Global ShearVector

A biomechanical metric that quantifies the cumulative shear force on the spine during upright posture -- combining magnitude and direction into a single resultant vector.

Ghaith Habboub, MD MS, Arpan Patel, MD, Jason Savage, MD, Dominic Pelle, MD, Thomas Mroz, MD, Michael Steinmetz, MD, Mohamed Macki, MD, Emily Hu, MD, Osama Kashlan, MD, Bilal Butt, MD & Seth Meade, MD MS

The Problem with Scalar Metrics

To understand why the GSV is needed, consider the tools we already have. Measures like Sagittal Vertical Axis (SVA) and Pelvic Incidence--Lumbar Lordosis mismatch (PI-LL) reduce complex 3D biomechanics to a single number. They tell you how far the spine is out of alignment, but not what forces result from that malalignment.

The missing ingredient: gravity. Every existing metric -- SVA, PI-LL, thoracic kyphosis, lumbar lordosis -- is purely geometric. Schwab et al. (2006) demonstrated that the gravity line shifts with age and pelvic parameters, yet these metrics measure angles or distances without accounting for the dominant physiological load on the spine: the force of gravity acting through body weight. A 10-degree tilt at L3 produces a very different shear force than the same 10 degrees at T4, because L3 carries far more cumulative weight above it. Without incorporating gravity, we are measuring shape but ignoring the forces that shape produces.

Balanced

Balanced Spine

Head centered over pelvis. Gravitational forces are mainly compressive -- minimal shear.

Imbalanced

Forward Imbalance

Head anterior to pelvis. Gravity generates large anterior shear forces that paraspinal muscles must continuously oppose. Protopsaltis et al. (2014) showed that T1 pelvic angle -- which captures both spinal inclination and pelvic tilt -- correlates with health-related quality of life, confirming that the force burden of imbalance translates directly to patient disability.

Dubousset's Cone of Economy: Dubousset described the "cone of economy" in the early 1970s -- a narrow postural envelope within which the body maintains upright balance with minimal energy expenditure (Hasegawa and Dubousset, 2022). When alignment falls outside this cone, compensatory mechanisms activate: pelvic retroversion, knee flexion, and increased paraspinal muscle recruitment. Abelin-Genevois (2021) defined sagittal balance as the trade-off between external forces and the trunk muscle response under sensorineural regulation. The GSV quantifies this trade-off directly, measuring the actual shear forces at every vertebral level that these compensatory mechanisms must resist.

Part I

Building the Metric

From spine geometry to gravitational force at every vertebral level

Modeling the Spine as a Spline

To move beyond angle-only metrics and calculate the actual gravitational forces on the spine, we first need to know the orientation of every vertebral endplate. From a standing lateral radiograph, the center of each vertebral body (C2 through S1 -- 24 points) is plotted and fitted with a natural cubic spline, creating a smooth continuous curve that captures the true shape of the spine. This approach aligns with Roussouly and Pinheiro-Franco's (2011) biomechanical framework, which emphasizes that the spino-pelvic organization must be understood as a continuous mechanical chain rather than a set of isolated angular measurements.

C2C5T3T7T11L2L5

Why a Cubic Spline?

A cubic spline passes exactly through every data point while maintaining continuous first and second derivatives. This means the curve -- and its slope -- change smoothly between vertebrae, just like the real spine.

Natural boundary conditions (zero second derivative at the endpoints) prevent artificial curvature at C2 and S1.

Weight Distribution

The GSV accounts for gravity acting through body weight at each vertebral level. Understanding how mass is distributed across body segments -- and why only a fraction reaches the spine above S1 -- is essential for interpreting force magnitudes.

50 kg120 kg

Segment Mass Distribution

S1
Segment% BWN
Head + Neck8.156
Upper Extremities10.069
Thorax21.6148
Abdomen13.995
Pelvis *14.298
Thighs20.0137
Legs + Feet12.284
Total100686
* Pelvis (14.2% BW) sits below S1 and is not transmitted up the spine
Dempster WT. Space requirements of the seated operator. Wright-Patterson Air Force Base; 1955. As compiled by Winter DA. Biomechanics and Motor Control of Human Movement. 4th ed. Wiley; 2009:82-106. Segment parameters adjusted per de Leva P (1996, PMID:8872282), whose corrections to Zatsiorsky-Seluyanov's original data remain the standard reference for segment inertia parameters. J Biomech. 1996;29(9):1223-1230.
Above S1: Head + neck + upper extremities + trunk = 65% of body weight (446 N at 70 kg). The pelvis and legs hang below.

