Send letter to journal:
Re: Re: Comment on age determination in adolescent male football players: It does not work!

Yes, it does work Prof. Malina

We thank Prof. Malina for his interest in our paper. In most sports, the performance of adolescent athletes is determined by their physical maturity and thus related to age. In order to guarantee equal chances for different age groups, age-related tournaments for male and female players have been established in football. However, due to the fact that registration at birth is not compulsory in some African and Asian countries, other methods of age determination are needed to prevent participation in the incorrect age group.

The determination of skeletal maturity using standard radiographs of the left wrist is an established method in pediatrics, whatever grading method is used (Greulich-Pyle, Tanner- Whitehouse, Fels). However, in sports the use of X-rays to determine players of over age is not allowed because of the exposure to radiation following the guidelines of the IAEA (International Atomic Energy Agency) which regulates the use and possible abuse of X-rays under the title: The International Basic Safety Standards for Protection against Ionizing Radiation and the Safety of Radiation Sources. The objective and aims of our presented papers (Dvorak et al. 2007a,b) were to develop a grading system for epiphysial fusion based on MRI of the wrist, to evaluate the reliability and validity of the grading system in 14-19 year-old football players, to compare male football players from different ethnic groups, and finally to examine football players of FIFA and AFC U-17 competitions. In both papers we concluded that MRI of the wrist is a reliable, valid, and non-invasive method for age determination in the groups of 14-19 year-old male football players (to establish the normative values we used young football players with absolute certainty of birth certification).

Based on the results, it seems that U-17 players are more mature than football players of the same age from the normative group from Switzerland, Malaysia, Algeria, and Argentina. However, the lack of correlation between age category and degree of fusion in U-17 players supports the suspicion that the age stated in the official documents of the examined U-17 players might not be correct in all cases. Furthermore, a drop in the rate of completely fused players was observed between the first and second competition under investigation (FIFA: from 2003 to 2005, AFC: from 2004 to 2006). It can only be hypothesized that this decrease is due to the fact that the team managers of the national associations were aware the controls were going to be carried out and more careful selection was consequently carried out. The authors are of the opinion that the method proposed by using MRI is not only reliable from a scientific point of view but that it also works as an educational tool to combat cheating in sport.

Unfortunately, relying on the honesty of trainers and players as suggested by Professor Malina does not work. To the contrary, the desirable influence of the MR method on player selection has been demonstrated by our companion paper (Dvorak et al. 2007b).

J. Dvorak, J. George, A. Junge, J. Hodler

References

Dvorak J, George J, Junge A, Hodler J (2007a) Age Determination by Magnetic Resonance Imaging of Wrist in Adolescent Male Football Players. Br J Sports Med 41(1):45-52

Send letter to journal:
Re: Comment on age determination in adolescent male football players: It does not work!

13 March 2007 Comment on Age Determination in Adolescent Male Football Players: It Does Not Work!

Dear Editor

The recent paper of Dvorak and colleagues (2007) has as one of its objectives “…the possible use of magnetic resonance imaging (MRI), which has no radiation risk, in estimating the age of healthy adolescent football players” (p. 45). It also proposed to “…evaluate the reliability and validity of the grading system of MRI examinations of the wrist to determine skeletal bone age in the age group 14-19 years” (p. 46). The focus is exclusively on the radius which is the last bone of the hand-wrist complex to fuse in adolescence.

One of the rationales for the paper is the medicolegal issue of chronological age (CA) estimation for individuals who do not have birth certificates or for whom such a record is not available. In the context of football, verification of CA among over age players in international tournaments is an issue. “Bone age” or skeletal age (SA) has been proposed for use (Tritrakarn and Tansuphasiri, 1991) and has been used (www.CNNSI.com, 2001) as an estimate of CA in international football competitions (see Malina, 2005). The paper of Dvorak and colleagues falls in the same genre. The use of SA to estimate CA medically and legally has been proposed for almost two generations (Webster and de Saram, 1954, Cameron, 1982). As a matter of fact, use of “anatomic age” based on hand-wrist radiographs for grouping in education, child labor and sport was proposed about 100 years ago (Rotch, 1908).

The paper of Dvorak et al. is based on an invalid premise – bone age or SA as assessed from MRI of the distal radius can provide an accurate estimate of CA. Though correlated, there simply is too much variation in SA within an age group to provide a valid estimate CA. Standard deviations for SA within half-year CA groups in the Fels sample of boys 12 through 16 years range from 0.94 to 1.26 years (Roche et al. 1988). The range of variation in SA for a given CA likely exceeds two years.

