JBJS welcomes reader comments on published articles. Letters to the Editor are reviewed by JBJS editors but are not peer-reviewed. To submit your letter, please follow the "submit a response" link that appears in the content box at the upper right of the full text of the article.

Letters to the Editor to:

Scientific Articles:
Alain Diméglio, Yann Philippe Charles, Jean-Pierre Daures, Vincenzo de Rosa, and Boniface Kaboré
Accuracy of the Sauvegrain Method in Determining Skeletal Age During Puberty
J Bone Joint Surg Am 2005; 87: 1689-1696 [Abstract] [Full text] [PDF]
*Letters to the Editor: Submit a response to this article

Electronic letters published:

[Read Letter to the Editor] Dr. Dimeglio and colleagues respond to Dr. Kosuwon
Alain Dimeglio, M.D., Yann Phillippe Charles, M.D., Jean-Pierre Daures, Ph.D., M.D.   (5 October 2005)
[Read Letter to the Editor] Determining Skeletal Age During Puberty
Weerachai Kosuwon, M.D.   (26 September 2005)

Dr. Dimeglio and colleagues respond to Dr. Kosuwon 5 October 2005
Previous Letter to the Editor Top
Alain Dimeglio, M.D.
Dept. of Pediatric Orthopaedics, University of Montpellier, FRANCE,
Yann Phillippe Charles, M.D., Jean-Pierre Daures, Ph.D., M.D.

Send letter to journal:
Re: Dr. Dimeglio and colleagues respond to Dr. Kosuwon

a-dimeglio{at}chu-montpellier.fr Alain Dimeglio, M.D., et al.

We thank Dr. Kosowon for his letter.

Average values and standard deviations of scores as well as skeletal ages determined by the method of Sauvegrain, et al, and by the Greulich and Pyle atlas are summarized in tables below for each evaluator, girls and boys, first and second review:

Girls
Review 1

Score Sauvegrain

Age Sauvegrain

Age Greulich and Pyle

Mean

Std. dev.

Mean

Std. dev.

Mean

Std. dev.

Observer 1

21.90

4.08

11.47

0.79

11.53

1.28

Observer 2

22.17

4.12

11.51

0.79

11.83

0.87

Observer 3

22.49

3.38

11.44

0.65

11.91

0.89

Girls
Review 2

Score Sauvegrain

Age Sauvegrain

Age Greulich and Pyle

Mean

Std. dev.

Mean

Std. dev.

Mean

Std. dev.

Observer 1

21.79

4.10

11.44

0.83

11.73

1.15

Observer 2

21.99

4.22

11.51

0.83

11.82

0.91

Observer 3

22.28

3.39

11.43

0.66

11.77

1.01

Boys
Review 1

Score Sauvegrain

Age Sauvegrain

Age Greulich and Pyle

Mean

Std. dev.

Mean

Std. dev.

Mean

Std. dev.

Observer 1

20.69

3.58

13.68

0.73

13.53

0.80

Observer 2

20.92

3.42

13.77

0.74

13.61

0.71

Observer 3

21.40

3.46

13.90

0.82

13.69

0.93

Boys
Review 2

Score Sauvegrain

Age Sauvegrain

Age Greulich and Pyle

Mean

Std. dev.

Mean

Std. dev.

Mean

Std. dev.

Observer 1

20.50

3.80

13.64

0.86

13.47

0.83

Observer 2

21.00

3.56

13.77

0.79

13.72

0.77

Observer 3

21.12

3.64

13.81

0.80

13.49

0.85

The Bland and Altman method is a graphical method using means and standard deviations of two compared values. The data of our study has been reevaluated by Prof. J.-P. Daures at the Department of Clinical Research, University of Montpellier, France. Overall this data confirmed the results assessed by the use of intraclass correlation coefficients. Because of the huge amount of data, we would not recommend to publish all of it. Nevertheless, we can send you some of these results if you specify which part of the evaluation in our study represents a major interest for the use of the Bland and Altman method.

Alain Diméglio, M.D.

Yann Philippe Charles, M.D.

Jean-Pierre Daures, M.D.

Department of Pediatric Orthopaedics

University of Montpellier, France
Determining Skeletal Age During Puberty 26 September 2005
Next Letter to the Editor Top
Weerachai Kosuwon, M.D.,
Government service
Orthopedic Dept., Faculty of Medicine, Khon Kaen University, Khon Kaen, THAILAND

Send letter to journal:
Re: Determining Skeletal Age During Puberty

weera_ko{at}kku.ac.th Weerachai Kosuwon, M.D.

To The Editor:

I read with interest "Accuracy of The Sauvegrain Method in Determining Skeletal Age During Puberty" (87:1689-1692,2005),by Dimeglio, et al. This well conducted study aimed to evaluate the reliability of the scoring systems for assessment of skeletal age from radiographs of the elbow. The authors analyzed the data by using intraclass correlation coefficients. However, they did not provide the scores of each evaluator which are crucial for verification of the variability of the two methods of skeletal age assessment.

Moreover, the intraclass correlation coefficient is often used to quantify the strength of relationship between two varibles. I think the Bland and Altman method(1) is more appropriate to assess the accuracy or repeatability of duplicate observations in this article. I would suggest that the authors recalculate the agreement of the three observers again by using the Bland and Altman method(2).

Weerachai Kosuwon, M.D.

MSc (Clinical Epidemiology)

References:

1. Stanton A. Glantz, Primer of Biostatistics 5th edition. Chicago: McGrawHill;2002, p230-283.

2. Bland MJ, Altman DG, Statistical Methods for Assessing Agreement Between Two Measures of Clinical Measurement. Lancet1(8476):307-310,1986.