Intraoral retrieval of iatrogenically displaced mandibular third molars from the sublingual and submandibular spaces using a lingual mucoperiosteal flap: a retrospective study of 18 patients

Background Iatrogenic displacement of a mandibular third molar or its root fragment into the floor of the mouth is an uncommon but potentially serious complication of third molar extraction. Published experience is largely limited to isolated case reports, and management remains unstandardised. The fragment-size threshold for retrieval, the optimal timing of retrieval, and the diagnostic value of the Y-sign on panoramic radiographs remain unclear. This study reports the outcomes of a lingual mucoperiosteal flap (LMF) approach used for retrieval on the day of presentation in all patients, including those with small fragments, and the frequency of the Y-sign and preoperative anxiety in this cohort. Methods This was a retrospective study of 18 consecutive patients treated for iatrogenic displacement of mandibular third molars or root fragments into the floor of the mouth at a private oral surgery clinic (Art Dentist Dental Clinic, Hanoi, Vietnam). Treatment and follow-up took place between January 2023 and December 2025. Patients were included if displacement was confirmed clinically and by cone-beam computed tomography, and all were managed with the same LMF approach with a minimum follow-up of 6 months. Results Mean patient age was 26.72 ± 5.88 years, with a slight male predominance (55.56%). Most displaced fragments were roots (94.44%), with a median size of 4.00 mm, and 88.89% of accidents occurred during extraction by general dental practitioners. The Y-sign was present in all 7 patients with available pre-extraction panoramic radiographs (100%). Retrieval was successful in all 18 patients (100%, 95% confidence interval [CI]: 81.47–100), with a median operating time of 22.50 min. No new neurosensory deficit was detected on non-blinded pin-prick testing (0%, 95% CI: 0.00–18.53), and both pre-existing deficits resolved within 5 months. Conclusions In this study, retrieval through an LMF was successful in all 18 patients, and no new neurosensory deficit was detected on non-blinded assessment. The Y-sign is an exploratory observation that merits further study. These findings derive from a single-centre, retrospective study without a comparison group and should be interpreted as hypothesis-generating. Larger prospective studies are needed before broader clinical adoption.