Spinopelvic Motion Lab

How to read the two classifications

The literature uses two related but distinct taxonomies, and mapping one onto the other is a common source of confusion:

• Hip–Spine groups (1A / 1B / 2A / 2B) — a 2×2 grid of exactly two axes: deformity (the number, from PI–LL mismatch) × stiffness (the letter, from ΔSS < 10°).
• Descriptive mobility patterns — normal, hypermobile, stiff, stuck standing, stuck sitting, fused/kyphotic — defined by pelvic behaviour and sacral-slope values.

They do not correspond one-to-one. Stuck-standing and stuck-sitting are both stiff (both "B"), yet they dislocate in opposite directions — the letter cannot distinguish them. A stiff pattern may be 1B or 2B depending on whether a deformity is also present. In this tool the preset name is a mobility-pattern label, while the badge is derived live from the sliders — change a slider and the badge can change independently of the preset name. That decoupling is the point, not a bug.

Teaching model

The model separates published measurements from teaching simplifications. Published criteria: the PI−LL mismatch threshold of 10°, the ΔSS stiffness cut-off of 10°, and the functional-anteversion gradient of roughly 0.7–0.8° per degree of pelvic tilt. Derived values: sacral slope from PI = PT + SS, functional cup orientation, and the neck-on-rim contact angle. Editable assumptions, exposed under Teaching assumptions in the control panel: the anteversion and inclination gradients, the low seated anteversion warning, and the external rotation assumed in the at-risk position. The impingement-free arc is an illustrative model region, not a validated patient-specific safe zone.

• Deformity uses the published criterion: PI − LL > 10°, measured on the standing frame. Pelvic incidence and standing lordosis are per-patient inputs, and the identity PI = PT + SS holds at every frame.
• Lordosis drives the drawing: the figure represents a 50° reference lordosis and distributes any difference evenly across the five lumbar joints. Real lordosis is not even — about two-thirds sits at L4–S1.
• PI−LL is itself correlated with PI, which limits it across the full PI range; relative lumbar lordosis has been proposed as more precise.
• Sagittal (2-D) projection only; true anteversion is a 3-D axial angle.
• Anterior dislocation needs external rotation. Both failure modes are the same event — the neck meeting the rim — read off opposite ends of one arc, so anteversion trades flexion range for extension range degree for degree. But in the pure sagittal plane the posterior rim sits near 50° of hip extension, beyond what any hip achieves: real anterior dislocation is an extension and external-rotation event. ER swings the neck toward that same rim, so it is modelled as an explicit, editable allowance (External rotation in the at-risk position) rather than being buried in a fitted constant. Set it to 0° to see the pure sagittal case. Adduction, a further contributor, is not modelled.
• Combined anteversion. Stability and impingement-free range depend on the combined (McKibbin) anteversion — functional cup AV plus stem AV — which is reported as its own metric and drives the instability verdict. The flexion impingement animation is keyed to the cup, so it shows the sagittal component; stem version acts on the combined value.
• Cup inclination changes functionally with pelvic tilt (reported as functional inclination). Its independent contribution to the 3-D impingement-free "safe zone" is a coronal-plane effect outside this sagittal view, so it is not folded into the sagittal impingement trigger.
• Head-size benefit is modelled continuously, though the real ROM benefit plateaus near 38 mm.
• Motion split, impingement threshold and functional-orientation factors are heuristic and editable under Teaching assumptions.

Skeletal illustration drawn in-house for this module (hip-spine.svg, Inkscape); the bone paths in anatomy.js are taken from that file. No third-party datasets, models or illustrations are used. Typefaces are IBM Plex Sans, IBM Plex Mono and Source Serif 4, served from Google Fonts under the SIL Open Font Licence. No other runtime dependencies.