Dentomaxillofacial Radiology (2004) 33, 233–235 q 2004 The British Institute of Radiology http://dmfr.birjournals.org
RESEARCH
Conventional and digital radiographic methods in the detection of simulated external root resorptions: a comparative study VPD Westphalen1, I Gomes de Moraes2, FH Westphalen1, WD Martins*,1 and PH Couto Souza1 1
Course of Dentistry, PUC – PR, Curitiba, Brasil; 2Faculty of Dentistry, USP, Bauru, Sa˜o Paulo, Brasil
Objectives: The purpose of this study was to evaluate and to compare the efficacy of a conventional and a digital radiographic method in diagnosing simulated external root resorption cavities. Methods: Human dry mandibles containing teeth were covered with bovine muscle slices in order to simulate the soft tissues. Nine teeth from each dental group were investigated. Three periapical radiographs of each tooth were taken in an orthoradial, mesioradial and distoradial aspect using conventional film (Insight Kodak F-speed; Eastman Kodak, Rochester, NY) and a digital sensor (DRS Gnatus System; Gnatus, Ribeira˜o Preto, Brasil). The teeth were extracted using a forceps and had 0.7 mm and 1.0 mm deep cavities prepared on their vestibular, mesial and distal surfaces at the cervical, middle and apical thirds. Following preparation, each tooth was replaced in its alveolus and new radiographs were taken. Three dental professionals, an endodontist, a radiologist and a general practioner, evaluated the images. Results: A larger number of cavities (P , 0.05) were detected by the digital method when compared with the conventional method, for all depths of lesions. Conclusions: The results of this study suggest that the digital radiographic method is more sensitive than conventional radiography to detect simulated external root resorption cavities. Dentomaxillofacial Radiology (2004) 33, 233–235. doi: 10.1259/dmfr/65487937 Keywords: external root resorption; direct digital radiography; dental radiography Introduction External root resorption is a condition associated with a physiological or pathological process, resulting in loss of mineralized tissues like dentin, cementum and alveolar bone.1 Usually this pathology does not present clinical symptoms and it is occasionaly detected during radiographic examination. When the root resorption is detected clinically it could be in such an advanced phase that treatment is impossible, causing loss of the tooth. Early diagnosis is based on the radiographic image of the resorption cavity. However, there is a difference between the radiographic aspect and the histological condition of the affected teeth.2 – 5 Also the accuracy of the radiographic image is limited owing to the superposition of the mineralized structures on the pathological processes and to inherent technical difficulties. The root lesions are radiographically detectable only with a specific diameter and depth.6 *Correspondence to: Dr WD Martins, Av. Repu´blica do Lı´bano, 462 – CEP 82520500 – Jardim Social, Curitiba – PR – Brazil; E-mail:
[email protected] Received 16 June 2003; revised 20 May 2004; accepted 6 June 2004
The present study evaluates and compares the efficacy of conventional periapical radiographs with digital images in the diagnosis of the external root resorptions, considering the size of the cavities when analysed by different examiners. Materials and methods Six dentate dry mandibles were obtained from the Morphology Department, PUC-PR-Brasil. Thirty-six teeth (nine incisors, some of which were central and some lateral, nine cuspids, nine bicuspids and nine molars) were selected by radiographic examination in order to exclude periapical pathoses. The number of selected teeth per mandible varied. Soft tissue simulation was obtained by covering the bone with bovine muscle. The muscle was cut in 1.5 cm strips to simulate the vestibular and lingual gingival tissues, and in 10 mm strips to simulate the cheeks. The following equipment and materials were used for the study.
Comparison of conventional and digital radiographs VPD Westphalen et al
234
(1)
(2) (3)
X-ray generator: Gnatus IntraOs (Gnatus, Sa˜o Paulo, Brasil), 70 kVp; 120 VAC ^ 10%; 50/60 Hz ^ 2 Hz; 6 A; X-ray tube with three electrodes; focal point 0.8 mm; Periapical films: Insightw (Eastman Kodak Co, Rochester, NY), E/F sensitivity, size 2; Digital device: DRS Digital Radiographic Sensor (Gnatus, Ribeira˜o Preto, Brasil and Cygnus Technologies LLC, Arizona, USA), with charge-coupled device (CCD) sensor, size 1, 312 000 points, resolution 10 – 12 lines mm21, 44 mm pixel size, active area 604 mm2.
