Surg 1984;200(3):297–309. [24] Dominioni L, Trocki O, Fang C-H, Mochizuki H, et al. Prevention of severe postburn hypermetabolism and catabolism by immediate intragastric feeding. J Burn Care Rehabil 1984;5(2):106–12. [25] Van Elburg RM, Uil JJ, De Monchy JGR, et al. Intestinal permeability in pediatric gastroenterology. Scand J Gastroenterol 1992;27(Suppl 194):19–24. [26] Deitch EA. Intestinal permeability is increased in burn patients shortly after injury. Surgery 1990;107: 411–6. [27] Herndon DN, Rutan TC. Management of the pediatric patient with burns. J Burn Care Rehabil 1993;14(1):3–8. [28] Garrel DR, Davignon I, Lopez D, et al. Length of care in patients with severe burns with or without early enteral nutritional support. J Burn Care Rehabil 1991;12(1):85–90. [29] Saito H, Trocki O, Alexander W, et al. The effect of route of nutrient administration on the nutritional state, catabolic secretion, and gut mucosal integrity after burn injury. J Parent Enter Nutr 1987;11(1):1–7. [30] Wilmore DW, Long JM, Mason AD, et al. Catecholamines: mediator of the hypermetabolic response to thermal injury. Ann Surg 1974;180(4):653–69. [31] Jenkins M, Gottschlich M, Alexander JW, et al. An evaluation of the effect of immediate enteral feeding on the hypermetabolic response following severe burn injury. J Burn Care Rehabil 1994;5:106. [32] Tietz NW. Clinical guide to laboratory tests, 3rd ed., Philadelphia: Saunders; 1995. [33] Painter PC, Cope JY, Smith JL. In: Burtis CA, Ashwood E, editors. Tietz textbook of clinical chemistry. 3rd ed., Philadelphia: Saunders; 1999. p. 2161–217. Chapter 41. [34] Zilva JF, Pannall PR, Mayne PD. The hypothalamus and pituitary gland. In: Zilva JF, Pannall PR, Mayne PD, editors. Clinical chemistry in diagnosis and treatment. 6th ed., London: Edward Arnold; 1994. p. 112. [35] Solomon SM, Kirby DF. The refeeding syndrome: a review. JPEN 1990;14(1):90–7. [36] Gilpin DA, Barrow RE, Rutan RL, et al. Recombinant human growth hormone accelerates wound healing in children with large cutaneous burns. Ann Surg 1994; 220:19–24. [37] Roe CF, Kinky J. The influence of human growth hormone on energy sources in convalescence. Surg Forum 1962;13:369–71. [38] Fleming RYD, Rutan LR, Jahoor F. Effect of recombinant human growth hormone on catabolic hormones and free fatty acids following thermal injury. J Trauma 1992;32(6):698–703. [39] Hildreth M, Herndon DN, Desai MH, et al. Current treatment reduces calories required to maintain weight in pediatric patients with burns. J Burn Care Rehabil 1990;11:405–9. [40] Holland KA, Gillespie RW, Lewis NM. Estimating energy needs of pediatric patients with burns. J Burn Care Rehabil 1995;16:458–60. [41] Gottschlich MM. Nutrition in the burned pediatric patient. In: Queen PM, Lang CE, editors. Handbook of pediatric nutrition. Gaithersburg, Md: Aspen Publishers; 1993. p. 536–59.
burns 33 (2007) 464–471
[42] Isolauri E, Gronlund MM, Salminen S, et al. Why don’t we bud? J Pediatr Gastroenterol Nutr 2000;30:214–6. [43] Zopf D, Roth S. Oligosaccharides anti-infective agents. Lancet 1996;347:1017–21. [44] Trier E, Wells JCK, Thomas AG. Effects of a multifibre supplemented enteral feed on gastrointestinal function. J Pediatr Gastroenterol Nutr 1999;27:595.
471
[45] Brewster DR, Manary MJ, Menzies IS, et al. Intestinal permeability in kwashiorkor. Arch Dis Child 1997; 76:236–41. [46] Carr CS, Ling KDE, Boulos P. Randomized trial of safety and efficacy of immediate postoperative enteral feeding in patients undergoing gastrointestinal resection. BMJ 1996;312:869–71.
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/burns
Enteral resuscitation and early enteral feeding in children with major burns—Effect on McFarlane response to stress M. Venter *, H. Rode, A. Sive, M. Visser Red Cross War Memorial Children’s Hospital, Paediatric Surgery, Klipfontein Road, Rondebosch, Cape Town 7700, South Africa
article info
abstract
Article history:
Aim: Early enteral feeding has become standard practice for burned patients. The aim of this
Accepted 14 August 2006
study was to determine whether early enteral feeding could be used as an avenue for resuscitation and feeding and the effect it would have on the induction/amelioration of the
Keywords:
hormonal stress response.
