- Email: [email protected]
Combined determination of selected radiological and morphological variables relevant for dental age estimation of young adults
ARTICLE IN PRESS l l HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
www.elsevier.de/jchb
Combined determination of selected radiological and morphological variables relevant for dental age estimation of young adults A. Olzea, A. Mahlowa, S. Schmidta, K.-D. Werneckeb, G. Gesericka, A. Schmelinga, a
Institut fu¨r Rechtsmedizin, Charite´—Universita¨tsmedizin Berlin, 10115 Berlin, Germany Institut fu¨r Medizinische Biometrie, Charite´—Universita¨tsmedizin Berlin, 10117 Berlin, Germany
b
Received 29 July 2004; accepted 16 November 2004
Abstract The mineralisation of third molars is one of the main criteria for dental age estimation of living subjects in criminal proceedings. Since the mineralisation of third molars is usually completed by the age of 19 or 20 years, this feature cannot be used to ascertain whether a person has attained the forensically relevant age of 21 years. This study examined whether determination based on an orthopantomogram of a combination of features relevant to dental age estimation of adults, supplies forensically useful information for ascertaining whether a person has attained 21 years of age. The features considered include the DMFT index of all permanent teeth, the DMFT index of all permanent teeth excluding third molars, the DFT index of third molars projecting beyond the occlusal plane, the eruption of third molars and the periodontal recession of second molars. A total of 650 conventional orthopantomograms of German subjects aged 18–30 years were evaluated. The statistical evaluation of our results was based on a discriminant analysis. It was carried out with two classes defined by two age groups: p 21 years and421 years. The probability of correct classification was 69.7% for males and 71.4% for females. Corresponding author. Tel.: +49 30 450525050; fax: +49 30 450525903.
E-mail address: [email protected] (A. Schmeling). 0018-442X/$ - see front matter r 2005 Elsevier GmbH. All rights reserved. doi:10.1016/j.jchb.2004.11.001
ARTICLE IN PRESS 134
A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
It can be concluded that an evaluation of the variables reviewed in this paper does not by itself yield sufficient data to determine with the accuracy required in criminal proceedings whether a person has attained 21 years of age. The additional radiographic examination of the medial clavicular epiphyseal cartilage is recommended. r 2005 Elsevier GmbH. All rights reserved.
Introduction As a result of increasing cross-border migration in recent years, many European countries have seen a growing number of foreigners who cannot provide documentary evidence for their date of birth. Because of this development, forensic age estimation of living subjects has become an integral part of forensic practice. The number of age diagnoses performed every year in the German-speaking area has soared from 185 in 1996 to approximately 500 at present (Schmeling et al., 2001b). The persons under examination are foreigners who are suspected of giving false statements about their age and whose genuine age is relevant in criminal proceedings for deciding whether they have reached the age of criminal responsibility and whether general criminal law in force for adults should be applied. German law defines four thresholds of relevance to criminal and civil prosecution: 14, 16, 18 and 21 years (Kaatsch, 2001). This is similar to numerous other European countries where the legally relevant age limits range between 14 and 21 years (Du¨nkel, 1997). In line with recommendations drawn up by the international and interdisciplinary study group on forensic age diagnostics (http://www.charite.de/rechtsmedizin/agfad/ index.htm), an age estimate for the purpose of criminal investigation should consist of a clinical examination, including the recording of body measurements and an evaluation of sexual maturity, a radiographic examination of the left hand and a dental examination which records dentition status and evaluates an orthopantomogram. In addition, a radiographic or tomographic examination of the collarbones is recommended to establish whether a person has attained 21 years of age (Schmeling et al., 2001a). The main criterion for dental age estimation is the evaluation of third molar mineralisation. Since the development of third molars is usually completed by the age of 19 or 20 years (Gunst et al., 2003; Kahl and Schwarze, 1988; Ko¨hler et al., 1994; Olze et al., 2003, 2004), this feature cannot be used to ascertain whether a person has attained the forensically relevant age of 21 years. The aim of this study is to examine whether the combined evaluation of various radiological and morphologic characteristics relevant for dental age estimation of adults, supplies forensically useful information allowing conclusions to be drawn about whether a person has reached the age of 21. The variables examined include the DMFT index of all permanent teeth, the DMFT index of all permanent teeth excluding third molars, the DFT index of third molars projecting beyond the occlusal plane, the eruption of third molars and the periodontal recession of second molars.
ARTICLE IN PRESS A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
135
Materials and methods In total 650 conventional orthopantomograms of German subjects aged 18–30 years were evaluated in total. Within each of the two groups 25 radiographs each of male and female subjects were analysed. The age groups were chosen in accordance with the legal objective, i.e. the age threshold of 21 years. The orthopantomograms, which date back to the period of 1988–1996, were taken from the radiography archive of the Charite´’s centre for dental medicine. Information recorded included name, sex, date of birth and date of orthopantomogram. The age of each subject was calculated from the difference between date of birth and date of orthopantomogram. The majority of the examined persons live in Berlin. The DMFT index of all permanent teeth, the DMFT index of all permanent teeth excluding third molars, and the DFT index of third molars projecting beyond the occlusal plane were determined. The DMFT index describes the rate of decayed (D), missing (M) and filled (F) permanent teeth (T) in a subject. The DFT index describes the total number of decayed and filled teeth of a person. The following classification of stages was used to evaluate the eruption of third molars:
Stage 0: The occlusal plane is covered by alveolar bone. Stage 1: Alveolar eruption; third molar has not yet developed to half the crown length of the second molar. Stage 2: Third molar has developed to at least half the crown length of the second molar, but has not reached the occlusal plane. Stage 3: Occlusal plane is reached, no elongation. Stage 4: Elongation.
