Our aim was to analyze the correlation between growth status in height and chronological age, carpal maturation, cervical maturation, and dental maturation, and assess the diagnostic performance of Demirjian’s dental maturation as an indicator of the pubertal growth spurt, through a retrospective longitudinal study.
Records of 60 Canadian patients obtained from the Burlington Growth Centre, which included height and weight charts and a set of x-rays at 6 points in time, were analyzed. The images at each point in time included 1 hand and wrist radiograph, a lateral cephalometric x-ray, and one 45° oblique cephalometric radiograph of each side, which were analyzed using the methods of Fishman, Baccetti, and Demirjian on the mandibular left and right second molars, respectively. The onset of the pubertal growth peak in height (distance to growth peak [DGP]) was identified, and the correlation between methods with DGP was assessed.
High levels of correlation were obtained between the methods of Fishman, Baccetti, and Demirjian with DGP. The cutoff point between prepubertal and postpubertal stages was F stage for women and G stage for men, with statistically significant levels of sensitivity and specificity for the test.
The use of the method of Demirjian applied to mandibular second molars is plausible as a predictor of the occurrence of the DGP for the studied population.
Serial records of orthodontically untreated patients from Burlington Growth Centre were assessed.
We assessed dental maturation as a growth indicator relative to other indicators.
Indicators were analyzed according to age and distance in years to pubertal growth peak in height.
High levels of correlation were observed between pubertal growth peak in height and all analyzed growth indicators.
The practice of dentofacial orthopedic or orthodontic treatments in growing patients usually involves the assessment of the developmental stage of the patient to be treated. It is considered an important aspect regarding the efficiency and effectiveness of dentofacial orthopedic procedures. Treatment modalities may vary according to the developmental stage of a patient. Prepubertal intervention aiming at midfacial orthopedic treatment allows favorable outcomes in Class II and Class III malocclusions as well as in transverse maxillary deficiency. , Moreover, it has been described that clinicians may take advantage of the pubertal growth spurt when dentofacial orthopedic treatments are carried out in Class II malocclusions.
Many techniques focus on the assessment of development and particularly, on the detection of the pubertal growth peak, including the observation of secondary sexual characteristics, and serial height and weight recordings. , An accurate method determining the occurrence of the pubertal growth spurt is the use of growth and height charts. These charts represent concrete records of the developmental process throughout time, which allows a retrospective detection of the occurrence of the pubertal growth spurt. In addition, age has been reported as a valuable developmental indicator, in association with other maturation indicators in the context of facial growth, but it becomes less valid in early- or late-maturing people.
Among the radiologic indicators of skeletal maturation, hand and wrist radiographs with carpal x-ray analysis are considered a reliable method for maturational assessment of an individual regarding his growth status, which can be analyzed by means of several methods. Cervical maturation evaluation on lateral cephalometric x-rays is another method , that appears to be efficient regarding the detection of the peak of mandibular development. It has a good correlation with carpal analysis, and is readily available among regular orthodontic records.
Dental maturation has been presented as an alternative for skeletal maturation assessment. , It can be assessed through the analysis of eruption, or tooth calcification process. It displays a good correlation with other growth indicators like hand-wrist and cervical vertebral maturation , , , in cross-sectional studies. However, when evaluated as a diagnostic tool relative to the actual peak in mandibular growth in a longitudinal study, its performance was not as strong to recommend its use for orthopedic treatment planning if mandibular growth peak is to be identified.
Currently, the most common maturity indicators used in orthodontics are the hand and wrist radiograph and the assessment of the cervical vertebral maturation with a cephalometric x-ray, but they are not usually carried out by the general dental practitioner or pedodontists.
It has been proposed that dental maturation can be measured using panoramic radiographs , or retroalveolar x-rays for single tooth evaluations, which are widely used in dentistry, potentially facilitating timely referral of a patient to a specialist with malocclusions of skeletal origin. However, longitudinal studies that correlate dental maturation with skeletal development measured through growth in standing height have not been reported. This information could allow not only timely referral in an orthodontic context but also in general health care for those conditions requiring growth monitoring.
Although some studies report a correlation between dental maturation, and hand-wrist or cervical maturation as development indicators, , , , , , the performance of dental maturation has been assessed based on its correlation with other growth indicators, and relative to the occurrence of mandibular growth peak, but not regarding the actual occurrence of the peak in somatic growth. It then becomes relevant to identify the level of correlation between dental development and the occurrence of the growth peak in height, as well as its diagnostic performance identifying its onset.