Enteral resuscitation
Method: Eighteen children with <20% TBSA were randomly assigned to either early enteral
Early enteral feeding
feeding and resuscitation, or intravenous resuscitation with the induction of enteral feeding
Metabolic response
delayed. The enteral fluid volume was incrementally increased every 3 h with a simulta-
Children
neous equal reduction in the intravenous volume until all the calculated intravenous fluid requirements for resuscitation and maintenance could be administered enterally. In the second group, intravenous resuscitation continued for 48 h when enteral feeding was introduced. Parameters measured were the clinical responses and outcome as well as the concentrations of insulin, insulin-like growth factor 1, glucagon, cortisone and growth hormone. The estimated and calculated energy expenditure was measured calorimetrically and bowel permeability was assessed using a dual sugar absorption test. Results: Three children were excluded from the study because of early death from organ failure or carbon monoxide poisoning. Early enteral resuscitation and feeding (ER/EEF) was initiated within a median of 10.7 h post-burn in nine children and late enteral feeding introduced on an average 54 h post-burn. The ER/EEF group showed an anabolic response with significantly higher insulin concentrations ( p = 0.008) and insulin: glucagon ratios ( p = 0.043). Although blood glucose concentrations were initially slightly elevated (EEF: 10.3 g/l, LEF: 8.1 g/l), they rapidly returned to within the normal range. The cortisol and IGF1 concentrations did not differ significantly between the two treatment groups. Growth hormone concentrations were significantly higher in the late enteral feeding (LEF) group ( p = 0.03). The estimated energy expenditure was not different amongst the groups. Small bowel permeability [lactulose:rhamnose (L:R) ratios] decreased significantly over time ( p = 0.02) in both study groups. No pulmonary aspiration was found. Diarrhoea in the ER/ EEF settled quickly (2–4 days), whereas in the LEF group it persisted for longer than a week. The LEF group lost a median of 7.75% (acceptable range = 5%) of admission body weight, whereas the ER/EEF group lost a median of 3.01%. Patients in the LEF group required antibiotic treatment for a longer period ( p = 0.08) and their hospital stay was longer, though not significant. Conclusions: Enteral resuscitation and early enteral feeding is a safe and effective method and particularly suited for children in developing countries. It resulted in the amelioration of
* Corresponding author. E-mail address: [email protected] (M. Venter). 0305-4179/$32.00 # 2006 Elsevier Ltd and ISBI. All rights reserved. doi:10.1016/j.burns.2006.08.008
burns 33 (2007) 464–471
465
the hormonal stress response and improved outcome. Enteral resuscitation should not be introduced in a patient in shock or with existing gastrointestinal disease. Complications were minimal. # 2006 Elsevier Ltd and ISBI. All rights reserved.
Numerous fluid resuscitation formulae have been developed to counterbalance thermally induced fluid losses, hypovolaemia, poor tissue perfusion and shock. Collectively they include crystalloid fluids (2–4 ml/kg/%burn), colloids or hypertonic saline solutions, aiming at maintaining haemodynamic stability and an hourly urine output of 0.5–1.5 ml/kg body weight [1–3]. However, during the same period two important additional advances were made: enteral resuscitation and early enteral feeding. Oral rehydration therapy (ORT) which combines the use of a balanced oral solution of electrolytes and carbohydrates, with early refeeding in children with acute diarrhoea was introduced during the early sixties [4,5]. This process has conclusively been shown by numerous studies to satisfy all principles of fluid replacement, restoration of electrolytes and acid based deficiencies and reducing the need for intravenous resuscitation by 70% [6–8]. In addition to the safety and efficacy of ORT, enteral feeding has become the preferred route of nutrient delivery provided the GIT is functioning. There is progressive evidence that early enteral feeding (<12 h post-burn) in burnt children can be administered safely and effectively in increasing volumes, to meet calculated energy requirements within 2–3 days following severe burn [9–12]. This process preserves gut integrity, restores cell mediated immunity, reduces bacterial translocation, decreases the hypermetabolic response, improves nitrogen retention, increases intestinal blood flow, stimulates the secretion of gut trophic hormones and generally decreases morbidity and mortality [9–12]. To date, few studies document the clinical and metabolic responses in children who have received early enteral feeding (EEF) following major burns [12,13]. Moreover, there are no studies documenting the progress of children where enteral feeding was used as an adjunct to intravenous fluids in resuscitating children with burns. This prospective study reports on the clinical course and outcome of children who have received EEF, in comparison to a cohort who were kept nil per mouth for 48 h during the resuscitation phase. In addition, stress hormones, involved in intermediary metabolism, energy expenditure and intestinal permeability were measured.
