ORIGINAL RESEARCH Year : 2020  Volume : 31  Issue : 2  Page : 186190 Assessment of validity and reliability of Kvaal's method for age estimation among a population sample – A retrospective study Prashant Kumar Chandan^{1}, Karandeep Singh Arora^{2}, Mayank Das^{3}, Prabhpreet Kaur^{4}, Shreeyam Mohaptra^{5}, Shubhangi Pareek^{6}, ^{1} Department of Oral Medicine and Radiology, Rajasthan Dental College and Hospital, Jaipur, Rajasthan, India ^{2} Department of Oral Medicine and Radiology, Bhojia Dental College and Hospital, Bhud, Baddi, HP, India ^{3} Department of Oral Medicine and Radiology, NIMS Dental College and Hospital, Jaipur, Rajasthan, India ^{4} Department of Oral Pathology and Microbiology, Genesis Institute of Dental Sciences and Research, Ferozepur, Punjab, India ^{5} Department of Oral Medicine and Radiology, SCB Dental College, Cuttack, Odisha, India ^{6} Department of Oral Pathology and Microbiology, Institute of Dental Sciences, Bhubaneswar, Odisha, India Correspondence Address: Objectives: The present research was carried out with an aim to estimate and correlate chronological age and estimated age by Kvaal's Method in North Indian population using digital panoramic radiography. Further, the study was an attempt to evaluate the reliability of Kvaal's method for age estimation and to arrive at a populationspecific regression equation. Materials and Method: One hundred digital orthopantomograms of participants aged between 20 and 70 years of age were selected. The evaluation of 6 teeth according to Kvaal's method was carried out using measure tool of Sidexis Software (provided by the manufacturer). Correlation coefficient was carried out between chronological age and estimated age and further regression analysis was carried out for obtaining a population specific regression equation. Results: It was observed that coefficient of determination (R^{2}) is highest (0.223) for mandibular canine which indicates that age can be better estimated with this particular tooth. Conclusion: It was observed that large variations between the chronological and estimated age were not found. Also, Kvaal's formula proved accurate for estimating age using all the six selected teeth. Thus, it was concluded that mandibular canine would be the better tooth for age estimation, followed by maxillary second premolar and maxillary three teeth taken together.
Introduction Age is one of the essential factor in establishing the identity of a person. Age can be estimated in different ways such as chronological age, skeletal age, and dental age.[1] Age estimation has become increasingly important in forensic sciences, and even from the earliest times of human history, teeth were considered to be an indicator of age.[2] Dental age estimation has gained acceptance because it is less variable when compared with other skeletal and sexual maturity indicators.[3] Dental age estimation is based on morphological, histological, biochemical, and radiological assessment of teeth.[4] Age can be estimated by determining the degenerative changes, in adult dentition these includes color, attrition, and periodontal attachment level. Color change is highly variable and is closely related to diet and oral hygiene.[5] In 1995, Kvaal et al. reported a method of age estimation from intraoral periapical radiographs using the radiographic parameters. Study conducted by various authors on age estimation was reported using the same method on the orthopantomographs as well.[2],[3] The present study aimed to estimate and correlate chronological age and estimated age by Kvaal's Method in North Indian population using digital panoramic radiography. Further the study was an attempt to evaluate the reliability of Kvaal's method for age estimation and to derive populationspecific regression equations. Materials and Method The study was undertaken after prior permission from Ethical committee of the institute. A total of One hundred orthopantomograms (OPGs) were selected from the database of the department of Oral Medicine and Radiology, acquired during the period between December 2013 and October 2015. The OPGs selected were those who had complement of total six teeth either right or left side i.e. maxillary central incisor, maxillary lateral incisor, maxillary second bicuspid, mandibular lateral incisor, mandibular canine, and mandibular first bicuspid that are free from morphological abnormalities and were completely erupted and also the detailed history and authentic date of birth proof was available. OPGs having one of the six required teeth missing or impacted or the history of any case revealed systemic disorders like hormonal deficiencies, renal diseases, blood dyscrasias, cardiovascular diseases, and syndromeassociated diseases were excluded from the study. Also, if the teeth to be analyzed were carious, restored either in the crown or root or had prosthesis, pulp stones or pulpal pathologies, malaligned teeth or teeth with wasting diseases (attrition/abrasion/erosion) were excluded from the study. The radiographs those fulfilled the above criteria included 62 males and 38 females aged between 20 and 70 years of age. The OPGs were acquired using Sirona Orthophos XG (Manufactured: Sirona, Germany). Two observers carried out evaluations of radiographs, one