A positioning device was constructed using two plastic sheets, forming a 908 angle, to keep the radiographic films parallel and close to the radiographed area. The focus – film distance was 30 cm. The exposure times were 0.12 s (conventional X-ray) and 0.10 s (digital). With the films and sensor positioned, initial conventional and digital radiographs were obtained in an orthoradial, mesioradial and distoradial view with a 158 horizontal angle change for the mesial and distal projections. The selected teeth were extracted from the mandibles and three 0.7 mm diameter/0.7 mm depth holes were drilled in each root, using cylindric burs and a drilling machine, with variations in faces and radicular thirds, following a pattern: cervical, middle and apical thirds and buccal, mesial and distal faces. Each tooth type (incisor, cuspic, bicuspid) had a total of 27 holes (9 teeth £ 3 holes per tooth), except for molars, which had 54 holes. The teeth were repositioned in their respective alveolus and radiographs were taken again. The teeth were removed once more and the holes were extended to 1.0 diameter/1.0 mm depth. After reposition in the alveolus, a third series of radiographs was taken. As a result, 27 small and 27 medium cavities were obtained for incisor, cuspid and bicuspid teeth and 54 for molars. The films were processed manually by the time/ temperature method. Three professionals, a radiologist, an endodontist and a general practicioner, examined the digital and conventional images for the presence of the root surface cavities. They were instructed to not manipulate the digital images, but were allowed to enlarge them up to 36%. The conventional images were analysed with a 4 £ magnifying glass in an image viewer with the films collimated by opaque frames. Statistical analysis The proportion of correctly identified lesions was calculated for all variables (imaging type, lesion size, observer). In addition, the difference between the proportions for conventional and digital imaging was also calculated.
Table 1 Comparison between the total number (and proportion) of cavities observed using conventional and digital methods for all observers combined Conventional Correct detection
The total number of images of the small and medium cavities observed by the three examiners can be seen in Table 1. The results showed a higher number of cavities Dentomaxillofacial Radiology
Digital
P-value
653 (0.81)
,0.0001
N ¼ 810 cavities for each imaging type
observed by the examiners when using the digital method (P , 0.0001). The possible false positive results were not considered. Both small and medium cavities were better observed when the digital method was used (P , 0.0001). The number of medium cavities observed was higher than the small for both radiographic methods (P , 0.0001) (Table 2). Table 3 shows the number of radiographic images of the cavities observed by each examiner individually. The results show a significant difference in the total number of cavities observed between the two imaging techniques by the endodontist (P , 0.0001) and by the general practitioner (P , 0.0001), but no significant difference between the techniques for the radiologist. For conventional imaging only, the radiologist saw a significantly higher number of lesions than the other two examiners (P , 0.0001), but there was no difference between the three for the digital images. The number of small cavities observed by the radiologist was significantly higher in the conventional images. On the other hand, the number of small and medium cavities observed by the endodontist and the general practitioner were significantly higher in the digital images (Table 4). Table 2 Comparison between the images of small and medium cavities observed using conventional and digital methods for all observers combined Cavity (number, proportion)
Conventional
Small Medium
235 (0.58) 307 (0.76)
Digital
P-value
291 (0.72) 362 (0.89)
,0.0001 ,0.0001
N ¼ 405 cavities of each size for each imaging type
Table 3 Total number (and proportion) of cavities observed by individual examiners, for conventional and digital methods Examiner
Conventional
Radiologist Endodontist General Practitioner
221 (0.82) 154 (0.57) 167 (0.62)
Digital
P-value
207 (0.77) 221 (0.82) 225 (0.83)
0.1507 ,0.0001 ,0.0001
N ¼ 270 cavities for each imaging type
Table 4 Comparison between the number (and proportion) of cavities observed in each method, considering the variables sizes and examiners Examiner Radiologist
Results
542 (0.67)
Endodontist General Practioner
Cavity Small Medium Small Medium Small Medium
Conventional 100 121 64 90 71 96
(0.74) (0.90) (0.47) (0.67) (0.53) (0.71)
Digital 82 125 104 117 105 120
N ¼ 135 cavities/size category for each imaging type
(0.61) (0.93) (0.77) (0.87) (0.78) (0.89)
