Those Eyes Are Not a Horizon.

Half the population may be walking around with an interpupillary line that is not actually level with the real horizon, and digital dentistry keeps behaving as though the eyes were factory-calibrated. In Observational Analysis of Discrepancy Between the Interpupillary Line and the Horizon Line in Natural Head Position, Germán Albertini and colleagues show why that assumption deserves less confidence. If your software aligns a facial photograph to the pupils and calls it horizontal, you may already have built a cant into the plan before you have arranged a tooth or tilted an occlusal plane.

This sounds like the sort of problem only a very committed photographer or a mildly haunted prosthodontist would care about. Then you remember that 2 degrees is enough for the human eye to notice, particularly once it migrates into a smile line or restorative plane. At that point the issue stops being theoretical and starts looking expensive.

The Data Anchor

This was an observational, cross-sectional clinical study including 235 participants aged over 20 years who were fully dentate and free from congenital or traumatic facial anomalies or ophthalmic pathology. The authors obtained calibrated facial photographs in natural head position using an external vertical reference, then drew two horizontal lines on each image: the interpupillary line (IL) and the true horizon (HOR). The angle between them was measured and grouped from 0 to 5 degrees.

The frequency distribution is the part worth pinning to the mental noticeboard. Only 20.4% of participants showed 0° discrepancy. Another 30.6% were within ±1°. But 25.9% showed ±2°, 15.3% showed ±3°, 5.9% showed ±4°, and 1.7% showed ±5°. Taken together, 49% of the sample had a discrepancy of 2° or more between IL and HOR in natural head position, which the paper notes is perceptible to the human eye.

Key Findings

• Perfect alignment was the minority outcome. Only 20.4% of participants had no discrepancy at all between the interpupillary line and the true horizon.
• Nearly half crossed the perceptibility threshold. 49% of the sample showed 2° or more discrepancy, the level the authors identify as visible to the human eye.
• Mild mismatch was common, not exotic. The distribution was broad, with 30.6% at ±1°, 25.9% at ±2°, and another 22.9% sitting between ±3° and ±5°.
• The clinical risk sits in the software assumption. If photographs are rotated so that the pupils are treated as horizontal by default, the resulting digital plan may inherit an inaccurate occlusal or aesthetic plane.
• The limitation is one of scope. This is a prevalence and measurement study, not an outcomes paper, so it shows the scale of the alignment problem rather than proving how often restorations fail because of it.

💡 The Clinical Bottom Line

For facially driven planning, particularly in full-arch or anterior aesthetic work, the practical move is simple: stop treating the interpupillary line as a universal spirit level. Record a true horizon or at least validate the facial photograph against an external vertical reference before you let the software start tidying reality on your behalf.

This is not an argument against digital planning. It is an argument against lazy alignment. The eyes may be expressive; they are not an obliging horizon.

Dr Samuel Rosehill is a general dentist with a prosthodontic focus, practising at Ethical Dental in Coffs Harbour, NSW. He holds a BDSc (Hons) from the University of Queensland, an MBA, an MMktg, and an MClinDent in Fixed & Removable Prosthodontics (Distinction) from King’s College London.

Reference: Albertini G, Bechelli D, Capusotto A et al. Observational Analysis of Discrepancy Between the Interpupillary Line and the Horizon Line in Natural Head Position. International Journal of Periodontics and Restorative Dentistry, 2025. DOI: 10.11607/prd.7501