being a senior dentomaxillofacial radiologist and second being a junior/trainee dentomaxillofacial radiologist of the same department. The measurements were carried out on the OPGs for all six types of teeth using Sidexis digital software (provided by the manufacturer) using the measurementaiding tool, which gave options of adjusting the brightness and contrast along with enlargement. Using the mousedriven cursor, the reference points on the images of the teeth were defined [Figure 1]. The unit of all measurements was set to be calculated in millimeters. First, the measurements were obtained for the maximum: tooth length, pulp length, and root length on mesial surface from cementenamel junction to root apex. Second, the measurement were obtained for the pulp and root width: at level CEJ (i.e. a); at level mid root length (i.e. c); and, at level of midpoint between (a) and (c) [i.e. b] [Figure 1].{Figure 1} Ratios between the length and width measurements of the same tooth were calculated in order to avoid measurement errors due to differences in magnification of the image on the radiograph. The ratios calculated were: tooth/root length (T); pulp/root length (P); pulp/tooth length (R); and, pulp/root width at three different levels a, b, and c. Mean values of all ratios (M) as first predictor and mean value of width ratios from levels b and c (W); mean value of length ratios P and R (L) were calculated. The difference (WL) being the second predictor. Regression formulae based on statistical analysis were calculated and different regression formulae for all six teeth i.e. three maxillary teeth and three mandibular teeth were derived and age was estimated. The estimated age was then corelated with the chronological age of the patient. Statistical analysis Age was estimated for all the participants by regression using two predictors, where “M” was the first predictor and “WL” was the second predictor. Prior to running the regression, corelation was carried out to find the relationship between age and the variables [Table 1]. After corelation, regression analysis [Table 2] was carried out which helped to estimate change in one of the variables (i.e. age) when there is a change in the other variables (i.e. M or WL). The chronological age was compared to the estimated age using student's ttest and the regression equation was derived using regression analysis. All the statistical analysis was performed using SPSS software (v19).{Table 1}{Table 2} Results The mean age range of participants was 34.35 years for male and 30.08 years for female. The correlation coefficient between all the six teeth and the ratio of measurement i.e. P, T, R, A, B, C, M, W, L, and WL, was carried out for determining the relation between the variable which has been listed in [Table 1]. After comparison of the chronological age and estimated age using paired ttest it was observed that there was no significant difference between the mean chronological age and mean estimated age in upper central incisor, upper lateral incisor, upper second premolar, lower lateral incisor, lower canine, lower first premolar, upper three teeth, lower three teeth, and all six teeth taken together. (P > 0.05) [Table 3].{Table 3} From the result of regression analysis, it was found that coefficient of determination (R2) is highest (0.223) for mandibular canine that means age could be better estimated with this particular tooth when “M” [Figure 2] and “WL” [Figure 3] are considered as first and second predictor for age. Here, both “M” and “WL” were found to be significant predictor.{Figure 2}{Figure 3} Next, coefficient of determination (R2) was 0.102 for maxillary second premolar which indicates the age could be next better estimated with this particular tooth. Here, only “WL” [Figure 4] were found to be significant predictor. Coefficient of determination (R2) for maxillary three teeth (0.080) was higher when compared to mandibular three teeth (0.046), which indicates that age could be next better estimated with maxillary three teeth. Here, “M” [Figure 5] and “WL” [Figure 6] were found to be significant predictor. Coefficient of determination (R2) was 0.048 for all six teeth. Here, both “M” and “WL” were found to be significant predictor. Coefficient of determination (R2) was very low for mandibular lateral incisor (0.029) and maxillary central incisor (0.016).{Figure 4}{Figure 5}{Figure 6} Therefore, from the present study it was observed that all the teeth are a reliable indicator of age and mandibular canine being the better tooth followed by maxillary second premolar and maxillary three teeth taken together. Discussion In our study, we had used digital OPGs for estimation of age. Studies conducted by Bosman N et al.[5] and Limdiwala PG et al.[1] Singaraju S et al.[6] also used OPGs for obtaining the regression formulae to estimate age whereas studies conducted by Patil SK et al.,[7] Kvaal et al.[8] and Sharma R et al.[9] used intraoral periapical radiographs. In our study specifically OPGs were selected because of the possibility of evaluation of all the teeth along with alveolar bone in both jaws and also that multiple measurements were possible on the same radiograph at the same time. Furthermore, panoramic radiography is a standard technique with high reproducibility, while the acceptability of intraoral radiographs is dependent on the techniques used and the practical training of the personnel. KanchanTalrejaPet al.