95% CI 8.75–20.45%
95% CI 7.07–13.80%
What is being pooled?
Mandible-level denominator
One mandible is the reporting unit. Presence on either or both sides contributes to a mandible-level prevalence estimate.
I² = 90%: substantial between-study heterogeneity.
Side-level denominator
Each mandibular side is the reporting unit. This estimate must not be treated as interchangeable with the mandible-level value.
I² = 93.1%: marked heterogeneity remains.
Core methodological point. The meta-analysis deliberately separates mandible-level and side-level reporting units. Mixing them would produce a denominator error rather than a more precise estimate.
Explore the main findings
Not rare, but not a single universal number
The pooled estimates place the mylohyoid bridge in a frequency range that is clinically worth recognising. Yet the heterogeneity is high, so population, sampling and reporting methods matter.
Anatomical logic
Mylohyoid groove
Medial mandibular surface; transmits the mylohyoid neurovascular bundle.
Ossified roof / bridge
Partial or complete osseous coverage converts part of the open groove into a more constrained corridor.
Potential anaesthetic relevance
The altered osseous corridor may be relevant to accessory mylohyoid innervation and has been proposed as one contributor to mandibular local-anaesthetic failure.
Potential surgical relevance
Recognition matters around the medial mandibular ramus because the bridge may change the expected surface anatomy and relationship of the mylohyoid neurovascular bundle.
Evidence boundary
The review does not establish a clinical effect size. The included evidence was entirely osteological. The anatomical variant is real and the proposed anaesthetic and surgical relevance is plausible, but direct outcome-based clinical evidence remains limited.
Supported directly
Prevalence, laterality, sex comparisons, unilateral/bilateral presentation and morphological observations in skeletal material.
Requires clinical validation
How often the bridge actually causes nerve entrapment, alters anaesthetic success, changes symptoms, or modifies surgical outcomes in living patients.
Methods at a glance
Search
MEDLINE/PubMed, Scopus and Web of Science through May 2026, supplemented by reference-list and targeted journal searching.
Meta-analysis
Random-effects models with inverse-variance weighting, Freeman–Tukey double-arcsine transformation and DerSimonian–Laird estimation, with reporting units analysed separately.
Original publication
PubMed record (PMID 42832164) ↗
This interactive page is based on the published abstract and bibliographic record available on 6 October 2026. It does not replace the peer-reviewed article.