A NEW OPEN AND CLOSED RESPIRATION CHAMBER

Transcription

1 Quaterly Journal of Experimental Physiology (1983) 68, Printed in Great Britain A NEW OPEN AND CLOSED RESPIRATION CHAMBER L. HOWARD AULICK, H. ARNHOLD, E. H. HANDER AND A. D. MASON, JR United States Army Institute of Surgical Research, Brooke Army Medical Center, Fort Sam Houston, TX 78234, U.S.A. (RECEIVED FOR PUBLICATION 2 DECEMBER 1982) SUMMARY A new, fully automated, open and closed respiration chamber has been constructed, which is large enough to accommodate man or comparable sized animals. Metabolic rate is calculated from measured changes in respiratory gas volumes while the system is hermetically sealed. When CO2 concentration reaches a prescribed level, large ports are opened and the chamber is ventilated. By repeating this process, a series of metabolic determinations can be conducted over an extended period. Temperature, humidity and air velocity are fixed, but chamber volume and pressure vary with changes in the respiratory exchange of the animal and/or barometric pressure. Small, rapid changes in gas volumes are monitored by a low-resistance spirometer and large changes accommodated in a second, high-volume, motorized spirometer. Chamber operations, data aquisition and metabolic calculations are performed by a small computer. System calibration with methanol combustion indicates that estimates of respiratory gas exchange are highly accurate over a temperature range from 1-4 'C. INTRODUCTION In 1972, Sir Kenneth Blaxter and colleagues at the Rowett Research Institute in Scotland built a large open and closed respiration chamber which combined the simplicity and accuracy of closed systems with the long-term measurement capabilities of open, flow through chambers (Blaxter, Brockway & Boyne, 1972). The instrument to be described in this communication is patterned after Blaxter's 'open and shut' chamber with one distinct feature. In the Scottish design, chamber volume is fixed and air temperature manipulated to keep internal pressure at atmospheric levels. The new chamber has a flexible volume, so that temperature can be held constant and chamber gas volume permitted to fluctuate with ambient pressure. This provides a level of environmental control not available in the original design. This new instrument was developed to study the metabolic response to burn injury - a response which is extremely sensitive to changes in the thermal environment. This relatively simple, constant temperature, variable volume instrument has proven to be a very accurate way to estimate heat production in large unrestrained animals. This paper describes the design, operation and validation of this new system. METHODS Design and operation This new open and closed system consists of a 2 4 x 1 2 x 1 2 m chamber and two attached spirometers (Fig. 1). The chamber is made of -48 cm steel plate and covered with a 2-54 cm layer of thermal insulation. It has a large door and a small window, both made of 1 25 cm acrylic plastic. The chamber is divided into an animal space and a narrow air treatment corridor running the length of the back wall. Air is continuously circulated through the chamber by four 2 3 cm fans located

2 352 L. H. AULICK AND OTHERS Low, g resistance,it spirometer r Temp. - Cr r > Mass spectrometer - ven tiao I J vniato Motor.-H2seall ~-Temp. 2 i Tissot UIH A V spirometer Fan s-- Intslation dornv } - Temp. Window -4 Insulation Temp. --s- Fig. 1. Schematic diagram of chamber. at one end of the air treatment corridor. Air velocity in the centre of the animal space averages 24 m/min when all four fans are operating and 1 m/min when only two fans are running. Two ports in the roof above the air treatment corridor can be opened to ventilate the chamber or closed to make it air tight. Each port is 25 4 cm in diameter and surrounded by a circular trough containing light-weight mineral oil. When the chamber is closed, these ports are covered by heavy, cylindrical caps which seat in the mineral oil and prevent air leaks across the closed ports. Shutters in the air-treatment corridor open when the ports close and air travels from the fans through a set of cold coils where water vapour condenses. Cold, drier air then passes through heated coils where it is returned to the desired temperature. Chamber humidity ranges between 4-5% during normal operating conditions. Chamber temperature is controlled by varying the water temperature in the heated coils. Water temperature is regulated by a proportional controller (Model 72, Yellow Springs Instruments, Yellow Springs, OH) which senses air temperature in the animal space and varies the output of a 115 W immersion heater in the external reservoir. When the ventilating ports open, the shutters close and chamber air leaves through one port to be replaced by room air entering through the other. Under normal operating conditions, chamber 2 and CO2 concentrations return to normal within 5 min. Volume flexibility is provided by the two attached spirometers. The piston of the smaller, lowresistance spirometer (11, Model 84 Ohio Medical Products, Madison, WI) moves whenever the pressure differential between the surrounding atmosphere and the chamber exceeds 5 mmh2o. By minimizing the pressure difference, this low resistance response significantly reduces the potential for air leaks. In a chamber of this size, however, the small spirometer cannot accommodate air changes accompanying daily fluctuations in barometric pressure. (A change in barometric pressure of only 2 Torr will result in a 1 1 change in chamber air volume. Daily fluctuations in barometric pressure may reach 2 Torr.) For this reason, the larger, Tissot spirometer (12 1 Gasometer, W. E. Collins, Inc., Braintree, MA) is required to serve as a high-volume reservoir. When either limit of the small spirometer is reached, an electric motor attached to the pulley of the Tissot moves the bell in a compensatory direction until the piston of the small spirometer is returned to mid-position.