Retained wisdom teeth were excluded from examination. Mesioangulated or distoangulated third molars and those with a vestibulo-oral position were considered to be retained (Archer, 1955; Wolf and Haunfelder, 1960). Those third molars where the direction of eruption could not be clearly determined were not included. Periodontal recession was determined for the second premolars of all four quadrants. To this end, the following four stages were defined:
Stage 0: No recession. Stage 1: Onset of periodontal recession, affecting less than half of the first third of the root. Stage 2: Advanced periodontal recession, affecting up to a third of the root. Stage 3: Considerable periodontal recession, affecting more than a third of the root.
Only teeth, which showed no signs of caries or restoration, were included in the determination of periodontal recession. Discriminant analysis was used for statistical processing. It was based on two classes defined by the two age groups of p 21 years and421 years. In order to take
ARTICLE IN PRESS 136
A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
account of possible misclassification, we performed a leaving-one-out (jack-knife) estimation of the error rate, which supplies an estimate of error for an external subject of unknown class identity (Wernecke, 1995). The following software was used for statistical processing: SAS Version 8.2, SPSS Version 11 and S-PLUS 2000 Release 2.
Results Due to the exclusion criteria mentioned above (missing values), 275 of the 650 examined cases were suitable for evaluation by discriminant analysis. Tables 1 and 2 show the classification results separately for the two sexes. In 69.7% of the males the classification was correct; in other words, this is the probability of correct classification. In 71.4% of the females the classification was correct.
Discussion In a recent survey, Ro¨sing and Kvaal (1998) discuss 11 sufficiently quantifiable features for dental age estimation of adults. These variables include number of teeth, tooth colour, attrition, periodontal recession, cement apposition, root resorption, Table 1. Classification results for males Presumed class identity Correct class
r 21 years
4 21 years
Total
Number r 21 years Number 4 21 years Percentage r 21 years Percentage 4 21 years
21 30 61.8 27.8
13 78 38.2 72.2
34 108 100 100
Table 2. Classification results for females Presumed class identity Correct class
r 21 years
4 21 years
Total
Number r 21 years Number 4 21 years Percentage r 21 years Percentage 4 21 years
19 31 73.1 29.0
7 76 26.9 71.0
26 107 100 100
ARTICLE IN PRESS A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
137
secondary dentine, root translucency, peritubular dentine, racemisation, and cement annulation. According to Ro¨sing and Kvaal (1998), the value of a method for age estimation is best described by the standard error in the regression of a variable to time. This represents the dispersion of the individual observations around the regression line. The single error encompasses 68% of the observations in the reference sample, the double error roughly 95%, and the triple error 99.72%. Although the processes of racemisation and cement annulation are the most accurate methods due to their minimal ranges of scatter, they do not lend themselves to forensic age estimation of living people involved in criminal proceedings, because they require extraction of a tooth or part of a tooth from the subject concerned. Similarly, the root translucency method can only be applied. The standard errors for the remaining methods range between 6.2 and 23.4 years (Ro¨sing and Kvaal, 1998). For most methods, standard error averages around 10 years. This means that if an age diagnosis yields an age of 50 years, then the single range is 40–60 years. However, there is only a probability of 68% that the true age is within this interval. For this reason, the triple error range is often used for forensic purposes. In the above example, only a span of 20–80 years would yield a satisfactory probability of correctness – and this is the entire normal adult life span. These single features are, therefore, not suitable for forensic age determination in the context of criminal proceedings (Ro¨sing and Kvaal, 1998). Friedrich et al. (2003 a, b) have recently published OPTG studies on the correlation between a person’s age and the number of decayed and filled wisdom teeth. They examined orthopantomograms of 1053 subjects aged 14–24 years. The statistical analysis of their results did not reveal a satisfactory correlation between the number of decayed teeth and chronological age for the examined age group. Highly positive predictive values, however, were derived from evidence of filled third molars. According to the authors, the correctness probability was 95.56% if tooth 28 was filled and 100% when tooth 38 was filled. Due to the fact that filled wisdom teeth were observed only in very rare cases in the examined sample (4.7–6.1%), this result should be interpreted with caution, all the more so since it is logically difficult to understand why filled third molars indicate a higher probability that a person is 18 years old than decayed third molars, considering that caries is a ‘prerequisite’ for the restoration of teeth. Matsikidis and Schulz (1982) developed a combined radiographic method. They evaluated 40 radiographs including 870 teeth in blind determination. The following variables were assigned grades ranging from 0 to 3: attrition, secondary dentine, periodontal recession, cement apposition and root resorption. The computed standard error for this combination of variables was 5.6 years. While Matsikidis and Schulz (1982) used dental films for their study, the study presented here evaluated orthopantomograms. Because images obtained by the orthopantomography technique are of poorer quality than dental film images (Jervoe, 1991), we were only able to determine periodontal recession among the variables proposed by Matsikidis and Schulz (1982). Insufficient sharpness of the images also impeded determination of the pulp height and pulp width indices as proposed by Kvaal et al. (1995).