Cumulative Weight at Each Level

GSV Model -- Per Level

C27.7%C47.7%C77.7%T112.8%T316.8%T521.0%T725.3%T930.3%T1137.0%T1241.0%L144.9%L352.4%L560.8%S165.0%Pelvis79.2% (+14.2%)Legs100% (+20.8%)
CervicalThoracicLumbar

Winter/Dempster -- Segment Regions

Head + Neck8.1%56 N
Upper Extremities10.0%69 N
Thorax21.6%148 N
Abdomen13.9%95 N
S1
Pelvis14.2%97 N
Thighs20.0%137 N
Legs + Feet12.2%84 N
Why 65% vs 53.6%?

Winter segments above S1 sum to 53.6% BW (head 8.1% + arms 10% + thorax 21.6% + abdomen 13.9%). The GSV model uses 65% to include tissue masses distributed along the upper lumbar spine (intervertebral discs, soft tissue) not captured by Winter's limb-based segmentation, and to match validated cadaveric disc load measurements.

Force Decomposition at Each Level

With the spline in hand, we know the local orientation of each endplate. Now we can do what purely geometric metrics cannot: incorporate gravity. At every level, the cumulative body weight above pulls straight down. Using the endplate tilt from the spline, this gravitational force is decomposed into two orthogonal components:

WShearNormal20 deg

Shear Force (parallel to endplate)

Fshear = g × W × sin(angle) × level_proportion

This force tends to slide the vertebra forward or backward. Paraspinal muscles and ligaments must resist it.

Normal Force (perpendicular to endplate)

Fnormal = g × W × cos(angle) × level_proportion

This compressive load acts along the spine axis and is borne by the intervertebral disc and vertebral body. By Newton's third law, the disc pushes back with an equal and opposite reaction force, effectively canceling the normal component -- no additional muscles or ligaments are needed.

Weight Distribution

Each level bears a different proportion of body weight. The head contributes ~7%, cervical region ~4%, and the trunk and upper extremities ~54%. Lower vertebrae bear cumulatively more weight.

Why the spine is an S-curve, not a straight column. If the spine were perfectly vertical, all gravitational force would be normal (compressive) -- maximizing the load on every intervertebral disc. The natural S-shaped curvature (cervical lordosis, thoracic kyphosis, lumbar lordosis) tilts each endplate slightly, redirecting a portion of the compressive load into shear. This reduces the normal force at each level, protecting the discs from excessive compression. The trade-off is that shear must now be resisted by muscles, facets, and ligaments. In a healthy spine this balance is efficient. In deformity, the curvatures become exaggerated or lost, and shear increases dramatically -- which is exactly what the GSV measures: the total cost of that imbalance.

Newton's Third Law and the Spine

Every action has an equal and opposite reaction. This principle determines which spinal structures must resist which forces -- and explains why the GSV focuses on shear, not compression.

Normal forces are self-canceling

The compressive (normal) component acts perpendicular to the endplate and is transmitted axially through each vertebral body and intervertebral disc. At every level, the disc and bone push back with an equal and opposite reaction force -- Newton's third law in action. Wilke et al. (1999) confirmed this compressive chain in vivo, measuring intradiscal pressures of 0.1-2.3 MPa across daily activities. This chain of action-reaction pairs continues down the entire column until the sacrum transfers the load to the pelvis, and the ground reaction force ultimately cancels it. No muscles or ligaments are needed to resist normal forces.

Shear forces require active resistance

The shear component acts parallel to the endplate and tends to slide one vertebra on the next. The intervertebral disc alone cannot generate an adequate reaction force in this direction -- it is not rigid enough. Instead, shear must be resisted by the facet joints (bony contact), spinal ligaments (passive tension), and paraspinal muscles (active contraction). The greater the endplate tilt, the larger the shear fraction, and the harder these structures must work. Schmidt et al. (2013) used finite element modeling to quantify lumbar shear response, showing that the facet joints bear 20-40% of the shear load in neutral posture, with the percentage increasing during combined loading. Cornaz et al. (2021) demonstrated that spinal ligament stiffness degrades with disc degeneration, compounding the loss of passive shear restraint at each segment.

