Int J Legal Med DOI 10.1007/s00414-015-1172-8
ORIGINAL ARTICLE
Toward the adoption of cementochronology in forensic context T. Colard & B. Bertrand & S. Naji & Y. Delannoy & A. Bécart
Received: 21 May 2014 / Accepted: 2 March 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Because acellular dental cementum is considered to be formed continually throughout life and to not undergo remodeling processes, cementochronology is considered to be a method with the potential for directly assessing chronological age. Considering that most previous studies on humans have assumed the superior performance of this method, it is surprising that this technique is not more widely adopted in anthropology. To understand this controversy, we highlight that there is no standardized procedure for sample preparation. The numerous technical approaches that exist impact the reliability of the method, and the recent creation of an international work group (Cementochronology Research Program) demonstrates the need for researchers to share their experience to overcome these obstacles. This paper aims to address this paradox by debating the aspects that contribute to the limited use of this method and by illustrating its potential through an application on forensic cases. A protocol, which was recently certified according to the ISO-9001, was applied to nine anthropological cases from the Forensic Medicine Institute of Lille (northern France) and compared with routine osteological and dental methods. The results show that traditional methods matched the known age due to the wide extent of their range, while the
T. Colard and B. Bertrand contributed equally to this work. T. Colard : B. Bertrand : Y. Delannoy : A. Bécart Unité de Taphonomie Médico-Légale, Institut de Médecine Légale, Université de Lille 2, Lille, France B. Bertrand (*) Laboratoire d’Anthropologie, Direction de l’Archéologie, Communauté d’Agglomération du Douaisis, Douai, France e-mail:
[email protected] S. Naji CNRS-PACEA-UMR 5199, Université de Bordeaux, Bordeaux, France
accuracy and precision of cementochronological estimates was also notable. This paper establishes that cementochronology may serve as a particularly important tool for age estimation for forensic anthropologists and should, at least, be used in addition to other methods. Keywords Forensic anthropology . Cementochronology . Age-at-death estimation . Dental cementum
Introduction Originally developed for nonhuman mammalian species [1, 2], cementochronology involves the counting of incremental lines in tooth root cementum to estimate an individual’s age at death (Fig. 1). The chemical composition of cementum is similar to that of bone [3]. However, unlike bone, cementum does not undergo a remodeling process and is continually formed throughout life [4–7], two characteristics that can be extremely beneficial to the development of methods to determine age at death. With rare exceptions [8–10], all previous studies have assumed the method’s superior performance for age determination in adults [11, 12]. However, a recent study on a representative sample of forensic anthropologists [13] revealed that the pubic symphysis, sternal rib ends, and auricular surface methods still are the most popular techniques for estimating age at death and that when postcranial material is not preserved, the cranial sutures are favored. This paper aims to address this paradoxical disregard for cementochronology by debating the aspects that may contribute to the limited use of this method and by illustrating its potential utility in routine forensic cases. Cementochronology clearly suffers from the fact that the physiological and structural biological background of cementum is not well elucidated [14]. While speculations have been made about these lines being hyper- and hypomineralized
Int J Legal Med
Fig. 1 Schematic representation of thin section and counting for cementochronology procedure
areas or being caused by an abrupt change in the direction of Sharpey’s fibers [15], we still do not know the biological cause of annual periodicity, nor do we understand the nature of this optical phenomenon, which is commonly named Btooth cementum annulation.^ All aspects considered, more questions than answers exist regarding the nature of cementum appositions and regarding the mechanisms producing these appositions and regarding their variability. Nonetheless, the studies performed on tooth cementum annulation have shown that the number of incremental lines in the cementum can be correlated with known age at death in as high as 98 % [11] and have considered cementochronology as a feasible method for estimating the age at death of an individual. Some (rare) skeptical studies that were not able to find any connection or that presented only a moderate correlation between the age and the number of lines noted the technical difficulties that made the preparation of series of sections in all samples impossible [14]. To understand why there is still a controversy, despite the apparent agreement that exists, we must note that, as previously analyzed by Grosskopf and McGlynn [16] and noted by Charles et al. [17], the abundance in methodologies and the effect on the ability to produce scorable and reliable sections are problematic. Thirty years after its first application [18], there is still no standardized procedure for evaluating and counting cementum lines nor for preparing specimens. Selected samples of control studies are teeth from anatomical departments or dental practices [11, 19], or even archaeological specimens [20–22]. Moreover, there is still disagreement as to which tooth class provides the most correct estimates and on the best histological preparation protocol. Some investigations are conducted on mineralized, demineralized, or acid-etched slices, on longitudinal or transversal sections, on different cutting angles, from 5 to 200 μm thickness, on stained or unstained preparation, embedded in plastic or without embedding, examined each or some sections, considered the maximum number of lines for further statistical analysis and ignored the other counted line numbers or took an average of different counts in different areas of cementum. All these different preparation and treatment methods, which often reflect the adaptability of the anthropologist to the available material and that betray the need for the adoption of a validated procedure, make it difficult to compare the results from different studies.