ARTICLE IN PRESS 138
A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
In addition to periodontal recession, we analysed three DMFT index variations, which allowed us to take account of missing, decayed and filled teeth. The fifth variable included in our study was the eruption of wisdom teeth. Haavikko (1970) studied the eruption of teeth based on orthopantomograms of 615 Finnish males and 547 Finnish females aged 2–21 years. Alveolar eruption of third molars took place at a mean age of 17.2–18.1 years; the standard deviation ranged between 3.9 and 6.3 years. By including additional eruption stages we sought to cover young adulthood as well. The probability of correct classification by discriminant analysis was a mere 69.7% for males and 71.4% for females. The weak genetic or other systemic determination of the epidemiologic features examined in our study is suggested as a possible explanation for this relatively low probability. Both individual caries activity, which determines the DMFT index, and periodontal recession are strongly influenced by environmental factors. Dental caries is an exogenetic process and is caused by specific impacts in the direct environment of teeth. Caries develops when microorganisms in the plaque exert a prolonged influence on the teeth, supported by a sugar-rich diet. In addition, disposing influences such as composition of saliva and resistance of dental matter promote or inhibit the likelihood of pathological changes. Diet patterns in western industrialised countries have changed in the last 150 years. Increasing sugar consumption has correlated with a considerable increase in the prevalence of caries. In addition, constant and regular ingestion of acidic food is problematic with regard to dental health. Today we observe a preference for acidic or sour food and beverages among teenagers and young adults (Willershausen et al., 2003). This applies in particular to the enormous increase in the consumption of ‘‘soft’’ drinks which contain various fruit acids as flavour enhancers, ascorbic acid as an antioxidant, benzoic acid as a food preservative and carbonic acid for a sparkling effect. Moreover, dental health also strongly depends on people’s behaviour after consuming acidic beverages. If, shortly after contact with acidic food, teeth are exposed to mechanical oral hygiene procedures, such as thorough tooth brushing or heavy chewing of abrasive food or chewing gum, those areas of the teeth superficially dissolved by the acids are at risk of erosion. In addition to genetically determined immunodeficiencies, systemic diseases and viruses, there are various other important factors in the aetiology of periodontal diseases, including a person’s general attitude to health, habits such as smoking, drinking alcohol or the use of medical drugs, the social environment and psychological factors (Rateitschak et al., 2004). Tobacco consumption ranks first among the avoidable risk factors for periodontal disease. Several controlled studies agree that smokers show higher values than nonsmokers for clinical parameters such as pocket depth measured with a dental probe, decreased attachment level, pus and furcation involvement (Bergstro¨m and Preber, 1994; Brochut and Cimasoni, 1997; Brunnemann and Hoffmann, 1981; Reichart, 2000; Salvi et al., 1997). Personal oral hygiene plays an important role in preventing periodontal disease. The daily thorough but gentle plaque removal through brushing the teeth helps to
ARTICLE IN PRESS A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
139
reduce the risk of periodontitis (Lang et al., 1998). All these influences and dependencies are well known in northern societies; the variable use of measures results in a variable incidence of disease. Drawing on the results of our study, we can summarise by observing that a combined determination, based on orthopantomograms, of variations on the DMFT index, periodontal recession and third molar eruption appears to provide a suitable additional criterion for forensic age estimation of young adults. By contrast, the evaluation of single variables does not yield sufficient data to determine with the accuracy required in criminal proceedings whether a person has attained 21 years of age. If forensic age estimation is required to ascertain whether a person is 21 or over, an additional X-ray examination of the medial clavicular epiphyseal cartilage is, therefore, recommended. If the medial cartilage of the clavicle is fully ossified, it can be assumed that the examined person is at least 21 years old (Kreitner et al., 1997, 1998; Schmeling et al., 2004).
Acknowledgements The authors wish to thank Prof. K-P Lange, director of the centre for dental medicine at Charite´, for the kind provision of the orthopantomograms. We also thank the reviewers Professors Winfried Henke and Thomas Riepert.
References Archer, W.H., 1955. Die Chirurgie des Mundes und der Za¨hne. Medica, Stuttgart. Bergstro¨m, J., Preber, H., 1994. Tobacco use as a risk factor. J. Periodontol. 65, 545–550. Brochut, P.F., Cimasoni, G., 1997. Auswirkungen des Rauchens auf das Parodont. Schweiz Monatsschr. Zahnmed. 107, 672–686. Brunnemann, K.D., Hoffmann, D., 1981. Assessment of the carcinogenic N-nitrosodietholamine in Tobacco products and tobacco smoke. Carcinogenesis 2, 1123–1127. Du¨nkel, F., 1997. Jugendstrafrecht in Europa—Entwicklungstendenzen und Strategien. In: Du¨nkel, F., van Kalmthout, A., Schu¨ler-Springorum, H. (Eds.), Entwicklungstendenzen und Reformstrategien im Jugendstrafrecht im europa¨ischen Vergleich. Forum, Mo¨nchengladbach, pp. 565–650. Friedrich, R.E., Ulbricht, C., von Maydell, L.A., 2003a. Karies und Fu¨llungen an Weisheitsza¨hnen als forensisch-odontologisches Hilfsmittel der Bestimmung eines chronologischen Alters 18 u¨ber Jahren. Radiologische Untersuchungen an Panoramaschichtaufnahmen von Kindern und Jugendlichen. Arch. Kriminol. 212, 74–82. Friedrich, R.E., Ulbricht, C., von Maydell, L.A., 2003b. Altersbestimmung in Abha¨ngigkeit von der Anzahl kario¨ser oder konservierend versorgter Za¨hne. Quintessenz 54, 863–874. Gunst, K., Mesotten, K., Carbonez, A., Willems, G., 2003. Third molar root development in relation to chronological age: a large sample sized retrospective study. Forensic Sci. Int. 136, 52–57. Haavikko, K., 1970. The formation and the alveolar and clinical eruption of the permanent teeth. Suom Hammasla¨a¨k Toim. 66, 103–170. Jervoe, P., 1991. OPG und Status im Vergleich: Ergebnisse einer Auswertung von periodisch-identischen Ro¨ntgenaufnahmen. Diss, Ko¨ln. Kaatsch, H.-J., 2001. Juristische Aspekte der Altersscha¨tzung. In: Oehmichen, M., Geserick, G. (Eds.), Osteologische Identifikation und Altersscha¨tzung. Schmidt-Ro¨mhild, Lu¨beck, pp. 243–254. Kahl, B., Schwarze, C.W., 1988. Aktualisierung der Dentitionstabelle von I Schour und M Massler von 1941. Fortschr. Kieferorthop. 49, 432–443.
