Differential Diagnosis Between Granuloma and Radicular Cyst: Effectiveness of Magnetic Resonance Imaging.
Study Details
Study Description
Brief Summary
The radiolucent periapical jaw lesions of 34 patients, which were surgically enucleated, were investigated by two radiologists using MRI, based on the same six criteria, to categorize the lesions as granulomas or radicular cysts. After apicoectomies, two oral pathologists (blinded to the radiologist's diagnoses) analyzed all specimens by referring to seven specific parameters and diagnosed the specimens as granulomas or radicular cysts. The inter-rater agreements between the radiologists and pathologists in terms of MRI and histological diagnoses, respectively, along with the discriminant power of the adopted criteria and the accuracy of the MRI assessments compared with the histopathologic results, were calculated.
A strong inter-rater reliability was observed between the two radiologists (k-statistic = 0.86, p = 0.0001) and the two pathologists (k-statistic = 0.88, p = 0.0001). Reliability was higher for the radiological (Guttmann's lambda lower bound > 0.6) than histopathological criteria. The accuracy (true positives plus true negatives) of the radiologists was higher than that of the pathologists based on receiver operator characteristic analysis (area under the curve = 0.87 and 0.91, respectively). MRI reliability and accuracy were high and comparable to histopathological reliability, highlighting the usefulness of this non-invasive exam as a pre-treatment diagnostic method for periapical endodontic lesions.
| Condition or Disease | Intervention/Treatment | Phase |
|---|---|---|
|
Study Design
Outcome Measures
Primary Outcome Measures
- The diagnostic accuracy of the resonance magnetic imaging, calculated in comparison with the pathologists' diagnosis with a ROC analysis [5 years]
To determine the diagnostic accuracy of the radiologists compared with the pathologists (considered the gold standard), receiver operator characteristic (ROC) analysis was performed assuming that the diagnostic test is dichotomous. This method for computing the area under curve (AUC) and its error is in agreement with fitting procedures used for diagnostic tests.
Secondary Outcome Measures
- inter-pathologist diagnoses , inter-radiologist diagnoses and internal consistency of the items used for the differential diagnosis by radiologists and pathologists. [5 years]
To examine inter-rater agreement between the two radiologists and two pathologists, we calculated Cohen's kappa. The sensitivity and specificity were determined to assess the ability of MRI to differentially diagnose cysts and granulomas. Sensitivity was calculated as the proportion of true-positive test results (cyst diagnosed consistently by both MRI and biopsy) among the patients with the "disease" (cyst diagnosed by biopsy). Specificity was calculated as the proportion of true-negative results (granuloma diagnosed consistently by both MRI and biopsy) among the "disease-free" subjects (granuloma diagnosed by biopsy) . Guttman's lambda coefficient was computed to account for the reliability of the used criteria.
Eligibility Criteria
Criteria
Inclusion Criteria:
-
patient age ≥ 18 years
-
the diameter of the lesion ≥ 5 mm, assessed based on periapical x-ray,
-
an acceptable endodontic treatment for the involved tooth.
Exclusion Criteria:
-
general medical contraindications for classification of the patient as level III or higher according to the American Society of Anesthesiologists,
-
teeth involved in the lesion, the function of which could not be salvaged after enucleation of the lesion and apicoectomy. The radiological exclusion criteria were
-
contraindication undergoing magnetic resonance: presence of a pacemaker or any metallic devices that could interfere with the magnetic field;
-
severe claustrophobia,
-
weight ≥ 130 kg
-
intolerance to gadolinium-DTPA as a contrast agent;
Contacts and Locations
Locations
No locations specified.Sponsors and Collaborators
- University of Bologna
Investigators
None specified.Study Documents (Full-Text)
None provided.More Information
Publications
None provided.- 15034