Accordingly, this variability contributes to dissimilar judgments on its applicability and may discourage novice users. Furthermore, while cementochronology is continuously being published in journals, it has failed to build its reputation as a traditional method, and despite the 20 years of validation studies, only a few anthropologists feel the required familiarity with its original study. Even if there is no harmonized method in cementochronology, there is a strong trend for 10 years that supports convergent key parameters and that is federating physical anthropologists. These determinant criteria need to be fulfilled in order to enable the reproducibility and to minimize inter- and intra-observer errors. These concerns about the variability in methodological processes and the lack of academic recognition led authors to create an international work group, the Cementochronology Research Program (www.cementochronology.com), and to assign importance to quality standards in applying for the certification according to ISO-9001. This international standard obviously does not guarantee the reliability of the estimation but is a recognition of the conformity and continuous improvement of the preparation process. A protocol, certified in 2013 at the Direction de l’Archéologie [23], has been applied to nine anthropological cases from the Anthropology Department of Lille 2 University (northern France). Considering that most forensic anthropologists vary their age-at-death methods case by case and combine information from multiple methods [13] and considering that in this field of expertise, often both narrow and broad age ranges are presented, we propose to compare known ages to estimates from osteological and dental methods for each individual to evaluate the level of differences among different techniques.
Materials and methods Teeth and bone samples were obtained from nine skeletonized individuals of known age. The ages at death, sex, and postmortem interval (PMI) were documented in all cases and are presented in Table 1. Anthropology protocols for skeletal aging were based on the following methods depending on the available anatomical region: pubic symphysis [24, 25], auricular surface [26, 27], first [28] and fourth sternal rib ends [29–31], epiphysis of the first sternal head [28], and ectocranial suture synostosis [32]. Dental methods included the use of the two-criteria dental method, the periodontosis height and the transparency of the root [33, 34]. Anterior teeth and uppermost canines, often less affected by pathologies and easier to position for sectioning, were selected for cementochronology. Teeth with pathological conditions were avoided to not bias the observations. For histological preparation, the teeth were processed at the BPhysical Analysis and Materials Characterization Laboratory – Direction de
54 (31–57) 53 (2.8) 56 55 (42–68) 60 (3.6) 64 51 (38–64) 62.3 (4.1) 64 39 (26–52) 40.6 (2.4) 46 52 (39–65) 66.3 (3.8) 66 50 (37–63) 52.9 (3.1) 55 CS complete skeleton, IS incomplete skeleton, MM maxilla and mandible, na not available
51 (38–64) 62.4 (5) 64 53 (40–66) 73.5 (5.1) 76
34 (21–47) 40.3 (4) 35
na na na na na na V (35–55) 6 (45–49) 40–49 40–55 35–50 43 V (35–55) 7 (50–60) 50–59 50–60 50–60 46 na na na na na na V (35–55) 6 (45–49) na na na 52 V (35–55) 6 (45–49) 40–49 40–55 35–50 47 na na na na na na V (35–55) 7 (50–60) 60–69 60–70 50–60 43
III (22–35) 3 (30–34) 20–29 20–30 30–35 32
CRO F Days MM ROD M 17 years CS
Cases Sex PMI Human remains Osteological methods Pubic symphysis (Suchey and Brooks, 1990) Auricular surface (Lovejoy, 1985) Fourth sternal rib end (Krogman and Iscan, 1986) First sternal rib end (Kunos et al., 1999) Epiphysis of the fist sternal head (Kunos et al, 1999) Ectocranial sutures synostosis (Masset, 1989) Dental methods Two-criteria dental method Cementochronology (Standard deviation) Documented age