ARTICLE IN PRESS 140
A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
Ko¨hler, S., Schmelzle, R., Loitz, C., Pu¨schel, K., 1994. Die Entwicklung des Weisheitszahnes als Kriterium der Lebensalterbestimmung. Ann. Anat. 176, 339–345. Kreitner, K.-F., Schweden, F., Schild, H.H., Riepert, T., Nafe, B., 1997. Die computertomographisch bestimmte Ausreifung der medialen Klavikulaepiphyse—eine additive Methode zur Altersbestimmung im Adoleszentenalter und in der dritten Lebensdekade? Fortschr. Ro¨ntgenstr. 166, 481–486. Kreitner, K.-F., Schweden, F.J., Riepert, T., Nafe, B., Thelen, M., 1998. Bone age determination based on the study of the medial extremity of the clavicle. Eur. Radiol. 8, 1116–1122. Kvaal, S.I., Kolltveit, K.M., Thompsen, I.O., Solheim, T., 1995. Age determination of adults from radiographs. Forensic Sci. Int. 74, 175–185. Lang, N.P., Attstro¨m, R., Lo¨e, H., 1998. Proceedings of the European Workshop on Mechanical Plaque Control. Quintessenz, Chicago. Matsikidis, G., Schulz, P., 1982. Altersbestimmung nach dem Gebiss mit Hilfe des Zahnfilms. Zahna¨rztl. Mitt. 72, 2524–2528. Olze, A., Schmeling, A., Rieger, K., Kalb, G., Geserick, G., 2003. Untersuchungen zum zeitlichen Verlauf der Weisheitszahnmineralisation bei einer deutschen Population. Rechtsmedizin 13, 5–10. Olze, A., Schmeling, A., Taniguchi, M., Maeda, H., van Niekerk, P., Wernecke, K.-D., Geserick, G., 2004. Forensic age estimation in living subjects: the ethnic factor in wisdom tooth mineralization. Int. J. Legal Med. 118, 170–173. Rateitschak, H., Rateitschak, E.M., Wolf, H.F. (Eds.), 2004. Parodontologie. Georg Thieme, Stuttgart. Reichart, P.A., 2000. Tabak und Mundgesundheit. Dtsch. Zahna¨rztl. Z. 55, 223–229. Ro¨sing, F.W., Kvaal, S.I., 1998. Dental age in adults – a review of estimation methods. In: Alt, K.W., Ro¨sing, F.W., Teschler, N. (Eds.), Dental Anthropology: Fundamentals, Limits and prospects. Springer, Wien, pp. 443–468. Salvi, G.E., Lawrence, H.P., Offenbacher, S., Beck, J.D., 1997. Influence of risk factors on the pathogenesis of periodontitis. Periodontol 2000 14, 173–201. Schmeling, A., Kaatsch, H.-J., Marre´, B., Reisinger, W., Riepert, T., Ritz-Timme, S., Ro¨sing, F.W., Ro¨tzscher, K., Geserick, G., 2001a. Empfehlungen fu¨r die Altersdiagnostik bei Lebenden im Strafverfahren. Rechtsmedizin 11, 1–3. Schmeling, A., Olze, A., Reisinger, W., Geserick, G., 2001b. Age estimation of living people undergoing criminal proceedings. Lancet 358, 89–90. Schmeling, A., Schulz, R., Reisinger, W., Mu¨hler, M., Wernecke, K.-D., Geserick, G., 2004. Studies of the time frame for ossification of medial clavicular epiphyseal cartilage in conventional radiography. Int. J. Legal Med. 118, 5–8. Wernecke, K.-D., 1995. Angewandte Statistik fu¨r die Praxis. Addison-Wesley, Bonn, Paris. Willershausen, B., Ernst, C.-P., Pistorius, A., Brandenbusch, M., 2003. In-Getra¨nke und ihre Folgen am Zahnschmelz. Zahna¨rztl. Mitt. 12, 38–44. Wolf, H., Haunfelder, D., 1960. Zahna¨rztliche Mundchirurgie fu¨r Studierende der zahnheilkunde, vol. 5. Berlinische Verlagsanstalt, Berlin, pp. 59-67.