P-value 0.0234 0.3776 ,0.0001 ,0.0001 ,0.0001 0.0003
Comparison of conventional and digital radiographs VPD Westphalen et al
Discussion The results of this study showed that the number of cavities detected was higher when the images were observed with the digital radiographic method in comparison with the conventional method. Our findings agree with those of Andreasen et al,2 Chapnick,7 and Goldberg et al,8 who studied the conventional radiographic method, and those of Levander et al,9 Borg et al10 and Clasen,11 comparing conventional and digital images. Borg et al10 produced 1.2 mm diameter/0.6 – 0.9 mm depth defects and compared the conventional radiographic and digital images, using only orthoradial projections. Their results did not show significant differences between the methods. The observers were allowed to modify brightness and density of the digital images. Levander et al,9 using only lower bicuspids and orthogonal and eccentric projections, found no differences between digital and conventional images. In our study, both the small and medium cavities were better observed with the digital method. A possible factor in this result could be the fact that the digital images presented a 36% enlargement in comparison with the conventional images. This enlargement was allowed in order to simulate the real clinical conditions. On the other hand, other types of alterations of the digital image that could interfere with the results were not allowed (contrast,
235
brightness, texture). Clasen11 found similar results, studying central incisors with 0.5 mm, 0.8 mm and 1 mm cavities simulating root resorption. This author also commented on the enlargement of the digital image and observed that the only possible auxiliary to conventional radiographic interpretation are the image viewer and the magnifying glass. The endodontist and the general practitioner found more cavities than the radiologist using the digital method. The results of the radiologist were not significant between the two methods but, when compared with the endodontist and the general practitioner, the radiologist discovered more cavities with the conventional method.
Conclusions Considering the methodology and the results of this study, it was concluded that: (1) the total number of cavities observed with the digital method was higher than with the conventional method, both for the small and medium cavities (P , 0.05); and (2) using the conventional method, the radiologist presented a better capacity of observation, followed by the general practitioner and the endodontist.
References 1. Tronstad L. Root resorption-etiology, terminology and clinical manifestations. Endod Dent Traumatol 1988; 4: 241 – 252. 2. Andreasen FM, Sewerin I, Mandel U, Andreasen JO. Radiographic assessment of simulated root resorption cavities. Endod Dent Traumatol 1987; 3: 21 – 27. 3. Holtzmann DJ, Johnson WT, Southard TE, Khademi JA, Chang PJ, Rivera EM. Storage-phosphor computed radiography versus film radiography in the detection of pathologic periradicular bone loss in cadavers. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998; 86: 90 –97. 4. Ferlini Filho J. Estudo radiogra´fico e microsco´pico das reabsorc¸o˜es radiculares na presenc¸a de periodontites apicais croˆnicas (microscopia o´ptica e eletroˆnica de varredura) Bauru: 1999 (Tese de Doutorado – Faculdade de Odontologia de Bauru, Universidade de Sa˜o Paulo). 5. Laux M, Abbot PV, Pajarola G, Nair PN. Apical inflammatory root resorption: a correlative radiography and histological assessment. Int J Endod 2000; 33: 483– 493. 6. Westphalen VPD. Estudo radiogra´fico das reabsorc¸o˜es radiculares externas em cavidades artificiais de dentes humanos. Floriano´polis:
7. 8. 9. 10.
11.
1998 (Dissertac¸a˜o de Mestrado em Endodontia – Universidade Federal de Santa Catarina). Chapnick L. External root resorption: an experimental radiographic evaluation. Oral Surg Oral Med Oral Pathol 1989; 5: 578 – 582. Goldberg F, De Silvio A, Dreyer C. Radiographic assessment of simulated external root resorption cavities in maxillary incisors. Endod Dent Traumatol 1998; 14: 133 – 136. Levander E, Bajka R, Malmgren O. Early radiographic diagnosis of apical root resorption during orthodontic treatment: a study of maxillary incisors. Eur J Orthod 1998; 20: 57 –63. Borg E, Ka¨llqvist A, Gro¨ndahl K, Gro¨ndahl HG. Film and digital radiography for detection of simulated root resorption cavities. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998; 86: 110 – 114. Clasen FN. Comparac¸a˜o in vitro da efica´cia da radiologia digital direta e da convencional na detecc¸a˜o de reabsorc¸o˜es radiculares externas, em dentes unirradiculares humanos. Sa˜o Paulo: 2001. (Dissertac¸a˜o de Mestrado em Endodontia – Faculdade de Odontologia, Universidade de Sa˜o Paulo).
Dentomaxillofacial Radiology