Why this matters clinically

Because normal forces are passively canceled through the bony column and ground reaction, they do not create ongoing muscular demand. Shear forces, however, require continuous active effort from the paraspinal muscles and integrity of the facet joints and ligaments. In spinal deformity, increased endplate tilt shifts more of the body weight into shear, raising muscle energy expenditure and accelerating facet and ligament degeneration. The GSV sums these shear forces across every level -- quantifying the total biomechanical burden that muscles, facets, and ligaments must bear. Meakin et al. (2013) provided direct evidence for this relationship, showing that extensor muscle volume in the lumbar spine scales with the degree of lordosis -- spines with greater curvature (and therefore greater shear demand) develop proportionally larger muscles to resist it.

Local Self-Cancellation: Why We Sum Per-Level Shears

A natural question arises: if 65 % of body weight rests on L5, why not calculate a single enormous shear force at the lumbosacral junction? Why decompose the spine into per-level shear components and then sum them? The answer lies in how shear is resisted — locally, at every motion segment, by the tissues that cross it.

Shear is resisted at the level it is generated

At L4–L5, the shear produced by the tilt of the L4 endplate is absorbed by the structures crossing that segment: the multifidus and other paraspinal muscles inserting on L4 and L5, the facet capsules of the L4–L5 zygapophyseal joints, the interspinous and supraspinous ligaments spanning those spinous processes, and the annulus fibrosus of the L4–L5 disc. These local tissues generate an equal-and-opposite reaction that cancels the shear before it can propagate to the next level.

Only the normal force propagates downward

What is passed from L4 to L5 is the compressive (normal) component — transmitted axially through the vertebral body and disc, and cancelled at each level by Newton's third law. Shear does not accumulate as a growing horizontal force travelling down the column; it is handed off to connective tissue at every segment and disappears into muscle work, ligament tension, and facet contact.

Why we compute segmental shear, not a single L5 force

If shear propagated like a fluid, we could lump everything into one calculation at L5: 65 % of body weight multiplied by the sine of the L5 endplate tilt. But because shear is locally cancelled, the shear demand at L5 is governed only by the L5 endplate angle acting on the weight above it — not by any "leftover" shear from T4 or L1. Each segment is a small, largely independent problem: cumulative weight above it times the sine of its own tilt. The GSV adds these segmental demands together not because the forces physically stack in a chain, but because their sum measures the total biomechanical work that the spine's muscles, facets, and ligaments must perform to hold the body upright.

The GSV as a global tissue-work integral

Read this way, the GSV is not a single resultant force travelling through the spine — it is the vector sum of locally generated, locally cancelled shear demands. A healthy S-curve cancels shear efficiently at every level, and the resulting GSV is small. In deformity, several adjacent levels tilt in the same direction and their local demands add constructively, producing a large GSV. That magnitude reflects the cumulative muscular and ligamentous effort required segment by segment — a biomechanical cost that shows up clinically as fatigue, pain, facet arthropathy, and progressive degeneration.

The Full Picture: 6 Degrees of Freedom

Each vertebra can move in six independent ways -- three translational and three rotational. The diagrams below show L4 moving relative to a fixed L5 in each anatomical plane. Drag the slider to explore each degree of freedom and observe the disc space deformation.

Lateral View

CranialCaudalAntPostL5L4

0.0°

Sagittal rotation -- primary driver of shear forces the GSV quantifies.

Coronal View

CranialCaudalRtLtL5L4

0.0°

Coronal rotation. Contributes to scoliotic deformity and asymmetric disc loading.

Axial View

AnteriorPosteriorRtLtL5L4

0.0°

Rotation about the vertical axis. Torsional stress on the disc annulus.

Translational DOFs

Normal / Axial Compression

Perpendicular to the endplate along the spine axis. Borne by the vertebral body and intervertebral disc. Increases caudally as body weight accumulates above each level.

Anterior-Posterior Translation

Sagittal-plane translation parallel to the endplate -- the primary component the GSV captures from a lateral radiograph. Unlike normal forces, which are carried by the intervertebral disc and canceled by Newton's third law, shear forces from this translation require paraspinal muscles, facet joints, and ligaments to resist.

