Why plan in 3D before a bone graft and an implant?

What cone beam CT shows, and a panoramic X-ray does not

Cone beam CT (CBCT) reconstructs the jaws in three dimensions. The clinician can scroll through slices in every plane, including cross-sections perpendicular to the ridge that show the bone “in section”.

On these slices, the following are measured:

• the available bone height, from the top of the ridge to the nerve or sinus;
• the width of the ridge at several levels (a ridge can be wide at the base and knife-edged at the top);
• the angle of the ridge relative to the opposing teeth;
• the apparent density of the bone and any lesions (retained root, cyst, infection).

A review of recommendations considers cone beam CT justified for diagnosis, planning and intraoperative transfer in implant dentistry ^[1]. After an extraction, the ridge loses on average 3.8 mm of width in six months ^[2]: it is precisely this loss of width, invisible on a panoramic X-ray, that 3D imaging reveals.

Two areas to respect: the nerve and the sinus

In the lower jaw, the inferior alveolar nerve runs through the bone in a canal, below the roots of the premolars and molars. It supplies sensation to the lower lip and chin. Injury can cause loss of sensation or abnormal sensations. On the cone beam CT, the canal is identified and traced along its full length, together with its exit (the mental foramen) and any loops.

In the upper jaw, the sinus is an air-filled cavity lined with a thin membrane, just above the roots of the molars. Cone beam CT shows the bone height beneath the sinus, the thickness of the membrane, any internal walls (septa) and the condition of the sinus lining. These findings point towards a crestal or lateral sinus lift. During a lateral sinus lift, membrane perforation is the most common complication, occurring in 19.5% of cases on average in a systematic review ^[3]: knowing the anatomy before opening helps to anticipate it.

Planning from the tooth, not from the bone

For a long time, the implant was placed wherever bone remained, and the crown had to make do, sometimes at the cost of an oblique axis or a neck that was hard to clean. Prosthetically driven planning reverses the order of decisions. It proceeds in four steps:

• a digital impression of the teeth (taken with an intraoral scanner) is superimposed on the cone beam CT, so that the visible teeth and the invisible bone appear in the same space;
• the future crown is modelled in the place that chewing, appearance and cleaning require;
• the virtual implant is slipped beneath that crown, along the axis of chewing forces;
• only then is it checked whether bone surrounds this implant along its whole length. Where it is lacking, the volume to rebuild is defined around the implant as it should be, not the other way round.

Transferring the plan to the mouth: guide and mesh

A plan is only worth something if it is reproduced in theatre. Two tools act as a bridge between the computer and the mouth.

The surgical guide is a splint that directs the drills along the planned axis. Its accuracy has been measured: on average, the placed implant deviates from the plan by a little over a millimetre (1.2 mm at the entry point, 1.4 mm at the tip) and by 3.5° ^[4]. Guiding every step rather than only the first drill reduces these deviations further: compared with freehand placement, the angular deviation falls from 7.5° to 2.6° and the deviation at the tip from 2.2 mm to 0.9 mm ^[5]. The comparison table of the three techniques and the principle of placement are on the page Placing an implant and guided surgery.

For large reconstructions, the custom titanium mesh is designed on the same file: it gives physical form to the bone volume drawn around the implants. A 2025 meta-analysis reports, with these meshes, a mean horizontal bone gain of 6.4 mm and vertical gain of 5.1 mm ^[7]. Design and manufacturing are explained in the article on 3D printing, and the treatment itself on the Custom titanium mesh page.

What 3D does not solve: the limitations

• Dose. Cone beam CT delivers more radiation than a panoramic X-ray, and the dose varies widely between machines: from the equivalent of 2 to 200 panoramic images ^[1]. The field of view is therefore limited to what is needed, and the scan is only repeated if it will change the decision.
• Artefacts. Metal crowns, old implants or patient movement create streaks that hide the adjacent bone ^[1].
• Measurement accuracy. It can reach 0.2 mm on the highest-performing machines, but be several times worse depending on the machine and settings ^[1]. Figures read on screen are not measurements to the tenth of a millimetre.
• Transfer. Even with a guide, a deviation of about 1 to 1.5 mm remains on average ^[4]; the guide must be perfectly stable in the mouth.
• Biology. A perfect plan ensures neither gum healing nor bone formation. With custom meshes, exposure still occurs in about one case in five ^[7].

Finally, greater accuracy does not automatically translate into superior clinical outcomes: when the available systematic reviews are pooled, implant survival is the same after guided surgery and after freehand placement; the contribution of guidance would mainly be to reduce surgical risk in difficult cases ^[8]. 3D planning is a tool for decision-making and safety, not a promise.

In practice, when is 3D planning useful?

In a simple situation with plenty of bone, planning is still useful but can be lighter. At Dr Hazout’s practice in Levallois-Perret, the level of planning is matched to the difficulty of the case: it is neither systematically maximal, nor reduced when the nerve or sinus are close.

Frequently asked questions