Attachment planning and visualisation

An attachment is a small composite body, commonly in the region of 3 mm along its longest dimension and around 1 mm thick, bonded to enamel and reproduced as an undercut in the aligner. What varies is its geometry, and geometry is chosen from the mechanics. A flat-faced rectangular attachment set with its long axis horizontal gives the aligner a broad surface to push vertically. The same rectangle set vertically gives a surface to push horizontally and a lever arm for rotation. A bevelled face turns a push into a directed force, and which face is bevelled decides which way the tooth is driven.

The tooth's own shape sets the constraints. A conical crown such as a canine offers almost nothing for the aligner to grip, which is why canines carry attachments in most cases with any rotation or extrusion in that segment. A premolar with a short clinical crown limits how large an attachment can be without approaching the margin or the occlusal surface. A molar with a broad buccal surface can take a larger attachment, which matters when it is the anchor for movement elsewhere.

Because the geometry follows the mechanics, the same tooth in two different plans can carry two different attachments. The bonding list therefore states the movement each attachment exists for, not only its shape, so a clinician replacing a debonded attachment knows what it was doing.

Rotation requires a couple: two opposing forces separated by a distance. An aligner can only generate that if it has a surface offset from the tooth's centre of rotation to push on, so a rotation attachment is placed away from the long axis and oriented so the aligner face meets it squarely. The further the rotation has to go, the more the attachment has to be able to keep engaging as the tooth turns, which is one reason large rotations are staged slowly rather than given a larger attachment.

Extrusion is the movement an aligner is least able to produce, because pulling a tooth out of its socket needs the plastic to grip rather than push. An extrusion attachment presents an undercut with an active surface facing gingivally, so the aligner engages beneath it and lifts. Anterior open bite closure and extruding a single depressed tooth both depend on this, and both are staged conservatively because the aligner loses grip the moment it deforms.

Root control is different again. Torque and root uprighting need a moment applied to the crown large enough to move the root while the crown is held, which typically means an attachment providing a force couple together with a stiff aligner fit, sometimes with power ridges or pressure areas in the aligner itself. Root movement expresses less readily than crown tipping, which is why plans that ask for substantial root correction carry more stages, more attachments, and sometimes planned overcorrection.

Patients notice attachments on upper anterior teeth, and clinicians are often asked to avoid them. That is a legitimate request and it has a cost, because the upper anteriors are exactly where rotation and extrusion control are usually needed. The plan's job is to price the request rather than to refuse or silently accept it.

There are several ways to reduce visibility. Smaller attachments can be used where the movement is modest. Attachments can be placed on the lingual surface where the mechanics allow, at the cost of a bulkier fit against the tongue and less predictable engagement. The work can sometimes be shifted distally, using canines and premolars as the active teeth and staging the anterior movements more slowly. Where a tooth genuinely cannot be controlled without an attachment, the plan says which movement will be compromised and by how much.

The same care applies to the bonding site itself. An attachment planned over a composite restoration, a veneer or an area of hypomineralised enamel will not bond reliably, and one that debonds mid-treatment stops the movement it was placed for. Where the records show restorative work at a planned site, the site is flagged so you can move it or accept the risk.

An attachment plan is only as good as the bonding. The aligner's undercut is manufactured to the planned position, so an attachment bonded a millimetre away from where it was planned meets an aligner surface that does not match it. The result is not a slightly weaker force but often the wrong force in the wrong direction, or an aligner that will not seat at all in that segment.

For that reason the bonding list records the position in terms that can be checked at the chair: the surface, the distance from the gingival margin, and the orientation of the long axis, alongside the 3D view. Where you bond from a template, the attachment geometry is carried into the production models so the template can be produced against the same data the aligners are made from.

Timing is part of the plan too. Not every attachment belongs at stage one. Bonding an attachment only when the movement it serves begins keeps the earlier aligners simpler and reduces how long a visible attachment sits on an anterior tooth. Equally, attachments are given a removal stage rather than being left in place by default, so the case does not finish with composite that has done its work several stages earlier.