Energy expenditure was measured by indirect open circuit calorimetry, using a portable calorimeter (MMC Horizon Beckman Sensor Medics). This calorimeter is able to analyse inspired and expired gas by a polygraphic sensor for oxygen and an infrared sensor for carbon dioxide. Indirect open circuit calorimetry by ventilated canopy is the best method for long-term measurements (from 20 minutes to many hours) (3). The patient's head is in a see-through canopy connected with the air source by a flexible tissue tube. An aspirator fixed on the extraction tube assures continuous air change in the canopy. The air flow is regulated at six times the expiration volume of the patient, avoiding in this way excessive C02 concentration in the canopy and also maintaining a slightly negative pressure to prevent air loss. On exit from the canopy, part of the expired air is analysed for the determination Of its 02 and C02 level (knowing the air flow, one can calculate the total volume Of 02 consumed and C02 produced) (Fig. 2).

Fig. 2

At the end of the test we collected the urine for the determination of urine urea nitrogen (UUN) (4).
In our work we performed two types of measurements: 1) the basal metabolic rate (BMR), or energy expenditure of a patient at rest, in the morning and without food for at least 12 hours and in conditions of thermic neutrality; and 2) the resting metabolic rate (RMR), or energy expenditure of a patient at rest and with continuous alimentation for at least 8 hours. We also performed 8 measurements not at rest in order to evaluate the metabolic effect of medication.
During the calorimetric tests we monitored heart rate (HR) and body temperature ('C).
There was daily registration of the quantity of calories given constantly to the patient and after the fourth post-injury day calories were calculated according to the 13MR (BN1R x 1.3). Every day we determined the quantity of UUN in the 24-hours-urine collection and during calorimetric measurements by spectrophotometric method (5). The patients were weighed at least twice a week.
Our results were statistically processed by one-way Anova analysis.

During the first and second week of study the patients' metabolisms had an average increase of 150% compared to that expected in the normal population, based on age and body surface area (6), decreasing slowly to 120% during the fourth week (Tab. II). The measured caloric requirements were lower than those calculated by Curreri'(2) (Tab. III).
In Fig. 3 we illustrate the pattern of BMR and RNIR of one burned patient; there is an evident two-phase pattern with higher values at the end of the second week and a progressive metabolic decrease during the following weeks. The same pattern can be noted evaluating the BNIR of all burned patients, with the basal metabolism expressed in square meters (Fig. 5). There is a significant correlation between BNIR, burned surface area (2nd week p<0.005) and weight (P < 0.00 1).
In our observations the measured BMR was 1169 +/- 45 Kcal/day/sqm. There is also an important variation (p < 0.0 1) in the BNIR values of the first and the fourth week of study.

Fig. 3 Fig. 4

Fig. 5 Fig. 6

The 4th week predicted BMR was 9% lower t.han the 1 st week BMR and in the 4th week of study the measured BMR (BMR x 1.3) was 15% higher than the predicted BN1R.
The average BMR values of the first four weeks of study is comparable to the average RMR values of the same period (Fig. 6). The variance analysis shows no significant RMR modifications during these weeks; there is however a positive correlation between RMR and time (p< 0.05).
In stress situation measurements, or "not at rest-, during medication, the average energy expenditure was 1280 +/- 118 Kcal/day/sqm, a 9.49% increase compared to the BMR. Fig. 6 shows all measurements of energy expenditure for body surface area/sqm. The average calories, calculated on BMR, in constant administration to the patients, were 2531 Kcal/day. The weekly results are shown in Tab. IV.
There is an important difference (p<0.02) comparing statistically the mean BMR and RMR values of the 40 days of study. There is no difference comparing the single week values.
Diet-induced thermogenesis (DIT) expressed as a percentage of BMR in the first week was 12.3 5.5, in the second week 8.9 +/- 3.8, and overall 8.6 2.4; expressed in diet calories the percentage was in the first week 11.9 +/- 3.9, in the second week 5.2 +/- 3. 1, and overall 6.6 +/- 1.7.
The mean respiratory quotient (RQ), measured in basal situations (BMR), was 0.82 +/- 0.01, and dillered significantly (p<0.01) from RQ at rest during alimentation (average values 0.86 +/- 0.01). RQs during 13MR and RMR were significantly different during the first (p<0.001) and the second week (p< 0.005), but not during the following weeks (Fig. 7). The RQ in BMR, correlated to time, had a significant variation (p< 0.0 1), while the RQ in RMR had none. The weekly average is shown in Tab. V.
The weekly average body temperature values are shown in Tab. VI. The average axillary temperature during the first week was 37.9 'C, and in the second week 38.0 'C. The values normalized within 30 days. Temperature correlation with energy expenditure/sq m (p < 0.00 1) and HR (p < 0.00 1) was significant (Fig. 8). The weekly average values of HR are shown in Tab. VII. During calorimetric tests at rest HR was 103 +/- 2 beats/min and had a significant correlation with time (p<0.001) and with energy expenditure (p<0.001) (Fig. 9). UUN increased in the first and second week without any significance compared to normal values. The total weight loss in the 30 days after injury was 5.9 +/- 2.2%.

