Predicting fractures using trabecular patterns on panoramic radiographs.

OBJECTIVES: The observer score of the trabecular pattern on panoramic radiographs is known to be a strong predictor of bone fractures. The aim of this study was to enhance the predictive power of panoramic radiographs by means of texture analysis methods. MATERIAL AND METHODS: The study followed 304 postmenopausal women during 26 years. At the beginning of the study, panoramic radiographs were obtained. One observer assessed the trabecular pattern in the premolar region as dense, sparse, or alternating dense and sparse. In addition, on each radiograph, a region of interest was selected in the molar/premolar region and analyzed with texture analysis procedures. During 26 years of follow-up, 115 women suffered a fracture of the hip, spine, leg, or arm. Logistic regression was applied to test the predictive power of various variables with respect to fractures. RESULTS: Of all variables, the observer score of the trabecular pattern correlated strongest with the occurrence of fractures. By itself, the score yielded an ROC curve with an area of 0.80 under the curve. Combining the observer score with the texture analysis features increased the area under the ROC curve to 0.85. CONCLUSIONS: The trabecular pattern on panoramic radiographs provides a strong predictor of fractures, at least for postmenopausal women. The assessment by an observer combined with texture analysis procedures yields a predictive power that parallels best known predictions in literature. CLINICAL RELEVANCE: This study illustrates that panoramic radiographs are state of the art predictors of postcranial fractures.

A new classification of peri-implant bone morphology: a radiographic study of patients with lower implant-supported mandibular overdentures.

OBJECTIVE: This study aimed to classify peri-implant bone defects (PIBDs) on the basis of their radiographic appearance in a cohort of patients with lower implant-supported overdentures. MATERIALS AND METHODS: Eighty-three patients with lower implant-supported overdentures were recruited to participate in the study, which was approved by the ethics committee of Ignatius Teaching Hospital. Details about the patients' smoking habits were recorded. The participants had a total of 224 implants involving 3214 implant sites. The mean observation time of the subjects was 10.7 years. Panoramic radiographs of all sites were evaluated in duplicate (first evaluation [t1], second evaluation [t2]) during 2 weeks by one experienced observer. PIBDs were classified into the following types: saucer-shaped, wedge-shaped, flat, undercut, and slit-like bone defects. Intra-examiner agreement was tested using crosstabs and Cohen's kappa coefficient. The association of PIBD type with gender, time after implant placement, smoking, and treatment strategy was investigated using multivariate test of independence on the basis of spatial signs. RESULTS: Intra-observer reliability was moderate (κ = 0.51). Saucer-shaped defects were the most frequent (42.8% [t1] and 44.6% [t2]), followed by wedge-shaped (26.0% [t1] and 27.4% [t2]), flat (10.7% [t2] and 17.7% [t1]), undercut (8.8% [t1] and 11.9% [t2]), and finally slit-like defects (4.7% [t1] and 5.4% [t2]). Peri-implant bone defects morphology was significantly associated with gender, smoking, and treatment strategy. CONCLUSION: The morphology of PIBDs can be classified into five meaningful classes, as opposed to the two described in the literature.

Detecting bone loss along dental implants by subtraction of panoramic radiographs.

OBJECTIVES: To test an extended subtraction method for measuring changes in bone along dental implants depicted on non-standardized panoramic radiographs. MATERIAL AND METHODS: The measuring method consisted of a subtraction module extended with modules correcting gray values, correcting geometry and defining the region of interest mesial and distal of dental implants. It was applied to an archive of panoramic radiographs of implant patients who had been monitored up to 16 years. RESULTS: Significant loss of gray value was demonstrated at a constant rate of 0.6 units per month throughout the study. This indicates gradual bone loss on the mesial and/or distal sides of the implants. Females were found to lose bone at a higher rate than males. Smokers and non-smokers, and various implant strategies yielded the same rate of bone loss. CONCLUSIONS: The measuring method is a helpful tool to monitor changes around implants even when non-standardized radiographs are being used. Whereas changes of marginal bone level are increased five to eight times during the first year after surgery, the present method shows gradual loss of bone during 15 years after surgery.

Changing trabecular patterns in panoramic radiographs of Swedish women during 25 years of follow-up.