3 RESPIRATION CHAMBER 353 Two fans continuously force well-mixed chamber air through the large spirometer. These fans are arranged and flow rate is high enough so that the composition of the air moving through the Tissot equals that entering the animal space. This arrangement eliminates the need for separate analyses of chamber and spirometer air. Gas samples are collected at a point where air enters the Tissot and the analysis is performed by a mass spectrometer (Model 11, Medical Gas Analyzer, Perkin-Elmer, Pomona, CA). This instrument records the relative fractions of 2, CG2, N2, CH4, A and H,O in chamber air. Chamber temperature is measured by five two-terminal, monolithic, integrated circuit transducers (Model AD 59, Analog Devices, Norwood, MA). Two transducers measure air temperature at each end of the treatment corridor, while the fifth monitors wall temperature in the animal space, The average of these five temperatures provides an estimate of over-all chamber temperature. Air temperature in the large spirometer is measured separately by a sixth transducer. System pressure is measured by an electronic aneroid barometer (B242 Analog Output Barometer, Weather Measure Corporation, Sacramento, CA). Chamber operation, data acquisition and metabolic calculations are performed automatically by a small desk top computer (Model 85, Hewlett Packard Company, Corvallis, OR). After a 5 min ventilation period, the large spirometer bell is positioned and the ventilating ports close. The actual measurement interval begins 1 min later when chamber temperatures, pressure and gas composition data are recorded and the initial volume of each respiratory gas calculated. Data acquisition is repeated at 15 min intervals throughout the closed phase. Following each measurement, the current gas volumes are determined and the respiratory gas exchange for the entire run computed. This process continues until chamber CO2 concentration reaches -85%. At this point, final calculations are performed and the chamber ports open. The computer provides a printout of chamber status and metabolic calculations following each 15 min sampling sequence and sounds an alarm at the end of each run. All phases of computer management can be modified through changes in the software. Calculations Metabolic heat production can be calculated from measured changes in respiratory gas volumes while the subject is confined in the closed chamber. For man and other monogastric animals, only 2 and CO2 volumes are involved in these calculations (Weir, 1949), but, in ruminants, CH4 production must also be included (Brouwer, 1965). Regardless of which formula is required, the rate of volume change of each gas (Vgas) is the absolute value of the difference between initial and final volumes divided by time. The basic equation is: V (VI X Igag) -( VF X Fgas) gas - time Where VI and VF are the initial and final volumes of all the air in the closed system (chamber+ two spirometers -animal) and Iga, and Fgas refer to the initial and final fractional concentrations of each gas. All gas volumes are corrected to standard conditions. RESULTS System validation The accuracy of respiratory gas exchange measurements by this system depends on (1) the chamber being hermetically sealed when closed; (2) an accurate assessment of the fixed, ambient chamber volume; and (3), well-calibrated analytical equipment. An air leak can be identified by the failure to maintain pressure and/or gas concentration gradients across the chamber wall. Such checks are performed weekly, and in 6 months of operation, no leaks have developed. The volume of air in each spirometer can be determined to within a few millilitres, but the fixed volume of the chamber itself is more difficult to assess. Since this represents the largest single volume of air, the accuracy of this assessment is extremely critical. For this reason, three independent estimates were employed. First, the best estimate by careful physical measurement was The second approach was to inject known volumes of pure argon into the chamber and calculate volume from its dilution in chamber air. Using