www.elsevier.de/jchb
Combined determination of selected radiological and morphological variables relevant for dental age estimation of young adults A. Olzea, A. Mahlowa, S. Schmidta, K.-D. Werneckeb, G. Gesericka, A. Schmelinga, a
Institut fu¨r Rechtsmedizin, Charite´—Universita¨tsmedizin Berlin, 10115 Berlin, Germany Institut fu¨r Medizinische Biometrie, Charite´—Universita¨tsmedizin Berlin, 10117 Berlin, Germany
b
Received 29 July 2004; accepted 16 November 2004
Abstract The mineralisation of third molars is one of the main criteria for dental age estimation of living subjects in criminal proceedings. Since the mineralisation of third molars is usually completed by the age of 19 or 20 years, this feature cannot be used to ascertain whether a person has attained the forensically relevant age of 21 years. This study examined whether determination based on an orthopantomogram of a combination of features relevant to dental age estimation of adults, supplies forensically useful information for ascertaining whether a person has attained 21 years of age. The features considered include the DMFT index of all permanent teeth, the DMFT index of all permanent teeth excluding third molars, the DFT index of third molars projecting beyond the occlusal plane, the eruption of third molars and the periodontal recession of second molars. A total of 650 conventional orthopantomograms of German subjects aged 18–30 years were evaluated. The statistical evaluation of our results was based on a discriminant analysis. It was carried out with two classes defined by two age groups: p 21 years and421 years. The probability of correct classification was 69.7% for males and 71.4% for females. Corresponding author. Tel.: +49 30 450525050; fax: +49 30 450525903.
E-mail address: [email protected] (A. Schmeling). 0018-442X/$ - see front matter r 2005 Elsevier GmbH. All rights reserved. doi:10.1016/j.jchb.2004.11.001
ARTICLE IN PRESS 134
A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
It can be concluded that an evaluation of the variables reviewed in this paper does not by itself yield sufficient data to determine with the accuracy required in criminal proceedings whether a person has attained 21 years of age. The additional radiographic examination of the medial clavicular epiphyseal cartilage is recommended. r 2005 Elsevier GmbH. All rights reserved.
Introduction As a result of increasing cross-border migration in recent years, many European countries have seen a growing number of foreigners who cannot provide documentary evidence for their date of birth. Because of this development, forensic age estimation of living subjects has become an integral part of forensic practice. The number of age diagnoses performed every year in the German-speaking area has soared from 185 in 1996 to approximately 500 at present (Schmeling et al., 2001b). The persons under examination are foreigners who are suspected of giving false statements about their age and whose genuine age is relevant in criminal proceedings for deciding whether they have reached the age of criminal responsibility and whether general criminal law in force for adults should be applied. German law defines four thresholds of relevance to criminal and civil prosecution: 14, 16, 18 and 21 years (Kaatsch, 2001). This is similar to numerous other European countries where the legally relevant age limits range between 14 and 21 years (Du¨nkel, 1997). In line with recommendations drawn up by the international and interdisciplinary study group on forensic age diagnostics (http://www.charite.de/rechtsmedizin/agfad/ index.htm), an age estimate for the purpose of criminal investigation should consist of a clinical examination, including the recording of body measurements and an evaluation of sexual maturity, a radiographic examination of the left hand and a dental examination which records dentition status and evaluates an orthopantomogram. In addition, a radiographic or tomographic examination of the collarbones is recommended to establish whether a person has attained 21 years of age (Schmeling et al., 2001a). The main criterion for dental age estimation is the evaluation of third molar mineralisation. Since the development of third molars is usually completed by the age of 19 or 20 years (Gunst et al., 2003; Kahl and Schwarze, 1988; Ko¨hler et al., 1994; Olze et al., 2003, 2004), this feature cannot be used to ascertain whether a person has attained the forensically relevant age of 21 years. The aim of this study is to examine whether the combined evaluation of various radiological and morphologic characteristics relevant for dental age estimation of adults, supplies forensically useful information allowing conclusions to be drawn about whether a person has reached the age of 21. The variables examined include the DMFT index of all permanent teeth, the DMFT index of all permanent teeth excluding third molars, the DFT index of third molars projecting beyond the occlusal plane, the eruption of third molars and the periodontal recession of second molars.
ARTICLE IN PRESS A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
135
Materials and methods In total 650 conventional orthopantomograms of German subjects aged 18–30 years were evaluated in total. Within each of the two groups 25 radiographs each of male and female subjects were analysed. The age groups were chosen in accordance with the legal objective, i.e. the age threshold of 21 years. The orthopantomograms, which date back to the period of 1988–1996, were taken from the radiography archive of the Charite´’s centre for dental medicine. Information recorded included name, sex, date of birth and date of orthopantomogram. The age of each subject was calculated from the difference between date of birth and date of orthopantomogram. The majority of the examined persons live in Berlin. The DMFT index of all permanent teeth, the DMFT index of all permanent teeth excluding third molars, and the DFT index of third molars projecting beyond the occlusal plane were determined. The DMFT index describes the rate of decayed (D), missing (M) and filled (F) permanent teeth (T) in a subject. The DFT index describes the total number of decayed and filled teeth of a person. The following classification of stages was used to evaluate the eruption of third molars:
Stage 0: The occlusal plane is covered by alveolar bone. Stage 1: Alveolar eruption; third molar has not yet developed to half the crown length of the second molar. Stage 2: Third molar has developed to at least half the crown length of the second molar, but has not reached the occlusal plane. Stage 3: Occlusal plane is reached, no elongation. Stage 4: Elongation.
