Impact of splinting implant-supported crowns on the performance of adjacent posterior implants: A systematic review and meta-analysis

Statement of problem. While rehabilitation with implant-supported crowns in the posterior region is widely practiced, guidance on clinical considerations when choosing between splinted or nonsplinted implant-supported crowns is limited. Purpose. The purpose of this systematic review with meta-analysis was to assess whether adjacent implants with internal connections in the posterior region should be treated with splinted or nonsplinted crowns. Material and methods. The systematic review was registered on the prospective register of systematic reviews (PROSPERO) under the code, CRD42021248525. Two authors searched 4 electronic databases (PubMed, Cochrane Library, Scopus, and Web of Science) independently for articles published up to December 2023. Additionally, a hand search was conducted in the nonpeer-reviewed literature and the reference lists of included articles. Only clinical studies comparing splinted and nonsplinted hexagonal or conical internal connection adjacent implants in the posterior region were considered. A single-arm meta-analysis was conducted using RevMan. Risk of bias and quality were assessed using the Cochrane Collaboration's Tool for Assessing Risk and the Newcastle-Ottawa scale. Results. Qualitative synthesis included 17 studies, and quantitative synthesis involved 13 studies. A total of 2085 implants (1244 splinted and 841 nonsplinted) were inserted in 1027 participants, with observation times ranging from 5 months to 16 years. Thirteen studies related the type of crown retention (cemented or screw-retained). Only 2 studies related annual bone loss results considering the type of crown retention, but whether they were splinted or nonsplinted crowns and significant differences were not specified. The meta-analysis revealed no statistical differences between splinted and nonsplinted implants for biological complications (P=.95, I-2=0%). However, the nonsplinted implants exhibited significantly more mechanical complications than the splinted implants (P<.001, I-2=1%, RR 0.37; 95% IC [0.26-0.54]). The studies included demonstrated a low risk of bias and were of good quality. Conclusions. Splinting was not found to affect the incidence of biological complications in posterior adjacent implants. However, more mechanical complications occurred in nonsplinted crowns than splinted ones.