The diagnosis of peri‐implantitis: A systematic review on the predictive value of bleeding on probing

OBJECTIVES: Bleeding on gentle probing (BOP) is the key parameter to the diagnosis of mucositis, while changes in crestal bone levels, along with clinical signs of inflammation, are required for the diagnosis of peri-implantitis. This systematic review and meta-analysis focused on the evaluation of BOP as a predictive measure for peri-implantitis. MATERIALS AND METHODS: An electronic search was performed through Medline and EMBASE databases, followed by a hand search through previous reviews and reference lists. Screening, study selection, data extraction and evaluation of publication bias were conducted by two independent examiners. Clinical studies reporting on the prevalence of peri-implantitis, BOP and/or suppuration (SUP) after more than 1 year of functional loading were selected. Meta-analyses were conducted to combine the proportions of peri-implantitis among BOP- and/or SUP-positive subjects and implants across studies. Subgroups were created and compared to investigate potential sources of heterogeneity. RESULTS: Thirty-one studies were selected for analysis. Inconsistent definitions of peri-implantitis were reported across the studies. Twenty-nine studies reported data on implant-level and twenty publications reported on subject-level. The combined proportion of peri-implantitis was 24.1% (95% CI 19.3-29.7) in BOP-positive implants and 33.8% (95% CI 26.7-41.6) for BOP-positive cases. However, the degree of variability among studies was high; the prediction intervals were 10.3-69.3 and 6.9-57.8, respectively. Evidence of asymmetry or publication bias could not be statistically detected. Short observation periods were significantly associated with lower proportions of peri-implantitis among BOP-positive implants. CONCLUSIONS: For BOP-positive implants, there was a 24.1% chance to be diagnosed with peri-implantitis; while for BOP-positive patients, there was a 33.8% probability to be diagnosed with peri-implantitis. This probability varied across study populations. Clinicians should be aware of the considerable false-positive rate of BOP to diagnose peri-implantitis.