Lateral Translation

Coronal-plane translation parallel to the endplate. Captured by the coronal GSV (C-GSVh) from an AP radiograph. The resulting shear forces must be resisted by muscles, facets, and ligaments -- the disc alone cannot cancel them. Present in scoliosis, lateral listhesis, and asymmetric loading.

Rotational DOFs

Pitch -- Flexion / Extension

Rotation about the lateral axis. Forward pitch (flexion) is the dominant sagittal-plane moment -- it drives kyphosis and the shear forces the GSV quantifies. Resisted by the posterior tension band (muscles, ligaments, facet capsules).

Roll -- Lateral Bending

Rotation about the AP axis. Produces asymmetric loading on the disc and contributes to coronal deformity. The intertransverse ligaments and lateral trunk muscles resist this moment.

Yaw -- Axial Rotation

Rotation about the vertical axis. Generates torsional stress on the disc annulus. Particularly relevant in rotatory scoliosis and degenerative subluxation. The facet joints are the primary restraint.

Why the GSV focuses on shear: As Newton's third law shows, normal forces are passively canceled through the bony column down to the ground reaction force -- they cost the body nothing. Shear forces, however, must be actively resisted by facet joints, ligaments, and paraspinal muscles. The rotational moments (pitch, roll, yaw) are coupled to these shear forces through the spine's geometry -- a tilted endplate that generates shear also creates a flexion moment. The GSV captures this cumulative shear burden across all levels, directly quantifying the demand on these structures.

Summing Into the Global Shear Vector

Now that we can decompose gravity into forces at each level -- something no angle-based metric does -- we add up all the shear components. The GSV is the vector sum of all per-level shear forces -- a single arrow that captures both the total magnitude and the net direction of gravitational shear across the entire spine.

Each yellow arrow represents the shear force at one vertebral level. The direction depends on the local tilt of the endplate, and the length depends on the force magnitude (weight x sin(angle)).

Adding all these vectors tip-to-tail yields a single resultant vector -- the GSV -- with two components:

S-GSVS1-m
Magnitude (Newtons) -- the total shear load
S-GSVS1-a
Angle (degrees) -- 0 deg = straight down (optimal)
Part II

Interpreting the GSV

What the magnitude and angle tell us, and the formal mathematics behind them

Reading the GSV

We now have a single resultant vector. How do we interpret it? The GSV has two components that together provide a complete picture of spinal shear loading.

+900-905

Well-balanced. The GSV points nearly straight down -- minimal net shear. Paraspinal muscles are under low demand.

S-GSVS1-a (Angle)

0 degrees = shear directed straight down (caudal). This is optimal -- the spine is balanced and gravitational forces are primarily compressive. The needle points downward on the gauge.

Positive angles (left on gauge) = anterior shear (forward lean, left on lateral X-ray). Negative angles (right on gauge) = posterior shear. Larger absolute values indicate more imbalance.

S-GSVS1-m (Magnitude)

Measured in Newtons (N). Higher values mean more total shear force that muscles and ligaments must actively resist.

Patients with SVA > 10cm have roughly twice the continuous shear force compared to balanced spines.

Regional Sub-Vectors

The GSV also breaks down into cervical (RSV-C), thoracic (RSV-T), and lumbar (RSV-L) components, showing how each region contributes to the total.

Labeling Convention

Different publications, datasets, and software tools refer to GSV measurements inconsistently — sometimes sGSV, sometimes GSV-A, sometimes just “GSV.” This canonical convention separates the metric definition (plane and component) from the measurement anchor (anatomical reference) so that every reported value carries its own unambiguous label.

The form
[Plane]-GSV[Reference]-[Component]
Read as: planemetric → (anchored at) referencecomponent.

Slot definitions

SlotDelimiterRendering & values
PlaneS sagittal · C coronal · T total. Normal text.
GSV-Core token; always present.
Reference(none)Subscript. Anatomical anchor: S1, L5, h (hip), etc.
Component-Optional. a angle, m magnitude. Omitting refers to the vector form.