The purpose of our work was to evaluate a protocol for the exact assessment of the nutritional state of burned patients. Until today the caloric requirement was determined by Curreri's formula, based on the patient's weight and the percentage of burned surface. The formula is suitable for the indirect calculation of the caloric requirement. The first part of our work regards the study of the usefulness of the formula for energy expenditure, referring to indirect calometric values. This method

Fig. 7 Fig. 8 Fig. 9

of energy expenditure and also evaluates variations of BMR, RQ and RMR in burned patients, both in acute phase and recovery. Our data show that the basal metabolism increases in the first and second week, and then decreases. In the first and second week the increase was 150% compared to the predicted BMR, based on body surface and age; in the fourth week it decreased to 120%. Therefore, four weeks after injury, * RMR in the recovery phase, there is still hypermetabolism (measured BMR is 15% higher than predicted BMR).
The fourth week predicted BMR is 9% lower than the first week BMR, due to weight loss, showing that the metabolic needs decrease with time, due to the decrease of lean body mass causing a lower caloric requirement than pre-burn, although hypermetabolism persists. Together with an initial BMR increase there was also a significant HR and temperature increase. The variation of these data is significatively correlated, showing the possibility of using HR measurements as a prediction for the actual individual energy expenditure.
Body temperature is also correlated to basal apidly gives information about individual energy expenditure, with very little disturbance to the patient; it is also very easy to perform. The results of our study show, as other Authors refer (7, 8, 9), that the average value of 24-hour energy requirements in the 6 patients, according to Curreri's formula, is higher than that predicted (Tab. 111). Indirect calorimetry allows an exact measurement metabolism, raising the question whether the basal metabolic increase is due to hyperthermia. We observed also hypermetabolism in patients without fever during recovery.
We calculated that the RQ, whose values allow us to know both the oxidative processes and utilized substrates, is significantly different during fasting from the RQ during constant alimentation. The average fasting RQ in the burned patients studied is 0.81, corresponding to normal subjects. Apparently this shows a normal substrate use in the short fasting period necessary for BMR measurement.
Burned patients with negative energy balance show a lean body mass loss with proteolysis due to gluconeogenesis. Despite the weight loss and increase of UUN excretion of the patients, the average RQ value was 0.80, indicating gluconeogenesis activation. The resulting RQ value may therefore depend on nutritional glucose oxidation processes, on glycogencilysis or on gluconeogenesis in previous days.
Determination of the UUN permits the evaluation of the actual protemic oxidation process entity, and therefore makes it possible to calculate, depending on the RQ, the proportions between the various substrates.
In constant nutritional conditions (RNIR) the average RQ was 0.86, corresponding to the food RQ (relation between total C02 produced and total 02 consumed).
According to the observation of other Authors (10, 11), the more revealed RQ and food correspond, the better the utilisation of all given substrates.
The RQ never exceeded 0.95, which was revealed in the fourth week, possibly depending on hypernutrition.
These studies on burned patients are only preliminary preparations and verifications for a protocol of the nutritional conditions and needs of such patients. Further research is still necessary to confirm our results.

