SNA, SNB & ANB Angles in Cephalometric Analysis: Tracing, Norms & Clinical Significance

Two patients land on the same ANB of 5.5°. On paper they are both skeletal Class II. One has a normally positioned maxilla and a genuinely retrusive mandible. The other has a slightly protrusive maxilla, a normal mandible, and a nasion sitting a few millimetres high. Same number, different diagnosis, different mechanics, different retention risk. SNA, SNB, and ANB are the most quoted measurements in cephalometrics and among the easiest to misread — not because the arithmetic is hard, but because a single angle collapses a lot of anatomy into one figure. This is a working reference for what the three angles actually represent, how to trace them cleanly, and how much of "normal" is really a property of the population your norms came from.

The three angles, defined

All three are read at nasion, against the anterior cranial base.

SNA is the angle between the S–N line and the N–A line. It describes the anteroposterior position of the maxilla — specifically point A — relative to the anterior cranial base. A larger SNA means the maxilla sits more forward (prognathic); a smaller SNA means it is retrusive. The commonly used reference value is 82° ± 2°.

SNB is the angle between the S–N line and the N–B line. It describes the anteroposterior position of the mandible (point B) against the same cranial base. Larger means a more forward mandible; smaller means retrognathia. The reference value is 80° ± 2°.

ANB is the difference between SNA and SNB, or read directly as the angle A–N–B. It expresses the sagittal relationship between the maxillary and mandibular apical bases — the skeletal class, compressed into one number. The reference value is roughly 2° ± 2°. An ANB above about 4° points to a skeletal Class II tendency; below 0° (a negative value) points to Class III. Steiner considered ANB the measurement of real interest, because it captures the relationship the clinician is usually trying to correct.

If you want the wider frame these three angles sit inside, the cephalometric analysis overview covers where they fit among the other skeletal, dental, and soft-tissue measurements.

Where the angles come from: four landmarks

Everything above rests on four points. Get these wrong and every downstream angle inherits the error.

• Sella (S) — the geometric centre of the sella turcica (the pituitary fossa). It is a constructed midpoint, not an edge.
• Nasion (N) — the most anterior point of the frontonasal suture in the midsagittal plane.
• A point / subspinale — the deepest point on the concavity of the anterior maxillary contour, between the anterior nasal spine and the alveolar crest above the upper incisor (prosthion).
• B point / supramentale — the deepest point on the concavity of the anterior mandibular symphysis, between infradentale (the alveolar crest below the lower incisor) and pogonion.

S and N together define the SN line, which stands in for the anterior cranial base. Steiner deliberately chose SN over the Frankfort horizontal because porion and orbitale are hard to identify consistently on a lateral film, whereas SN lies in the midsagittal plane and is easy to locate. That choice is also SN's main weakness, which we come back to below.

How to trace them without stacking error

The construction itself is short:

1. Identify and mark S, N, A, and B.
2. Draw the SN line.
3. Draw N–A. The angle between SN and N–A is SNA.
4. Draw N–B. The angle between SN and N–B is SNB.
5. Read ANB directly as A–N–B, or subtract SNB from SNA. When A sits ahead of B relative to nasion, ANB is positive.

The cautions matter more than the steps:

A point and B point are not fixed bony landmarks. They are defined as the deepest concavity, and that concavity remodels with alveolar changes, incisor movement, and treatment. Heavy proclination or retraction of the incisors will drag A and B with it, which means ANB is never a purely skeletal number — it carries a dentoalveolar component. This is exactly why several population studies report an enlarged ANB alongside compensatory incisor proclination.

Nasion is a growth site and is not stable early in life. A nasion placed too high or too low tilts the whole SN line and shifts SNA and SNB together. In the anteroposterior direction the effect is sharper still: a posteriorly positioned nasion inflates ANB, while an anteriorly positioned one deflates it. Because ANB is a difference of two angles, any slip at nasion — or at A or B — propagates into both SNA and SNB and then compounds into ANB. Small landmark errors do not stay small.

Reading the numbers: clinical significance

The three angles earn their keep by localising a discrepancy and grading it.

A high SNA signals maxillary protrusion; a low SNA signals maxillary retrusion. That reading feeds directly into whether the maxilla is the problem — and therefore whether headgear, maxillary protraction, or a Le Fort procedure belongs in the conversation.

A high SNB signals a protrusive mandible; a low SNB signals retrognathia. That reading drives functional appliance timing and mandibular advancement or setback decisions.

ANB is the summary line. It tells you a discrepancy exists and how large it is, and it is the single most-cited driver of the Class II versus Class III label, of extraction versus non-extraction reasoning, and of how much incisor compensation is acceptable before you are camouflaging rather than correcting.

