PURPOSE: Several panoramic indices have been suggested to assess bone quality from the morphology and width of mandibular cortex on panoramic radiography. The purpose of this study was to compare dental implant failure group with control group in panoramic mandibular index (PMI), mandibular cortical index (MCI), and gonion index (GI) and to determine the effect of these panoramic indices on dental implant failure. MATERIALS AND METHODS: A case-control study was designed. Test group (n = 42) consisted of the patients who had their implants extracted because of peri-implantitis. Control group (n = 139) consisted of the patients who retained their implants over one year without any pathologic changes and had been followed up periodically. They had dental implants installed in their mandibles without bone augmentation surgery from 1991 to 2001. The following measures were collected for each patients: 1) PMI, MCI, and GI were measured twice at one-week interval on preoperative panoramic views; and 2) age, sex, implant length, implant type, installed location, occluding dentition state, and complication were investigated from the chart record. RESULTS: The PMI showed moderate level of repeatability. The intra-observer agreement of MCI and GI were good. There was statistically significant difference in PMI between two groups. There were significant different patterns of distribution of MCI and GI between two groups. Among the panoramic indices, PMI and MCI showed significant correlation with dental implant failure. CONCLUSION: Panoramic indices can be used as reference data in estimating bone quality of edentulous patients who are to have implants installed in their mandibles.