1.
Materials and methods
Children admitted to the Paediatric Burns Unit of Red Cross War Memorial Children’s Hospital were studied. For inclusion, the children were between 6 months and 13 years; have burns equal to or greater than 20% of total body surface area (TBSA) and have sustained the injury within 24 h of admission. On admission, the children were randomised to receive either conventional IV resuscitation and introduction of enteral feeding after 48 h [late enteral feeding] (LEF), or enteral
resuscitation and early enteral feeding (EEF) within 24 h of having sustained the injury. Enteral feeding was administered via a nasojejunal feeding tube. These were placed either under fluoroscopic guidance [14] or bed-side-technique; the latter’s position confirmed by abdominal radiograph. The radiological service was only available between 8:00 and 23:00 h and therefore for reasons of safety, the last three children admitted outside of these hours were assigned to the LEF group. The total volume of enterally administered fluid consisted of the following: Resuscitation fluid: Day 1, 3.5 ml/kg/% burn; Day 2, 1.5 ml/kg/% burn. Maintenance fluid: 0–1 years, 120 ml/kg/day; 1–2 years, 100 ml/ kg/day; 2–5 years, 80 ml/kg/day; 5–10 years, 60 ml/kg/day; 1–12 years, 50 ml/kg/day. The total volume, i.e. resuscitation and maintenance fluid, was divided by 24 which constituted the hourly administered volume. Resuscitation was initiated intravenously with a balanced crystalloid solution (Plasmalyte B) at the calculated volume with simultaneous commencement of enteral resuscitation (ER) at a volume of 1 ml/kg/h. The enteral feeding consisted of commercial polymeric formula (Abbott Laboratories). The enteral fluid volume was incrementally increased every 3 h by 50% (2 ml/kg), 33% (3 ml/kg), 25% (4 ml/kg), 20% (5 ml/kg), 17% (6 ml/kg), respectively until the calculated hourly volume was reached. At the same time the intravenous volume, resuscitation and maintenance was correspondingly decreased until the total calculated hourly volume was administered enterally. The intravenous line was kept open with 5 ml/h until the end of the 48 h resuscitation period for purposes of analgesia and if additional intravenous therapy should be required. During the study period there was no need to deviate from this regime, none of the children were shocked on admission and all maintained a urine output between 0.5 and 1.5 ml/kg/h. The children were all admitted to a high care area and routinely monitored to evaluate the progress of the resuscitation, i.e. BP, pulse, pulse-oximetry, urine output and routine haematological and biochemical investigations. A nasogastric tube was aspirated every hour. If the residual volume would have exceeded the previous hourly intake by 50%, the ER would be temporarily stopped and intravenous fluid increased to the full hourly volume. However, this scenario remained absent during the study period. Routine burn wound management was administered to all children. The carbohydrate and protein concentrations of the feed were supplemented from Day 3 and Day 5 onwards in the EEF and FEF patients, respectively using glucose polymer and hydrolysed protein (Abbott Laboratories). The final feed
466
burns 33 (2007) 464–471
composition provided 17–20% protein, 22–28% fat and 55–58% carbohydrate. In both groups, once full enteral nutrition was established, solid meals were offered as an adjunct to the enteral feed. In addition to blood drawn for immediate clinical management, venous blood was sampled for purposes of stress hormone bends determinations on admission, daily for 5 days, and then weekly on days 12, 19 and 26 after admission. Serum and plasma were separated and aliquots stored at 20 8C until hormone concentrations were measured in batches. Routine chemistry was performed immediately. Serum IGF1 was measured using the Medgnix ‘‘IGF-D-RIA’’ kit assay [15]; plasma glucagons [16], serum insulin [17] and serum growth hormone [18] by radioimmunoassay (RIA). Serum cortisol by a ‘heterogenous competitive magnetic separation assay (MSA) [19]. The intra-assay coefficients of variation (CV) for the concentrations of cortisol, glucagon, insulin, IGF1 and GH were 3.1, 3.9, 4.9, 3.9 and 3.5%, respectively. In each patient, resting energy expenditure (REE) was measured daily using a Paediatric Calorimeter (Delta Trac). The machine was calibrated before use with 100% alcohol and a combination of 95% oxygen and 5% carbon dioxide. Daily barometric pressure was obtained from the meteorological office. Full calibration was repeated before daily measurements. The estimate energy requirements (EEE) were extrapolated from these measurements by adding 30% for daily activity [20] and compared to the Recommended Daily Allowances (RDA) [21] for energy and to energy requirements derived by Solomons [22] and Galveston [23] formulae, respectively for children with burns.
