| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Obesity Research Center, St. Luke's-Roosevelt Hospital, New York 10025; and 2 Department of Radiology, College of Physicians and Surgeons, Columbia University, New York, New York 10032
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Magnetic resonance imaging (MRI) has the ability to discriminate between various soft tissues in vivo. Whole body, specific organ, total adipose tissue (TAT), intra-abdominal adipose tissue (IAAT), and skeletal muscle (SM) weights determined by MRI were compared with weights determined by dissection and chemical analysis in two studies with male Sprague-Dawley rats. A 4.2-T MRI machine acquired high-resolution, in vivo, longitudinal whole body images of rats as they developed obesity or aged. Weights of the whole body and specific tissues were determined using computer image analysis software, including semiautomatic segmentation algorithms for volume calculations. High correlations were found for body weight (r = 0.98), TAT (r = 0.99), and IAAT (r = 0.98) between MRI and dissection and chemical analyses. MRI estimated the weight of the brain, kidneys, and spleen with high accuracy (r > 0.9), but overestimated IAAT, SM, and liver volumes. No differences were detected in organ weights using MRI and dissection measurements. Longitudinal MRI measurements made during the development of obesity and aging accurately represented changes in organ and tissue mass.
longitudinal body composition; in vivo; obesity; high-fat diet; small animal
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
THE ACCURATE MEASUREMENT OF body compartments is important in the study of obesity and other diseases. There are many techniques available for body composition measurement, including hydrodensitometry, anthropometry, computerized tomography (CT), and dual-energy X-ray absorptiometry (14). Magnetic resonance imaging (MRI) and CT have been applied in the measurement of total and regional human and animal body composition (10, 22, 25). However, limitations associated with ionizing radiation in CT have made MRI a more desirable method for human body composition assessment, since it provides a safe, noninvasive, in vivo measurement approach (3).
To evaluate the accuracy and validate the viability of MRI as a means of measuring body composition in rodents, Ross et al. (22) tested a model using MRI to measure adipose tissue (AT) and skeletal muscle (SM) volumes in rats. MRI-derived AT volumes were compared with those obtained by CT and chemical analysis in rats. The results (22) showed highly significant correlation coefficients among these methods without systematic differences, thereby demonstrating that MRI and CT assess AT volume with a comparable level of accuracy compared with chemical analysis in rats. Fowler et al. (10) reported that MRI accurately measured AT in lean and obese pigs in vivo, and Ishikawa and Koga (13) proposed an animal model of non-insulin-dependent diabetes mellitus for the accurate measurement of abdominal fat in Otsuka Long-Evans Tokushima fatty rats using MRI. Mystkowski et al. (19) evaluated proton magnetic resonance spectroscopy (MRS) for murine body composition. Compared with chemical carcass analysis, MRS measures of total body fat, total body water, fat free mass, and total lean mass were highly correlated with chemical determinations. Finally, two studies (5, 13) have demonstrated the ability of MRI to accurately quantify specific organs or tissues in rats.
The noninvasive nature of MRI also makes it an attractive method for longitudinal studies (11). However, there has been no previous report of MRI used for the longitudinal measurement of tissue and organ volumes in vivo in rats. The aim of the current study was to validate high-resolution MRI as a measurement tool for the study of body composition in vivo, in two rat experiments designed to assess the following conditions: normal growth, the development of obesity, and aging. These conditions were selected because they are known to result in changes in the proportion of organ and tissue volumes over time. Results from high-resolution MRI were compared with carcass analysis and dissection measures obtained from the same animals.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Experimental Design
Experiment 1. Eighteen 9-mo-old male Sprague-Dawley rats (Taconic Farms, Germantown, NY) were individually housed in hanging stainless steel cages and fed standard laboratory chow (Purina 5001, 3.54 kcal/g) ad libitum. The rats were maintained in a temperature-controlled room (22-24°C) with a 12:12-h light-dark cycle (0500-1700). The body weight and caloric intake of the rats were measured at 48-h intervals throughout the study. All rats were imaged at the start of the experiment. Subsequently, the following three groups of equal body weight were created: the baseline group (n = 5), in which rats were euthanized and dissected immediately after the initial imaging, and the carcasses were frozen for chemical analysis; the control group (n = 6), in which rats were maintained on laboratory chow; and the obese group (n = 7), in which rats were fed a diet with 60% of calories as fat (diet D12492, 5.24 kcal/g, Research Diets, New Brunswick, NJ). The control and obese groups were imaged at 4-wk intervals over the following 16 wk, after which the animals were euthanized and dissected, and the carcasses were frozen for chemical analysis
Experiment 2. Ten 18-mo-old male Sprague-Dawley rats (Taconic Farms) were housed as for experiment 1 and maintained on laboratory chow ad libitum. The body weight and caloric intake of the rats were measured at 48-h intervals throughout the study. All rats were imaged at the start of the experiment. Subsequently, the following two groups of equal body weight were created: the aging baseline group (n = 5), in which rats were euthanized and dissected immediately after the initial imaging, and the carcasses were frozen for chemical analysis; and the aging group (n = 5), in which rats were continued on laboratory chow ad libitum and imaged at 4-wk intervals over the following 16 wk. After final imaging, animals in the aging group were euthanized and dissected, and the carcasses were frozen for chemical analysis.
