10.5005/jp-journals-10011-1060 JIAOMR Comparison of Cephalometric Measurements Obtained with Conventional and Digital Methods and their Reproducibility
ORIGINAL ARTICLE
Comparison of Cephalometric Measurements Obtained with Conventional and Digital Methods and their Reproducibility 1
Anshu Kalra, 2Shirish Goel, 1Manish Thadani, 3RM Shetty, 4Divya Kalra, 5 Savita Lodam
1
Associate Professor, Department of Orthodontics, Chhattisgarh Dental College and Research Institute, Rajnandgaon Chhattisgarh, India 2 Senior Lecturer, Department of Orthodontics, Chhattisgarh Dental College and Research Institute, Rajnandgaon Chhattisgarh, India 3
Reader, Department of Pedodontics, Chhattisgarh Dental College and Research Institute, Rajnandgaon, Chhattisgarh, India
4
Associate Professor and Head, Department of Prosthodontics, Government Dental College, Rohtak, Haryana, India
5
Postgraduate Student, Department of Oral Medicine and Radiology, Chhattisgarh Dental College and Research Institute Rajnandgaon, Chhattisgarh, India Correspondence: Anshu Kalra, Associate Professor, Department of Orthodontics, Chhattisgarh Dental College and Research Institute, Rajnandgaon, Chhattisgarh-491441, India, e-mail:
[email protected] ABSTRACT Aims and objectives: The purpose of the study was to compare the intraobserver reproducibility of cephalometric measurements obtained with digital cephalograms traced with software and equivalent hand-traced conventional cephalograms. Further, the cephalometric measurements obtained with both methods were compared to know any significant differences. Materials and methods: The sample consisted of pretreatment digital and conventional cephalograms of 40 patients (20 males and 20 females). Eleven cephalometric landmarks were identified and 10 measurements calculated by one operator, both manually and with digital tracing software. Intraobserver reliability was assessed for both methods by duplicating the tracings at two weeks interval and using Pearson’s correlation coefficient. Further paired “t” test was used to compare the conventional and digital methods. Results: The analysis of error (correlation coefficient) on both methods showed a high correlation of repeated measures. Results indicate that the reliability of repeated measurements appears to be slightly better with conventional radiographs. In the comparison between two methods, statistically significant differences (p < 0.005) were detected for 5 of the 10 measurements evaluated (Wits, Sn-GoGn, PP-GoGn, U1-L1, L1-GoGn). However, three of these statistically significant results were highly significant (p < 0.001) of Wits, U1-L1, L1-GoGn. Conclusion: Intraobserver reproducibility was found to be better with conventional cephalometric tracings than with monitor displayed digital image tracings. The differences, however, were clinically insignificant. Therefore, computerized cephalometric measurement using direct digital imaging is inherently preferable for its user-friendly and time saving characteristics. Keywords: Digital cephalogram, Conventional cephelogram, Intraobserver reliability, Cephalometric tracings.
INTRODUCTION For a long time cephalometrics has been a very important tool in the hands of orthodontist for diagnosis, treatment planning, treatment evaluation, growth prediction and for research work. Traditional cephalometric analysis performed by tracing radiographic landmarks on acetate overlays and measuring with the use of protractor is very time consuming and has several drawbacks, including a high risk of errors during hand tracing, landmark identification and measurement.1,2 Cephalometric errors can be divided into acquisition, identification and technical measurement errors. Research carried out on conventional cephalometrics proved landmark identification to be the main source of error. 3,4 Among the factors contributing to the identification error include observer experience, landmark identification, density and sharpness of image.5 The rapid evolution of digital radiographic system and digital tracing software is having a deep impact on cephalometrics, slowly replacing traditional hand-tracing methods on hard copies of radiographic films. First generation computer-based analysis systems used digitizer pads for tracing conventional cephalometric
films and software programs to compute the measurements, whereas second generation systems use scanner or digital cameras to export cephalometric images to measurement programs. Recently, third generation systems have been introduced that transmit digital radiographs directly to a computer database through the use of photostimulable phosphor plates, charge-coupled device receptor or direct digital systems. Digital imaging offers several advantages over conventional film-based radiography: a. Faster data processing b. Elimination of chemical and associated environmental hazards c. Elimination of technique-sensitive developing processes d. Reduction of radiation dose e. Digital radiographic images are easy to store and facilitate communication between providers f. Digital archiving is also a valuable method for overcoming the problem of film deterioration. Reproducibility of measurements is a prerequisite for determining the accuracy of any method of analysis. The use of computers in treatment planning is expected to reduce the incidence
Journal of Indian Academy of Oral Medicine and Radiology, October-December 2010;22(4):S9-12
S9
Anshu Kalra et al
of personal errors due to fatigue and provides standardized, fast and effective evaluation with a high rate of reproducibility. Several studies have been undertaken to compare the accuracy of scanned, digitized and digital radiographs with conventional radiographs.6-11 No clear consensus is there as to which of the three methods is best but the first two generation methods are more time consuming and may also introduce magnification errors. This can be overcome through the use of direct digital images. The literature contains only a few studies comparing the accuracy of digital cephalometric measurements with the handtracing method. Therefore, this study was done to evaluate the accuracy of cephalometric measurements obtained with digital tracing software compared with equivalent hand-tracing measurements. Cephalometric measurements instead of landmarks were the focus of this investigation in order to acquire data with direct clinical applications. The literature would benefit from more data with direct clinical applications and an answer to the question of whether a computerized cephalometric analysis generates a diagnostic product equivalent to the conventional hand traced one. MATERIALS AND METHODS Forty patients who reported to the Department of Orthodontics Chhattisgarh Dental College and Research Institute for treatment were selected for study. Their mean age was 17 years and 5 months (range—14 years 2 months to 20 years 1 month). Males and females were equally distributed in the sample.