Retained wisdom teeth were excluded from examination. Mesioangulated or distoangulated third molars and those with a vestibulo-oral position were considered to be retained (Archer, 1955; Wolf and Haunfelder, 1960). Those third molars where the direction of eruption could not be clearly determined were not included. Periodontal recession was determined for the second premolars of all four quadrants. To this end, the following four stages were defined:
Stage 0: No recession. Stage 1: Onset of periodontal recession, affecting less than half of the first third of the root. Stage 2: Advanced periodontal recession, affecting up to a third of the root. Stage 3: Considerable periodontal recession, affecting more than a third of the root.
Only teeth, which showed no signs of caries or restoration, were included in the determination of periodontal recession. Discriminant analysis was used for statistical processing. It was based on two classes defined by the two age groups of p 21 years and421 years. In order to take
ARTICLE IN PRESS 136
A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
account of possible misclassification, we performed a leaving-one-out (jack-knife) estimation of the error rate, which supplies an estimate of error for an external subject of unknown class identity (Wernecke, 1995). The following software was used for statistical processing: SAS Version 8.2, SPSS Version 11 and S-PLUS 2000 Release 2.
Results Due to the exclusion criteria mentioned above (missing values), 275 of the 650 examined cases were suitable for evaluation by discriminant analysis. Tables 1 and 2 show the classification results separately for the two sexes. In 69.7% of the males the classification was correct; in other words, this is the probability of correct classification. In 71.4% of the females the classification was correct.
Discussion In a recent survey, Ro¨sing and Kvaal (1998) discuss 11 sufficiently quantifiable features for dental age estimation of adults. These variables include number of teeth, tooth colour, attrition, periodontal recession, cement apposition, root resorption, Table 1. Classification results for males Presumed class identity Correct class
r 21 years
4 21 years
Total
Number r 21 years Number 4 21 years Percentage r 21 years Percentage 4 21 years
21 30 61.8 27.8
13 78 38.2 72.2
34 108 100 100
Table 2. Classification results for females Presumed class identity Correct class
r 21 years
4 21 years
Total
Number r 21 years Number 4 21 years Percentage r 21 years Percentage 4 21 years
19 31 73.1 29.0
7 76 26.9 71.0
26 107 100 100
ARTICLE IN PRESS A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
137
secondary dentine, root translucency, peritubular dentine, racemisation, and cement annulation. According to Ro¨sing and Kvaal (1998), the value of a method for age estimation is best described by the standard error in the regression of a variable to time. This represents the dispersion of the individual observations around the regression line. The single error encompasses 68% of the observations in the reference sample, the double error roughly 95%, and the triple error 99.72%. Although the processes of racemisation and cement annulation are the most accurate methods due to their minimal ranges of scatter, they do not lend themselves to forensic age estimation of living people involved in criminal proceedings, because they require extraction of a tooth or part of a tooth from the subject concerned. Similarly, the root translucency method can only be applied. The standard errors for the remaining methods range between 6.2 and 23.4 years (Ro¨sing and Kvaal, 1998). For most methods, standard error averages around 10 years. This means that if an age diagnosis yields an age of 50 years, then the single range is 40–60 years. However, there is only a probability of 68% that the true age is within this interval. For this reason, the triple error range is often used for forensic purposes. In the above example, only a span of 20–80 years would yield a satisfactory probability of correctness – and this is the entire normal adult life span. These single features are, therefore, not suitable for forensic age determination in the context of criminal proceedings (Ro¨sing and Kvaal, 1998). Friedrich et al. (2003 a, b) have recently published OPTG studies on the correlation between a person’s age and the number of decayed and filled wisdom teeth. They examined orthopantomograms of 1053 subjects aged 14–24 years. The statistical analysis of their results did not reveal a satisfactory correlation between the number of decayed teeth and chronological age for the examined age group. Highly positive predictive values, however, were derived from evidence of filled third molars. According to the authors, the correctness probability was 95.56% if tooth 28 was filled and 100% when tooth 38 was filled. Due to the fact that filled wisdom teeth were observed only in very rare cases in the examined sample (4.7–6.1%), this result should be interpreted with caution, all the more so since it is logically difficult to understand why filled third molars indicate a higher probability that a person is 18 years old than decayed third molars, considering that caries is a ‘prerequisite’ for the restoration of teeth. Matsikidis and Schulz (1982) developed a combined radiographic method. They evaluated 40 radiographs including 870 teeth in blind determination. The following variables were assigned grades ranging from 0 to 3: attrition, secondary dentine, periodontal recession, cement apposition and root resorption. The computed standard error for this combination of variables was 5.6 years. While Matsikidis and Schulz (1982) used dental films for their study, the study presented here evaluated orthopantomograms. Because images obtained by the orthopantomography technique are of poorer quality than dental film images (Jervoe, 1991), we were only able to determine periodontal recession among the variables proposed by Matsikidis and Schulz (1982). Insufficient sharpness of the images also impeded determination of the pulp height and pulp width indices as proposed by Kvaal et al. (1995).
