HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Koo, W. W. K. Articles by Ellis, K. J. Search for Related Content PubMed PubMed Citation Articles by Koo, W. W. K. Articles by Ellis, K. J.

Bone and Body Composition Measurements of Small Subjects: Discrepancies from Software for Fan-Beam Dual Energy X-Ray Absorptiometry

Carmen and Ann Adams Department of Pediatrics, Obstetrics and Gynecology, Wayne State University, Detroit, Michigan (W.W.K.K., M.H.)
Children’s Nutrition Research Center, Baylor College of Medicine, Houston, Texas (R.J.S., K.J.E.)

Address correspondence to: Dr. Winston Koo, Hutzel Hospital, Department of Pediatrics, 4707 St. Antoine Blvd, Detroit MI 48201. E-mail wkoo{at}wayne.edu

	ABSTRACT

TOP FOOTNOTES ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ADDENDUM ACKNOWLEDGMENTS REFERENCES

 
Objectives: A piglet model was used to determine the variations in measurements from different software algorithms used in the same type of dual energy X ray absorptiometry (DXA) instruments from the same manufacturer.

Methods: Forty-one piglets (6190 +/– 5856g, mean +/– SD) were scanned in duplicate with a fan-beam densitometer (Hologic QDR4500A, Hologic Inc, Bedford, MA) in the infant whole body scan mode. The same scans were analyzed with two software versions: vKH6 (validated with carcass chemical measurement) and v11.2 (commercial software from the same densitometer manufacturer).

Results: All analysis values were highly correlated (r = 0.90 to 1.00) and DXA values for total weights were almost identical. However, v11.2 results consistently overestimated bone mineral content (49.3 +/– 23.4%, mean +/– SD), bone area (21.1 +/– 8.2%), bone mineral density (24.1 +/– 22.2%), and fat mass (160.9 +/– 71.7%) but underestimated lean mass (–14.3 +/– 5.5%) when compared to the values from vKH6. Differences between software versions increased with heavier piglets.

Conclusion: The commercial software for fan-beam DXA measurement of piglets, matched for the size of human infants and young children, has major inaccuracies for bone mineral and body composition that become further exaggerated with increasing weight of the subject.

Key words: one, body composition, pig, infant, dual energy X-ray absorptiometry

Abbreviations: DXA = dual energy X-ray absorptiometry

	INTRODUCTION

TOP FOOTNOTES ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ADDENDUM ACKNOWLEDGMENTS REFERENCES

 
Piglets are frequently used as a research model to study infant nutrition and growth [1–4]. The development of dual energy X-ray absorptiometry (DXA) with its ability to provide simultaneous measurements of the bone, lean and fat mass components of body composition [3] has increased the ability to understand the interaction between nutrition and growth in research and clinical situations involving animals and humans.

With the older pencil beam DXA technique, different versions of software used for the study of infants and small animals of similar body mass provided different results [5,6]. The development and release of the newer fan beam DXA densitometer also has led to the availability of multiple software versions since the initial prototype (v8.26, Hologic Inc., Bedford, MA), but there are no data to document the validity of the recent software versions for the measurement of infants and small animals of similar body mass.

This study aims to determine the relationship between an independent software version that was validated with carcass chemical analysis with a recent commercial software released by the manufacturer of the same fan beam DXA densitometer. Comparisons were obtained for the measurements of bone mineral mass and soft tissue composition.

	METHODS

TOP FOOTNOTES ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ADDENDUM ACKNOWLEDGMENTS REFERENCES

 
Forty-one domestic swine piglets (J&M Farms, Lansing, MI) with a mean weight of 6190g (600g to 21100g) were scanned in duplicate with a fan-beam densitometer (Hologic QDR4500A, Hologic Inc., Bedford, MA) operated in the infant whole body scan mode. All animal scans and quality control scans were performed according to the manufacturer’s recommendations as described previously [4]. Each piglet scan was analyzed using software vKH6 validated by carcass chemical analysis [4], and the commercial software version v11.2, provided with the instrument. The animal study protocol was approved by the institutional review board for animal investigations at Wayne State University, Detroit, MI.