Animal Procedures
Rats were fasted overnight prior to imaging. For imaging, rats were injected with 50 mg/kg pentobarbital sodium ip and 20 min later placed in a cylindrical holder that permitted precise positioning of the animal in the MRI coil. For euthanasia, rats were anesthetized with CO2 and immediately decapitated. Carcass hair was removed, and individual organs were dissected, weighed to the nearest 0.01 g, and returned to the carcass. Omental and mesenteric fat was carefully dissected from the gastrointestinal tract, which was then emptied of its contents, washed, and returned to the carcass. The carcasses were stored at
20°C and thawed before chemical analysis.
Dissected organs included the brain, heart, lungs, liver, spleen,
kidneys, and testes. Dissected adipose depots included the epididymal,
retroperitoneal, visceral (omental + mesenteric), inguinal,
intrascapular, and forelimb axillary. The epididymal, retroperitoneal,
and visceral depots collectively were designated as intra-abdominal AT
(IAAT). The protocol was approved by the Institutional Animal Care and
Use Committees of St. Luke's-Roosevelt Hospital and Columbia University.
MRI Data Acquisition
A 4.2-T high-field whole body nuclear magnetic resonance system was used to acquire high-resolution three-dimensional images of the rats. The cylindrical holder containing the anesthetized rat was precisely placed within a 10-cm birdcage radio frequency coil and a high strength gradient coil (3 G/cm). A spin-echo sequence with a repetition time of 530 ms and echo time of 26 ms was used for all image acquisitions. The image matrix was 256 × 256 within a field of view of 102 × 102 mm, and 72 transverse slices were acquired every 4 mm over the entire body length, excluding the tail. The slice thickness was 3.7 mm with a 0.3-mm interslice gap.Volume and Weight Calculations
A single investigator, who was blinded to the experimental condition of the rat, analyzed images using a personal computer with self-developed computer image analysis software (24). Semiautomatic and interactive segmentation algorithms were employed to calculate the volumes of specific tissues of interest. To calculate and adjust for noise generated due to the nonuniform nature of MRI (4), a phantom tube (filled with water) was positioned along the side of the rat during image acquisition. Accordingly, the slice volume was verified. A mean error ranging from 0% to 2% was detected, possibly due to the nonuniformity of MRI and the interactive segmentation. Segmentation of the individual organs and tissues involved manual selection of the region of interest (ROI). The "threshold" determined from the gray-level histogram of ROI was used for segmentation. Figure 1 shows the segmentation method based on a threshold selection in an abdominal ROI.
|
Total body images were segmented into 10 regions, specifically subcutaneous AT (SAT), IAAT, SM, the brain, heart, liver, kidneys, spleen, testes, and bone. All non-IAAT was labeled as SAT, and total body AT (TAT) was expressed as the sum of the IAAT and SAT measures. Tissue volume was estimated from voxel number and size, where voxel size was calculated by multiplying the pixel size (from MRI parameter settings) by the slice thickness. Tissue weights were calculated using the respective densities of each organ or tissue (23). Additional semiautomatic segmentation methods, including two- and three-dimensional region growing and active contouring, were applied to connected and smoothed regions (24). To maximize speed and efficiency, when an ROI extended across more than one slice, a semiautomated multislice segmentation approach was employed. Rather than segmenting the same organ or tissue region in each individual slice, the same organ or tissue in contiguous slices was segmented and labeled simultaneously.