Examples

Display formMeaning
S-GSVS1-aSagittal GSV angle anchored at S1
S-GSVS1-mSagittal GSV magnitude anchored at S1
S-GSVS1Sagittal GSV vector at S1 (component unspecified)
C-GSVh-aCoronal GSV angle anchored at the hip
C-GSVh-mCoronal GSV magnitude anchored at the hip
T-GSVS1-mTotal GSV magnitude anchored at S1

Reference vocabulary

CodeTypeDefinition
S1Vertebral bodySacral S1 endplate
L5Vertebral bodyL5 endplate or centroid (specify in protocol)
C7Vertebral bodyC7 centroid
T1Vertebral bodyT1 centroid
hJoint axisBicoxofemoral / hip axis
FHJoint axisFemoral head center (reserved if distinguished from hip axis)

Rendering by medium

MediumExampleNote
Display (canonical)S-GSVS1-aSubscript on reference; baseline elsewhere. Use in publications, slides, and figures.
LaTeX\text{S-GSV}_{S1}\text{-a}Wrap text-mode tokens to preserve hyphens; subscript via _{...}.
Markdown / HTMLS-GSV<sub>S1</sub>-aUse the <sub> tag for the reference token only.
Word / PowerPointS-GSVS1-aApply subscript formatting to the reference token; leave plane, GSV, and component at baseline.
Plain-text fallbackS-GSV_S1-aWhen subscript cannot render (logs, CSV labels, email).
Code-safe (SQL, dict keys)S_GSV_S1_aHyphens replaced with underscores. Use as column names, variable names, and dict keys.

Try it

Canonical (display)
S-GSVS1-a
Plain-text
S-GSV_S1-a
Code-safe
S_GSV_S1_a

Mathematical Derivation of the GSV

The summation above gives the intuition. Here we formalize each step into precise equations, from raw landmark coordinates to the final GSV vector.

Part III

The Pelvis and Clinical Application

The spine does not end at S1 -- pelvic compensation, deformity simulation, and clinical evidence

Pelvic Parameters: The Foundation of Balance

The GSV tells us about forces in the spine -- but the spine does not exist in isolation. It sits atop the pelvis, and pelvic morphology fundamentally shapes how the spine must align to maintain balance. Boulay et al. (2006) established that pelvic incidence regulates the magnitude of lumbar lordosis, linking pelvic anatomy to spinal curvature. Three interconnected parameters describe this relationship.

FHS1PT=20PI=55SS=3591.6 N33.3 NPI = PT + SS (55 = 20 + 35)

Drag to see how PT and SS change while PI remains fixed at 55 deg

Pelvic Incidence (PI)

The angle between the line perpendicular to the sacral endplate at the midpoint of S1 and the line from S1 to the center of the femoral heads. PI is a fixed anatomical parameter -- it does not change with posture. It defines the "budget" of pelvic compensation available.

Normal range: 40-65 deg. Higher PI allows more lordosis. Roussouly and Pinheiro-Franco (2011) demonstrated that PI is the fundamental regulator of sagittal alignment: it determines the sacral slope, which in turn dictates the required lumbar lordosis. Patients with high PI (>65 deg) require substantially more lordosis to maintain balance, while low PI (<40 deg) limits the compensatory capacity of the pelvis.

Pelvic Tilt (PT)

The angle between the vertical and the line from the femoral head center to the midpoint of S1. PT is a positional parameter -- it changes as the pelvis rotates to compensate for spinal imbalance. Increased PT (pelvic retroversion) is a compensatory mechanism for sagittal malalignment.

Normal: 10-25 deg. Elevated PT (>25 deg) indicates compensation. Ryan et al. (2014) showed that T1 pelvic angle, which incorporates PT, is a more reliable marker of global sagittal deformity than SVA alone because it is less susceptible to postural compensation strategies like knee flexion.

Sacral Slope (SS)

The angle of the sacral endplate from horizontal. SS determines how much lordosis the lumbar spine must develop to stay balanced. The fundamental identity PI = PT + SS always holds -- as the pelvis tilts back (PT increases), the sacral slope decreases by the same amount.

Normal: 30-45 deg. Low SS with high PI suggests pelvic retroversion.

Pelvic Forces (70 kg patient, pelvis = 97.5 N)

33.3 N
Shear (posterior, perp. to S1-FH)
91.6 N
Hip compression (along S1-FH)

The pelvis weighs 14.2% of body weight (Winter/Dempster). For a 70 kg patient, the pelvis exerts 97.5 N of gravitational force. Pelvic tilt determines how this splits: shear is always posteriorly directed and must be actively resisted by hip extensors, while compression is passively carried by bone along the S1-FH axis. These values scale linearly with patient weight.