Table 1
Summary data for forensic evidence and age estimation
STE M 1 year CS
PIL M 7 months CS
JOL M 6 months IS
DEV M 6 months MM
CAR M 5 months CS
BLA M 4 months CS
DER M 2 months MM
Int J Legal Med
l’Archéologie^ in Douai, France, according to the ISO-9001 protocol [23] (Fig. 2). Each tooth was dried with acetone and the roots were embedded in a two-component epoxy resin (Araldite® 2020) and dried in a vacuum chamber. The crowns (not embedded in resin and allowing direct investigations for further study) and the upper third of the root were removed. Five sequential, 100-μm, non-decalcified cross sections were prepared for each tooth from the middle third of the root using a Low Speed Saw Buehler® IsoMet® fitted with a diamondcoated blade. The polishing of both faces was performed to control the thickness and remove cutting marks, and each slice was washed with distilled water. Unstained sections were mounted on slides with Canada balsam and covered with coverslips. Observation was conducted using a Leica® DMEP microscope at ×400 magnification. Segments that showed readable cementum layers were captured as JPEG images with a Leica® DFC 280 fitted on the microscope (Fig. 3). Readings were performed on selected segments in Adobe® Photoshop CS5. The estimated age was obtained by the sum of the average of the counting on each region of interest plus the age of the completion of the root [35]. Three observers (TC, BB, SN) were involved in counting the cementum increments. The multiple operators counted on the same photographs to increase concordance in counts, as advised by Grosskopf and McGlynn [16], who noted the fundamental problem of optical variability of incremental lines during acquisition. The civil age of the individuals was not known to the second and third observers. These data were revealed after the observers achieved the estimations. For cementochronology, an analysis of variance (ANOVA) was performed on StatPlus® (Table 2) for each case to determine the significance of the differences in means between the three observers’ counts. Student’s t test was also used to quantify the statistical significance in cases of divergence. Correlation between civil age and estimates, and standard error and mean absolute error, were calculated.
Results The results of estimated age at death obtained with osteological and dental methods that were made possible by the preservation of human remains are shown in Table 1. Five wellpreserved individuals allowed all the estimation approaches considered. One specimen allowed the authors to apply five techniques and three enabled only two methods, supporting the notion that for bodies that are not well preserved, it is necessary to adopt extra techniques to confirm the estimates. Of all the teeth prepared for histological preparation, the nine canines from the nine individuals, no teeth were discarded and each observer operated on the same five photographs from the same five regions of interest for each individual. Considering that main criticisms against
Int J Legal Med
Fig. 2 Cementochronology procedure for dental preparation certified according to the ISO‐9001 [23]
Int J Legal Med
Fig. 3 Photomicrograph showing cementum annulations in a transverse section of right lower canine from case STE. Twenty-seven incremental lines can be counted in this picture (C cementum, D dentin)
Table 2
cementochronology estimates are regarding the counting phase and consequent observer errors [36], a one-way analysis of variance technique was used to test the equality of the count means from the three observers to assess the homogeneity of the three operators. A value of P