ARTICLE IN PRESS 138
A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
In addition to periodontal recession, we analysed three DMFT index variations, which allowed us to take account of missing, decayed and filled teeth. The fifth variable included in our study was the eruption of wisdom teeth. Haavikko (1970) studied the eruption of teeth based on orthopantomograms of 615 Finnish males and 547 Finnish females aged 2–21 years. Alveolar eruption of third molars took place at a mean age of 17.2–18.1 years; the standard deviation ranged between 3.9 and 6.3 years. By including additional eruption stages we sought to cover young adulthood as well. The probability of correct classification by discriminant analysis was a mere 69.7% for males and 71.4% for females. The weak genetic or other systemic determination of the epidemiologic features examined in our study is suggested as a possible explanation for this relatively low probability. Both individual caries activity, which determines the DMFT index, and periodontal recession are strongly influenced by environmental factors. Dental caries is an exogenetic process and is caused by specific impacts in the direct environment of teeth. Caries develops when microorganisms in the plaque exert a prolonged influence on the teeth, supported by a sugar-rich diet. In addition, disposing influences such as composition of saliva and resistance of dental matter promote or inhibit the likelihood of pathological changes. Diet patterns in western industrialised countries have changed in the last 150 years. Increasing sugar consumption has correlated with a considerable increase in the prevalence of caries. In addition, constant and regular ingestion of acidic food is problematic with regard to dental health. Today we observe a preference for acidic or sour food and beverages among teenagers and young adults (Willershausen et al., 2003). This applies in particular to the enormous increase in the consumption of ‘‘soft’’ drinks which contain various fruit acids as flavour enhancers, ascorbic acid as an antioxidant, benzoic acid as a food preservative and carbonic acid for a sparkling effect. Moreover, dental health also strongly depends on people’s behaviour after consuming acidic beverages. If, shortly after contact with acidic food, teeth are exposed to mechanical oral hygiene procedures, such as thorough tooth brushing or heavy chewing of abrasive food or chewing gum, those areas of the teeth superficially dissolved by the acids are at risk of erosion. In addition to genetically determined immunodeficiencies, systemic diseases and viruses, there are various other important factors in the aetiology of periodontal diseases, including a person’s general attitude to health, habits such as smoking, drinking alcohol or the use of medical drugs, the social environment and psychological factors (Rateitschak et al., 2004). Tobacco consumption ranks first among the avoidable risk factors for periodontal disease. Several controlled studies agree that smokers show higher values than nonsmokers for clinical parameters such as pocket depth measured with a dental probe, decreased attachment level, pus and furcation involvement (Bergstro¨m and Preber, 1994; Brochut and Cimasoni, 1997; Brunnemann and Hoffmann, 1981; Reichart, 2000; Salvi et al., 1997). Personal oral hygiene plays an important role in preventing periodontal disease. The daily thorough but gentle plaque removal through brushing the teeth helps to
ARTICLE IN PRESS A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
139
reduce the risk of periodontitis (Lang et al., 1998). All these influences and dependencies are well known in northern societies; the variable use of measures results in a variable incidence of disease. Drawing on the results of our study, we can summarise by observing that a combined determination, based on orthopantomograms, of variations on the DMFT index, periodontal recession and third molar eruption appears to provide a suitable additional criterion for forensic age estimation of young adults. By contrast, the evaluation of single variables does not yield sufficient data to determine with the accuracy required in criminal proceedings whether a person has attained 21 years of age. If forensic age estimation is required to ascertain whether a person is 21 or over, an additional X-ray examination of the medial clavicular epiphyseal cartilage is, therefore, recommended. If the medial cartilage of the clavicle is fully ossified, it can be assumed that the examined person is at least 21 years old (Kreitner et al., 1997, 1998; Schmeling et al., 2004).
Acknowledgements The authors wish to thank Prof. K-P Lange, director of the centre for dental medicine at Charite´, for the kind provision of the orthopantomograms. We also thank the reviewers Professors Winfried Henke and Thomas Riepert.
References Archer, W.H., 1955. Die Chirurgie des Mundes und der Za¨hne. Medica, Stuttgart. Bergstro¨m, J., Preber, H., 1994. Tobacco use as a risk factor. J. Periodontol. 65, 545–550. Brochut, P.F., Cimasoni, G., 1997. Auswirkungen des Rauchens auf das Parodont. Schweiz Monatsschr. Zahnmed. 107, 672–686. Brunnemann, K.D., Hoffmann, D., 1981. Assessment of the carcinogenic N-nitrosodietholamine in Tobacco products and tobacco smoke. Carcinogenesis 2, 1123–1127. Du¨nkel, F., 1997. Jugendstrafrecht in Europa—Entwicklungstendenzen und Strategien. In: Du¨nkel, F., van Kalmthout, A., Schu¨ler-Springorum, H. (Eds.), Entwicklungstendenzen und Reformstrategien im Jugendstrafrecht im europa¨ischen Vergleich. Forum, Mo¨nchengladbach, pp. 565–650. Friedrich, R.E., Ulbricht, C., von Maydell, L.A., 2003a. Karies und Fu¨llungen an Weisheitsza¨hnen als forensisch-odontologisches Hilfsmittel der Bestimmung eines chronologischen Alters 18 u¨ber Jahren. Radiologische Untersuchungen an Panoramaschichtaufnahmen von Kindern und Jugendlichen. Arch. Kriminol. 212, 74–82. Friedrich, R.E., Ulbricht, C., von Maydell, L.A., 2003b. Altersbestimmung in Abha¨ngigkeit von der Anzahl kario¨ser oder konservierend versorgter Za¨hne. Quintessenz 54, 863–874. Gunst, K., Mesotten, K., Carbonez, A., Willems, G., 2003. Third molar root development in relation to chronological age: a large sample sized retrospective study. Forensic Sci. Int. 136, 52–57. Haavikko, K., 1970. The formation and the alveolar and clinical eruption of the permanent teeth. Suom Hammasla¨a¨k Toim. 66, 103–170. Jervoe, P., 1991. OPG und Status im Vergleich: Ergebnisse einer Auswertung von periodisch-identischen Ro¨ntgenaufnahmen. Diss, Ko¨ln. Kaatsch, H.-J., 2001. Juristische Aspekte der Altersscha¨tzung. In: Oehmichen, M., Geserick, G. (Eds.), Osteologische Identifikation und Altersscha¨tzung. Schmidt-Ro¨mhild, Lu¨beck, pp. 243–254. Kahl, B., Schwarze, C.W., 1988. Aktualisierung der Dentitionstabelle von I Schour und M Massler von 1941. Fortschr. Kieferorthop. 49, 432–443.
