J Appl Physiol 102: 1229-1234, 2007.
First published November 2, 2006; doi:10.1152/japplphysiol.00744.2006
8750-7587/07 $8.00
INNOVATIVE METHODOLOGY
Apparatus for measuring rat body volume: a methodological proposition
Rodrigo Hohl,1
Renato Buscariolli de Oliveira,1
Denise Vaz de Macedo,1 and
René Brenzikofer2
1Laboratório de Bioquímica do Exercício, Departamento de Bioquímica, Instituto de Biologia, and 2Laboratório de Instrumentação para Biomecânica, Faculdade de Educação Física, Universidade Estadual de Campinas, Campinas, Brasil
Submitted 4 July 2006
; accepted in final form 30 October 2006
 |
ABSTRACT
|
|---|
We propose a communicating-vessels system to measure body volume in live rats through water level detection by hydrostatic weighing. The reproducibility, accuracy, linearity, and reliability of this apparatus were evaluated in two tests using previously weighed water or six aluminum cylinders of known volume after proper system calibration. The applicability of this apparatus to measurement of live animals (Wistar rats) was tested in a transversal experiment with five rats, anesthetized and nonanesthetized. We took 18 measurements of the volume under each condition (anesthetized and nonanesthetized), totaling 90 measurements. The addition of water volumes (50–700 ml) produced a regression equation with a slope of 1.0006 ± 0.0017, intercept of 0.75 ± 0.81 (R2 = 0.99999, standard error of estimate = 0.58 ml), and bias of 
1 ml. The differences between cylinders of known volumes and volumes calculated by the system were <0.4 ml. Mean volume errors were 0.01–0.07%. Among the live models, the difference between the volumes obtained for anesthetized and nonanesthetized rats was 0.31 ± 2.34 (SD) ml (n = 90). These data showed that animal movement does not interfere with the volume measured by the proposed apparatus, and neither anesthesia nor fur shaving is needed for this procedure. Nevertheless, some effort should be taken to eliminate air bubbles trapped in the apparatus or the fur. The proposed apparatus for measuring rat body volume is inexpensive and may be useful for a range of scientific purposes.
density; live rat; hydrostatic weighing
THE DENSITY OR SPECIFIC GRAVITY of the body is a biological parameter determined by the body volume-to-body weight ratio. Density determination is important for evaluation of morphological changes due to age (19), use of drugs (24, 28), different diets (6, 16, 25), or training (28), because it is inversely related to the percentage of body fat in humans (1) and animals (6). Moreover, the determination of density may play a major role in the application of exhaustion tests in rats subjected to swimming effort (5, 11, 17, 20, 25), since the volume and mass of the animals account for their buoyancy in water. In 1947, Scheer et al. (25) showed that the reproducibility of performance tests was significantly increased in animals in which densities were homogenized before tests. However, effort intensity is still standardized by the addition of a load equivalent to a percentage of body mass, rather than animal volume (i.e., density), in rats subjected to exhaustion swimming tests (4, 10, 21, 29).
Other studies on the development and the validation of procedures for measuring volume in humans and small animals date from the beginning of the 20th century and comprise methods including air displacement, helium dilution, and hydrostatic weighing (1, 3, 6, 9, 13, 14, 16, 22, 23, 27, 30, 31).
The air-displacement procedure consists of partial evacuation of a standard chamber connected to another chamber of known volume that contains the body to be measured. After the two chambers are connected, pressure equilibrates, and the body volume is calculated from the temperature and pressure obtained in both chambers (9, 12, 15). According to Hix et al. (12), possible sources of error in this procedure are changes in temperature and air humidity and insufficient precision of pressure sensors. More recently, an air-displacement plethysmograph (BOD POD), for which isothermal conditions are not required since the air in the chambers is allowed to compress and expand adiabatically, has become commercially available for measurement of human body volume (7, 8, 18).
The helium dilution method is performed in a chamber of known volume with a fixed volume of helium (9, 12, 26, 30). The unknown volume of a body is determined by a conductance cell used to analyze helium concentration (12) or by the principle of dilution of gases (30). In the first case, determination of the volume by the conductance cell is affected by air humidity and CO2, since the cell is sensitive to both of these variables (12). Temperature, helium absorption by the body, and incomplete mixture of the gas interfere with the volume determination in the other method (30).