4 354 L. H. AULICK AND OTHERS Table 1. Summary offorty-one methanol combustion studies Range VO VC 2 R.Q. Predicted 53 to to (ml/min) Observed 53 to to to 7 (ml/min) Error* -8-4 to to to 4 6 Mean % error Observed -Predicted * x 1. Predicted this technique, calculated chamber volume was (mean+ S.E.M., n = 11). The third approach was to burn methyl alcohol at measured rates and calculate the volume of air necessary to yield the observed rate of either 2 disappearance or CO2 accumulation, assuming complete combustion and using predicted rates of 2 uptake and CO2 production. Chamber volume by this approach was (n = 39) when 2 disappearance was the standard and when CO2 accumulation was used. Based on the close agreement of all estimates, an average value of was selected. Analytical accuracy for the mass spectrometer is + -1% for 2 and N2,±+1 %for CO2, CH4 and A and + 1 Torr for water vapour. The barometer is calibrated to + 1 Torr and temperature transducers are accurate to + -1 C. In a chamber of this size, the standard volume of 2 is well over 6 1. A 1 % error in 2 concentration would cause the calculated total 2 volume to be off by as much as 3 1. The limits of accuracy in 2 measurement represent the major analytical weakness of this system, since possible errors in the temperature and pressure measurements become relatively insignificant when applied in the equations correcting gas volumes to standard conditions. A standard test of any system designed to measure respiratory gas exchange is to compare observed rates of 2 consumption (VO2) and CO2 production (Vco2) with those predicted from measured rates of alcohol combustion. A total of forty-one such studies were performed at three chamber temperatures: 1, 26, and 4 'C. The rate of combustion was varied to provide a range of V2 from 53 to 65 ml/min. Because each study lasted until CO2 concentration reached.85%, measurement periods ranged from 9 to 7 5 h. In nine other studies (three at each temperature), no alcohol was burned. These runs lasted 2 h. The greatest error in observed VO2 was 8.4% below that predicted and occurred when the chamber was heated (Table l).the average error, however, was only 1 %. Observed VC2 was off by as much as 5.4% in one study but, once again, the average error was less than 1 %. These error estimates compare favourably with those from other forms of direct and indirect calorimetry (Blaxter et al. 1972; Spencer, Zikria, Kinney, Broell, Mechailoff & Lee, 1972; Caldwell, Hammel & Dolan, 1966; Spinnler, Jequier, Favre, Dolivo & Vannotti, 1973; Pullar, Brockway & McDonald, 1967; Webb, Annis & Troutman, 198). The respiratory quotient (R.Q.) for methanol oxidation is -67. Observed values ranged from 62 to 7 but averaged 67. A stepwise multiple regression of the data from all fifty studies revealed that a small component of the variation is observed VO2 was a function of chamber temperature (Fig. 2). Since this effect only increased the index of determination (r2) from 9989 to 999. it was considered functionally insignificant. The rate of methanol combustion was the sole

5 RESPIRATION CHAMBER E t 2. 1 Observed = X predicted -*2 X temperature r2 = * Predicted oxygen consumption (m/min) Fig. 2. The observed rates of oxygen consumption were in close agreement with those predicted from measured rates of methanol combustion. Chamber temperature ranged from 1-4 'C. 4 3 t-2 Observed = x predicted r2 = / Predicted carbon dioxide production (ml/min) Fig. 3. The observed rates of carbon dioxide production were in close agreement with those predicted from measured rates of methanol combustion. Chamber temperature ranged from 1-4 C.