70 kg x 14.2% x 9.81 = 97.5 N | x sin(20 deg) = 33.3 N shear | x cos(20 deg) = 91.6 N compression

Pelvic Shear: The Cost of Compensation

The spine does not end at S1. Below it, the pelvis -- weighing 14.2% of body weight -- must transmit all spinal load to the femoral heads. Pelvic tilt determines how much of that weight becomes shear. Unlike the GSV, which can point anteriorly or posteriorly depending on spinal curvature, pelvic shear is always posteriorly directed -- a consequence of retroversion that hip extensors must actively resist.

Clinical States

The interplay between spine GSV (anterior or posterior) and pelvic shear (always posterior, 14.2% BW x sin(PT)) defines five clinical patterns. Combined shear -- the vector sum of both -- represents the total burden on the musculoskeletal system.

StateSpine GSVPelvic ShearCombined
BalancedLowModerateLow
CompensatedModerateHighModerate
Rigid PelvisHighLowHigh
DecompensatedVery HighVery HighVery High
Pelvic DominantLowHighModerate

Pelvic shear = 14.2% BW x sin(PT) per Winter/Dempster. Absolute forces scale with patient weight -- use the Interactive Calculator or Spine Deformity Simulation for patient-specific values.

The Pelvis as Shock Absorber

In a balanced spine, the GSV is small and anteriorly directed. The pelvis already contributes a larger posterior shear force from its own weight alone. As deformity increases anterior GSV, the pelvis retrovertes (increasing PT) to generate an opposing posterior shear that compensates. This retroversion further increases pelvic shear. The hip extensors (gluteals, hamstrings) must actively generate this posterior pelvic shear force.

When compensation capacity is exhausted, both spine GSV and pelvic shear remain high. The combined shear burden exceeds what the hip extensors can sustain, and the patient decompensates -- a state that Abelin-Genevois (2021) describes as the breakdown of the sensorineural regulation loop, where the trunk muscles can no longer maintain stable posture against external forces. The Spine Deformity Simulation above shows this progression in real time.

Spine Deformity Simulation

Beyond the traditional PI, PT, and SS described earlier, recent work has shown that measuring pelvic angles at specific vertebral levels provides a finer-grained view of spino-pelvic alignment. Each level pelvic angle is measured at the femoral head between the line to S1 and the line to that vertebra. The T4 and L1 pelvic angles are increasingly recognized as key indicators of sagittal alignment, capturing both thoracic deformity and thoracolumbar junction mechanics.

BalancedSagittal Imbalance

Spine Deformity Simulation

C7-SVA = 0.8 cmC2-C7 = 1.7 cmSS=39°GSV 32NCombined 59NPelv 114NC2C3C4C5C6C7T1T2T3T4T5T6T7T8T9T10T11T12L1L2L3L4L5S1FH

Hover to see pelvic angles

Display

Pelvic Parameters

55.7
PI (deg)
16.7
PT (deg)
39
SS (deg)
PI = PT + SS (55.7 = 16.7 + 39)
-0.8 cm
C7-SVA
-1.7 cm
C2-C7 SVA
-3 deg
PI-LL

Regional Curvature

1
CL (deg)
-30
TK (deg)
59
LL (deg)

Global Shear Vector

32 N
GSV Magnitude
-18 deg
GSV Angle

0° = caudal (balanced), positive = anterior shear

Combined Shear (GSV + Pelvic)

59.3 N
Magnitude
-9.6 deg
Angle

Spine GSV + pelvis weight shear (14.2% BW at PT = 16.7 deg)

Total Shear Burden

Spine GSV
32 N
Pelvic Shear (14.2% BW)
28 N
Combined Shear60 N

Anterior GSV dominates -- spinal deformity is the primary shear source

Key Level Pelvic Angles

C2
11.7 deg
T1
12.6 deg
T4
10.7 deg
L1
10.7 deg
T4-PA10.7 deg

Aligned

L1-PA10.7 deg

Aligned

Target: 7 deg (PI/2-21)

T4-PA captures global sagittal alignment from the upper thoracic spine to the pelvis, making it sensitive to thoracic hyperkyphosis and overall trunk inclination. Ryan et al. (2014) and Protopsaltis et al. (2014) demonstrated that T1 pelvic angle -- closely related to T4-PA -- correlates more strongly with health-related quality of life than SVA, because it accounts for both spinal inclination and pelvic compensation simultaneously. T4-PA has emerged as a key surgical planning target.