Hydrostatic weighing has been the main procedure used to determine the volume of animals. This technique, based on Archimedes' principle, compares weight in the air with the apparent weight in a liquid medium of known density. The main difficulty in using this procedure is the movement of the animal during measurement. To address this issue, anesthetized (14, 28) or dead (6, 22–24) animals have been used in most studies. Also, the animal's fur is shaved to decrease bubbles trapped among the hairs (6, 16, 22, 23). However, these procedures are not feasible in a longitudinal study evaluating the individual body density of live animals subjected to a specific training or nutrition regimen for several weeks.
This study presents an easily manufactured apparatus for measurement of body volume in small animals that can accurately and reproducibly measure body volume in live rats without anesthesia or fur shaving.
|
MATERIALS AND METHODS
|
|---|
Apparatus for Measurement of Rat Body Volume
The apparatus for measurement of rat body volume (AMV) consists of two communicating cylindrical vessels linked through a flexible hose (Fig. 1). The rat is placed in vessel C, while water level changes are measured by hydrostatic weighing of a cylindrical test tube with a lead ballast in vessel B. The connecting flexible hose is sized to attenuate the transmission of disturbances caused by animal movements, and the reinforced walls prevent pressure-induced alteration of the internal volume. The AMV is simple and dismountable. Caution should be taken to ensure that supports are tight and the scale is protected from circulation of air. Glass or other transparent material should be used for vessel C, so that the animal can be viewed and observation can be facilitated. To maximize the linearity of the system, those parts that might become wet when the water level rises must be cylindrical. The characteristics of the main components of the AMV are described in Table 1.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 1. Apparatus for measuring rat body volume (AMV): test tube with lead ballast (A), measurement vessel (B), glass vessel for submerging the animals (C), digital scale (D), flexible hose connecting the vessels (E), screen for feces retention (F), faucet (G), iron rod (K), and iron supports to fix the system (H, I, and J). h, Variation in water height before and after animal was introduced into the system.
|
|
The AMV setup protocol is as follows. 1) Vessels B and C are connected through the flexible hose. 2) Water (1.7 liters) is added to the communicating-vessels system, and air bubbles are removed from inside the flexible hose and vessels B and C by vigorous shaking (lifting and tilting) of the system and water flow from the faucet. 3) The ballasted test tube is suspended from the scale by a rod and centralized inside vessel B. 4) Water is added to vessel C until the bottom of the test tube is reached inside vessel B, which is signaled by the scale. 5) Water (150 ml) is added to vessel C until water surrounds the base of the test tube and reaches its cylindrical portion. 6) The scale is tared before initiation of calibration.
Volume Determination
The introduction of the animal inside the vessel increases water level in the system. Simulations in which the measured volume was related to the dimension of the components optimized the sensitivity of the apparatus. The rat volume (Vrat) is related to the variations in volume, 
VB and VC, from vessels B and C, respectively
where VA and VF are the variations in volume occupied by the test tube and the supporting rod of the retention screen, respectively.
The increase in the water level (h) around the test tube of diameter dA causes variation in the scale reading (Mscale) of
where Dwater indicates water specific mass (1 g/ml).
Hence, Vrat is directly proportional to the variation indicated in the scale
where dB and dC are the inner diameters of vessels B and C, dA is the outer diameter of the test tube, and dF is the diameter of the supporting rod of the retention screen.
Thus, Mscale varies according to the volume of the object inserted into the system. If the components (A, B, C, and F) are cylindrical in the range of water level variation, Vrat does not depend on the initial water level at tare time. In practice, the volume measure depends only on the value read in the scale (Mscale) and the calibration equation quantified after setup.
AMV Calibration
The purpose of the calibration procedure is to quantify the equation that relates the volume of the submersed body to the value displayed by the scale. A fast and precise method of calibration was developed using six cylindrical aluminum machined parts (5 cm diameter and 5 cm high) measured for a precise determination of individual volume. Calibration consisted of consecutive introduction of the parts into vessel C, with care taken to eliminate eventual air bubbles. After each part was introduced, the scale reading was registered after stabilization (20 s). This procedure was repeated until the last part was added. The values measured by the scale were related to the summation of the part volumes, and a linear least squares fit was applied to these data to determine the calibration equation. Thirteen of these calibrations were carried out over 5 mo.