The study is limited to adolescent football players 14-19 years of age. As a group, adolescent soccer players tend to be advanced in SA relative to CA, especially after 13-14 years of age. This has been documented in studies of skeletal maturation of youth soccer players which date to 1985 (Malina, 2003). Unfortunately, none of these studies are integrated into the discussion of Dvorak and colleagues. Football players also tend to be advanced in sexual maturation as noted in pubertal stages of genital and pubic hair development and testicular volume (Malina, 2003). Hence, to develop a system to estimate CA from stages of union of the distal epiphysis of the radius in adolescent football players who tend to be advanced in skeletal maturation is of limited utility. Rather, the protocol should have been initially developed on a random sample from the general population. To achieve the stated aim, it is essential to develop stages of epiphyseal union for the distal radius on a clinically normal sample of the general population and not on athletes who are select, especially in later adolescence. Given the long tradition of use of SA in clinical and human biology studies and to a large extent in studies of young athletes, it would make sense to validate the proposed MRI-based stages of union again radiographic data and perhaps skeletal data for individuals of known CA.

Dvorak and colleagues set their study in the context of the Greulich-Pyle and Tanner-Whitehouse methods for the assessment of SA. Unfortunately, the most recent version of the Tanner-White method (Tanner et al., 2001) and the Fels method (Roche et al., 1988) are not mentioned. The methods of assessment vary in the degree of specificity and discrimination of criteria, especially in the later stages of maturation of the radius, which has a significant effect on maturity status. The Fels method is more relevant to the issue under study as its criteria for final stages of maturation of the radius are more specific than those of the Greulich-Pyle and Tanner-Whitehouse methods (Malina et al., 2004, 2007).

The paper of Dvorak and colleagues concludes that “…MRI of the wrist offers an alternative as a non-invasive method of age determination in 14-19-year-old male adolescents” (p 45). Moreover, the authors emphasize that only one of 130, 16 year old players presented complete fusion. This is precisely the point in the context of the stated purpose of the paper. This boy would be misclassified if the protocol was used in an under 17 tournament (false positive). Further, 12 players, all of whom were less than 18.0 years of age, would be misclassified in an under 18 tournament; 22 less than 19.0 years of age would be misclassified in an under 19 tournament (see Table 3 in Dvorak et al.). I would suspect that the number of false positives would be greater among 15 and 16 year old players if a reference based on the general population of male adolescents would have been used (see Malina et al., 2007). Ethnic variation, correctly noted by the authors, is another issue which would likely contribute to false positives with any method.

In summary, normal variation in skeletal maturation among adolescents is considerable so that CA cannot be accurately estimated from SA (see Cameron, 1982). Moreover, elite football players represent a highly select group who tend to be advanced in skeletal maturation which further confounds issues relative to CA and SA in the sport.

What is the answer to the question of correct age in the sport of football (and other sports)? I believe it lies in the culture of the sport itself - honesty of trainers, coaches, administrators and others associated with the sport, including parents and agents, is probably the only method! Unfortunately, talented young players are a major commodity developed for and used by the system.

Robert M. Malina, PhD, FACSM Professor Emeritus, University of Texas, Austin, Texas Research Professor, Tarleton State University, Stephenville, Texas, USA rmalina@skyconnect.net

Gaston P. Beunen, PhD, FACSM Professor Emeritus, Department of Biomedical Kinesiology, Katholieke Universiteit Leuven, Leuven, Belgium Gaston.Beunen@faber.kuleuven.be

References

Cameron R. 1982. Age estimation: estimation of chronological age in children. Sci Public Policy, February, pp 20-27.

CNN Sports Illustrated. 2001. ‘Cheating does not pay’: Asia bans teams, players for over-age infractions. http://sportsillustrated.cnn.com/soccer/news/2001/05/10/Asia accessed 14 June 2001.

Dvorak J, J George, A Junge, J Hodler. 2006. Age determination by magnetic resonance imaging of the wrist in adolescent male football players. Br J Sports Med. 41:45-52.

Malina RM. 2003. Growth and maturity status of young soccer players. In T Reilly, AM Williams, Science and Soccer, 2nd edition. London: Routledge, pp 287-306.

Malina RM. 2005. Estimating passport age from bone age: Fallacy. Insight: The F.A. Coaches Association Journal 8(Autumn/Winter):23-27.

Malina RM, C Bouchard, O Bar-Or. 2004. Growth, Maturation, and Physical Activity, 2nd edition. Champaign, IL: Human Kinetics.

Malina RM, Chamorro M, Serratosa L, Morate F. 2007. TW 3 and Fels skeletal ages in youth soccer players. Ann Hum Biol. 34:in press.

Roche AF, CW Chumlea, D Thissen. 1988. Assessing the Skeletal Maturity of the Hand-Wrist: Fels Method. Springfield, IL: C.C. Thomas. Rotch TM. 1908. Chronological and anatomical age early in life. J Am Med Assoc 51:1197-1205

Tanner JM, MJR Healy, H Goldstein, N Cameron. 2001. Assessment of Skeletal Maturity and Prediction of Adult Height (TW3 Method), 3rd edition. London: WB Saunders.

Tritrakarn A, V Tansuphasiri. 1991. Roentgengraphic assessment of skeletal ages of Asian junior youth football players. J Med Assoc Thai 74:459-464.