One investigator (MH) supervised all aspects of scan acquisition and analysis using vKH6, and determined that all scans were free from movement artifacts [7,8]. Analysis of the same scans using v11.2 was independently performed by one investigator (RS) without knowledge of the previous scan results. Of the 82 scans performed for this study, one scan was inadvertently deleted before analysis. For each DXA parameter (total weight, bone mineral content, bone area, bone mineral density, lean mass and fat mass), the values from duplicate total body scans were highly significantly correlated (r = 0.98 to 1.00, p < 0.001 for all comparisons) regardless of the software used for scan analysis. The precision of DXA measurements calculated with the method of Gluer et al [9] for total weight, bone mineral content, bone area, bone mineral density, lean mass and fat mass using the infant whole body vKH6 software were 0.2, 1.9, 1.4, 1.7, 1.0 and 13.1% respectively. The precision results for the corresponding parameters using the scanner’s infant whole body software v11.2 were 0.2, 2.3, 1.1, 1.5, 1.3 and 6.9% respectively.

Data analysis employed paired t test, and the limits of agreement between the commercial and validated softwares were determined with the Bland and Altman technique [10]. Except for the piglet with a single scan, all analyses were based on the average of the duplicate measurements.

	RESULTS

TOP FOOTNOTES ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ADDENDUM ACKNOWLEDGMENTS REFERENCES

 
All DXA-derived body composition (lean and fat mass), bone parameters (area, mineral content and density) and total weight from each of the two softwares were highly correlated (r = 0.90 for bone mineral density, r = 0.99 to 1.00 for all other parameters). However, there were significant differences in each of the body composition and bone parameters (Table 1). The results of the limits of agreement analysis were poor with the values for v11.2 software consistently overestimating bone mineral content (49.3 +/– 23.4%, mean +/– SD), bone area (21.1 +/– 8.2%), bone mineral density (24.1 +/– 22.2%), and fat mass (160.9 +/– 71.7%) but underestimating lean mass (–14.3 +/– 5.5%) compared to the values obtained from version vKH6 which had been previously validated with carcass chemical analysis [4]. The extent of over- or under-estimation increased with increasing weight of the piglets except for bone mineral density (Fig. 1). The over-estimation of bone mineral density using software v11.2 tends to decrease with increased weight of the piglet.

View larger version (30K): [in this window] [in a new window]   Fig. 1. Bland-Altman plots for the fan beam dual energy X-ray absorptiometry measured bone and body composition values for piglets using software version vKH6 and v11.2. X axis = average of the values for both software versions. Y axis = difference in the values between commercial and modified software. Solid line represents zero difference and dash lines represent +/– one and two standard deviation from zero difference.

	DISCUSSION

Similar to the observations with pencil beam DXA measurements in small subjects [5,6], this study demonstrated that the commercial software used with the newer generation fan-beam DXA technique to examine piglets of a size comparable with human infants and young children results in grossly discrepant bone mineral mass and body composition measurements. In an earlier study, we demonstrated that the prototype commercial software version v8.26 could not provide results comparable to the chemical carcass measurements [14]. Thus taken together with the findings in this study, inaccuracies continue to exist in commercial software versions for fan beam DXA measurements of small subjects, although the relative differences among the various upgrades from the prototype v8.26 to the recent v11.2 are not known. Since the physics of the DXA technique is not a priori based on knowing the species being examined, the concerns for bone and soft tissue results are applicable to both animals and humans. Furthermore, the DXA data in the present study were generated under optimal conditions with sedated piglets thereby eliminating any spurious data from movement artifacts [7,8].