The proportion of AT as lipid was assumed to be 0.87 (1, 23). To compare chemical total body lipid (TBL) with MRI TAT, we converted MRI TAT to grams of lipid using the factor of 0.87. When this conversion has been used, MRI TAT is designated MRI TBL. This conversion was not made for comparisons of other AT depots or weights. The assumed density of AT was 0.92 g/cm3. The skin was not included in the measurement of SAT. Additional assumed densities were as follows: SM, 1.04 g/cm3; brain and heart, 1.03 g/cm3; liver, kidney, and spleen, 1.05 g/cm3; and bone, 2.2 g/cm3 (23).
MRI Precision and Reliability
The precision and reliability of MRI measurements were determined by calculating the mean coefficient of variation (CV) for repeated MRI scans and for repeated same-operator segmentation of body images, respectively. The precision of the MRI scanner was determined by scanning the same three rats three times each with repositioning between scans. The CV was calculated for total body weight (0.63%), TAT (2.15%), IAAT (2.96%), SM (0.27%), and individual organs (ranging from a low of 1.99% for the brain to a high of 9.59% for the heart, see Table 1). Intraoperator reliability, for which the same operator segmented three rats each three times (on different days), was also determined by calculating the CV for total body weight (0.27%), TAT (1.58%), IAAT (1.86%), SM (1.84%), and individual organs (ranged from a low of 0.88% for the brain to a high of 5.39% for the spleen, see Table 1).
|
Carcass Analysis
All carcasses were individually analyzed at the completion of the experiments. Carcasses were autoclaved for 24 h in large beakers covered with aluminum foil to prevent evaporation, and then homogenized by a large-bore Polytron (PT 6000; Brinkman Instruments, Westbury, NY). Water was added before autoclaving. The amount was ~1.5 times the weight of the animal. After weighing, homogenized aliquots were dried to a stable weight to determine water content. The carcass, water, and beaker were weighed together and then the individual weights of the carcass and beaker were subtracted to determine how much water had been added. Subsequent moisture determinations on aliquots of sample were then corrected for water added to determine moisture in the sample. Additional homogenized aliquots were used to determine body lipid and nitrogen content. TBL was measured by the method of Folch et al. (9). Carcass nitrogen content was determined by an adaptation of the Kjeldahl method (8). Total body protein was derived by multiplying nitrogen content by 6.25 (16).Statistical Analysis
Between group differences in baseline demographic and body composition measurements were tested using one-way ANOVA or independent t-tests as appropriate. Within group differences were compared using one-way ANOVA with repeated measures or paired t-tests. Comparisons of group means following significant main effects in ANOVA were made using the Student-Newman-Keuls post hoc test. Pearson correlation coefficients were used to quantify the univariate relationships between imaging-derived variables and corresponding values determined by chemical and dissection analyses. Agreement among techniques (MRI, chemical, and dissection) was tested with the method of Bland and Altman (2). Differences were considered statistically significant with P < 0.05. Data were analyzed using Microsoft Excel version 5.0, and group results are presented as means ± SD.| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Experiment 1: Obese and Control Groups
Chemical body composition and dissection measurements.
The body composition results from chemical carcass analysis for the
baseline, control, and obese groups are shown in Table 2. Weights for the whole carcass, lipid,
protein, and water are given as well as the percentage of the carcass
as lipid, protein, and water. There were no significant differences
between the baseline and control groups for any variables. The obese
group had higher values than the baseline and control groups for whole
carcass (P < 0.001) and lipid (P < 0.001) weights and percent lipid (P < 0.001) and lower
values for percent carcass protein (P < 0.001) and
water (P < 0.001). The whole body and organ weights
obtained at dissection for the three groups are shown in Table
3. The total organ weight is the sum of
the weights of the heart, lungs, liver, kidneys, spleen, and testes.
With two exceptions (lungs and gastrocnemius muscle), there were no
differences between baseline and control groups for any of the
variables. The obese group had significantly higher values than the
baseline (P < 0.001) and control (P < 0.05) groups for whole body weight and significant increases of heart,
lungs, and gastrocnemius muscle compared with the baseline group
(P < 0.05). Fat depot weights obtained at dissection are shown in Table 4. There were no
differences between the baseline and control groups for any of the
variables, whereas the obese group had significantly higher values for
all depots and IAAT (P < 0.001) compared with the
former groups.
|
|
|
Comparisons of body composition among MRI, chemical, and dissection
methods.
