Guidance Before Reading
Critical Axes and Publication Limitations
- Article Type: Exposure/Epidemiology. The topic is organized across five clinical decision axes, not by a global judgment.
- Evidence Basis: moderate / moderate / fully evaluated. 9 studies, exclusively Systematic Reviews and Meta-analyses.
- Bias Risk: low to moderate. Source integrity: clean. No single study claims.
- Key Message: Every cited prevalence number is only as strong as the case definition and the cohort behind it.
Clinical Question
How common is peri-implantitis really, what role does case definition play in reported prevalence, and which risk factors are reliably proven?
Executive Summary
The local body of literature for this topic consists of nine systematic reviews and meta-analyses, published between 2015 and 2024. DDJ reads the strength of conclusion as moderate and classifies the main direction as an association signal, not an intervention outcome.
The central finding: The enormous variation in reported peri-implantitis numbers is not primarily biological, but definition-driven. Depending on the threshold for bone loss, probing depth, and BOP criteria, the prevalence changes manifold. Restrictive Sanz-Chapple-like criteria converge in independent meta-analyses on an approximate patient prevalence of 15 to 20 percent. [4,5,7]
Structured maintenance, smoking, diabetes, history of periodontitis, and lack of keratinized mucosa are consistently proven risk factors. BOP alone is not a sufficient diagnostic criterion. The generalizability of study numbers to one's own practice always requires a collective review.
How DDJ Reads This Topic
DDJ treats this topic as an Exposure/Epidemiology article. This means: it is not about the efficacy of a single intervention, but about what we actually know regarding the frequency, diagnostics, and risk profile of a disease, and where the literature is systematically biased.
The core conflict lies not in whether peri-implantitis exists, but how frequently it actually occurs depending on the definition and cohort. Anyone who treats a single prevalence number as universal ignores the strongest confounder: the case definition itself.
DDJ organizes the topic along five clinical decision axes: case definition, maintenance, BOP status, risk factors, and cohort selection. Strong language is only permissible where multiple independent meta-analyses converge.
Claim Clusters and Decision Axes
Claim Cluster 1 · ddj_0011_c01
Pocket Definition and Threshold Dependency
Clinical Axis: How much does pocket definition influence reported peri-implantitis prevalence?
Why this axis matters: The enormous variability in prevalence data in the literature cannot be explained by biology. Depending on the threshold for bone loss (2 mm versus 3 mm), probing depth, and BOP criteria, reports in primary literature fluctuate between under 5 and over 60 percent at the patient level.
Evidence Status: Meta-analyses using restrictive, Sanz-Chapple-like criteria converge on estimates of 15 to 20 percent patient prevalence. Nevertheless, heterogeneity remains high, with I-squared values regularly above 75 percent. [4,5,7,8]
Where the signal is stable: The convergence on 15 to 20 percent is consistent across several independent reviews and is considered robust.
Where the uncertainty begins: Non-standardized thresholds generate prevalence ranges that complicate any clinical comparison. The high I-squared heterogeneity signals that significant variability remains even within restrictive definitions.
Clinical Implication: Every cited prevalence number must be read in conjunction with the underlying pocket definition. A number without a defined criteria is worthless for clinical classification.
Claim Cluster 2 · ddj_0011_c02
Maintenance and Cohort
Clinical Axis: Is structured implant maintenance associated with a lower peri-implantitis burden?
Why this axis matters: The maintenance effect is the most consistently modifiable factor in the study material. Patients in regular recall programs show a significantly lower frequency of peri-implantitis in several systematic reviews compared to patients without structured follow-up.
Evidence Status: The difference is consistently demonstrated, even if it is not formally causal. Dreyer et al. 2018 report a median prevalence of 9 percent with recall versus 18.8 percent without structured recall. [4,5,7]
Where the signal is stable: The association between recall attendance and a lower peri-implantitis rate is robust across multiple reviews.
Where the uncertainty begins: Patients who participate in recall programs may systematically have different risk profiles than non-participating patients. The selection is not controlled.