Background: Although there are multiple studies comparing the efficacy of different interdental hygiene aids on natural teeth, little or no data exists to show what method of interdental cleaning may be most effective on dental implants. The aim of this study was to compare the effectiveness of two interdental hygiene aids (an interdental brush v. dental floss) on dental implants, assessing their effects on clinical parameters that impact on impact tissue health, as well as cytokines in peri-implant crevicular fluid. Method: This was a single blind, randomized controlled trial in a single centre. Thirty-two implant patients currently under maintenance was evaluated at baseline, 3 and 6 months. One implant was studied per participant. Sixteen patients used dental floss while the other 16 used interdental brushes. All hygiene aids used were supplied to participants after randomized allocation. Clinical parameters evaluated include full-mouth plaque scores and full-mouth bleeding on probing, width of implant keratinized mucosa, distance of papilla from the occlusal point of prosthesis, implant probing depths, implant plaque levels and implant bleeding on probing. Peri-implant crevicular fluid samples were also taken and IL-2, IL-4, IL-6, IL-10, TNF-alpha and IFN-gamma levels were measured. Results: The mean and median levels of all clinical variables and cytokines for both groups were calculated and compared analytically. A repeated measures analysis to compare group-time interaction was also carried out. There were reductions from baseline to 6 months in most variables for both groups. While most results were not statistically significant, there was an increase in mean probing depth of interdental brush group at 6 months for the distobuccal (0.4mm) and lingual (0.25mm) implant surfaces that was statistically significant. Conclusion: Although floss slightly outperformed interdental brushes in implant plaque index and full mouth bleeding scores, other indices showed relatively comparative values between the two groups. We were unable to reject the null hypothesis as determined within the limits of these 6-months results.