Mean whole body or carcass weights and organ and tissue weights for all
groups combined (N = 18), as measured by MRI, chemical, and dissection methods, are given in Table
5, with corresponding univariate
correlation coefficients. Body and carcass weights obtained by imaging
and chemical analyses significantly underestimated whole body weight
due to tissue volumes not included in the analyses (e.g., tail, hair,
blood, and intestinal contents). However, body weights determined by
all three methods were highly correlated (Table 5). Comparisons among
the methods using Bland and Altman's analysis (2) showed
no significant bias between methods. Figure 2 shows regression lines obtained by
comparing MRI weight with whole body weight and carcass weight. MRI
weight systematically underestimated whole body weight (Fig.
2A), but this was not the case when MRI weight was compared
with carcass weight (Fig. 2B). MRI IAAT was highly
correlated with dissection IAAT (Fig.
3A), and MRI TBL was highly
correlated with chemical TBL (Table 5). No systematic difference was
observed between MRI TBL and chemical TBL weight estimations (Fig.
3B). However, imaging overestimated IAAT (P < 0.001), brain (P < 0.001), kidney
(P < 0.05), spleen (P < 0.001), and
liver (P < 0.001) volumes (Table 5). Using the assumption that the protein content of SM is 17.1% (23),
SM mass in these rats represents 40% of whole body protein. A high correlation (r = 0.92) was found between MRI-derived
whole body SM mass and total body protein weight.
|
|
|
Longitudinal results determined by MRI.
Figure 4, A-C,
shows changes in body weight and body components for the groups in
experiment 1 as determined by MRI over the 16-wk
experimental period. Compared with the body weight of the control
group, which showed only a modest increase (of 40.6 ± 36.6 g, P < 0.05), the body weight of the obese group
increased dramatically over the 16-wk experimental period (increase of
207.4 ± 46.7 g, P < 0.001, Fig.
4A). Body components of the control group determined by MRI
(Fig. 4B) remained relatively constant over this period,
with only mild increases observed in SAT, total organ, and SM weights
(P < 0.05). In contrast, highly significant increases
in all MRI-determined body components, with the exception of total
organ weight (P < 0.001), were observed in the obese group (Fig. 4C). Thus incremental changes in specific body
tissues in rats developing obesity are clearly detectable by the MRI
technique.
|
Validity of MRI longitudinal results.
Figure 5A demonstrates that
only nonsignificant changes were seen in several dissection and
chemical body component measures between the baseline group, which was
measured at the start of the experiment, and the control group, which
was measured 16 wk later. These results are consistent with those of
Fig. 4B showing minor or no changes in body components as
determined by MRI for the control group, at intervals over the 16-wk
experimental period. Figure 5 compares differences determined by
chemical and dissection analysis (Fig. 5A) with differences
determined by MRI analysis (Fig. 5B) for the baseline vs.
obese groups at start and completion of the experimental
period. Dramatic changes in body components were seen in the
obese group, when body components were analyzed by either chemical and
dissection techniques or the MRI technique. Highly significant
increases of IAAT and TBL (TAT) were seen by these methods
(P < 0.001), and the MRI technique also detected an
increase in SM mass (P < 0.001, Fig.
5B). No changes in organ mass were detected by
either method. Although the weights of the body components are not
precisely comparable between the chemical and dissection and MRI
measures, the patterns of changes seen in these components are
virtually identical and are consistent with longitudinal changes in
body components detected at intervals during the experimental period by
MRI (Fig. 4C).
|
Experiment 2: Aging Groups
Chemical body composition and dissection measurements.
Body composition results from chemical analysis for aging baseline and
aging groups are shown in Table 6. At the
end of the 16-wk experimental period, the aging group had significantly
lower values for total carcass, lipid, protein, and water weights
(P < 0.01) than the aging baseline group, measured at
the start of the period. Although the weight of all carcass components
was markedly decreased, the largest relative decrease was in TBL, which
was reduced by ~50%. Whole body and organ weights and AT depot
weights determined at dissection are shown in Tables
7 and 8,
respectively. Whole body weight was significantly lower in the aging
vs. aging baseline group (P < 0.01). Significant reductions were observed in the weights of the testes and gastrocnemius muscle of the aging group (P < 0.05), with the weight
of the gastrocnemius muscle decreasing by 45%. Despite this large
decrease in representative SM weight in the aging group, there was no
difference in total organ weight between groups at the end of the 16-wk
experimental period (Table 7). However, heart weight increased
significantly in the aging group (P < 0.05),
reflecting cardiomyopathic enlargement of the heart muscle. Among the
adipose depots, reductions in the weight of all depots measured were
observed in the aging group, with the reductions being significant for
the intrascapular, retroperitoneal, and visceral depots
(P < 0.05). Total IAAT weight shrank to less than half
that of the aging baseline group (P < 0.05).