[10] conducted a study on digital radiographs of 100 Indians, which were acquired using the conventional paralleling technique (n = 47) and bisecting angle technique (n = 53), the latter being the prevalent method of periapical radiography in India. Pulp and tooth lengths and widths were measured (using commercially available computer software) and their ratios substituted in Kvaal's formulae; also, populationspecific formulae were developed by them using principal component regression analyses. The average errors of age estimation were found to be 18–20 years for the paralleling and 19–21 years for the bisecting angle technique; estimates in both samples of radiographs were significantly different from actual age (P < 0.001). The Indian formulae produced smaller errors for both samples (11–14 years), an improvement over Kvaal's formulae. Large errors from Kvaal's formulae may owe primarily to variation in the rate of secondary dentinal deposition in Indians influenced both by environmental and genetic variation. In our study, mandibular canine was the strongest indicator for age estimation because of highest coefficient of determination (R2 = 0.223). Thus, age could be estimated with reliability using this particular tooth followed by maxillary second premolar (R2 = 0.102) and maxillary three teeth taken together (R2 = 0.080). In Kvaal et al.,[8] regression formula derived for all the six teeth taken together, substituting “M” and “WL” showed significant results with coefficient of determination being the strongest (R2 = 0.76) followed by maxillary three teeth (R2 = 0.74) and mandibular three teeth (R2 = 0.71) taken together. A study conducted by Limdiwala PG et al.,[11] concluded that three maxillary teeth together were the strongest predictor (R2 = 0.3) followed by all six teeth (R2 = 0.29) and mandibular three teeth taken together (R2 = 0.14). The reason behind such variations between studies could be that both length and width measurements have significant influence on age estimations. Stronger correlation with age is found by employing the mean values of all ratios, as it is an expression of the overall size of the pulp. The size of the pulp is governed by a lot of factors including the rate of secondary dentin deposition, hereditary, and genetic influences. So, every tooth has different coefficient of determination (R2) as seen in similar studies. Our study also revealed that maxillary central incisor was the weakest indicator for age because of low coefficient of determination (R2 = 0.016) whereas M and WL were found to be nonsignificant (P > 0.05) followed by mandibular lateral incisor (R2 = 0.029) and mandibular three teeth (R2 = 0.046). In a study conducted by Chandramala R et al.[3] which included three mandibular teeth, observed that they were the weakest indicator for age because of low coefficient of determination (R2 = 0.011) followed by three maxillary three (R2 = 0.020) and maxillary second premolar (R2 = 0.031). Another study conducted by KanchanTalrejaPet al.[10] in which mandibular first premolar (R2 = 0.18) was the weakest indicator for age followed by maxillary lateral incisor (R2 = 0.19), mandibular lateral incisor (R2 = 0.21). In our study, we observed that age could be better estimated with mandibular canine when “M” and “WL” are considered as first and second predictor for age. Here, both “M” and “WL” were found to be significant predictor (P < 0.05) whereas a study conducted by Chandramala R et al. both “M” and “WL” were found to be significant predictor (P < 0.05) using maxillary second premolar. In our study, mean chronological age was compared with mean estimated age and the results showed no significant difference between mean actual age and mean estimated age (P > 0.05). Similar result was found between chronological age and estimated age in studies conducted by Bosmans N et al.[5] and Agarwal N et al.[12] (P > 0.05). The result of our study reveals that Kvaal's technique is a conformable and reproducible method of age estimation in adults. The estimated age was found to be close to the chronological age for all the study participants thus pronouncing it an ideal technique. However, a few studies on Indian population estimated age using Kavaal's technique with a large error margin, which necessitates the need of more studies with a larger sample size and diverse ethnic and geographical origin (owing to the vast diverse ethnic groups in India) which may help in arriving at a region wise accurate population specific formulae. Conclusion The present study was an attempt to assess the accuracy of Kvaal's method for age estimation of North Indian population and arriving at populationspecific regression equations in the event the original formulae resulted in large errors. However, the study results revealed that large variations between the chronological and estimated age were not found. Also, Kvaal's formula proved accurate for estimating age using all the six selected teeth. This validates the use of this simple method for age estimation in adults. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest. References