RÉSUMÉ. L'organisme humain répond aux divers stress (traumatismes, brûlures) avec une augmentation de la dépense énergétique et du besoin calorique. Chez les brûlés le taux métabolique, en condition de repos, augmente par rapport à la surface corporelle brûlée; il y a aussi une réduction pondérale due au catabolisme augmenté, aux pertes protémiques des surfaces brûlées et à l'immobilisation. Il est donc indispensable d'évaluer la dépense énergétique des brûlés afin de régler l'administration nutritive soit quantitativement soit qualitativement. A cette fin, au Centre des Brûlés de Padoue, nous avons étudié pour 60 jours 6 patients brûlés (2 mâles, 4 femelles, surface corporelle brûlée 20-60%, âge moyen 35,4 +/- 5,2). Pendant cette période nous avons administré les nutriments suivants i.v.: le glucose 10%, les aminoacides (Freamine 111) et les lipides (Intralipid 10%); et d'autres nutriments avec le tube nasogastrique: les nutriments semiliquides prêts à l'emploi (Nutrisond Nutricia-Pierrel 1 Kcal/mL), et en cas de besoin: les solutions polysalines, le plasma, le sang, l'albumine.
La dépense énergétique a été mesurée avec la calorimétrie indirecte à circuit ouvert (MMC Horizon Sensor Medics). Nous avons exécuté deux types d'évaluation: Taux Métabolique Basal (TMB), c'est-à-dire la dépense énergétique du sujet en condition de repos (le matin, après le jeûne du sujet pendant au moins 12 heures à température constante), et le Taux Métabolique à Repos (TMR), c'est-à-dire la dépense énergétique du sujet à repos qui est alimenté avec le tube nasogastrique pendant au moins 8 heures. Pendant les tests calorimétriques nous avons suivi au moniteur la fréquence cardiaque (FQ et la température corporelle axillaire ('C).
Le TMB des patients était 2090 +/- 148,5 Kcal. Le cours du TMB était biphasique, avec des valeurs plus élevées à la fin de la deuxième semaine après la lésion et une diminution progressive pendant les semaines successives. Nous avons observé des corrélations significatives entre le TMB, la surface corporelle brûlée (P< 0,005) et le poids corporel (P< 0,001). Nous avons observé une variation significative entre les valeurs du TMB à la première semaine d'étude et celles de la neuvième semaine (P 0,01), mais aucune variation des valeurs du TMR. La thermogenèse postprandiale (la thermogenèse déterminée par la diète (TDD), exprimée en % du TMR, était 12,3 +/- 5,5 pendant la première semaine, 8,9 +/- 3,8 pendant la teuxième semaine, et 8,6 +/- 2,4 pendant l'entière période d'observation. Le métabolisme, pendant la première semaine, montrait une augmentation moyenne de 150% par rapport à ce qu'on peut attendre chez une population normale, avec référence à l'âge et la surface corporelle brûlée. Le quotient respiratoire (QR = C02/02) du TMR pendant les premiers 60 jours était 0,82 +/-0,0 1) (valeur moyenne) et nous avons observé une différence significative (P < 0,0 1) par rapport au QR en condition de repos pendant l'alimentation (QR = 0,76 +/- 0,01). Les QRs pendant le TMB et le TMR étaient significativement différents pendant la première semaine (P<0,001) et la deuxième (P<0,005), mais non plus pendant les semaines successives. Les valuers moyennes de la température corporelle étaient 37,9 'C pendant la première semaine, et 38,0 OC pendant la deuxième, avec une réduction successive et le rétablissement des valeurs normales dans les semaines suivantes. La corrélation de la température avec la dépense énergétique/mètre carré (P < 0,00 1) et avec la FC (P < 0,00 1) était significative. La FC pendant le TMB avait une valeur moyenne de 103 +/- 2 battements par minute, avec une corrélation significative avec la dépense énergétique (P< 0,001). Ces résultats nous donnent une évaluation quantitative du métabolisme augmenté, des besoins nutritifs, de la variation de la thermogenèse postprandiale et des QRs, grâce à la grande fiabilité des méthodes de mesurage employées pendant les diverses fases des brûlures.

1. Kinney J.M.: "Energy Metabolism. Surgical nutrition". Fisher J.E. ed., Little Brown & Co., Boston, Toronto, 1983.
2. Curreri P.W., Richmond D., Maruen J.A., et al: Dietary requirements of patients with major burns. J. Am. Diet, Assoc., 65: 415-17, 1974.
3. Jequier E.: Long term measurement of energy expenditure in man: direct or indirect calorimetry? In: Recent advances in obesity research 111. Biontorp P., Cairella M., Howard AX, eds., 130, John Libley, London, 1981.
4. Jequier E.: Métabolisme energetique. Nutrition. EMC. Paris, 11, 1980.
5. Kaplan A.: Urea nitrogen and urinary ammonia. In: Meites S., ed. Standard methods of clinical chemistry, 245-56, Academic Press, New York, 1965.
6. Harris J.A., Benedict F.G.: A biometric study of basal metabolism in man. Washington D.C., Carnegie Institute of Washington. Publication n' 279, 1919.
7. Schane J, Goede M, Silverstein P.: Comparison of energy expenditure measurement techniques in severely burned patients. JBCR, 8: 366, 1987.
8. Turner W.W. Jr, Ireton C.S., Hunt J.L. and Baxter C.R.: Predicting energy expenditures in burned patients. J Trauma, 25: 11, 1985.
9. Saffle J.R., Medina E., Raymond J., Westenskow D., Kravitz M., Warden G.D.: Use of indirect calorimetry in the nutritional management of burned patients. J Trauma, 25: 32, 1985.
10. Flatt J.P.: The biochemistry of energy expenditure. In: Recent advances in obesity research. 11. Bray G.A., ed., 211, Newman Publishing Co., London, 1978.
11. Zurlo F, Schutz Y., Frasearolo P., Enzi G, Deriaz 0., Jequier E: Variability of resting energy expenditure in healthy volunteers during fasting and continuous enteral feeding. Crit. Care. Med., 14: 535, 1986.