|
|
|
Longitudinal results determined by MRI. A significant decrease in body weight was detected by MRI over the 16-wk experimental period in the aging group (Fig. 4A). In addition, significant decreases paralleling this body weight loss in the aging group were detected in several body compartments by MRI (Fig. 4D), including TAT, IAAT, SAT, and SM (P < 0.05). Thus MRI measures appeared to be as sensitive to carcass composition changes over time in the aging group as respective measures made by chemical analysis and/or dissection (Tables 6-8).
Validity of MRI longitudinal results.
Figure 6 compares longitudinal changes
determined by chemical and dissection analysis with changes determined
by MRI analysis for the groups. Figure 6A demonstrates that
chemical analysis detected a significant decrease of body protein
content in the aging group after 16 wk of aging compared with protein
content of the aging baseline group (P < 0.05).
Significant decreases of TBL and IAAT in the aging group were also
revealed by chemical and dissection analysis (P < 0.05, Fig. 6A). Body component weights of the aging baseline
and aging groups over the 16-wk period as determined by MRI are seen in
Fig. 6B. MRI detected significant decreases of TAT, IAAT,
and SM in the aging group (P < 0.05). Again, no
changes in organ weight were detected by any method. Although the
weights of respective body components determined by the three
techniques are not precisely comparable, a virtually identical pattern
of body component changes is seen (Fig. 6) in the aging group over
time. These results are consistent with changes in body components
detected at intervals during the experimental period by MRI (Fig.
4D). Thus MRI measurements made over the course of normal
aging in the rat accurately represent relative changes in body organ
and tissue mass.
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Carcass composition analysis continues to be the reference method for the measurement of body composition in animals. Although accurate, this method is costly in terms of animal lives due to the need to euthanize animals periodically to assess longitudinal body composition changes. There is an increasing need to accurately monitor tissue and organ changes during aging and disease progression and in response to pharmacological and nutritional interventions. Imaging modalities are now being developed that allow for more sensitive assessment of tissues in vivo. The purpose of the current study was to test an in vivo technique for measuring longitudinal body composition changes using high-resolution MRI. The data herein report the validation of high-resolution MRI compared with chemical or dissection analysis for the measurement of whole body and regional soft tissues and selected organ masses in rats.
Treatment Effects on Body Composition
In experiment 1, a 60% high-fat diet induced a 45% increase in body weight (+240 g), of which 67% (162 g) was represented by body fat, as determined by carcass analysis. These body weight and fat increases are consistent with results obtained in similarly aged male rats fed a 60% high-fat diet for a comparable period (20). In contrast with the results of Levin and Dunn-Meynell (17), who observed that only half of a group of wild-type Sprague-Dawley rats responded to a high-energy diet with marked weight gain, we found that all members of the obese group in the present study gained large amounts of body weight over the 16-wk experimental period, compared with members of the obese control group. This difference may be due to different compositions of the high-energy diets used in the two studies or different ages of the rats at the start of high-energy diet feeding (growing vs. mature). In confirmation of the accuracy of the imaging technique, MRI detected an increase of 161.7 g in TAT in the obese group over the 16-wk treatment period (Fig. 4C). Similarly, dissection measures revealed an increase of 68.1 g in IAAT in the obese group, compared with the baseline group, whereas MRI detected an increase of approximately the same size (66.2 g) for IAAT in the obese group over the 16-wk treatment period. No change in overall organ mass was detected by MRI in the obese group (Fig. 5B), and the weights of the brain, kidneys, spleen, liver, and testes determined by dissection did not change significantly over time. These results are in good agreement with those of Pitts and Bull (21), who detected no changes of organ weights in similarly aged Sprague-Dawley rats made obese on a high-fat diet.Age-induced changes in body composition are well recognized. In experiment 2, the observed decrease in body weight in the aging group reflected a loss of both lean body and AT mass, as evidenced by significant reductions of total body protein and lipid (Table 6) and gastrocnemius muscle and AT depot weights (Tables 7-8). These changes were accurately tracked by MRI over time in the aging group (Fig. 4D). Significant reductions of AT observed in this study in the aging group are consistent with previous results (15) obtained in similarly aged Sprague-Dawley rats. However, significant reductions of lean body mass, also seen in the aging group in this study, were not observed in three previous studies (6, 7, 12) of age-related changes of body composition in similarly aged rats. This may be due to the fact that different strains of rat were used in each of the studies considered. Together, these data suggest that in rats, reductions of SM mass do not necessarily parallel losses of body fat with aging. It is interesting to note that with the exception of the testes, which decreased significantly in size, organ size as determined by dissection did not change or was increased (heart) in the aging group in the present study. This effect is clearly seen in the longitudinal measurements made by MRI (Fig. 4D). In this regard, our results are in general agreement with those of earlier studies (6-7), which reported no change or increases of organ size with aging in rats.