Testing for permeability of the small intestine was undertaken in 15 children using a dual sugar absorption test (SAT). Each child was tested on two occasions, once on Day 3 after admission and again 48 h later. The children were fasted for 2 h before and isotonic solution, of lactulose 5 g and L-rhamnose 1 g in 100 ml of water, was administered via feeding tube. The nasojejunal tube was clamped during the procedure. Fasting was maintained for a further 5 h during which time all urine was collected. Hydration and plasma glucose was maintained using the intravenous route. Feeds withheld during this period were subsequently made up by increasing the rate of the nasojejunal feeds. Urinary sugar excretion was measured using thin layer chromatography [24,25] and expressed as a ration of the disaccharide to the monosaccharide. Descriptive statistics for the groups are provided. Where appropriate, normally distributed inter-group data are compared using the Student’s t-test. MANOVA was used for the analysis of repeated measures. Log-transformation was applied to the growth hormone, which were not normally distributed, before comparison. The data were analysed using Statistica (Version Z.34, 1998) and Microsoft Excell packages, 2002. Significance was accepted at the 5% level. The Ethics Research Committee of the University Of Cape Town approved the study. Each child’s legal caregiver signed written consent.
2.
Results
Nine patients in each group completed the study. Twenty-one children were originally recruited for the study, 11 in the EEF
Table 1 – Clinical characteristics of the patients in the study Subject
Age (yr, years, m, months)
TBSAB (%)
Admission weight (kg)
Cause (HWB, hot water burn; FB, fire burn)
Inhalation injury (U) ventilated (V)
EEF group (n = 11) 1 2 3 4 5 6 7 8 9 *10 *11 Medians
8m 2 yr 2 yr, 9 m 4 yr, 7 m 5 yr, 4 m 6 yr, 4 m 7 yr 7 yr 11 yr, 3 m 1 yr, 10 m 1 yr, 4 m 4.54
30 23 30 31 23 35 20 33 30 75 30 29.5
8.72 12.3 13 15.5 18 22 24.5 24.5 30 13.8 11 14.38
HWB HWB HWB HWB FB FB FB FB FB FB HWB
U (V) U (V) U (V)
LEF group (n = 10) 12 13 14 15 16 17 18 19 20 *21 Medians
8 yr, 11 m 3 yr, 8 m 1 yr 1 yr, 1 m 9 yr, 1 m 3 yr, 5 m 1 yr, 9 m 4 yr, 6 m 1 yr *9 yr 4.45
20 25 25 20 28 32 29 30 62 32 30.00
25 13.25 10.5 10.95 15 13 11 19 8 28 14.25
HWB FB HWB HWB HWB FB FB FB FB FB
U (V) U (V) U (V)
burns 33 (2007) 464–471
Fig. 1 – Median serum cortisol concentrations (nmol/l) over time in the EEF group and in the LEF group (normal range: 5.52–800 nmol/l).
and 10 in the LEF group. Two patients in the EEF group died from organ failure on day 3 and 19, respectively, and one patient in the LEF group died from monoxide poisoning and brain ischaemia on day 5 post-burn. The groups were similar in terms of age, body weight and total body surface area of the burn (Table 1). There were no significant difference in the groups with respect to the initial time to debridement, skin grafts or the total number of grafts performed. Feeding in the EEF group was started at a median of 10.7 h (range: 5.5–23 h) after sustaining the burn and at 16 h (range: 13–24 h) all were on full enteral feeds. In the LEF group, enteral nutrition was started at a median of 54 h (range: 51–58 h) post-burn and full feeds were achieved at a median of 10 h (range: 7–10 h) later. There is a shorter time range in the LEF group, since intravenous resuscitation was completed. Although the LEF group consumed fewer calories over the entire period of 26 days; (excluding the first 2 days) there was no statistical significance ( p = 0.7). Serum glucose was equivalently elevated in both groups on admission (median: 10.3 mmol/l in EEF and median: 8.1 mmol/l) in LEF). For the remainder of the study, the glucose concentrations for both groups were within the normal range.
467
Fig. 3 – Median serum insulin concentrations (mU/l) over time in the EEF group and the LEF group (normal range: 0– 30 mU/l).