Because calibration must be determined with the test tube wetted, we suggest that metallic parts be inserted together in the system before calibration begins.
AMV Tests With Weighed Water and Inanimate Objects
The AMV and its calibration were tested for agreement, linearity, and accuracy. Inasmuch as there are no animals with accurately known volume, standard volumes of water measured by weighing (50.00 ± 0.05 ml) were consecutively introduced into vessel C. The values obtained on the scale were registered, and the calibration equation (obtained with the aluminum cylinders) was applied to evaluate the measured volumes of water.
We also evaluated the ability of the AMV to determine with accuracy and reproducibility the known volume of inanimate objects. We used the 13 calibration lines to determine the volumes of 6 aluminum cylinders throughout 5 mo.
Transversal Evaluation of the Volume of Anesthetized and Nonanesthetized Animals
The experimental protocols using animals were previously approved by the Animal Experimentation Ethics Committee (638-1). To test the reproducibility of volume measurement in live animals, five male Wistar rats of different sizes and ages (Table 2) were used in a transversal study. Each animal was introduced six times into the system, either anesthetized or nonanesthetized. The experiments were carried out on different days, always at the same time. One rat was used per day, and the system was previously calibrated with the aluminum cylinders.
A nonanesthetized rat was first introduced into vessel C. Measurements were taken only after the animal was calm and acclimated to the water (35 ± 1°C) and container conditions. The rat was kept almost submerged by a screen similar to that used for feces retention, only its muzzle was not under water to allow breathing. Three volume determinations were measured each time an animal was introduced into the system. Each volume determination resulted from the mean of 20 measurements taken over 30 s and calculated automatically by the scale software. Because rats were introduced into the system 6 times, 18 measurements were taken for each animal. A total of 90 volume measurements were made. Each apparent weight measurement lasted 3 min. After each weighing session, the animal was dried with a towel and dryer. Before another trial, the volume of water trapped in the fur of the rats and removed from the system was replaced. The scale was then tared.
The same procedure was repeated when the animals were anesthetized with thiopental sodium (40 mg/kg body wt ip). Each rat was introduced into the system by a stick. Dental floss was tied in the incisive teeth of the rat to keep its muzzle above water and, thus, ensure that the animal was breathing while it was weighed. The animals were agitated with a stick introduced into the water that passed between the walls of the vessel and the rat, so that the air bubbles trapped in the fur were released.
To evaluate the effect of air bubbles trapped in the fur of the anesthetized animals on the volume, complementary volume measurements were taken from rat 1. Six weight sessions were conducted: in each session, three measures were taken before and three after the rat was agitated with a stick, resulting in 18 weight measures in each condition. The volume of the animal was obtained in all the cases, with the value read on the scale applied to the calibration equation obtained in the same experimental trial.
Statistical Analysis
For comparison of the measured volumes with the standard volumes, two types of analysis were performed. A linear regression between the two variables related the line of identity to the test of the agreement of the results. To show the bias across the range of measured inanimate volumes, a Bland-Altman (2) analysis, which also permits the evaluation of the precision and accuracy of the methodology, was performed. A parametric ANOVA for independent groups was used to evaluate rat 1 anesthetized or nonanesthetized, with or without agitation. The significance level was set at P < 0.05.
|
RESULTS
|
|---|
Behavior of AMV Measuring Previous Weighted Water and Inanimate Objects
A typical linear regression between the volumes of the aluminum cylinders and the scale displays resulted in a calibration line with slope of 6.87 ± 0.04 ml/g and an intercept of 1.70 ± 2.1 ml [R2 = 0.99998, standard error of the estimate (SEE) = 0.22 ml]. This calibration was performed 13 times over 5 mo, and a specific equation for adjustment was calculated for each calibration.