MRI Accuracy and Reproducibility
MRI-derived whole body weight agreed closely with chemical carcass weight (r = 0.98), with a regression slope = 1.01 (Fig. 2B), thereby suggesting that MRI is an accurate alternative to chemical analysis at the whole body level. A similar high correlation was found between MRI-derived whole body weight and whole body land weight (r = 0.995); however, MRI underestimated whole body weight by 6% (Fig. 2A). The latter may be explained by combined factors including nonuniformity of MRI, noninclusion of tail, intestinal contents, and hair in the image analysis procedure. Tail and hair weights combined were found to contribute 3% of whole body weight (Vasselli, unpublished data). No systematic difference was observed between MRI and chemical methods for TBL weight estimation. MRI overestimated the weights of the IAAT, kidneys, spleen, liver, and brain and underestimated heart weight. In explanation, organ volumes may be increased by 2-5% in blood perfused states, and motion artifact of the heart reduces the ability of MRI to clearly delineate edges. These effects are not present at the time of chemical and dissection analysis.Because of the technical difficulties associated with dissecting whole body SM mass, we were unable to determine how accurately MRI assesses muscle mass due to the lack of a reference standard. Using the assumptions that carcass nitrogen content is reflective of whole body protein (16) and that ~40% of whole body protein is found in SM mass (23), a high correlation (r = 0.92) was found between MRI-derived SM mass and nitrogen and protein content. It is noteworthy that in adults, MRI possesses the ability to measure individual muscle groups (18). Whether this is possible in small animals where the muscle groups are so much smaller was not addressed in this study.
Ross et al. (22), using a similar approach, reported that MRI (1.5 T) accurately assessed whole body TAT and visceral AT volumes compared with CT and chemical analysis in rats. Their (22) approach differed, however, from the current study in using an AT density factor of 0.90 g/cm3 and a data acquisition protocol with 3-mm slice thickness and 18-mm interslice gaps to produce a total of 12 slices for TAT estimation (22). To compare the two protocols, lipid values were calculated using slices chosen every 12, 20, and 28 mm from the current scans (every 4 mm). When applying the AT volume calculation equation developed by Ross et al. (22), to the current data, the regression results derived for MRI TBL compared with chemical carcass analysis suggest that the smaller the interslice gap, the closer the regression line is to the line of identity.
MRI Precision and Reliability
MRI-derived total body weight showed the highest precision. MRI estimated IAAT, TBL, SM, brain, kidneys, and liver with greatest precision since these tissues are displayed in good intensity contrast. Lower precision was found in the estimation of the spleen, due likely to operator difficulty in visually differentiating the spleen from other organs. MRI estimated the heart and lungs with the lowest precision due to the confounding effects of motion artifacts.Advantages and Limitations
The lack of radiation exposure associated with MRI provides a distinct advantage over CT, especially when repeated imaging is required. MRI allows for the acquisition of superb high-resolution anatomic images, including the reconstruction and visualization of three-dimensional images, with the appropriate software (24). The imaging protocol and software described herein could be applied to the study of body composition in many small and large animals, ranging in size from mice to bears. Low availability, high cost, and the need for trained MRI technicians make high-resolution MRI unaffordable and impractical for many researchers. The anesthetization of the animal to ensure nonmovement during image acquisition increases the risk of death due to likelihood of overdose.Conclusions
The use of the MRI analysis technique has considerable appeal as a method of body composition analysis, particularly when multiple in vivo measures are required. The procedure is highly accurate, nonradioactive, noninvasive, and minimally disruptive to animals in long research protocols. The present study demonstrates that the MRI technique is capable of tracking changes in body composition, in vivo, with precision and accuracy in both aging and developing obesity models. The technique is highly sensitive to incremental organ and tissue changes, and with the exception of the heart, estimates major organ weights precisely. This technique has significant application for studies of growth, aging, disease progression, and the effects of nutritional and pharmacological interventions.| |
ACKNOWLEDGEMENTS |
|---|
We thank L. Basilio for chemical analyses and K. Hess for anesthetizing the animals.
