Detection of misfits at the abutment-prosthesis interface in the esthetic zone: Implications of the radiographic technique and the magnitude of the misfit

Abstract

Misfits in the abutment-prosthesis interface represent a setback for implant-supported prostheses. Periapical radiographs have been used as an auxiliary method in the evaluation of prosthesis fit to the abutments; however, the evidence supporting the use of this method is still restricted to studies of low to moderate quality. Furthermore, studies on the diagnostic accuracy of different periapical techniques used to detect misfits in the abutment-prosthesis interface, especially in the esthetic zone, are lacking. The purpose of this in vitro study was to assess the discrimination power of 3 periapical radiographic techniques in detecting misfits at the abutment-prosthesis interface in the esthetic zone and evaluate whether the magnitude of the misfit influenced the diagnosis. A total of 15 implants with an internal conical connection were installed in the central incisor region in polyamide jaws. Custom ceramic copings for cemented crowns were made by using a computer-aided design and computer-aided manufacturing (CAD-CAM) system. Misfits of 50, 100, and 150µm were simulated by interposing 1, 2, or 3 50-μm-thick polyester strips at the abutment-prosthesis interface; the absence of the strip represented the control group. Digital radiographs were obtained by using film holders for the following periapical techniques: bisecting angle (PBA), standard paralleling (PSP), and modified paralleling (PMP). Two radiologists and 1 prosthodontist evaluated a total of 180 radiographs. The values of the area under the receiver operating characteristic curve (Az) were regarded as a measure of diagnostic accuracy and subjected to the Friedman test with post hoc Durbin-Conover (α=.05). The PSP (Az=0.873) had higher Az values than the PBA (Az=0.753) for the 50-μm misfits (P<.05). Larger misfits resulted in greater accuracy than smaller misfits (P<.05). Interactions between the factors radiographic technique and misfit magnitude resulted in statistically significant differences for all comparisons (P<.05), except between the PSP for the 100-μm misfits (Az=0.976) and the PMP for the 150-μm misfits (Az=0.998). The PSP was more accurate than the PBA in detecting the 50-µm misfits at the abutment-prosthesis interface; larger misfits resulted in more accurate diagnoses, regardless of the technique used.