Revision and refinement planning

A revision is a targeted change to a plan that has not been manufactured. It exists because review is meant to be productive: you should be able to say that tooth 13 should finish more distally, that the IPR at 32 and 42 should come out, or that the attachment on 24 is unacceptable, and get a version that does exactly that and nothing else.

What is preserved matters as much as what changes. The case number, the original records, the earlier versions and their comments all remain, so the case reads as a history rather than a single current state. That protects both parties: you can see what you asked for and what was delivered, and a later refinement can be planned against the original intent rather than against a plan whose reasoning has been lost.

Some requests are not local. Asking for a different arch form, a different midline priority or a change from a non-extraction to an extraction approach changes the space analysis, and therefore the IPR, the staging and often the attachments. Those are rebuilt rather than patched, and the change log says so, because a plan that has been partly rebuilt and partly retained is the kind of plan that produces surprises at stage twenty.

A refinement is a new plan built on new records. The patient has worn a series of aligners, some movements have expressed fully, some partially, and some not at all, and the only reliable description of where the teeth are is a fresh scan taken with the aligners out and the teeth settled. Planning a second series from the original file rather than from a new scan means planning from a position the patient is not in.

The first task is measurement, not movement. Each tooth's current position is compared with the position the original plan intended at that stage, and the difference is the residual. Reading the residuals as a pattern is more informative than reading them one at a time: rotations that all under-expressed point to attachment or tracking problems, an arch that under-expressed uniformly points to wear time, and a single tooth that did nothing points to a debonded attachment or a collision that was never resolved.

The second series is then staged for what is left. It is usually shorter than the first, and its per-stage limits are often more conservative on exactly the movements that under-expressed, because repeating the same figure that failed the first time is unlikely to produce a different result. Where a movement has proved unachievable with aligners alone, the plan says so and leaves the decision about auxiliaries to the treating clinician.

The most common error in refinement planning is scheduling interproximal reduction at a contact that has already been stripped. The enamel is finite, the first series may have removed most of what was available, and a plan built from the new scan alone cannot tell the difference between a contact that is naturally narrow and one that has been reduced. That is why the actual IPR performed is asked for as an input rather than inferred.

Attachments need the same treatment. An attachment that is still bonded and still in the right position can often be retained, which saves chair time and avoids re-etching enamel. One that has debonded, one that was never placed, and one whose planned movement is now complete are three different situations with three different answers. The refinement lists each existing attachment as kept, removed or replaced, and adds new ones only where the remaining movements need them.

Where the first series' record is incomplete, the plan works from what the new scan shows and says which assumptions it had to make. An assumption that is written down can be corrected at review; one that is buried in the geometry cannot.

Every version of a plan is numbered and kept. Approving version three does not delete versions one and two, and requesting a change to version three produces version four rather than editing it in place. This is deliberate: aligner cases run for months, staff change, and a case that can only show its current state cannot answer the question of why a decision was made.

Comments are attached to the thing they are about. A comment on tooth 21 stays with tooth 21 across versions; a comment on stage 14 stays with that stage; a comment on an IPR contact stays with the contact. That is what makes a change log meaningful rather than a list of file dates, and what allows a refinement planned a year later to be read against the original reasoning.

The approval remains the boundary throughout. A revision or a refinement is a proposal until the treating clinician approves it, and no version, however many times a case has been replanned, moves to production file preparation without that approval.