6 356 L. H. AULICK AND OTHERS determinant of observed Vco2 over this same temperature range (Fig. 3). These results clearly indicate that under normal operating conditions, this new open and closed respiration chamber accurately measures gas exchange. Temperature control Unlike its Scottish predecessor, this instrument was designed to provide a stable internal temperature during the closed phase of operation. In a total of thirty-six 24 h studies on large goats, mean chamber temperature (TCH) averaged IC at the beginning of the closed phase and C at the end (n = 32). Under these experimental conditions, it is rare for mean TCH to vary more than 3 'C. Each time the chamber is opened, its thermal equilibrium is transiently disturbed by any difference in the heat content of the outside air (room temperature ranges from 25-3 'C). In the methanol lamp studies, the average change in TCH was.1 'C (range --6 to 5 'C; ATCH = initial TCH -final TCH) when the chamber was heated to 4 'C as compared to-- 5 'C in the 1 'C chamber (range -2- to -5 IC). These variations are considered extreme, however, since unlike the animal studies, the chamber door had to be opened at the end of each run in order to refill the methanol lamp. The results of these studies indicate that even the more extreme fluctuations in TCH have little or no effect on the accuracy of respiratory gas exchange measurements. Animals studied Goats, pigs and rats have been studied in this instrument. The 2-4 kg goats never adapted to chamber confinement but became progressively more active with repeated trials. The pigs (4-8kg, Pitman Moore), however, quickly accepted the chamber and after eight to ten conditioning runs, they maintained resting metabolic rates of ±1 5 W/m2 (mean + S.E.M., n = 28). The rats were studied in a group of thirty-nine animals, each housed individually in its own cage. They, like the pigs, adjusted quickly to chamber living and soon reached a resting oxygen consumption of ml/h. g. The capacity to study groups of small animals in this fashion provides an excellent way to characterize the average metabolism of a relatively large population in a reasonable period of time and under identical experimental conditions. While not the origirtal intent, such an application demonstrates the flexibility and great utility of this large chamber. DISCUSSION In summary, this new open and closed respiration chamber offers two advantages over the fixed volume, variable temperature system developed earlier in Scotland. First, the flexible xolume design eliminates the need to control chamber pressure, a process which is slower and must tolerate larger pressure gradients than are permitted by this instrument. (The elimination of the latter is most important in that it will reduce the risk of leaks in a system designed to be airtight.) Secondly, with both volume and pressure free to follow ambient conditions, greater control of internal temperature can be achieved. In the strictest sense, this instrument is not an environmental chamber, but temperature control is very good, and the observed variations in temperature do not affect measurement accuracy. Like its predecessor, this new design has proven to be a simple, relatively inexpensive* and extremely accurate way to measure respiratory gas exchange of unrestrained animals for an extended period of time. * The chamber was constructed at the Institute for less than $6,. (including labour). About half of the total cost went for the purchase of the mass spectrometer. Operational expenses are extremely low, consisting mainly of the cost of electrical power. Maintenance costs are also minimal, for in almost a year of continuous operation, we have had only to replace one immersion heater and to clean the cooling coils on one occasion.

7 RESPIRATION CHAMBER 351 The authors gratefully acknowledge the thoughtful advice and encouragement provided by Sir Kenneth Blaxter. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defence. In conducting the research described in this report the investigators adhered to the 'Guide for Laboratory Animal Facilities and Care', as promulgated by the Committee on the Guide for Laboratory Animal Facilities and Care of the Institute of Laboratory Animal Resources, National Academy of Sciences National Research Council. REFERENCES BLAXTER, K. L., BROCKWAY, J. M. & BOYNE, A. W. (1972). A new method for estimating the heat production of animals. Quarterly Journal of Experim-eital Physiology 57, BROUWER, E. (1965). In Energy Metabolism, ed. BLAXqER, K. L., p London: Academic Press. CALDWELL, F. T., HAMMEL, H. T. & DOLAN, F. (1966). A calorimeter for simultaneous determination of heat production and heat loss in the rat. Journal of Applied Physiology 21, PULLAR, J. E., BROCKWAY, J. M. & MCDONALD, J. D. (1967). A comparison of direct and indirect calorimetry. Proceedings of the 4th Symposium on Energy Metabolism, Warsaw, European Association for Animal Production, Publication No. 12, pp SPENCER, J. L., ZIKRIA, B. A., KINNEY, J. M., BROELL, J. R., MECHAILOFF, T. M. & LEE A. B., (1972). A system for continuous measurement of gas exchange and respiratory functions. Journal ofapplied Physiology, 33, SPINNLER, G., JEQUIER, E., FAVRE, R., DOLIVO, M., & VANNOTTI, A. (197J). Human calorimeter with a new type of gradient layer. Journal of Applied Physiology 35, WEBB, P., ANNIS, J. F. & TROUTMAN, S. J. JR (198). Energy balance in man measured by direct and indirect calorimetry. American Journal of Clinical Nutrition 33, WEIR, J. B. DE V. (1949). New method for calculating metabolic rate with special reference to protein metabolism. Journal of Physiology 19, 1-9.