L1-PA isolates the thoracolumbar junction contribution and is particularly useful for detecting proximal junctional kyphosis (PJK) and thoracolumbar malalignment that T4-PA alone may mask. Together, T4-PA and L1-PA provide a more complete picture than any single global metric.

SRS-Schwab targets: SVA < 4 cm, PT < 20 deg, PI-LL < 10 deg, validated by Liu et al. (2013) across 492 ASD patients. Target L1-PA = PI/2 - 21 deg, T4-L1PA mismatch< 4 deg indicates balanced regional alignment (Lafage et al.).

Clinical Significance

With the biomechanical framework and pelvic foundation in place, what does the evidence show? Research from the Cleveland Clinic Center for Spine Health demonstrates that the GSV correlates strongly with established alignment measures, patient outcomes, and paraspinal muscle morphology. Unlike purely geometric metrics, the GSV integrates gravitational loading with spinal geometry -- which is why it captures variance in patient-reported outcomes that SVA and PI-LL cannot explain independently (Schwab et al., 2006; Abelin-Genevois, 2021).

R = -0.82
S-GSV_S1-a vs SVA
Strong inverse correlation with sagittal vertical axis
R = 0.50
S-GSV_S1-a vs PROMIS-PH
Correlates with patient-reported physical health outcomes
R² = 0.86
GSV vs Multifidus pCSA
Strongly predicts paraspinal muscle cross-sectional area
R² = 0.90
GSV vs Erector Spinae pCSA
Near-perfect correlation with erector spinae muscle size

Key Findings

  • S-GSVS1-a explains ~9% more variability in quality-of-life scores (PROMIS-PH) than SVA alone, providing additional clinical insight beyond traditional metrics.
  • The GSV captures forces that paraspinal muscles must oppose. Patients with higher shear demands develop larger paraspinal muscles (particularly multifidus, erector spinae, and psoas major).
  • The multifidus muscle is uniquely sensitive to global shear loading, underscoring its role as a deep spinal stabilizer and its vulnerability in sagittal imbalance. Sun et al. (2019) showed that multifidus atrophy correlates with spinopelvic parameter severity in adult degenerative scoliosis, and Panjabi (1992, 2003) identified the muscular subsystem as the primary active stabilizer whose failure leads to clinical instability.
  • The GSV may help predict proximal junctional kyphosis (PJK) risk after adult spinal deformity surgery by quantifying the mismatch between shear forces and the musculoligamentous complex's capacity to resist them.

SVA Group Comparison

SVA GroupMean S-GSVS1-aMean S-GSVS1-m (N)
< 0 cm~85~15
0 - 5 cm~75~25
5 - 10 cm~55~45
> 10 cm~35~75

Roussouly Classification of Sagittal Alignment

The Roussouly classification (2005) was the first system to describe normal variation in sagittal alignment of the asymptomatic spine. Studying 160 volunteers, Roussouly classified spines into four types based on sacral slope and the location of the lordosis apex -- demonstrating that there is no single "normal" alignment, but rather distinct morphological patterns determined by pelvic anatomy.

The Four Types (+ Type 3 AP)

TypeSacral SlopeLordosis Apex
Type 1< 35Center of L5
Type 2< 35Base of L4
Type 335 - 45Center of L4
Type 4> 45L3 or higher
Type 3 APLow PIVery low / negative PT

PI determines the theoretical (ideal) type: Type 1/2 correspond to PI < 45 deg, Type 3 to PI 45-60 deg, Type 4 to PI > 60 deg. Laouissat et al. (2018) added Type 3 AP -- an anteverted pelvis variant with hyperlordosis and low or negative PT -- refining the original classification.

GSV by Roussouly Type

Each Roussouly type produces a distinct shear force pattern. Select a type below to run it through the GSV calculator -- the same engine used throughout this application. All models use 70 kg body weight.

Type 3

Apex at L4 | SS 35 - 45 deg
S1FHC2C5T1T4T7T10L1L4SSApexL4GSV 37N
Type 3SS 35 - 45 deg | Apex at L4

Well-balanced lordosis and kyphosis with harmonious curves. Considered the optimal morphotype. Best biomechanical stability and shear cancellation.