| |
FOOTNOTES |
|---|
This study was supported in part by National Institute on Aging Grant R29-AG14715 and by a Pilot and Feasibility award from the New York Obesity Research Center.
Address for reprint requests and other correspondence: D. Gallagher, Obesity Research Center, 1090 Amsterdam Ave., New York, New York 10025 (E-mail: dg108{at}columbia.edu).
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.
10.1152/ajpregu.00527.2001
Received 31 August 2001; accepted in final form 7 November 2001.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Behnke, AR.
Role of fat in gross body composition and configuration.
In: Fat as a Tissue, edited by Rodahl K.. New York: McGraw-Hill, 1964, p. 285-313.
2.
Bland JM and Altman DG. Statistical methods for assessing
agreement between two methods of clinical measurement.
Lancet: 307-310, 1986.
3.
Callaghan, PT.
Principles of Nuclear Magnetic Resonance Microscopy. Oxford: Clarendon, 1993.
4.
Clarke, LP,
and
Velthuizen KP.
MRI segmentation: methods and applications.
Magn Reson Imaging
13:
343-368,
1995[ISI][Medline].
5.
Cockman, MD,
Hayes DA,
and
Kuzmak BR.
Motion suppression improves quantification of rat liver volume in vivo by magnetic resonance imaging.
Magn Reson Med
30:
355-360,
1993[ISI][Medline].
6.
Delp, MD,
Evans MV,
and
Duan C.
Effects of aging on cardiac output, regional blood flow, and body composition in Fischer-344 rats.
J Appl Physiol
85:
1813-1822,
1998
7.
Even, PC,
Rolland V,
Roseau S,
Bouthegourd JC,
and
Tome D.
Prediction of basal metabolism from organ size in the rat: relationship to strain, feeding, age, and obesity.
Am J Physiol Regulatory Integrative Comp Physiol
280:
R1887-R1896,
2001
8.
Ferrari, A.
Nitrogen determination by a continuous digestion and analysis system.
Ann NY Acad Sci
87:
792-800,
1960.
9.
Folch, J,
Lees M,
and
Stanley GHS
A simple method for the isolation and purification of total lipids from animal tissues.
J Biol Chem
193:
497-509,
1951
10.
Fowler, PA,
Fuller MF,
Glasbey CA,
Cameron GG,
and
Foster MA.
Validation of the in-vivo measurement of adipose tissue by magnetic resonance imaging of lean and obese pigs.
Am J Clin Nutr
56:
7-13,
1992
11.
Fuller, MF,
Fowler PA,
Mcneil G,
and
Foster MA.
Body composition: the precision and accuracy of new methods and their suitability for longitudinal studies.
Proc Nutr Soc
49:
423-436,
1990[ISI][Medline].
12.
Gupta, G,
She L,
Ma XH,
Yang XM,
Hu M,
Cases JA,
Vuguin P,
Rossetti L,
and
Barzilai N.
Aging does not contribute to the decline in insulin action on storage of muscle glycogen in rats.
Am J Physiol Regulatory Integrative Comp Physiol
278:
R111-R117,
2000
13.
Ishikawa, M,
and
Koga K.
Measurement of abdominal fat by magnetic resonance imaging of OLETF rats, an animal model of NIDDM.
Magn Reson Imaging
16:
45-53,
1998[ISI][Medline].
14.
Jensen, MD.
Research techniques for body composition assessment.
J Am Diet Assoc
92:
454-460,
1992[ISI][Medline].
15.
Johnson, JA,
Yu CHY,
and
Pi-Sunyer FXP
Effect of age on protein conservation during very-low-energy diet in obese Spague-Dawley rats.
Obes Res
6:
448-457,
1998[ISI][Medline].
16.
Kleiber, M.