The point of reporting all three, rather than ANB alone, is that ANB does not tell you which jaw is at fault. An ANB of 6° is a Class II discrepancy — but SNA and SNB decide whether you are treating a protrusive maxilla, a retrusive mandible, or some of both. Read in isolation, ANB will send you toward the wrong appliance often enough to matter.

SNA, SNB, and ANB across different analyses

These three angles are essentially analysis-agnostic. What changes between systems is the reference-plane philosophy and the dental and soft-tissue measurements built around them.

• Steiner (1953) is where most clinicians first meet SNA, SNB, and ANB. Steiner built his analysis on the SN plane and made these three the skeletal backbone. The Steiner analysis reference walks through the full set of measurements and norms.
• Riedel (1952) actually introduced the SNA/SNB/ANB norms that Steiner then popularised, so the values you use every day trace back to Riedel's work.
• Downs (1948) predates Steiner and uses the Frankfort horizontal rather than SN, reporting the facial angle and the A–B plane. Downs' own normal sample produced a mean SNA of about 81° and a mean SNB of about 78° — a useful reminder that even the "classic" values shifted depending on whose small sample you looked at. See the Downs analysis reference for the full set.
• McNamara (1984) keeps these skeletal angles in view but adds nasion-perpendicular linear measures and effective jaw lengths, so a discrepancy can be read as a size problem, a position problem, or both. The McNamara analysis reference covers that linear approach.
• Eastman analysis, widely taught in the UK, keeps SNA, SNB, and ANB unchanged but references the incisors to the maxillary and mandibular basal planes instead of the NA and NB lines. Same three skeletal angles, different dental philosophy. The Eastman analysis reference lays out that distinction.

Because ANB has known weaknesses, it has also spawned a family of "second opinion" measurements — which is the next section.

The problem with ANB (and what to read alongside it)

No serious discussion of these angles is complete without ANB's documented distortions. The literature has catalogued them repeatedly:

• The position of nasion, both anteroposteriorly and vertically.
• Rotation of the jaws and of the SN line itself.
• The vertical distance of points A and B from the cranial base — taller faces read differently for the same true sagittal relationship.
• Cranial base length.
• Age. ANB tends to decrease with growth as the mandible catches up to the maxilla.

Ferrazzini's critical evaluation of the ANB angle and Hussels and Nanda's analysis of the geometric factors affecting it both formalised how much these variables can move the number without any real change in jaw relationship. In response, several alternatives were designed to sidestep the cranial-base dependence:

• The Wits appraisal (Jacobson, 1975) drops perpendiculars from A and B onto the functional occlusal plane and measures the distance between the two feet (AO minus BO). It removes the nasion and cranial-base problems entirely, but it introduces its own — it is misled by occlusal plane orientation, erupting tooth buds, a heavy curve of Spee, and open bites. See the Wits appraisal reference.
• The Beta angle (Baik and Ververidou, 2004) uses points A, B, and the condyle axis, making it independent of cranial reference planes — though condylion is not an especially reproducible landmark.
• Newer indicators including APDI, the W angle, the Yen angle, and the Pi angle each try to lean on more stable constructed points.

The practical stance is that no single sagittal indicator is definitive. ANB remains the shared language every orthodontist reads, so it is worth keeping — but it should travel with a cranial-base-independent cross-check such as the Wits appraisal. When ANB and Wits disagree, the disagreement is itself diagnostic; it usually means a vertical or rotational effect is inflating or hiding the skeletal picture. Running Steiner, Wits, and an APDI-type reading from a single set of landmarks — which is how BCeph is built — turns that cross-check into one tracing session rather than three separate ones.

"Normal" is not one number: norms across populations

Steiner was explicit that his figures expressed his idea of "a normal average American child," and that they had to be modified for age, sex, race, growth potential, and individual variation. That caveat is routinely ignored. Most reference values still in daily clinical use descend from small mid-twentieth-century white North American samples (Riedel, Downs, Steiner). Applying them uncritically to patients of other ancestries produces systematic, predictable misdiagnosis.

The published record on population variation is large. A representative slice:

A note on the table: where a well-established mean exists we have given the figure; where the literature agrees on the direction of difference relative to Caucasian norms but not on a single pooled mean, we have summarised the finding and cited the study rather than invent precision. Means and standard deviations shift with sample size, digitisation method, and inclusion criteria, so these entries are orientation, not gospel.

Several themes recur across all of it. Bimaxillary dentoalveolar protrusion is common across African, Middle Eastern, and East and South Asian samples, which means ANB read against Caucasian norms will over-call some patients and under-call others. Incisor norms diverge even further between populations than the skeletal angles do — so borrowing a foreign norm set does the most damage on the dental side. And sex differences within a single population are frequently as large as some of the differences between populations, which is the next point.