The comparison between the water volume measurements by AMV (using a previous aluminum-cylinders calibration) and by weighing shows linearity and agreement. The linear regression indicates a slope of 1.0006 ± 0.0017 and an intercept of 0.75 ± 0.81 (R2 = 0.99999, SEE = 0.58 ml). The difference between these variables as a function of the volume is shown in the Bland-Altman plot (Fig. 2). The differences between the volumes show a mean of 0.996 ml and a standard deviation of 0.584 ml with a bias of 1 ml. In addition to the line for mean difference, Fig. 2 includes the experimentally observed ±2SD limit of the differences between volumes. Expressed as percentage of the measured volume, the values for the mean ±2SD are 0.262 ± 0.266%.
View larger version (9K):
[in this window]
[in a new window]
|
Fig. 2. Difference between water volume measured by the AMV and weighed water volume (standard) compared with the mean for both measurements. Horizontal lines, mean difference ± 2SD.
|
|
Table 3 shows the actual volumes of six cylinders and the mean ± 2SD of the 13 volumes of each respective cylinder measured in the AMV throughout 5 mo. The differences between the actual volume and the volume measured by AMV is <0.4 ml. The mean volume errors varied from 0.01 to 0.07%, except for the first cylinder (0.36%), showing good accuracy of the system. The reliability of the measurements was also good. The coefficient of variation (2SD AMV volume ÷ actual volume) was 0.16 ± 0.14% (SD).
Behavior of AMV Measuring Live Models
Transversal evaluation of body volume of anesthetized and nonanesthetized animals.
Table 4 shows the volume and density of the rats, either anesthetized or nonanesthetized. The volume variation of rat 1 in the three experimental conditions (18 measurements each) is shown in Fig. 3. The mean ± 2SD volume differences between anesthetized rat 1, which was not agitated for air bubble removal, and nonanesthetized rat 1 was 5.54 ± 3.92 ml, and that between agitated anesthetized rat 1 and nonanesthetized rat 1 was 0.07 ± 3.06 ml. Consequently, the apparent density of the anesthetized nonagitated rat 1 (1.008 ± 0.005 g/ml) was significantly lower than that of the agitated rat (1.028 ± 0.004 g/ml). In addition, the volume variability in anesthetized rat 1 was higher before (±3 ml) than after (±2 ml) agitation.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 3. Volumes calculated from a previous calibration of rat 1. A: nonanesthetized (solid bars) and anesthetized (shaded bars) rat after agitation. B: anesthetized rat before agitation (open bars).
|
|
The differences between the 90 anesthetized and nonanesthetized volumes of the five rats evaluated in the transversal study are shown in Fig. 4. A nearly normal distribution is observed, and the mean ± SD was 0.31 ± 2.34 ml.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 4. Volume differences between anesthetized and nonanesthetized rats (n = 5): 0.31 ± 2.34 (SD) ml (n = 90 measurements).
|
|
The coefficient of variation, expressed as a percentage of the volumes of the anesthetized and nonanesthetized rats was 0.45 ± 0.13% and 0.38 ± 0.04% (mean ± SD), respectively. The percent coefficient of variation and the small standard deviations (Table 4) indicate a good reliability over repeated measurements of the anesthetized and nonanesthetized rats.
|
DISCUSSION
|
|---|
In the present study, we propose a new apparatus for measuring Vrat that does not require animal death, anesthesia, or fur shaving. This method introduces important advantages relative to current alternative methods. The reproducibility, accuracy, and sensitivity of this method are shown in this study. Because this apparatus is easy to construct, it can be used for several scientific purposes.
The movements of the nonanesthetized animal have been reported to impair the determination of body volume by hydrostatic weighing (30). However, the results obtained by the AMV do not support this fact. The density values of the rats, anesthetized or nonanesthetized, were similar to those in other animals (Table 5), which is consistent with the fact that an animal does not need to be anesthetized to be measured by AMV, since its movements do not interfere with the process. In contrast, animal movements helped release air bubbles from the fur. Moreover, irrespective of the volume of the animals (289.9 ± 1.0 and 410.4 ± 1.5 ml for small and large animals, respectively), the standard deviation was low, which confirms the repeatability of AMV measurements. With respect to live animals, considering the standard deviation of ±2.34 ml (Fig. 4), we suggest measuring each animal twice in the AMV and using the mean value if the two measurements agree within 2 ml. If they disagree by >2 ml, then a third measurement must be made.