The conventional and digital lateral cephalometrics images were acquired for each patient. The radiographic unit used for conventional exposures was XTROPAN 2000. The radiographic unit used for digital exposures was ORTHOPHOS XG5 (SIRONA) with the recommended settings of 65 kV and 20 mAS for single exposure. The phosphor coated plate was processed in the image scanner after exposure. All digital images were then compressed in 8-bit jpeg-100 format and imported into the tracing software (NEMOCEPH). An automatic film processor, VELOPEX EXTRA-XE was used to develop the conventional films. All radiographs were traced by a single examiner (AK). A total of 11 cephalometric landmarks (Table 1) and 10 cephalometric measurements (Table 2) commonly used to assess dentofacial relationship were selected. In order to examine intraobserver reproducibility, all tracings were repeated at two weeks interval. Conventional radiographs were traced manually over a view box in dark room using the 4H pencil and cephalometric protractor. Measurements were adjusted to the nearest 0.5 unit. For digital radiographs, landmarks were identified by using a mousecontrolled cursor linked to the tracing software. Digital radiographs were calibrated by the use of ruler incorporated in the cephalostat at the time of radiographic exposure. This avoided any magnification errors in linear measurements. The image enhancement features of the software were used if needed (brightness and contrast). All measurements were automatically calculated by software and rounded to the nearest 0.1mm or 1mm.
Table 1: Definitions of cephalometric landmarks Landmarks
Definition
S N A B Occlusal plane Go Gn Palatal plane U1 L1 Pog
Sella: Center of pituitary fossa Nasion: Most anterior part of frononasal suture Point A: Deepest point of concavity of premaxilla; corresponds to the junction between denture bases and basal bone Point B: posterior point of concavity between chin and superior aspect of mandibular alveolar process A line passing through maximum intercuspation of first molars and second premolars Gonion: most posterior and inferior point on outline of mandibular angle Ganathion: point midway between pogonion and menton on outline of symphysis A line joining ANS and PNS Long axis of the most prominent maxillary incisor Long axis of the most prominent mandibular incisor The most prominent point on the symphysis
Table 2: Definitions of the various angular and linear measurements used in the study
S10
Measurements
Unit
Definition
SNA SNB ANB Wits SN-GoGn PP-GoGn U1-SN U1-L1 L1-GoGn L1-A Pog
Degree Degree Degree mm Degree Degree Degree Degree Degree mm
Angle between SN and NA Angle between SN and NB Angle between AN and NB Linear distance between Point A and Point B parallel to occlusal plane Angle between S-N and Go-Gn Angle between Palatal Plane and Go-Gn Angle between SN and long axis of upper incisor Internal angle between long axis of upper and lower incisors Angle between Go-Gn and long axis of lower incisor Linear perpendicular measurements from tip of lower incisor to A-Pog plane
JAYPEE
Comparison of Cephalometric Measurements Obtained with Conventional and Digital Methods and their Reproducibility Table 3: Intraobserver reproducibility of cephalometric measurements obtained by both methods Conventional radiographs
Digital radiographs
Measurements
Difference between two readings (mean ± SD)
Correlation coefficient
Difference between two readings (mean ± SD)
Correlation coefficient
SNA (0) SNB (0) ANB (0) Wits (mm) SN-GoGn (0) PP-GoGn (0) U1-SN (0) U1-LI (0) L1- GoGn (0) L1-APog (mm)
–0.12 ± 0.13 0.08 ± 0.12 0.10 ± 0.42 –0.32 ± 0.92 0.14 ± 1.04 –0.12 ± 1.24 –1.02 ± 1.86 0.54 ± 1.88 –0.22 ± 1.12 0.32 ± 1.16
0.98 0.98 0.96 0.87 0.96 0.95 0.97 0.92 0.99 0.98
–0.16 ± 0.45 0.12 ± 0.38 –0.14 ± 0.84 –0.42 ± 1.22 0.18 ± 1.24 0.34 ± 1.56 1.28 ± 1.92 –0.64 ± 2.28 –0.24 ± 1.28 0.68 ± 1.14
0.97 0.98 0.95 0.88 0.95 0.92 0.95 0.89 0.97 0.97