ARTICLE IN PRESS 140
A. Olze et al. / HOMO — Journal of Comparative Human Biology 56 (2005) 133–140
Ko¨hler, S., Schmelzle, R., Loitz, C., Pu¨schel, K., 1994. Die Entwicklung des Weisheitszahnes als Kriterium der Lebensalterbestimmung. Ann. Anat. 176, 339–345. Kreitner, K.-F., Schweden, F., Schild, H.H., Riepert, T., Nafe, B., 1997. Die computertomographisch bestimmte Ausreifung der medialen Klavikulaepiphyse—eine additive Methode zur Altersbestimmung im Adoleszentenalter und in der dritten Lebensdekade? Fortschr. Ro¨ntgenstr. 166, 481–486. Kreitner, K.-F., Schweden, F.J., Riepert, T., Nafe, B., Thelen, M., 1998. Bone age determination based on the study of the medial extremity of the clavicle. Eur. Radiol. 8, 1116–1122. Kvaal, S.I., Kolltveit, K.M., Thompsen, I.O., Solheim, T., 1995. Age determination of adults from radiographs. Forensic Sci. Int. 74, 175–185. Lang, N.P., Attstro¨m, R., Lo¨e, H., 1998. Proceedings of the European Workshop on Mechanical Plaque Control. Quintessenz, Chicago. Matsikidis, G., Schulz, P., 1982. Altersbestimmung nach dem Gebiss mit Hilfe des Zahnfilms. Zahna¨rztl. Mitt. 72, 2524–2528. Olze, A., Schmeling, A., Rieger, K., Kalb, G., Geserick, G., 2003. Untersuchungen zum zeitlichen Verlauf der Weisheitszahnmineralisation bei einer deutschen Population. Rechtsmedizin 13, 5–10. Olze, A., Schmeling, A., Taniguchi, M., Maeda, H., van Niekerk, P., Wernecke, K.-D., Geserick, G., 2004. Forensic age estimation in living subjects: the ethnic factor in wisdom tooth mineralization. Int. J. Legal Med. 118, 170–173. Rateitschak, H., Rateitschak, E.M., Wolf, H.F. (Eds.), 2004. Parodontologie. Georg Thieme, Stuttgart. Reichart, P.A., 2000. Tabak und Mundgesundheit. Dtsch. Zahna¨rztl. Z. 55, 223–229. Ro¨sing, F.W., Kvaal, S.I., 1998. Dental age in adults – a review of estimation methods. In: Alt, K.W., Ro¨sing, F.W., Teschler, N. (Eds.), Dental Anthropology: Fundamentals, Limits and prospects. Springer, Wien, pp. 443–468. Salvi, G.E., Lawrence, H.P., Offenbacher, S., Beck, J.D., 1997. Influence of risk factors on the pathogenesis of periodontitis. Periodontol 2000 14, 173–201. Schmeling, A., Kaatsch, H.-J., Marre´, B., Reisinger, W., Riepert, T., Ritz-Timme, S., Ro¨sing, F.W., Ro¨tzscher, K., Geserick, G., 2001a. Empfehlungen fu¨r die Altersdiagnostik bei Lebenden im Strafverfahren. Rechtsmedizin 11, 1–3. Schmeling, A., Olze, A., Reisinger, W., Geserick, G., 2001b. Age estimation of living people undergoing criminal proceedings. Lancet 358, 89–90. Schmeling, A., Schulz, R., Reisinger, W., Mu¨hler, M., Wernecke, K.-D., Geserick, G., 2004. Studies of the time frame for ossification of medial clavicular epiphyseal cartilage in conventional radiography. Int. J. Legal Med. 118, 5–8. Wernecke, K.-D., 1995. Angewandte Statistik fu¨r die Praxis. Addison-Wesley, Bonn, Paris. Willershausen, B., Ernst, C.-P., Pistorius, A., Brandenbusch, M., 2003. In-Getra¨nke und ihre Folgen am Zahnschmelz. Zahna¨rztl. Mitt. 12, 38–44. Wolf, H., Haunfelder, D., 1960. Zahna¨rztliche Mundchirurgie fu¨r Studierende der zahnheilkunde, vol. 5. Berlinische Verlagsanstalt, Berlin, pp. 59-67.