This study aimed at establishing the level of correlation between growth status in height and different growth indicators as the hand-wrist radiograph, cervical vertebrae, dental maturation methods, and chronological age, and to determine the diagnostic performance of radiographic dental maturation assessment using Demirjian’s method applied to the mandibular second molar as an indicator of the pubertal growth peak. A historical cohort study was carried out using a sample of subjects from the Burlington Growth Centre of the University of Toronto.
Material and methods
A longitudinal historic-cohort study was performed, using serial records from the C-6 group from the Burlington Growth Centre of the University of Toronto, Canada. These records included only orthodontically untreated subjects.
The sample size was defined based on a previous study by Cisternas et al. The sample size was calculated as 30 men and 30 women, aiming at an 80% of statistical power, and an alpha level of 5% to detect a Spearman’s correlation coefficient above 0.6 between cervical vertebral maturation and dental maturation observed on mandibular second molars, using 6 observations in each individual.
Serial records of 60 subjects (30 of each sex) were included. Each spanning 14 years, taken at age 6, 9, 12, 14, 16, 18, and 20 years between 1952 and 1970. One hand and wrist radiograph, 1 cephalometric radiograph, and one 45° oblique radiograph from each side were analyzed from each patient at every time point. The annual records of height and weight between the age of 3 and 18 years were also included.
Each image obtained from the samples was analyzed to assess skeletal or dental maturation. Hand and wrist radiographs were analyzed with the method of Fishman et al, cephalometric x-rays with that of Baccetti et al, and 45° oblique radiographs with that of Demirjian et al applied to the mandibular second molar ( Fig 1 ). If the crypt of a tooth was not observed, it was labeled as stage 0. Three researchers analyzed the radiographic records. One analyzed the hand and wrist radiographs (F.S.), another analyzed the lateral head films (J.M.), and the third researcher evaluated the oblique cephalometric radiographs (A.C.). Dental maturation was carried out with Demirjian’s method on the left side, and on the right side for the diagnostic performance analysis. Kappa values were recorded for each researcher.
For each patient, the historic height chart data was analyzed and tabulated on a Microsoft Excel spreadsheet, defining the onset of the pubertal peak in growth as the height record preceding the greatest increase in height between 2 height records in time. Once identified, the variable distance to growth peak (DGP) was established, to correlate the actual record of growth in height, with the different methods of developmental assessment.
Statistical analysis was performed using Stata software (version 12.1, Stata Corporation, College Station, TX).
Each indicator was analyzed based on its behavior in time, according to the patient’s age and distance in years to the pubertal growth spurt, for the sample in general and each sex. The correlation between them was analyzed using the Spearman correlation analysis, as well as the concordance between tooth maturation stages and the hand and wrist maturation.
A diagnostic analysis of tooth maturation was carried out with a multilevel logistic regression, to determine the probability that the pubertal growth spurt occurred, as the mandibular second molar of an individual was in a certain stage of development. In addition, receiver operating characteristic (ROC) curves for tooth maturation were plotted to obtain the discriminative ability of the test. The mandibular right and left second molars were compared to identify the most accurate side. Sensitivity, specificity, predictive values, and likelihood ratios for tooth maturation as a diagnostic test were calculated. From these values, the cutoff point of greatest specificity was determined, to identify the closest stage to the occurrence of the pubertal growth spurt.
The calibration process yielded kappa indexes for each maturational analysis. Calculated kappa values were 0.82 for hand and wrist skeletal maturation as well as for cervical maturation method and 1 for dental maturation.
The average onset of the pubertal growth spurt was observed at a mean age of 10.83 years in females, and 13.13 years in males. Distribution of the different stages of maturation indicators of the hand and wrist, cervical and dental development according to age are detailed in Tables I–III , respectively. In Figures 2-4 the distribution of the skeletal maturity stages can be observed for the 3 different methods. For carpal maturation, the closest stage to the peak was skeletal maturity indicator (SMI4) for both sexes with an average of 0.14 years postpeak in men and 0.66 years postpeak in women. For cervical maturation, the closest stage was a cervical vertebral stage (CVS3) for both sexes, with an average of 0.56 years postpeak in women and 0.12 years postpeak in men.
|Stage||Sex||Mean||SD||Median||Interquartile range||Minimum||Maximum||Frequency||Percentage, %|
|Stage||Sex||Mean||SD||Median||Interquartile range||Minimum||Maximum||Frequency||Percentage, %|