Our data also demonstrated that inaccuracies of the v11.2 software version for the fan beam DXA tend to worsen with increasing body weight, thus raising concerns for longitudinal growth studies. The trend towards similar bone mineral density values at larger weights may indicate the improved ability to more accurately detect the bone edge and therefore delineation of bone area. Alternately, a difference in the threshold value for bone detection may result in less noticeable difference as bone mineralization increased in larger piglets. In any case, any improvement in bone mineral density measurements is of minor clinical importance since this measurement in growing subjects is of questionable value [15,16].

With the older but most widely used pencil beam DXA technique in the study of small subjects, the uncritical acceptance by some investigators of the earlier unvalidated versions of the infant software continues to contribute to the conflicting data in the literature [3,7,8]. Unfortunately, subsequent improvements in the software may not be re-applied to the earlier scans because of differences in scan acquisition technique and non-compatibility of software. For the newer fan-beam DXA technique, an examination of our data show that despite the multiple upgrades in software versions from the prototype v8.26 to the v11. 2, there has been no apparent improvement in the analysis algorithm for the infant mode. It is therefore critical that the validity of the various softwares used with the newer generation of fan-beam DXA instruments must be adequately tested before extensive publications of research and clinical data for bone or soft tissue body composition. Furthermore, if updated versions of the software are automatically loaded onto instruments during routine maintenance or repairs, then adequate documentation of the impact of the changes on bone or body composition must be included.

	CONCLUSION

TOP FOOTNOTES ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ADDENDUM ACKNOWLEDGMENTS REFERENCES

 
The commercial infant software provided by the manufacturer for the fan-beam DXA measurement of bone mineral mass and body composition has major deficiencies when used to examine piglets of a size comparable with human infants and young children. It is recommended that investigators avoid uncritical acceptance of results obtained using commercial software and to remain attentive to any changes in scan acquisition and analysis that might affect the validity of the DXA measurements. Manufacturers need to provide validation of their software before commercial release, and to clearly identify any major changes in updated version of DXA software that could substantially alter bone or body composition values.

	ADDENDUM

TOP FOOTNOTES ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ADDENDUM ACKNOWLEDGMENTS REFERENCES

 
Another software version v12.1 is now available from the manufacturer. Analysis based on the same piglets showed the limits of agreement for DXA measurements between v11.2 and v12.1 with one standard deviation of difference as a percentage of mean (10) for bone mineral content, bone mineral density, bone area, fat, lean, fat percent are 33.7, 22.4, 28.7, 80.5, 12.9 and 116.9% respectively, although the total weight obtained from both versions remained almost identical.

	ACKNOWLEDGMENTS

TOP FOOTNOTES ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ADDENDUM ACKNOWLEDGMENTS REFERENCES

 
Supported in part by a USDA/ARS cooperative agreement (#58-6250-6-001) with Baylor College of Medicine.

	FOOTNOTES

TOP FOOTNOTES ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ADDENDUM ACKNOWLEDGMENTS REFERENCES

 
There is no financial or contractual conflict of interest associated with this submission.

Received April 14, 2004. Accepted June 11, 2004.

	REFERENCES

TOP FOOTNOTES ABSTRACT INTRODUCTION METHODS RESULTS CONCLUSION ADDENDUM ACKNOWLEDGMENTS REFERENCES

 