RESULTS Intraobserver reproducibility of measurements was evaluated for both conventional and digital radiographs separately by subjecting the readings to the Pearson, correlation coefficient. The results are tabulated in Table 3. Differences in measurements obtained by both methods, i.e. conventional cephalometric measurement and digital cephalometric measurement were compared for statistical significance by subjecting the result to paired “t” test based on equality of variance between two samples. A “p” value of 0.005 was used as the minimal level of statistical significance. Results obtained are tabulated in Table 4. Table 4: Comparison of measurements obtained by conventional cephalometrics and digital cephalometrics Measurements
Difference (Mean + SE)
SNA (0) SNB (0) ANB (0) Wits (mm) SN-GoGn (0) PP-GoGn (0) U1-SN (0) U1-L1 (0) L1-GoGn (0) L1-APog (mm)
0.18 ± 0.16 0.12 ± 0.18 0.26 ± 0.17 0.48 ± 0.21 0.42 ± 0.25 0.56 ± 0.28 0.19 ± 0.26 0.59 ± 0.39 0.89 ± 0.63 0.14 ± 0.19
p-value 0.32 0.38 0.08 0.0005** 0.003* 0.002* 0.23 0.0003** < 0.0001** 0.42
* Statistically significant at p < 0.005 ** Statistically significant at p < 0.001
DISCUSSION Landmark identification is greatly affected by operator’s experience as well as the tracing method. Intraobserver errors are generally found to be less than interobserver errors, so to minimize errors, all tracings and measurements were made by one observer. Results obtained for intraobserver reproducibility for conventional
hand-traced cephalograms are similar to those obtained in other studies. 12-14 The analysis of error (correlation coefficient, Table 3) on conventional radiographs showed a high correlation of repeated measures, meaning that the landmarks were readily identifiable on conventional cephalograms. For the conventional radiographs, the correlation coefficients (r2) of most variables were above 0.95 (high correlation means good reproducibility) with the exception of Wits (0.87) and upper incisor to lower incisor angle (0.92). For the digital method, three readings were below an r2 value of 0.95, but the values were within 1 standard error (SE). The correlation coefficient of Wits (0.88), Palatal plane to Gonion Gnathion plane angle (0.92) and upper incisor to lower incisor plane (0.89) were below 0.95. Most of these readings are in accordance with the study by Santoro, Jayoma and Cangialosi.15 The low value of correlation coefficient for Wits measurement obtained with both methods suggests that the reproducibility problem is inherent in the type of measurement, i.e. a combination of angular and linear reference structures. The unit’s measurement is affected by double images of occlusal surfaces of teeth, interfering with identification of the occlusal plane and also by its small absolute value making it more sensitive. Correlation coefficient values for upper incisor–lower incisor angle with both methods were also low. This is probably because of tracing difficulties of incisor position and variation in incisor angular measurements as described by various authors in their studies.14,16,17 With the digital method, correlation coefficient ratio for palatal plane to Gonion Gnathion plane angle (0.92) was also lower than 0.95. This can be explained by difficulty in locating gonion. This point corresponds to a poorly-defined outline associated with bilateral anatomic structures, located away from mid-sagittal plane and often projecting as a double image on the film. Results indicate that the reliability of repeated measurements appears to be slightly better with conventional radiographs than with monitor displayed images and tracing software, corroborating the findings of Geelen et al13 and Chen et al.12 Since the magnitude of differences was small with both methods, as confirmed in our study, the differences had little clinical significance. Both methods can be safely regarded as reliable in daily orthodontic routine.