Growth, age, and sex move the target

Age. ANB decreases through childhood in most patients, because the mandible grows forward and downward more than the maxilla does. A 5° ANB at age eight does not carry the same prognosis as a 5° ANB at sixteen. SNA and SNB both tend to creep up slightly as the face translates forward relative to the cranial base, but SNB catches up faster, which is what pulls ANB down over time.

Sex. Males generally show larger linear craniofacial dimensions and, in many populations, slightly different angular norms. Several of the population studies above found significant sex differences in anterior and posterior facial heights even where the SNA, SNB, and ANB differences between the sexes were smaller. When your source provides sex-specific norms, use them.

Sources of error and reproducibility

The dominant source of variability in these three angles is landmark identification, not the drawing of the lines. A point, B point, and nasion all carry moderate identification error, and because ANB is a difference of two angles, that error compounds rather than averages out.

Layered on top of landmark judgement are radiographic magnification and calibration differences between machines, head positioning during exposure, and the fact that a lateral cephalogram superimposes bilateral structures onto a single two-dimensional plane.

The manual-versus-digital question is often misunderstood. Digital tracing removes protractor and drawing error, but it does not remove the judgement of where the landmark sits — that is still a clinical decision. AI auto-landmarking is improving quickly but still deviates by 1–3 mm on certain points and needs clinician verification before the numbers can be trusted.

Where this leaves you

SNA, SNB, and ANB earn their central place because three numbers localise a skeletal discrepancy to a jaw and grade its severity in a language every orthodontist reads. They lose that value the moment they are read in isolation, measured against the wrong population's norms, or trusted without acknowledging that A, B, and nasion all move.

So: trace the four points carefully, report all three angles rather than ANB alone, cross-check ANB against a cranial-base-independent measure such as the Wits appraisal, and choose norms that match the patient in front of you instead of a 1950s reference sample. We built BCeph partly so those steps stop being a chore — trace once, and Steiner, Wits, Eastman, and the rest read from the same landmark set, so the skeletal picture and its cross-checks arrive together instead of one analysis at a time.

Frequently Asked Questions

References

1. Downs WB. Variations in facial relationships: their significance in treatment and prognosis. Am J Orthod. 1948;34(10):812–840.
2. Riedel RA. The relation of maxillary structures to cranium in malocclusion and in normal occlusion. Angle Orthod. 1952;22(3):142–145.
3. Steiner CC. Cephalometrics for you and me. Am J Orthod. 1953;39(10):729–755.
4. Drummond RA. A determination of cephalometric standards for the American Negro. Am J Orthod. 1968;54(9):670–682.
5. Jacobson A. The "Wits" appraisal of jaw disharmony. Am J Orthod. 1975;67(2):125–138.
6. Ferrazzini G. Critical evaluation of the ANB angle. Am J Orthod. 1976;69(6):620–626.
7. Hussels W, Nanda RS. Analysis of factors affecting angle ANB. Am J Orthod. 1984;85(5):411–423.
8. Baik CY, Ververidou M. A new approach of assessing sagittal discrepancies: the Beta angle. Am J Orthod Dentofacial Orthop. 2004;126(1):100–105.
9. Hassan AH. Cephalometric norms for Saudi adults living in the western region of Saudi Arabia. Angle Orthod. 2006;76(1):109–113.
10. Comparison of cephalometric norms between Japanese and Caucasian adults in the antero-posterior and vertical dimension. Eur J Orthod. 2007;29(5):493–499.
11. Cephalometric comparison of skeletal and dental characteristics between typical Arab and Chinese adults. AIMS Med Sci. 2017;4(4):413–426.
12. Cephalometric norms for African Americans with normal occlusion in the Greater Philadelphia region. J Orofac Orthop. 2025.
13. Lateral cephalometric norms for Saudi adults: a meta-analysis — pooling multiple Saudi samples (e.g., Al-Showial 2000; Namankani 2004; Hassan 2006).
14. Comparison of popular sagittal cephalometric analyses (ANB, Wits appraisal, Beta, Yen, W angle) for validity and reliability. Saudi Dent J. 2017.

Population figures above are drawn from these primary studies and standard reviews; several secondary studies (Bailey & Taylor; Richardson; Cooke & Wei; Başçiftçi and colleagues) are referenced in the text for direction rather than as primary sources. For any individual patient, consult the norm set derived from that patient's own population.

About BCeph: BCeph is a free browser-based cephalometric analysis tool built by the clinical and engineering team at Bayan Healthcare Analytics, used by orthodontists and dental educators worldwide.