The Fire of Life. New York: Wiley, 1961.
17.
Levin, BE,
and
Dunn-Meynell AA.
Defense of body weight against chronic caloric restriction in obesity-prone and -resistant rats.
Am J Physiol Regulatory Integrative Comp Physiol
278:
R231-R237,
2000
18.
Mitsiopoulos, N,
Baumgartner R,
Heymsfield SB,
Lyons D,
Gallagher D,
and
Ross R.
Cadaver validation of skeletal muscle measurement by magnetic resonance imaging and computerized tomography.
J Appl Physiol
85:
115-122,
1998
19.
Mystkowski, P,
Shankland E,
Schreyer SA,
Boeuf RL,
Schwartz RS,
Cummings DE,
and
Kushmerick M.
Validation of whole-body magnetic resonance spectroscopy as a tool to assess murine body composition.
Int J Obes
24:
719-724,
2000.
20.
Oscai, LB,
Brown MM,
and
Miller WC.
Effect of dietary fat on food intake, growth and body composition in rats.
Growth
48:
415-424,
1984[ISI][Medline].
21.
Pitts, GC,
and
Bull LS.
Exercise, dietary obesity, and growth in the rat.
Am J Physiol Regulatory Integrative Comp Physiol
232:
R38-R44,
1977.
22.
Ross, R,
Leger L,
Guardo R,
Guise JD,
and
Pike BG.
Adipose tissue volume measured by magnetic resonance imaging and computerized tomography in rats.
J Appl Physiol
70:
2164-2172,
1991
23.
Snyder, WS,
Cook MJ,
Nasset ES,
Karhausen LR,
Howells GP,
and
Tipton IH.
Report of the Task Group on Reference Men. International Commission on Radiological Protection No. 23. Oxford: Pergamon, 1975.
24.
Tang, H,
Vasselli J,
Wu EX,
and
Gallagher D.
In-vivo determination of body composition in rats using high resolution magnetic resonance imaging.
Ann NY Acad Sci
904:
32-41,
2000
25.
Thomas, EL,
Saeed N,
Hajnal JV,
Brynes A,
Goldstone AP,
Frost G,
and
Bell JD.
Magnetic resonance imaging of total body fat.
J Appl Physiol
85:
1778-1785,
1998
This article has been cited by other articles:
|
|
 |
|
  Z. Chen, Z. Wang, T. Lohman, S. B. Heymsfield, E. Outwater, J. S. Nicholas, T. Bassford, A. LaCroix, D. Sherrill, M. Punyanitya, et al. Dual-Energy X-Ray Absorptiometry Is a Valid Tool for Assessing Skeletal Muscle Mass in Older Women J. Nutr., December 1, 2007; 137(12): 2775 - 2780. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Y. Ruan, D. Gallagher, T. Harris, J. Albu, S. Heymsfield, P. Kuznia, and S. Heshka Estimating whole body intermuscular adipose tissue from single cross-sectional magnetic resonance images J Appl Physiol, February 1, 2007; 102(2): 748 - 754. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Garcia, I. Wernstedt, A. Berndtsson, M. Enge, M. Bell, O. Hultgren, M. Horn, B. Ahren, S. Enerback, C. Ohlsson, et al. Mature-onset obesity in interleukin-1 receptor I knockout mice. Diabetes, May 1, 2006; 55(5): 1205 - 1213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Xie, C. Audette, M. Hoffee, J. M. Lambert, and W. A. Blattler Pharmacokinetics and Biodistribution of the Antitumor Immunoconjugate, Cantuzumab Mertansine (huC242-DM1), and Its Two Components in Mice J. Pharmacol. Exp. Ther., March 1, 2004; 308(3): 1073 - 1082. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. W. Cline, K. Johnson, W. Regittnig, P. Perret, E. Tozzo, L. Xiao, C. Damico, and G. I. Shulman Effects of a Novel Glycogen Synthase Kinase-3 Inhibitor on Insulin-Stimulated Glucose Metabolism in Zucker Diabetic Fatty (fa/fa) Rats Diabetes, October 1, 2002; 51(10): 2903 - 2910. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Bentley, M. C. Ortiz, E. L. Ritman, and J. C. Romero The use of microcomputed tomography to study microvasculature in small rodents Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2002; 282(5): R1267 - R1279. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
P < 0.01, 
P < 0.001, within-group differences between
measurements 1 and 4.