Journal of Indian Academy of Oral Medicine and Radiology, October-December 2010;22(4):S9-12
S11
Anshu Kalra et al
Results of comparison between conventional tracing measurements and digital cephalometric measurements are tabulated in Table 4. The statistically significant differences were detected for 5 of the 10 measurements evaluated (Wits, Sn-GoGn, PP-GoGn, U1-L1, L1-GoGn). However, three of these statistically significant results were highly significant (p < 0.001) of Wits, U1-L1, L1-GoGn. These results are probably because of difficulty in landmark identification of occlusal plane and Gonion, as described earlier. Chen et al18 in an article on digitized and hand-traced cephalometric measurements, reported statistically significant differences for dental angular measurements. Accurate location of maxillary and mandibular root apices is sometimes difficult on digital films, but other factors can also cause inconsistencies between digital and manual tracing methods. This explains highly significant statistical results in measurements related to lower incisors. CONCLUSION Intraobserver reproducibility was found to be better with conventional cephalometric tracings than with monitor-displayed digital image tracings. The differences, however, were clinically insignificant because for most values, the standard error of the operator was within one standard unit, and the difference between the means was minimal and clinically acceptable. Therefore, computerized cephalometric measurements using direct digital imaging is inherently preferable for its user-friendly and time saving characteristics. When comparing two methods, the difference between means in individual measurements was below 1mm, 1o or 1%. Statistically significant differences were obtained for measurements of only difficult to locate landmarks. Therefore, direct digital cephalograms can reliably be used for clinical purposes and comparable to conventional cephalograms. REFERENCES 1. Baumrind S, Frantz RC. The reliability of head film measurements 2. Conventional, angular and linear measures. American Journal of Orthodontics 1971;60:505-17. 2. Sandler PJ. Reproducibility of cephalometric measurements. British Journal of Orthodontics 1988;15:105-10. 3. Houston WJB. The analysis of errors in orthodontic measurements. American Journal of Orthodontics 1983;83:382-90.
S12
4. Houston WJB, Maher RE, McElroy D, Sheriff M. Sources of error in measurements from cephalometric radiographs. European Journal of Orthodontics 1986;8:149-51. 5. Bjork A, Solow B. Measurments on radiographs. Journal of Dental Research 1961;41:672-83. 6. Oliver RG. Cephalometric analysis comparing five different methods. British Journal of Orthodontics 1991;18:277-83. 7. Macri V, Wenzel A. Reliability of landmark recording on film and digital lateral cephalograms. European Journal of Orthodontics 1993; 15:137-48. 8. Nimkarn Y, Miles PG. Reliability of computer-generated cephalometrics. International Journal of Adult Orthodontics and Orthognathic surgery 1995;10:43-52. 9. Geelen W, Wenzel A, Gotfredsen E, Kruger M, Hansson L-G. Reproducibility of cephalometric landmarks in conventional film, hardcopy and monitor-displayed images obtained by storage phosphor technique. European Journal of Orthodontics 1998;20:331-40. 10. Chen YJ, Chen SK, Chan HF, Chen KC. Comparison of landmark identification in traditional versus computer-aided digital cephalometry. Angle Orthodontics 2000;70:387-92. 11. Ongkosuwito EM, Katsaros C, van‘t Hof MA, Bodegom JC, KuijpersJagtman AM. The reproducibility of cephalometric measurements: A comparison of analogue and digital methods. European Journal of Orthodontics 2002;24:655-65. 12. Chen YJ, Chen SK, Chan HF, Chen KC. Comparison of landmark identification in traditional versus computer- aided digital cephalometry. Angle Orthodontics 2000;70:387-92. 13. Geelen W, Wenzel A, Gotfredsen E, Kruger M, Hansson LG. Reproducibility of cephalometric landmarks on conventional film, hardcopy, and monitor-displayed images obtained by the storage phosphor technique. European Journal of Orthodontics 1998;20:33140. 14. Gravely JF, Benzies PM. The clinical significance of tracing errors in cephalometry. British Journal of Orthodontics 1974;1:95-101. 15. Margherita Santoro, Karim Jarjoura, Thomas J. Cangialosi. Accuracy of digital and analogue cephalometric measurements assessed with the sandwich technique. American Journal of Orhtodontics and Dentofacial Orthopedics 2006;129:345-51. 16. Baumrind S, Frantz RC. The reliability of head film measurements. Landmark identification. American Journal of Orhtodontics 1971;60: 111-27. 17. Lim KF, Foong KW. Phosphor-stimulated computed cephalometry: Reliability of landmark identification. British Journal of Orthodontics 1997;24:301-08. 18. Chen YJ, Chen SK, Yao JC, Chan HF. The effects of differences in landmark identification on the cephalometric measurements in traditional versus digitized cephalometry. Angle Orthodontics 2004; 74:155-61.
JAYPEE
