Imagine a tooth erupting upright during adolescence, appearing perfectly normal, only to gradually “collapse” sideways until it is almost lying horizontally. Not because of trauma and not because it was extracted. It moves on its own, silently, over the course of several years.
This is known as a fainting wisdom tooth, a rare eruption phenomenon that was identified, analyzed, and scientifically described for the first time in a longitudinal study published in the International Journal of Dentistry in 2025. Behind this discovery is drg. Rellyca Sola Gracea, Sp.RKG. Subsp.RDP(K), a lecturer in the Department of Dental Radiology at the Faculty of Dentistry, Universitas Gadjah Mada, who is also pursuing doctoral research at KU Leuven, Belgium.
One in Every 250 Patients
From a combined database containing panoramic radiographic records of 11,932 patients at Folktandvården Stockholm, Sweden, only 50 cases met the criteria for this phenomenon. That translates to 0.4%, or approximately one in every 250 patients with a complete dentition.
The rarity of the phenomenon presented a challenge in itself. Manually identifying those 50 cases from nearly 12,000 radiographic records would be almost impossible within a reasonable amount of time. This is where artificial intelligence (AI) played a crucial role.
The research team used a deep-learning convolutional neural network (CNN) that had previously been clinically validated to automatically measure molar angulation on panoramic radiographs. The model was capable of measuring tooth angulation with 98.1% accuracy within a five-degree error margin. Using this tool, thousands of radiographs could be analyzed systematically, efficiently, and objectively, far surpassing the practicality of manual measurements.
Each patient had two panoramic radiographs: the first taken between the ages of 8 and 15 years (T1), and the second between the ages of 16 and 23 years (T2). This period encompasses a critical phase of growth and wisdom-tooth eruption.
From 29 Degrees to 83 Degrees
The results were surprising. In the “fainting” group, the median angulation of the wisdom teeth changed from 29 degrees at T1 to 83 degrees at T2. In contrast, the control group with normal eruption showed almost no change, from 18 degrees to 20 degrees.
A shift of more than 50 degrees over several years, without consistently noticeable symptoms, makes this phenomenon particularly concerning because of its silent nature.
Logistic regression analysis identified several major risk factors. The most prominent was wisdom-tooth depth, with an odds ratio of 0.1. This indicates that teeth positioned deeper in the jaw were actually more likely to erupt upright rather than “faint.” Other statistically significant factors included a narrower space-to-crown-width ratio, excessive proximity to the second molar, a higher CEJ/apical ratio, and an increase in the gonial angle.
“Wisdom teeth that initially appear to be well positioned can completely change direction. Early monitoring through periodic radiographs is therefore very important for detecting this pattern before complications occur.”
The findings became even more interesting in the split-mouth analysis, which compared the left and right sides in 16 patients who experienced the “fainting” phenomenon on only one side of the jaw. The result: the only variable that differed significantly between the “fainting” and normal sides was the tooth’s proximity to the occlusal plane, rather than its initial angulation. This suggests that the fate of a wisdom tooth is determined not simply by its initial angular position but also by the anatomical limitations of the surrounding space.
Why Does This Matter to Clinicians?
The clinical implications of these findings are significant. Until now, wisdom teeth that appeared upright during adolescence were often considered safe and not in need of special attention. This study suggests that such an assumption may be incorrect.
The researchers recommend several preventive strategies that could be implemented earlier: prophylactic extraction before root formation is complete if the “fainting” pattern can be predicted, orthodontic space management to expand the retromolar area, more frequent radiographic monitoring in high-risk cases, and early intervention when initial mesial tilting begins to appear.
The study also opens a discussion about other anatomical factors that have received less attention, including vertical mandibular growth patterns (dolichofacial patterns), which are associated with a high gonial angle and a narrower retromolar space. According to previous research, individuals with this facial growth pattern may have less potential for bone remodeling, resulting in more limited space for upright eruption.
AI Is Not a Replacement, but an Extension of the Clinician’s Eyes
One point is consistently emphasized in this study: AI is not intended to replace clinical judgment but to strengthen it. Without a deep-learning-based tool, manually identifying 50 cases from nearly 12,000 patients would take an unrealistic amount of time and be highly susceptible to inter-rater variability.
By automating angulation measurements, researchers were able to focus on interpretation and clinical decision-making. The CNN model used in this study could even work four times faster than manual measurement while providing considerably more consistent accuracy.
Nevertheless, the researchers acknowledged several limitations. All subjects were patients referred for orthodontic treatment, meaning they may not represent the general population. Panoramic radiography also has inherent limitations as a two-dimensional imaging modality. Prospective research using three-dimensional imaging and more diverse populations is therefore an important direction for future research.
The phenomenon of the “fainting wisdom tooth” may affect only a small proportion of patients. But for those who experience it without detection, the consequences can be much more serious: impaction, damage to the second molar, or infection that could otherwise have been prevented much earlier.
Source DOI: https://doi.org/10.1155/ijod/1932229
Authors: Anny Anggraini; drg. Achmad Zam Zam Aghasy, M.Kes.
Photo: Pexels