Fig. 4 – Median plasma glucagon concentrations (pg/ml) over time in the EEF group and the LEF group (normal range: 25–250 pg/ml).
Fig. 5 – Medians for the insulin: glucagon ratios over time in the EEF group and the LEF group.
Fig. 2 – Median serum insulin-like growth factor 1 concentrations over time in the EEF group and LEF group (normal range: 36–350 ng/ml).
There were no significant differences in the serum cortisol or IGF levels between the groups either on admission or subsequently (Figs. 1 and 2, respectively). The range of values for patients in the LEF group appeared to be higher than those in the EEF group, but was not significant. Insulin concentrations were significantly higher ( p = 0.008) in the EEF for the entire study period, as shown in (Fig. 3). Although the plasma glucagon concentrations (Fig. 4) appeared lower in the EEF
468
burns 33 (2007) 464–471
Fig. 6 – The log transformation of serum growth hormone concentrations (mU/l) over time in the EEF group and the LEF group (normal range: 0.5–150 mU/l).
Fig. 7 – Medians for the estimated energy expenditure (REE T 1.3), as measured by indirect calorimetry in the EEF vs. LEF groups.
group on days 3 and 4, the difference failed to reach significance. However, insulin:glucagon (IG) ratio (Fig. 5) were significantly ( p = 0.04) higher in the EEF group till day 4. From admission to day 12, the LEF group had significantly ( p = 0.03) higher hGH concentrations (Fig. 6). The estimated energy expenditure (EEE) for the groups is shown in Fig. 7. Although the median EEE was higher in the EEF
Fig. 8 – The median values of the lactulose:rhamnose ratios for the EEF and LEF groups on day 3 and day 5 post-burn.
group on certain days, the difference for the entire study period was not significantly different ( p = 0.2). The EEE derived from indirect calorimetry in this study was similar to that calculated using the Galveston equation. Small bowel permeability was abnormal in both groups on Day 3 with median lactose:rhamnose ratios (L:R) of 24.3 and 20.9 for the EEF and LEF groups, respectively. Fourty-eight h later, the mean L:R had returned to the normal range in the LEF group but remained elevated in the EEF group (Fig. 8). With respect to the clinical course of the patients, two children in the EEF and 1 child in the LEF group died on days 3, 19 and 5. The EEF patients spent a total of 222 days in hospital compared to a total of 298 days in the LEF group. The difference was not significant ( p = 0.9). The LEF group lost 7.75% of body weight between admission and discharge compared to 3.0% in the EEF group ( p = 0.1). Three patients in the EEF group but none in the LEF group experienced one to two vomiting episodes each during the period of PER. The vomiting did not have adverse consequences. Five of the children in the EEF group had loose stools for between 2 and 4 days after starting the enteral feed as did three in the LEF group. In LEF children, the loose stools lasted for longer than a week. Antibiotic treatment was only given after a positive blood culture, or positive surface swab culture. All 18 children received antibiotics during their hospital treatment, however, the EEF group required a median of 11 days per patient compared to 14 days in the LEF group ( p = 0.08). None of the 18 children who completed the study had blood-culture proven septicaemia.
3.
Discussion
The small bowel has become a central organ for resuscitation and early enteral feeding which came about because of a better understanding of the pathophysiological function of the gut during the early phases of acute illness and trauma. The discovery of the glucose/sodium co-transport system in the 1960s has provided the basis for the development of oral resuscitation methods [4,5]. Essentially sodium is absorbed from the lumen of the small intestine by three basic mechanisms. Sodium moves down an electrochemical gradient, and is exchanged for hydrogen. In addition chloride is exchanged for bicarbonate in an active coupled mechanism. Sodium absorption is also coupled to glucose and other small ions (e.g. amino acids). The co-transport mechanism remains intact during the early phases of hypotension and mucosal injury resulting from enteric infections. Intestinal sodium absorption is critical for fluid homeostasis, since water transfer is always passive and occurs secondarily to the osmotic gradient created by the movement of sodium ions during this process. This process has made possible the effective introduction of an oral rehydration method using a balanced sodium/glucose solution [6–8]. Adult literature has shown several advantages of EEF, and also provided a basis to point out some differences in the paediatric thermal patient’s response to EEF. Adults show a decreased catabolic response as shown by catecholamine concentrations which stayed within normal ranges [12], elevated cortisol concentrations which returned within
burns 33 (2007) 464–471