It seems that the volume of nonanesthetized live animals has been measured solely by a model proposed by Scheer et al. (25). Their model was similar to the AMV; i.e., it measured water displacement after introduction of a nonanesthetized live animal. In this study (25), the densities of the animals were obtained from the ratio of body weight to the weight of the water volume displaced by the animal through three transfers of water. However, the authors did not comment about residual water in the containers, which could lead to measurement error. In addition, the animal in their system was completely submerged and isolated during volume measurement (complete filling of the container, valve closure, opening of the emergency exit, and emptying of water content). This study (25) did not report calibration efforts when determining accuracy and/or care in handling air bubble interference; its reproducibility (1%) was mentioned without further specifications. We have shown that the AMV has some advantages over this system. Our data (Table 3) show that the maximal relative error in the AMV was 0.4% when 100- to 700-ml volumes were measured.
The densities of the five rats presented in this work (Table 4) are lower than the minimum values found by others in Sprague-Dawley and Wistar rats (Table 5). This could be attributed to the fact that the present work, as well as work by Scheer et al. (25), calculates the density of conscious living animals, where imprisoned air in the lungs, superior aerial ways, and the gastrointestinal tract can interfere with density, presenting lower values than the carcass of dead shaved animals (6). Thus it is not possible to use the same relation that exists for dead animals to indirectly define the percentage of body fat of conscious living animals through the density. To define the equation that relates the density of living animals and percentage of body fat was not the objective of this work.
In conclusion, our data show that the AMV is reliable, accurate, reproducible, and sensitive. The principle for volume measurement can be applied to any animal as long as the system has physical dimensions. AMV is inexpensive and can be used with live animals, allowing for the evaluation of individual body volumes over time in longitudinal experiments that may last many months.
|
GRANTS
|
|---|
This study was supported by the Foundation for Research Support of the State of Sao Paulo and the Technological and Scientific Development National Council (03/09923-2P and 523383-96-7). R. Hohl received grants from the Coordination for the Improvement of Higher Education Personnel, and R. Buscariolli de Oliveira received grants from the Technological and Scientific Initiation Program-Technological and Scientific Development National Council.
|
FOOTNOTES
|
|---|
Address for reprint requests and other correspondence: R. Brenzikofer, LIB, UNICAMP, Cidade Universitária Zeferino Vaz Cx. Postal 6134, Campinas, SP, Brasil (e-mail: rene{at}fef.unicamp.br)
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES
|
|---|
- Behnke AR, Feen BG, Welham WC. The specific gravity of healthy men. JAMA 118: 495–498, 1942.
- Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1: 307–310, 1986.[CrossRef][Web of Science][Medline]
- Bohnenkamp H, Schmah J. Untersuchungen zu den Grundlagen des Energie und Stoffwechsels. IV. Mitteilung Das Reinvolumem sowie die spezifische Dichte des Menschen und die Bestimmungsweise dieser Grossen. Arch Ges Physiol 228: 100–124, 1931.
- Braga LR, Mello MAR, Gobatto CA. Exercício contínuo e intermitente: efeitos do treinamento e do destreinamento sobre a gordura corporal de ratos obesos. Arch Latinoam Nutr 54: 58–65, 2004.[Web of Science][Medline]
- Dawson CA, Horvath SM. Swimming in small laboratory animals. Med Sci Sports 2: 51–78, 1970.[Medline]
- Dahms WT, Glass AR. Correlation of percent body fat with body specific gravity in rats. J Nutr 112: 398–400, 1982.[Abstract/Free Full Text]
- Dempster P, Aitkens S. A new air displacement method for the determination of human body composition. Med Sci Sports Exerc 27: 1692–1697, 1995.
- Fields AD, Goran MI, McCrory MA. Body-composition assessment via air-displacement plethysmography in adults and children. Am J Clin Nutr 75: 453–467, 2002.[Abstract/Free Full Text]
- Gnaedinger RH. Estimation of Density of Live Pigs by Air Displacement and Helium Dilution Techniques (Ph.D. thesis). East Lansing: Michigan State Univ., 1962.