OBJECTIVES: The radiographic trabecular pattern on dental radiographs may be used to predict fractures. The aim of this study was to analyze longitudinal changes in the mandibles of 145 females between 1980 and 2005. METHODS: Panoramic radiographs were obtained in 1980 and 2005. On 290 radiographs, regions of interest (ROIs) were selected in the ramus, angle and body. In all ROIs, the orientation was measured in 36 directions with the line frequency deviation method. The effects of ageing were analyzed for the fracture and the non-fracture groups separately. RESULTS: During the follow-up, 61 females suffered fractures of the hip, wrist, spine, leg or arm. The fracture and non-fracture groups displayed dissimilar age changes in each investigated ROI. All significant changes pertained to increasing values of line frequency deviation. With increasing age, the trabecular network in the mandible lost details and the trabeculae became more aligned in their main direction. In the "ramus", the alignment was to the 110-120˚ axis, parallel to the posterior and anterior ramus border. In the "angle", the alignment was to the 135-150˚ axis, parallel to the oblique line, and in the "body" ROI to the 150-175˚ direction, approximately parallel to the occlusal plane and inferior cortex. CONCLUSION: Most changes were consistent with the notion that the bone aged less severely in the non-fracture group. In the fracture group, the findings indicate that bone loss leads to redistribution of the remaining bone tissue in such a way that the trabeculae are accentuated perpendicular to the principal loading.

Association Between Peri-implant Bone Morphology and Marginal Bone Loss: A Retrospective Study on Implant-Supported Mandibular Overdentures.

PURPOSE: The present study aimed to explore the association between marginal bone loss and type of peri-implant bony defect determined using a new peri-implant bony defect classification system. MATERIALS AND METHODS: A total of 110 patients with implant-supported mandibular overdentures were involved. Clinical information was collected, including gender, age, smoking habit, and the overdenture attachment system used. Peri-implant bony defect types and marginal distances (ie, distance between the marginal bone level and the top of the implant shoulder) of all sites were identified on panoramic radiographs by a single experienced observer. The associations between marginal distance and peri-implant bony defect type, gender, age, smoking habit, attachment system, and time after implantation were investigated using marginal generalized linear models and regression analysis. RESULTS: A total of 83 participants were included in the final sample with a total of 224 implants involving 3,124 implant sites. The mean observation time was 10.7 years. All peri-implant bony defect types except Type 5 (slit-like) were significantly related to marginal distance in all models (P < .01). Smoking and time after implantation were significantly related to marginal distance while gender, age, and the overdenture attachment system used were not. CONCLUSIONS: The peri-implant bony defect type, determined using the new classification system, is associated with the extent of marginal bone loss.

Implant-supported mandibular overdentures can minimize mandibular bone resorption in edentulous patients: results of a long-term radiologic evaluation.

PURPOSE: It has been suggested that functional loading and light irritative stimuli could lead to changes in bone architecture, shape, and volume, and that by placing implants in the edentulous mandible and subsequently loading them, functional conditions could be created to limit bone resorption or even stimulate bone apposition (the latter was reported only for fixed implant-supported prosthetic reconstructions) in the distal area of the mandibular osseous crest. The aim of this study was to radiographically assess the bone height changes in the posterior area of the mandible after implant placement and loading with an overdenture on two or four implants over a mean follow-up period of 10.5 years. MATERIALS AND METHODS: Panoramic radiographs were taken of 82 totally edentulous patients before implant placement and at repeated follow-up intervals spread over a mean observation time of 10.5 years. All patients received an implant-supported overdenture as prosthetic treatment. The mandibular bone height in the distal part of the mandible was measured on each of the available radiographs and the initial, intermediate, and final values were compared. RESULTS: A mean mandibular bone height reduction of 0.5 mm was measured. CONCLUSION: No clinically relevant difference was found between the posterior mandible height before implant placement and at follow-up after functional loading with an implant-supported mandibular overdenture.

Radiographic features of mandibular trabecular bone structure in hypodontia.

PURPOSE: Radiographic parameters of mandibular trabecular bone structure between 67 subjects having hypodontia and those without were studied on digital panoramic radiographs. MATERIALS AND METHODS: Three regions of interest (ROI) were defined: the ascending ramus, apical of the mandibular molar and mesial of the first mandibular molar. The effects of the presence of hypodontia and the ROI on the mandibular trabecular bone structure were tested for statistical significance by means of multivariate analysis. RESULTS: Radiographic parameters of trabecular bone architecture were found to differ between various regions of the mandible (p = 0.000), but not between the group of hypodontia subjects and their controls (p = 0.23). There was no interaction effect between the ROIs and the two groups (p = 0.79). For people having hypodontia, some directional parameters of trabecular bone have a reverse correlation with the number of missing teeth. The fractal dimension and the number and perimeter of white segments in the binarized image correlate positively with the number of congenitally missing teeth. CONCLUSIONS: A limited number of parameters of radiographic mandibular trabecular bone structure correlate with the number of missing teeth. However, a markable difference in radiographic parameters of mandibular trabecular bone structure between hypodontia and non-hypodontia subjects could not be demonstrated.