The aim of this systematic review was to compare the crestal bone loss around splinted and nonsplinted adjacent implants.

To address the focused question, “Is crestal bone loss around adjacent implants different with splinted from that with nonsplinted restorations?,” indexed databases were searched from 1965 up to and including May 2016 using various combinations of the following keywords: “implant,” “splinted,” “nonsplinted,” “unsplinted,” “connected,” “unconnected,” “nonconnected,” and “bone loss.” Letters to the editor, commentaries, historic reviews, case reports, case series, animal studies, and studies on full‐arch rehabilitation were excluded.

Six studies were included with titanium implants ranging from 114 to 1187 implants. All studies had nonsplinted and splinted restorations that ranged from 20 to 234 restorations and from 60 to 970 restorations, respectively. In all the studies, the follow‐up period after the restoration placement ranged between 1 and 22 years, with a mean follow‐up ranging between 3 and 10.18 ± 3.18 years. In all studies, the mean crestal bone loss for implants restored with nonsplinted restorations ranged between 0.30 ± 0.65 and 1.3 ± 0.2 mm, whereas the mean crestal bone loss for implants restored with splinted restorations ranged between 0.50 ± 0.8 and 1.22 ± 0.95 mm.

Within the limitations of this review it is concluded that adjacent implants restored with splinted and nonsplinted fixed restorations did not exhibit a difference in crestal bone loss. The evidence from this systematic review suggests further investigation.