- Gobatto CA, Mello MAR, Sibuya CY, Azevedo JRM, Santos LA, Kokubum E. Maximal lactate steady state in rats submitted to swimming exercise. Comp Biochem Physiol A 130: 21–27, 2001.[CrossRef][Medline]
- Hardin D. Reliability of selected swimming endurance tests for laboratory rats. Res Q 39: 405–407, 1968.[Web of Science][Medline]
- Hix VM, Pearson AM, Reineke EP. Specific gravity of human subjects by air displacement and helium dilution. J Appl Physiol 19: 955–958, 1964.[Abstract/Free Full Text]
- Jongbloed J, Noyons AKM. Die Bestimmung des wahren Volumens und des Spezifischen. Arch Ges Physiol 240: 197–201, 1938.
- Kodama AM. In vivo and in vitro determinations of body fat and body water in the hamster. J Appl Physiol 31: 218–222, 1971.[Free Full Text]
- Kodama AM, Pace N. A simple decompression method for in vivo body fat in small animals. J Appl Physiol 18: 1272–1276, 1963.[Free Full Text]
- Kraybill HF, Bitter HL, Hankins OG. Body composition of cattle. II. Determination of fat and water content from measurement of body specific gravity. J Appl Physiol 4: 575–583, 1952.[Free Full Text]
- McArdle WD, Montoye HJ. Reliability of exhaustive swimming in the laboratory rat. J Appl Physiol 21: 1431–1434, 1966.[Free Full Text]
- McCrory MA, Gomez TD, Bernauer EM, Mole PA. Evaluation of a new air displacement plethysmograph for measuring human body composition. Med Sci Sports Exerc 27: 1686–1691, 1995.
- Myhre LG, Kessler WV. Body density and potassium 40 measurements of body composition as related to age. J Appl Physiol 21: 1251–1255, 1966.[Free Full Text]
- Naito HK, Griffith DR. Interaction of thyroxine, energy metabolism and swimming performance of rats. Life Sci 20: 1121–1130, 1977.[CrossRef][Web of Science][Medline]
- Raja G, Brau L, Palmer TN, Fournier PA. Repeated bouts of high-intensity exercise and muscle glycogen sparing in the rat. J Exp Biol 206: 2159–2166, 2003.[Abstract/Free Full Text]
- Rathbun EN, Pace N. The body water and chemically combined nitrogen content in relation to fat content. J Biol Chem 158: 685–691, 1945.[Free Full Text]
- Rathbun EN, Pace N. The determination of total body fat by means of the body specific gravity. J Biol Chem 158: 667–676, 1945.[Free Full Text]
- Rogers P, Webb GP. Estimation of body fat in normal and obese mice. Br J Nutr 43: 83–86, 1980.[CrossRef][Web of Science][Medline]
- Scheer BT, Dorst S, Codie JF, Soule DF. Physical capacity of rats in relation to energy and fat content of the diet. Am J Physiol 149: 194–203, 1947.[Free Full Text]
- Siri WE. Apparatus for measuring human body volume. Rev Sci Instrum 27: 729–738, 1955.
- Stern H. Investigations on corporeal specific gravity and on the value of this factor in physical diagnosis. Med Rec 59: 204–207, 1901.
- Stone MH, Rush ME, Lipner H. Responses to intensive training and methandrostenelone administration. II. Hormonal, organ weights, muscle weights and body composition. Pflügers Arch 375: 147–151, 1978.[CrossRef][Web of Science][Medline]
- Terada S, Yokozeki T, Kawanaka K, Ogawa K, Higuchi M, Ezaki O, Tabata I. Effects of high-intensity swimming training on Glut-4 and glucose transport activity in rat skeletal muscle. J Appl Physiol 90: 2019–2024, 2001.[Abstract/Free Full Text]
- Walser M, Stein SN. Determination of specific gravity of intact animals by helium displacement: comparison with water displacement. Proc Soc Exp Biol Med 82: 774–777, 1953.[CrossRef][Medline]
- Wedgewood RJ, Breckenridge JR, Newman RW. Measurement of body volume by air displacement (Abstract). Fed Proc 12: 151, 1953.
Copyright © 2007 by the American Physiological Society.
TOP
ABSTRACT
MATERIALS AND METHODS