normal ranges much faster [23,24] and glucagon concentrations which were significantly lower in the EEF group [12]. Anabolism was shown in significantly higher insulin concentrations on days 4 and 8 post-burn [12]. In addition EEF prevents mucosal breakdown, bacterial translocation and decreases small bowel permeability [25,26], maintains (within a 5% range) admission weight [27], with a shorter wound healing period and hospital stay [28], as well as decreased wasting and mortality [23,29]. This study showed that children who have suffered major burns can be resuscitated and nourished via the enteral route from the time of admission to hospital. None of the children who completed this study had any major complications as a result of early enteral feeding. A striking difference between the two groups is the significantly higher insulin response in the EEF group. In adults, delayed feeding was associated with low insulin concentrations, Wilmore et al. [30] Chiarelli et al. [12] found significantly higher insulin concentrations in adults receiving EEF on days 4 and 8 after burn. Insulin is an anabolic hormone and it is feasible that the higher concentrations were involved in earlier tissue repair and were partly responsible for the smaller decline in body weight in the EEF group. The higher I:G ratio in the EEF group resulted not only from higher insulin but also from lower glucagon in the EEF patients. Although the EEF group had lower glucagon concentrations, most patients in both groups had elevated levels for the entire study period. A similar glucacon pattern was documented by Jenkins et al. [31]. It was shown that elevated hGH concentrations during periods of stress, trauma and starvation, enhances gluconeogenesis and lipolysis [32–35]. Significantly higher hGH concentrations in the LEF group confirmed this. The catabolic properties of hGH during times of severe stress have been documented as the cause of increased protein breakdown, increased free fatty acid levels, increased catecholamine levels, and the induction of insulin release resulting in hyperglycaemia [36–38]. The higher hGH concentrations in the LEF group may also have reflected the need for increased glucose production. Taken together, the findings suggest that early feeds provide sufficient energy and that the patients do not have to rely on glycogenolyis or gluconeogenesis for their glucose requirements. Serum cortisol concentrations were moderately elevated in both groups on admission but settled rapidly. The fall in cortisol concentration may reflect appropriate pain control and fluid replacement in the children. IGF-I concentrations were similar in the two groups, also reflecting that early enteral feeding did not disadvantage the children. Unfortunately we were unable to measure nitrogen balance during the study to document an anabolic effect in the EEF patients. The estimated energy expenditure (EEE), measured by indirect calorimetry (EEE 1.3), is in keeping with that derived from the Galveston equation. Several authors found Galveston Equation to be the most reliable [39,40]. The EEE was also similar to that predicted using the RDA. It has been proposed (Gottschlich) [41] that the reduced physical activity in the patients offset the increased energy requirements imposed by the injury and the findings in this study accord with that supposition. Although the median EEE was higher in the EEF
469
group on days 5–7,9,10, 12 and 26, the difference for the entire study period was not significantly different ( p = 0.2). The higher energy expenditure initially did not result in greater weight loss and may have been offset by a greater but not significant energy intake in the EEF group, or possible energy spent on increased wound healing. Although not significant, the EEF group spent less days/patient in hospital than LEF group. Despite most of the children coming from poor socioeconomic backgrounds, none of the children were wasted or stunted on admission. The weight loss of 3% in the EEF group is in the acceptable limit (<5%) for children who have suffered major burns whereas the 7.7% experienced by the children in the LEF group is outside this limit [27]. Loss of the intestinal mucosal barrier results in increased permeability of the small bowel and increases the risk of bacterial translocation and infection [25,26]. The sugar absorption tests confirmed an abnormal small bowel permeability which was more than double normal using a L:R ratio of 10 as the upper limit on admission [45]. The L:R ratio returned to normal in the LEF group and improved in the EEF group. We were surprised that EEF seemed to delay the recovery of small bowel integrity because in other situations e.g. after intestinal surgery, early feeding enhances the recovery of intestinal integrity [46]. However, in spite of the abnormal small bowel permeability, an increased propensity to septicaemia, infection or diarrhoea was not seen. Antibiotics may have prevented infection but would have been more likely to be associated with prolonged