Digital Bite Registration: Buccal Scans, Interocclusal Records and When to Send More

The margins are sharp and the bite scan took a moment. Then the crown arrives out of contact, or the lab asks whether the patient really closes that way, because the scanned arches overlap on one side and float apart on the other.

A digital bite is a few buccal scans that tell the software where the lower arch sits. Below: how scanners use them, how they compare with silicone records, why scanned bites overlap, and which cases need more.

Quick Answer

For a crown or short bridge with teeth on both sides, two buccal bite scans, one per side, taken in firm habitual closure are as accurate as a silicone record or better. Add a physical record scanned in place when several posterior teeth are missing, support is one-sided or the arch is edentulous. Changing the vertical dimension or restoring in centric relation needs a dedicated record. Expect some overlap in scanned bites; the lab can realign it.

How Does a Scanner Put the Two Arches Together?

The arches are scanned separately, mouth open. A buccal scan of the closed teeth then gives the software shared surfaces to place one arch against the other, so the bite depends on where those scans are taken and how many there are.

One side, both sides or one long scan?

In a lab study of complete-arch scans, the alignment tilted toward the side of the buccal record, and the contacts changed with the record's location (Edher et al., 2018). Two separate records did better than one long one: aligning the arches with a scan on each side was off by 0.053 mm on average, while a single complete-arch bite scan was off by 0.265 mm, with larger errors posteriorly (Lee et al., 2024). More is not automatically better: in a cast study, adding a front record to left and right records was no more accurate than the two lateral records (Solaberrieta et al., 2016). How many teeth each side should cover is in our guide to why crowns come back high.

The bite is only as good as the arch scans

In a lab study, distortion in the full-arch scans themselves, 0.183 mm in upper intermolar width, hurt the bite more than bite-scan variation did (Osnes et al., 2021). Scan path is covered in Intraoral Scanning Tips That Prevent Remakes.

Are Digital Bites as Accurate as Silicone Records?

For dentate arches, the reviews say yes. A 2022 review of 28 studies, 9 clinical, found virtual articulation comparable to conventional methods with complete dentitions, stable contacts, a single prep or one missing posterior tooth, mostly on lab evidence (Shadid and Sadaqah, 2022). A 2024 meta-analysis judged scanner bite records acceptably accurate, with a pooled sensitivity of 0.76 and specificity of 0.80 in diagnostic accuracy studies (Morsy and El Kateb, 2024).

In patients there is no true reference, so clinical studies measure precision: how closely repeated records agree. In 8 patients, repeated scanner bites differed by 25 and 31 µm, against 128 and 154 µm for silicone records on mounted casts (Iwauchi et al., 2022). A randomized trial in 9 patients found the same pattern.

Why Do Scanned Bites Show Overlap or Open Contacts?

Teeth and the mandible move when the patient bites

The arch scans record unloaded teeth; the buccal scan records them clenched. Li et al. (2021) note that impressions cannot duplicate occlusion under force because of tooth displacement and mandibular deformation. In a clinical trial of 29 molar crowns, standard buccal bite registration left the opposing teeth overlapping by up to 79 to 136 µm per tooth group; a research method that registers each tooth separately reduced the overlap (Zhao et al., 2026).

Overlap is real in patients, and it can mean a low crown

In 30 patients, every method showed overlap between virtual arches: 48 to 79 mm² on partial-arch intraoral scans, about 2 to 3 mm² on scanned stone casts. The authors warn it can produce infra-occluded restorations (Beck et al., 2023).

How hard the patient bites

Clenching strength changed scanner bite records in 8 volunteers (Nishine et al., 2024). In 44 subjects, contact and near-contact areas recorded at 20% to 60% of maximum clenching differed from those at higher levels, which reached a plateau above 60% (Ruktanakit et al., 2025).

Mandibular flexure

The mandible flexes inward on opening and protrusion, from a few micrometres to more than 1 mm (Sivaraman et al., 2016), more in posterior regions (Gülsoy et al., 2022). One review advises keeping opening to about 10 to 20 mm where possible (Mijiritsky et al., 2022).

When Should You Send More Than Buccal Scans?

