Construction of a predictive model for oral infection complications in elderly patients with complete dental arch loss undergoing oral implant restoration and corresponding prevention and treatment strategies

DONG Diwen; SU Xu; WUFANBIEKE Baheti

doi:10.19749/j.cn.cjgd.1672-2973.2026.01.004

Abstract：Objective To explore the influencing factors of oral infection in elderly patients with complete dentition loss after oral implant restoration,construct a predictive model,and formulate targeted prevention and treatment strategies.Methods A total of 176 elderly patients with dentition loss admitted to the hospital from September 2021 to December 2023 were prospectively selected as the modeling set,and divided into an infected group(58 cases)and a non-infected group(118 cases)according to whether oral implant restoration was complicated by oral infection.A total of 44 elderly patients with dentition loss from January 2024 to April 2024 were selected as the validation set.The baseline data of the patients were collected and statistically significant differences were obtained by univariate Logistic analysis.A Nomogram prediction model for oral infection complicated by dental implant restoration in elderly patients with dental loss was established and verified by receiver operating curve(ROC).Results Univariate analysis showed that there were significant differences in smoking,diabetes mellitus,periodontal disease,bone density in the implant area,preoperative oral health index and gingival bleeding index between the two groups(all P<0.05).Multivariate analysis showed that smoking,diabetes mellitus,periodontal disease,bone mineral density in the implant area,preoperative oral hygiene index and gingival bleeding index were the influential factors for oral infection complicated by dental implant repair in elderly patients with missing dentition(all P<0.05).Internal verification showed that the area under ROC curve of the constructed Logistic model was 0.810(95%CI:0.740-0.880),the sensitivity was 67.2%,and the specificity was 84.7%.External verification showed that the area under the curve was 0.783(95%CI:0.646-920),the sensitivity was 86.7%,and the specificity was 65.5%.The average absolute error of theoretical value and actual value of calibration curve is 0.028,which has a good agreement.Conclusion Diabetes,smoking,reduced bone density,periodontal disease,poor preoperative oral hygiene,and elevated gingival bleeding index are significantly associated with increased risk of oral infection after implant restoration in edentulous elderly patients.The proposed nomogram model demonstrates favorable predictive accuracy and clinical applicability,supporting its use for preoperative risk stratification.It is recommended that clinicians implement targeted preventive measures,including glycemic control,smoking cessation programs,oral hygiene optimization,and bone augmentation when indicated,to reduce postoperative infection rates.