1. Fiorotto ML, Shulman RJ, Sheng HP, Garza C: The effects of different total parenteral nutrition fuel mixes on skeletal muscle composition of infant miniature pigs. Metabolism35 :354 –359,1986 .[Medline]
2. Morgan 3rd W, Yardley J, Luk G, Niemiec P, Dudgeon D: Total parenteral nutrition and intestinal development: a neonatal model. J Pediatr Surg22 :541 –545,1987 .[Medline]
3. Koo WWK: Body composition measurements during infancy. Ann N Y Acad Sci904 :383 –92,2000 .[Medline]
4. Chauhan S, Koo WWK, Hammami M: Fan beam dual energy X-ray absorptiometry body composition measurements in piglets. J Amer Coll Nutr22 :408 –4014,2003 .[Abstract/Free Full Text]
5. Picaud JC, Nyamugabo K, Braillon P, Lapillonne A, Claris O, Delmas P, Meunier P, Salle B, Rigo J: Dual-energy x-ray absorptiometry in small subjects: influence of dual-energy x-ray equipment on assessment of mineralization and body composition in newborn piglets. Pediatr Res46 :772 –777,1999 .[Medline]
6. Picaud JC, Lapillonne A, Pieltain C, Reygrobellet B, Claris O, Salle BL, Rigo J. Software and scan acquisition technique-related discrepancies in bone mineral assessment using dual-energy X-ray absorptiometry in neonates. Acta Paediatr91 :1189 –1193,2002[Medline]
7. Koo WWK, Walters J, Bush AJ: Technical considerations of dual energy x-ray absorptiometry-based bone mineral measurements for pediatric studies. J Bone Mineral Res10 :1998 –2004,1995 .[Medline]
8. Koo WWK, Hockman EM, Hammami M. Dual energy X ray absorptiometry measurements in small subjects: conditions affecting clinical measurements. J Am Coll Nutr23 :212 –219,2004 .[Abstract/Free Full Text]
9. Gluer CC, Blake G, Lu Y, Blunt BA, Jergas M, Genant HK: Accurate assessment of precision errors: how to measure the reproducibility of bone densitometry techniques. Osteoporos Int5 :262 –2670,1995 .[Medline]
10. Bland JM, Altman DG: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet1 :307 –310,1986 .[Medline]
11. Tothill P, Avenell A, Love J, Reid DM: Comparisons between Hologic, Lunar and Norland dual-energy X-ray absorptiometers and other techniques used for whole-body soft tissue measurements. Eur J Clin Nutr48 :781 –794,1994 .[Medline]
12. Tothill P, Avenell A, Reid DM: Precision and accuracy of measurements of whole-body bone mineral: comparisons between Hologic, Lunar and Norland dual-energy X-ray absorptiometers. Br J Radiol67 :1210 –1217,1994 .[Abstract]
13. Albanese CV, Diessal E, Genant HK: Clinical applications of body composition measurements using DXA. J Clin Densit6 :75 –85,2003 .
14. Koo WWK, Hammami M, Hockman EM: Use of fan beam dual energy X ray absorptiometry to measure body composition in piglets. J Nutr132 :1380 –1383,2002 .[Abstract/Free Full Text]
15. Nelson DA, Koo WW: Interpretation of absorptiometric bone mass measurements in the growing skeleton: issues and limitations. Calc Tiss Int65 :1 –3,1999 .[Medline]
16. Rauch F, Schoenau E: Changes in bone density during childhood and adolescence: an approach based on bone’s biological organization. J Bone Miner Res16 :597 –604,2001 .[Medline]

This article has been cited by other articles:

				K. J Ellis, M. Yao, R. J Shypailo, A. Urlando, W. W Wong, and W. C Heird Body-composition assessment in infancy: air-displacement plethysmography compared with a reference 4-compartment model Am. J. Clinical Nutrition, January 1, 2007; 85(1): 90 - 95. [Abstract] [Full Text] [PDF]

				K. J. Loud and C. M. Gordon Adolescent bone health. Arch Pediatr Adolesc Med, October 1, 2006; 160(10): 1026 - 1032. [Abstract] [Full Text] [PDF]

				J. E Williams, J. C. Wells, C. M Wilson, D. Haroun, A. Lucas, and M. S Fewtrell Evaluation of Lunar Prodigy dual-energy X-ray absorptiometry for assessing body composition in healthy persons and patients by comparison with the criterion 4-component model Am. J. Clinical Nutrition, May 1, 2006; 83(5): 1047 - 1054. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Koo, W. W. K. Articles by Ellis, K. J. Search for Related Content PubMed PubMed Citation Articles by Koo, W. W. K. Articles by Ellis, K. J.