Bone Grafting and Sinus Lifts: The Imaging Checklist Before Surgery
Ask a surgeon who has torn a Schneiderian membrane mid-procedure what they wish they had known, and the answer is rarely exotic. Usually a septum. Or a membrane thinner than it looked. Or a lateral wall that proved to be cortical bone where the panoramic radiograph suggested otherwise.
Most intraoperative surprises in sinus augmentation were not surprises. They existed before the incision, in a scan either never taken or never read closely enough. What follows is not a surgical protocol — it is the checklist our radiologists work through on every CBCT that arrives with “sinus lift planning” on the referral, and what each finding changes. It sits inside our wider implant planning protocol.
Why a panoramic radiograph will not do and CBCT will
A panoramic radiograph tells you roughly how much bone sits below the sinus floor. That is all it tells you reliably. It cannot give you buccopalatal width, membrane health, whether a septum crosses your field, where the posterior superior alveolar artery runs, or whether the sinus drains. Each changes the operation; two can end it.
The geometry is the problem, not image quality — a panoramic radiograph flattens a curved structure, with the worst distortion happening in the posterior maxilla. More on what an OPG cannot show in the posterior maxilla.
The seven-point imaging checklist
1. Residual bone height — the figure that picks your technique
Measure crest to sinus floor at the planned site, on the CBCT cross-section, not the panoramic reconstruction. Residual bone height most influences whether you approach crestally or through a lateral window. Current frameworks — including the 2023 International Journal of Oral Science review — group roughly as follows:
| Residual Bone Height | Usual Approach | What to Watch |
|---|---|---|
| Above 6 mm | Transcrestal | Predictable; confirm membrane health |
| 4–6 mm | Transcrestal if the sinus is healthy; lateral if not | Pathology moves you laterally |
| 2–4 mm | Lateral window if the lateral wall is thin; otherwise transcrestal | Perforation risk climbs sharply |
| Below 2 mm | Lateral window; assess membrane stretching potential | Below 3.5 mm is a recognised risk factor for perforation |
2. Membrane thickness — both extremes cause trouble
A healthy Schneiderian membrane is thin on CBCT, averaging around half a millimetre. This is one of the few places on the list where the relationship is not linear, and it catches people out.
Thin membranes tear — perforated cases consistently show thinner membranes than intact ones. So thicker looks safer. But thickening beyond about 2 mm usually means something: chronic inflammation, an odontogenic source, allergic disease. It carries a higher rate of postoperative sinusitis, and beyond roughly 5 mm most protocols treat the sinus as unsuitable without ENT input.
The comfortable window is narrow, around 1.5–2 mm. Both extremes are worth flagging, for different reasons.
Worth checking
Isolated thickening confined to one sinus? Look at the teeth beneath it before blaming allergy. Odontogenic sinusitis is common, frequently missed, and will not settle until the dental source is treated — which changes the sequencing of your graft.
3. Septa
Underwood's septa are bony walls projecting from the sinus floor, and they are the most consistently proven anatomical risk factor here. Prevalence runs at roughly one sinus in four to one in three. A 2024 meta-analysis put perforation at close to 40 per cent when a septum is present against roughly 12 per cent when it is not — an odds ratio around four.
Knowing in advance lets you design a two-window approach rather than find the septum with a bur. Orientation matters too: a transverse septum is usually manageable; one running longitudinally through your window is a different operation.
4. Lateral wall thickness and access angle
Measure the buccal wall where the antrostomy will go — typically one to two millimetres at the molars.
We would rather be honest than tidy: the evidence is contested. Several studies report much higher perforation above 2 mm — one found 55 per cent against 12 per cent below 1 mm. Others find no significant difference, and a recent meta-analysis concluded no firm conclusion can be drawn.
Less disputed is the buccopalatal angle where the walls converge. Below about 30 degrees there is little room to elevate the membrane without tension, and that group carries the highest perforation risk. Measure it — it takes seconds and tells you how much working space you have.
5. The posterior superior alveolar artery (PSAA)
The intraosseous branch of the PSAA runs within or against the lateral wall, and a window osteotomy can cross it. CBCT shows it in most cases, though detection varies with voxel size.
Typical position in the molar region is around 17 mm from the crest and 10 mm above the sinus floor — but the range is wide enough that the average is little help in an individual case, which is rather the point of scanning. Vessels near or above 2 mm will obscure your field.
Bleeding here is rarely dangerous. It is extremely inconvenient at the moment you need to see.
6. Sinus width
Measure the buccopalatal dimension at the graft site. A narrow sinus grafts more predictably — more bony walls mean better vascular supply and less distance for new bone to bridge. Wide sinuses are slower and may justify longer healing. Which classification you use matters less than recording the number.
7. Ostium patency and drainage
The item most often skipped, and the one behind the most miserable postoperative courses.
The ostium sits high on the medial wall and drains the sinus by mucociliary transport. Augmentation reduces sinus volume and increases mucosal oedema. If drainage was marginal, the graft can tip it into an obstructed, infected sinus — and an infected graft usually fails.
Scroll the CBCT to the ostiomeatal complex and confirm patency before you commit. If it is obstructed, opacified, or you are unsure, refer to ENT.
When to postpone rather than proceed
Some findings stop the plan rather than modify it:
Acute sinusitis or an air-fluid level
Complete or near-complete opacification
An obstructed ostiomeatal complex
Membrane thickening beyond roughly 5 mm without ENT input
An untreated odontogenic source beneath the graft site
None is a permanent contraindication. All are reasons to sequence differently — and sequencing is where most avoidable failures begin. More on the imaging mistakes behind implant failure.
What a good report should tell you
A CBCT report saying “maxillary sinuses are clear” has not answered your question. It should return the seven numbers above, site by site.
That is a reporting standard, not a scanning standard — one volume answers all of it if whoever reads it knows to look. Which is why it matters who reports the scan rather than who acquires it. Every scan at DMD Imaging is reported by an MDS-qualified dental radiologist, and you can speak to the person who read yours.
Practical note on field of view
A small field centred on the graft site will not include the ostiomeatal complex. If drainage is part of the question — and for augmentation it always is — the field must extend high enough to include it. Tell us it is a sinus lift when you refer and we will select accordingly.
Frequently Asked Questions
-
No body mandates it absolutely — every scan must still be justified individually. But the anatomy that determines the safety of the procedure is not assessable in two dimensions, so a planned augmentation without one is hard to defend. Background on the modality: our complete guide to CBCT scanning.
-
Close enough that it still reflects the anatomy. Bone changes slowly; sinus mucosa does not. A scan taken during rhinitis can show thickening that resolves weeks later. If months have passed, or there has been a sinus infection since, repeat rather than plan on stale data.
-
No. Mild thickening on CBCT is common and often incidental. The concern is marked thickening, thickening with an obstructed ostium, or an obvious odontogenic source. Mild thickening in a patent, otherwise healthy sinus usually does not change the plan.
-
Sometimes — and that is the difficulty. A septum may be visible, partly visible, or hidden entirely by superimposition, and its height and orientation cannot be judged on a panoramic. Absence on an OPG is not evidence of absence.
-
Pricing varies by centre and field of view, so rather than quote a figure that dates, confirm directly — and ask who will report the scan. See our scan services.
