• Dr. Shekhar Grover. Post Graduate student, Department of Public Health Dentistry,
  Sudha Rustagi College of Dental Sciences & Research, Faridabad.
  
  
• Dr. Charu Mohan Marya. Professor and Head, Department of Public Health Dentistry,
  Sudha Rustagi College of Dental Sciences & Research, Faridabad.
  
  
• Dr. Avinash J. Professor, Department of Public Health Dentistry, Sudha Rustagi College of Dental Sciences & Research, Faridabad.
  
  
• Dr. Nidhi Pruthi Senior Lecturer, Department of Public Health Dentistry, Sudha Rustagi College of Dental Sciences & Research, Faridabad.
  
  Correspondence concerning this article should be addressed to:
  Dr. Shekhar Grover
  Department of Public Health Dentistry,Sudha Rustagi College of Dental Sciences & Research,Kheri More, Village Bhopani, Old Faridabad,District Faridabad, Haryana (India).Pin - 121001Ph: +919899098070 Email: [email protected]

INTRODUCTION
Dental age determination is required in various clinical and scientific disciplines including paediatric dentistry, orthodontics, archaeology, palaeontology and forensic dentistry (1, 2). In certain communities, the chronological age of living people bears significant importance regarding social benefits, employment and marriage (3). Assessment of tooth development to estimate the age of living subjects has a long history. In industrial sociology, the presence of the first permanent molar was a sign that a child had attained six years of age and such children were condemned to working in the coal mines of the 19th century industrial revolution in England (4).

Individuals may not have accurate information about their date of birth, or they may choose to suppress such information. In such circumstances, age determination technique i.e. estimation of chronological age, may be required (5). The main criteria for forensic age determination in the relevant age group based on odontological examination are tooth eruption and tooth mineralization, both developmental biological features.

Tooth mineralization is evaluated based on what is known as an Orthopantomogram (OPG), a radiograph of the complete dentition. For the evaluation of tooth mineralization, various stages classifications have been put forward (6-8). The most widely used method for the comparison between different populations was first described in 1973 by Demirjian et al (6), although other methods have also been used (3). His method is based on the development of seven left permanent mandibular teeth. This method has been tested in different populations, and has been mostly reported to overestimate the age of an individual (9-16).

In 2001, Willems et al evaluated the accuracy of Demirjian's method in Belgian Caucasian population and modified the scoring system when a significant overestimation was reported (17). This modification has been evaluated among various communities and has been reported to be more accurate than the original method (10, 18, 19). This method has been tested a few times in North Indian children, so little is known about this method's applicability in the same (20). For that reason, the aim of this study was to evaluate the applicability of Willems method for dental age estimation in North Indian children, 6-15 years of age.


MATERIALS AND METHOD
A cross sectional study was carried out by estimating the development of teeth in the Mandibular left permanent teeth (central incisor to second molar) in panoramic radiographs of children aged 6-15 years (with North Indian descent and having parents of same ethnicity). The study involved 215 radiographs. The radiographs were estimated by tracing them by a pencil (by a single examiner) on a translucent paper against a light source and were assessed using Willems method (17) and compared with the chronological age of the child (as told by the parent).

Children showing congenital developmental abnormalities, physically/mentally challenged children, children having systemic diseases or having a gross malocclusion were not included in the study as they may possess tilted teeth, which may fall out the focal trough of the image, leading to bias in age estimation. For any subject with an absent left permanent mandibular tooth, the equivalent tooth on the subject's right was used. Tooth formation is divided into eight stages and criteria for these stages are given for each tooth separately. Each stage of the seven teeth is given score. The sum of score for the seven teeth is referred to a table giving the dental age.

The sample size was calculated to be around 190 from the pilot study. The pilot study was performed over 5 OPGs randomly selected from the Department of Orthodontics and Dentofacial Orthopaedics in Sudha Rustagi College of Dental Sciences & Research, Faridabad. After keeping the confidence interval limits as 95%, with the minimum marginal error to 1% (to estimate the difference of at least 1 year between the chronological and the dental age), the sample size came out to be 188, i.e. approximately 190.

A convenience sampling method was performed to select the panoramic radiographs available in the Department of Pedodontia and Preventive Dentistry and in the Department of Orthodontics and Dentofacial Orthopaedics in Sudha Rustagi College of Dental Sciences & Research, Faridabad. After obtaining all the OPGs, age stratification was performed to segregate the radiographs and the desired number was selected.


