Risk factors for food impaction following implant restoration and development of a nomogram prediction model
CAI Yuru
LI Hongqing
Abstract：Objective To analyze the risk factors for food impaction following implant restoration,construct a prediction model,and propose targeted preventive strategies.Methods Data were collected from 310 patients who underwent implant restoration for partial or complete edentulism in Hospital of Stomatology,Wenzhou Medical University between January 2023 and December 2024.A split-sample approach was employed to randomly divide patients into modeling(n=217)and validation(n=93)groups in a 7∶3 ratio.Clinical data were collected,and LASSO regression was used to screen candidate variables for inclusion in subsequent multivariable analysis to identify risk factors for food impaction following implant restoration.A nomogram was constructed based on these findings.Receiver operating characteristic(ROC)curves and calibration curves were used to assess predictive performance,while decision curve analysis(DCA)and clinical impact curve(CIC)were performed to evaluate clinical utility.Results The incidence of food impaction among enrolled patients was 40.00％(124/310).LASSO regression identified seven key variables based on λ1se for inclusion in multivariable analysis.Multivariable analysis revealed that proximal contact loss[OR(95％CI)=2.973(1.364,6.478)],distance from the proximal contact area to the alveolar ridge crest ≥7 mm[OR(95％CI)=3.058(1.519,6.158)],width of keratinized gingiva ≤ 1 mm at the implant site[OR(95％CI)=2.062(1.042,4.082)],alveolar bone resorption at the surgical site[OR(95％CI)=4.172(1.926,9.039)],frequency of daily tooth brushing＞2 times[OR(95％CI)=3.369(1.538,7.378)],frequency of daily oral hygiene maintenance＜1 time[OR(95％CI)=3.012(1.152,7.880)],and elevated plaque index at 6 months postoperatively[OR(95％CI)=2.297(1.106,4.769)]were independent risk factors for food impaction following implant restoration(all P＜0.05).ROC analysis demonstrated areas under the curve of 0.802[95％CI(0.740,0.864)]in the modeling group and 0.813[95％CI(0.716,0.910)]in the validation group.Hosmer-Lemeshow goodness-of-fit tests yielded x2 values of 5.475(P=0.706)and 5.332(P=0.720)for the modeling and validation groups,respectively.Calibration slopes were 0.88 and 0.97,with Brier scores of 0.11 and 0.06.DCA demonstrated net benefit of applying the nomogram for preventive intervention across threshold probability ranges of 0.10-0.83 and 0.20-0.92,respectively.CIC analysis indicated that,within the threshold probability range of 0.2-1.0,the nomogram-based predicted probability showed good agreement with the actual observed risk of food impaction.Conclusion Food impaction after implant restoration is associated with multiple factors.This nomogram accurately predicts postoperative food impaction risk and provides a basis for developing personalized preventive strategies in clinical practice.
Keywords：dental implantationpostoperative restorationfood impactionrisk factorsalveolar ridgekeratinized gingivaoral hygienepredictive models
Publication Date：2026-06-30
Online Publishing Date：2026-08-26(First online date of this platform, not the publication date of the document)
Pages：8( 889-895,902 )