diarrhoea, a complication not encountered in the EEF group of patients. Whether the return to normal ranges in intestinal permeability is purely a function of time or can be accelerated by altering the feed, for example, by adding multi-fibre [42–44], requires further study in paediatric patients. From a clinical standpoint, the children who received EEF as PER were not disadvantaged. There was concern about pulmonary aspiration but the recorded vomits were small and infrequent and resolved following revised placement of the nasojejunal tubes. The correct siting of the feeding tube is fundamental to this form of resuscitation and feeding. None of the children In the EEF group experienced ileus, intestinal obstruction or intestinal blood loss and absorption was not impaired as reflected by maintenance of weight. None of the children who completed the study had proven septicaemia. Antibiotic use and total period of hospitalisation was lower in the EEF group, but given the wide range and the small numbers in the study, failed to reach statistical significance. However, even in this study, the total of 76 days shorter hospitalisation and the lower antibiotic use, had cost-saving implications. A study designed to compare the costs of EEF with LEF in terms of hospitalisation, pharmaceuticals, nutrition and human resource utilisation is necessary to determine whether EEF is financially efficient in the long term. Contraindications for enteral feeding in thermal patients include hypovolaemic shock, instability and hypoperfusion, intolerance to oral glucose, pre-existing bowel disease, visible abdominal distention and established diarrhoea. These patients may experience difficulties in maintaining a positive fluid balance. In the presence of severe postburn dehydration or shock the patient should first be resuscitated parenterally
470
burns 33 (2007) 464–471
until circulatory parameters have returned to normal. A relative limitation may be persistent vomiting. The position of the NJT should be corrected immediately. We believe that this study demonstrated that enteral resuscitation and early enteral feeding in burns is safe, effective and not complicated by untoward effects. This approach is particularly suited to the management of children in developing countries, provided that nasojejunal tubes can be positioned correctly. In the paediatric burned population studied, EEF enhanced insulin secretion/anabolism, facilitated maintenance of admission weight, decreased antibiotic treatment days and shortened hospital stay.
references
[1] Muller MJ, Herndon DN. The challenge of burns. Lancet 1994;343:216–20. [2] Graves TA, Cioffi WG, McManus WF, Pruitt BA, et al. Fluid resuscitation of infants and children with massive thermal injury. J Trauma 1988;28:1656–9. [3] Merrel SW, Saffle JR, Warden GD, et al. Fluid resuscitation in thermally injured children. Am J Surg 1986;152:664–9. [4] Fischer RB, Parsons DS. Glucose movements across the wall of the rat small intestinal. J Physiol 1953;119:210–23. [5] Hirschborn N, Kinzie JL, Sachar DB, et al. Decrease in net stool output in cholera during intestinal perfusion with glucose-containing solutions. N Engl J Med 1968;297:176–81. [6] Schedl HP. Scientific rationale for oral rehydration therapy. Clin Ther 1990;12:14–21. [7] Santosham M, Greenough WB. Oral rehydration therapy. A global perspective. J Paediatr 1991;118:S44–51. [8] Farthing MJG. Oral rehydration therapy—past, present and future. Trans Coll Med S Afr 1994;82–9. [9] Scott McDonald W, Sharp CW, Deitch EA. Immediate enteral feeding in burn patients is safe and effective. Ann Surg 1991;213:177–83. [10] Moore FA, Moore EE. The benefits of enteric feeding. Adv Surg 1997;30:141–54. [11] Engelbrecht VJ, Clarke SM. Early enteral feeding of a severely burned pediatric patient. J Burn Care Rehabil 1994;15:293–7. [12] Chiarelli A, Enzi G, Casadei A, et al. Very early nutrition supplementation in burned patients. Am J Clin Nutr 1990;51:1035–9. [13] Klasen HJ, ten Duis HJ. Early oral feeding op patients with extensive burns. Burns 1987;13(1):49–52. [14] Chellis M, Sanders SV, Webster H, et al. Bedside transpyloric tube placement in the paediatric intensive care unit. J Paren Enter Nutr 1996;20(1):88–90. [15] Medgenix IGF-1-D-RIA-CT Kit, Catalog Number: 30 158 60, BioSource Europe S.A., Nivelles Belgium. [16] Glucagon-RIA Kit, Catalog Number: 07-152101, Separations, ICN Pharmaceuticals Inc. Diagnostic Division, 3300 Hyland Avenue, Costa Mesa, California. [17] Coat-A-Count Insulin, Catalog Number: 2 D01-20, Abbott Laboratories (Pty), 149 Samuel Evans Road, Aeroton, Johannesburg S.A. [18] Pharmacia hGH RIA, Pharmacia and Upjohn Diagnostics AB, Uppsala Sweden. [19] Cortisol (Cort), Technicon Immune 1 Septems Bayer Corporation, Business Group Diagnostics, Tarrytown, NY 1059-5097 USA. [20] Gore DC, Rutan RL, Hildreth M, et al. Comparison of resting energy expenditure and caloric intake in children with severe burns. J Burn Care Rehabil 1990;11:400–4.