STATISTICAL ANALYSIS:
The data was analyzed with the SPSS software (SPSS 11.5). Chronological age and estimated dental age was described by frequency distribution, means and standard deviation in both the genders and compared using Paired 't' test. The difference between the two genders was analyzed with Unpaired 't' test. The level of significance was set to be at 0.05 (*p < 0.05).


RESULTS
The subjects were divided into 10 age groups of 1 year each, from 6 years to 15 years, converting the chronological age (as told by the parent) into a whole number. Figure A gives a multiple bar diagram of distribution of study samples in every age group according to age and sex. Willems method produced a mean over-estimation of 0.36 years in boys and 0.24 years in girls (the mean over-estimation of whole sample was 0.30 years). Table 1 provides the comparison of boys and girls with respect to Chronological age and Dental Age by Unpaired t-test. The t-statistic value between chronological ages of boys and girls came out to be -1.3756 (p = 0.1704), the t-statistic value between Dental Ages of boys and girls came out to be -1.1006 (p = 0.2723), the t-statistic value for chronological age minus dental age for both boys and girls came out to be 2.0676 (p = 0.0399*) (significant). Figure B shows the correlation between chronological age and dental age minus chronological age among Boys. Figure C shows the correlation between chronological age and dental age minus chronological age among Girls. Table 2 shows the overall comparison between the Chronological age and the Dental Age (Willems method).

DISCUSSION
There are often seen large differences in growth and development rates among children of same chronological age. This is why biological age is defined, and it remarks different stages in the child's development and maturity, whereas chronological age only roughly estimates the child's maturity (21). Teeth are one of the key systems in the body, and their degree of development is used as one of the indices of biological age. Most studies conducted after 1960 base dental age on the degree of development of the whole tooth, and not upon tooth eruption, and are thus considered reliable in determining dental age. The study of morphological parameters of teeth on dental x-ray of children is more reliable than most other methods for age estimation and is most commonly used to determine age in living humans. OPGs are considered as the best tool for age estimation in children because intraoral radiography is difficult to obtain in children without image distortion (22). A total of 215 OPGs were investigated (aged 6-15 years) in the present study, 102 males (48%) and 113 females (52%). Ethnic uniformity of the study sample was considered as development of teeth may vary among populations (23).

The mild overestimation of age by Willems method seen in the present results is supported by studies in the past conducted over different populations owing to its potential applicability (10, 18, 19, 24). Being a modification of the Demirjian method, Willems method is more reliable, as Demirjian method was framed almost 30 years ago and there may be a difference in the present day comparisons due to positive secular trends. But statistically, these findings were significant (Table 2), supporting the need for revision of Willems Method for application in North Indian Children. When comparison among genders is done, Willems method is seen to be more accurate for girls (Table 1), significantly. However, this is contrary to other studies reported where Willems method favours boys (10, 24).

When scatter plots were created between overestimation of age and chronological age (among both the genders), there were linear lines showing a strong negative correlation between both the variables (Figure 2 and Figure 3). This implied that the younger age groups had a higher amount of overestimation than the older age groups. This may be explained by the fact that acceleration of growth reduces as the age advances and body development becomes gradually stabilized, as the growth reaches maturity. This finding is comparable to results of Leurs et al (study done in Dutch children, 2005) (25), where a significant difference was seen in the 5-10 years age groups. Some other studies predict a variable result, showing more overestimation in higher age groups (26) or in different age categories (11, 18). This varying degree of overestimation of age among both boys and girls indicate that dental growth is not a steady and uniform process, but is possibly associated with Para-pubertal speed fluctuations.

Assessing OPGs was a major strength of the study as it is a non-invasive approach and hence readily acceptable. Also, teeth were assessed for development and not for eruption which accounts for low variability due to local and environmental factors. But, comparison among age categories would have cleared the picture of growth trends in North Indian children.


CONCLUSION:
Dental age determination is required in various clinical and scientific disciplines: pediatric dentistry, orthodontics, archeology, paleodontology and forensic dentistry. The present study assessed tooth development in seven left mandibular teeth from 215 orthopantomograms (OPGs) belonging to healthy children aged 6 - 15 years and were staged according to Willems scale.

The Willems method produced a significant overestimation of age. The mean overestimation for was less for girls as compared to boys and the results were statistically significant. The younger age groups had a higher amount of overestimation than the older age groups, but overestimation of the actual chronological age supports the need for refinement of the population-specific standards in Willems method, for further application.