Missing teeth remove the landmarks the software matches, which compromises scanned bites (Chinam et al., 2023). In a lab study, one missing posterior tooth made no significant difference, but 3 or more missing posterior teeth, bilateral posterior gaps or six missing anterior teeth all reduced accuracy (Ren et al., 2020). If you use a physical record, its extent matters: with one-sided support, records covering the complete arch deviated 1.58 to 1.75 times more than records limited to the prepared teeth (Matta et al., 2026).

Edentulous arches are the weakest case: on edentulous implant models, only Primescan of four scanners reached clinically acceptable bite accuracy (Rutkūnas et al., 2024), and a silicone index performed comparably to a bite without added material (Pletkus et al., 2024). In a posterior implant gap, titanium posts reaching the opposing teeth gave the least error with a quadrant scan (Yu et al., 2024). When to skip the scanner is in Digital vs. Conventional Impressions.

Can You Record Centric Relation and a New Vertical Dimension Digitally?

Yes, but the evidence is still emerging. With a Kois deprogrammer and bilateral scans in one dentate volunteer, the trueness of the centric relation record ranged from 0.14 to 0.40 mm across four scanners and a jaw tracking system (Revilla-León et al., 2024). A 2026 review of 20 jaw tracking studies rated the risk of bias high, from small samples and no long-term data (Tafuri et al., 2026).

For a raised vertical dimension, a 40-participant clinical study recorded the new position with a leaf gauge and a scanner. Evaluation restorations from those records and from a jaw tracker's virtual records were judged comparable, though jaw tracking gave fewer false contacts (Ntovas et al., 2025).

What About Protrusive and Lateral Records?

Some scanner software records several bites, but not every position equally well. In a lab study of a multi-occlusion function, scanner records found 100% of true contacts in maximum intercuspation and protrusion, but only 28% in lateral positions (Morsy and Hammad, 2024). A 2026 review of 8 studies found virtual articulators comparable to mechanical ones for static occlusion, but less accurate for dynamic contacts (Jain and Abduo, 2026). A technique report shows protrusive and laterotrusive scans programming condylar inclination and Bennett angles with only a scanner (An and Bompolaki, 2025); it is not a comparative study.

What Can the Lab Correct, and What Can't It?

A technician can realign overlapping arches in 3D software (Beck et al., 2023) and compare the bite with the contacts you report; how collisions are removed matters, as the high-crown guide explains. What the lab cannot do is recover what was never recorded:

• A one-sided bite tilts the alignment, with nothing to check the other side against.
• Distortion across a full-arch scan carries into the bite.
• Centric relation and a new vertical dimension exist only if you recorded them.
• Lateral guidance from lateral scans alone is unreliable; describe the guidance you want on the Rx.

Digital Cases at Elegant Dental Laboratory

• Digital scans are reviewed before production, and every case is made in house in Brooklyn by more than 35 technicians.
• We accept STL and PLY files from all major intraoral scanners, with direct connections for iTero, 3Shape TRIOS, Dentsply Sirona, Medit, Carestream, Planmeca, Shining 3D, ORI and Panda (setup steps; more in How to Send a Scan to a Dental Lab).
• Screw-retained, cement-retained and full-arch implant restorations, and dentures, with intraoral scans accepted for complete dentures and design files kept.
• 5 business day standard turnaround on most restorative cases, rush options, and up to a 7-year warranty on final restorations (policies and warranties). Free pickup and delivery in all five New York City boroughs and 13 northern New Jersey counties, and UPS labels anywhere. Call 877-335-5221.

FAQ

Do I need a buccal bite scan on both sides for a single crown?

Yes. In lab studies, a one-sided record tilted the alignment, and two side scans were more accurate than one complete-arch bite scan.

Is a digital bite more accurate than a silicone bite?

For dentate arches with stable stops, reviews find it comparable or better.

Why do the arches overlap on my scan?

Teeth move and the mandible deforms under biting force, so the clenched buccal scan and the unloaded arch scans do not match exactly. Realigning the scans can correct it.

When should I add a physical bite record?

With 3 or more missing posterior teeth, one-sided support, an edentulous arch or a vertical dimension change. Keep it small and scan it in place.

Can I record centric relation with a scanner?

Yes, with a deprogrammer and bilateral scans, but the studies are small; treat it as emerging.

Reviewed by Gary Fingerman, founder and CEO of Elegant Dental Laboratory, Brooklyn, New York (founded 2007). Last reviewed: October 1, 2026. Most bite-registration evidence comes from lab studies or small clinical samples; follow your scanner's current instructions for bite scans.