ACP Classification System for Complete Edentulism Predicts Implant Overdenture Complications

Objectives: The purpose of this investigation was to assess prosthetic complications for patients classified with the ACP Classification System for Complete Edentulism. Methods: A total of 59 completely edentulous patients (mean age 58.3 years) received new complete dentures, and were classified using the Classification System. After prosthesis insertion and adjustment, two non-splinted microthreaded implants (ASTRA Tech, Lexington, MA) were surgically placed in the mandibular intraforaminal region. At 3 months, retentive ball anchors were inserted, reline impressions made, and laboratory relined prostheses with gold Dalla bona style retentive cap attachments inserted (baseline). Correlations of the number and severity of prosthetic complications were assessed based on the ACP Classification of each patient. Patients were evaluated for 8 undesirable outcome events: tighten retentive ball anchor abutment, tighten retentive gold cap attachment, replace retentive ball anchor abutment, replace retentive gold cap attachment, reline, repair, and remake of prosthesis, and adjustment of sore spots. Results: A total of 48 patients were available for the 60 month recall assessment. The patients were classified as follows: no patients were assessed as Class I (least severe), 11 patients were Class II, 19 were Class III, and 12 were Class IV. For all 8 categories, the mean number of undesirable outcome events per patient was: 9.02 + 6.66 for all patients, 6.27 + 3.58 for PDI Class II patients, 9.47 + 6.35 for PDI Class III patients, and 11.83 + 8.86 for Class IV patients. One way ANOVA on groups II, III, and IV revealed that, at the 95% confidence level (p = 0.046) there was significant differences between the groups. Conclusions: The ACP Classification System appears to be a valid instrument for assessing the potential for complications associated with treatment of the edentulous patient with implant retained overdenture prostheses. This research was funded by ASTRA Tech, Molndal, Sweden.