[21] Food and Nutrition Board. National Research Council Recommended Dietary Allowances. 9th ed. Washington DC: National Academy of Science, 1989. [22] Solomon JR. Nutrition in the severely burned child. Progr Pediatr Surg 1981;14:653–79. [23] Mochizuki H, Trocki O, Dominioni L, et al. Mechanism of prevention of postburn hypermetabolism and catabolism by early enteral feeding. Ann Surg 1984;200(3):297–309. [24] Dominioni L, Trocki O, Fang C-H, Mochizuki H, et al. Prevention of severe postburn hypermetabolism and catabolism by immediate intragastric feeding. J Burn Care Rehabil 1984;5(2):106–12. [25] Van Elburg RM, Uil JJ, De Monchy JGR, et al. Intestinal permeability in pediatric gastroenterology. Scand J Gastroenterol 1992;27(Suppl 194):19–24. [26] Deitch EA. Intestinal permeability is increased in burn patients shortly after injury. Surgery 1990;107: 411–6. [27] Herndon DN, Rutan TC. Management of the pediatric patient with burns. J Burn Care Rehabil 1993;14(1):3–8. [28] Garrel DR, Davignon I, Lopez D, et al. Length of care in patients with severe burns with or without early enteral nutritional support. J Burn Care Rehabil 1991;12(1):85–90. [29] Saito H, Trocki O, Alexander W, et al. The effect of route of nutrient administration on the nutritional state, catabolic secretion, and gut mucosal integrity after burn injury. J Parent Enter Nutr 1987;11(1):1–7. [30] Wilmore DW, Long JM, Mason AD, et al. Catecholamines: mediator of the hypermetabolic response to thermal injury. Ann Surg 1974;180(4):653–69. [31] Jenkins M, Gottschlich M, Alexander JW, et al. An evaluation of the effect of immediate enteral feeding on the hypermetabolic response following severe burn injury. J Burn Care Rehabil 1994;5:106. [32] Tietz NW. Clinical guide to laboratory tests, 3rd ed., Philadelphia: Saunders; 1995. [33] Painter PC, Cope JY, Smith JL. In: Burtis CA, Ashwood E, editors. Tietz textbook of clinical chemistry. 3rd ed., Philadelphia: Saunders; 1999. p. 2161–217. Chapter 41. [34] Zilva JF, Pannall PR, Mayne PD. The hypothalamus and pituitary gland. In: Zilva JF, Pannall PR, Mayne PD, editors. Clinical chemistry in diagnosis and treatment. 6th ed., London: Edward Arnold; 1994. p. 112. [35] Solomon SM, Kirby DF. The refeeding syndrome: a review. JPEN 1990;14(1):90–7. [36] Gilpin DA, Barrow RE, Rutan RL, et al. Recombinant human growth hormone accelerates wound healing in children with large cutaneous burns. Ann Surg 1994; 220:19–24. [37] Roe CF, Kinky J. The influence of human growth hormone on energy sources in convalescence. Surg Forum 1962;13:369–71. [38] Fleming RYD, Rutan LR, Jahoor F. Effect of recombinant human growth hormone on catabolic hormones and free fatty acids following thermal injury. J Trauma 1992;32(6):698–703. [39] Hildreth M, Herndon DN, Desai MH, et al. Current treatment reduces calories required to maintain weight in pediatric patients with burns. J Burn Care Rehabil 1990;11:405–9. [40] Holland KA, Gillespie RW, Lewis NM. Estimating energy needs of pediatric patients with burns. J Burn Care Rehabil 1995;16:458–60. [41] Gottschlich MM. Nutrition in the burned pediatric patient. In: Queen PM, Lang CE, editors. Handbook of pediatric nutrition. Gaithersburg, Md: Aspen Publishers; 1993. p. 536–59.
burns 33 (2007) 464–471
[42] Isolauri E, Gronlund MM, Salminen S, et al. Why don’t we bud? J Pediatr Gastroenterol Nutr 2000;30:214–6. [43] Zopf D, Roth S. Oligosaccharides anti-infective agents. Lancet 1996;347:1017–21. [44] Trier E, Wells JCK, Thomas AG. Effects of a multifibre supplemented enteral feed on gastrointestinal function. J Pediatr Gastroenterol Nutr 1999;27:595.
471
[45] Brewster DR, Manary MJ, Menzies IS, et al. Intestinal permeability in kwashiorkor. Arch Dis Child 1997; 76:236–41. [46] Carr CS, Ling KDE, Boulos P. Randomized trial of safety and efficacy of immediate postoperative enteral feeding in patients undergoing gastrointestinal resection. BMJ 1996;312:869–71.