GSV Analysis at 70 kg

37
S-GSVS1-m (N)
-18°
S-GSVS1-a
259
Normal (N)
Regional Curvature
CL
30°
TK
69°
LL

All Types Compared (70 kg)

Type 3 (balanced curves) achieves the lowest GSV through optimal shear cancellation. Types 1/2 (flat lordosis) have higher shear due to reduced opposing curvature. Type 4 (long lordosis) shows moderate GSV.

Why Classification Matters for Surgery

Matching postoperative alignment to the patient's ideal Roussouly type -- determined by their fixed PI -- reduces mechanical complications and improves clinical outcomes. Three lines of evidence support this.

Mechanical Complication Prevention

Bari et al. (2020) showed a 72% mechanical complication rate when postoperative alignment did NOT match the patient's ideal Roussouly type, versus 15% when matched.

Confirmed by systematic review and meta-analysis (Aoun et al. 2025): Roussouly type-matched surgical planning significantly reduces rod breakage, PJK, and revision rates.

Complementary to SRS-Schwab

Passias et al. (2021) demonstrated that patients matching BOTH their ideal Roussouly type AND improving SRS-Schwab modifiers had superior outcomes compared to either system alone.

Roussouly defines the target shape (what the spine should look like). SRS-Schwab quantifies severity (how far it deviates). Together they provide both direction and magnitude for correction.

Degenerative Patterns by Type

Types 1 and 2 (flat lordosis) predispose to disc degeneration, particularly at L4-L5. Types 3 and 4 predispose to facet joint degeneration due to higher compressive facet loads.

Type mismatch after fusion surgery predicts adjacent segment disease. Restoring the patient's native type minimizes abnormal loading at adjacent levels.

Shape + Force: Roussouly Meets the GSV

The Roussouly classification is categorical and shape-based: it identifies which morphological pattern a spine belongs to. The GSV is continuous and force-based: it quantifies the total shear burden in Newtons with a direction in degrees. Each system captures what the other misses.

Characteristic GSV Signatures by Roussouly Type

TypeShear CancellationS-GSVS1-mS-GSVS1-a
Types 1 / 2ReducedHigherAnterior
Type 3OptimalLowestNear zero
Type 4ModerateModeratePotentially posterior

Types 1/2 have reduced lordosis and less opposing shear between thoracic and lumbar regions, yielding higher GSV magnitude with anterior direction. Type 3 achieves near-perfect shear cancellation. Type 4, with long lordosis, may shift the GSV posteriorly.

Proposed Clinical Workflow

1. Classify
PI-based type
2. Quantify
S-GSV_S1-m and S-GSV_S1-a
3. Plan
Shape + force targets
4. Validate
GSV Success Zone
5. Monitor
Continuous GSV tracking

The GSV Success Zone

We can now read S-GSVS1-a and S-GSVS1-m independently. But clinical outcome is governed by their combination -- just as the SRS-Schwab classification (Lowe et al., 2006; Liu et al., 2013) uses multiple thresholds (SVA, PT, PI-LL) rather than a single number. A one-dimensional threshold on either GSV parameter alone can misclassify patients in both directions.

Patient A
S-GSVS1-a = 5°, S-GSVS1-m = 30 N

Small angle, low magnitude. Minimal shear in any direction. Both metrics agree: this spine is well-balanced.

Good outcome expected
Patient B
S-GSVS1-a = 5°, S-GSVS1-m = 180 N

Same small angle -- but 6x the shear magnitude. An angle-only threshold would classify this as "well-balanced," missing the enormous force burden.

Poor outcome likely
Patient C
S-GSVS1-a = 35°, S-GSVS1-m = 25 N

Large angle suggesting imbalance, but the magnitude is low -- the actual shear force is small. A magnitude-only threshold would miss the angular deviation.

Threshold case

These examples illustrate a fundamental limitation: no single threshold on angle or magnitude can correctly classify all three patients. We need a two-dimensional zone that captures their interaction -- where the tolerable angle narrows as magnitude increases, and vice versa.

Interactive Zone Map

-60°-40°-20°0°20°40°60°050100150200250S-GSV_S1-a (degrees)S-GSV_S1-m (N)SuccessModerate
achieved MCIDdid not achieve MCID
Success Zone

Both GSV direction and magnitude fall within favorable ranges. Low shear burden with well-directed forces -- the biomechanical conditions for a good clinical outcome.

Current position
Angle: 8° (anterior)
Magnitude: 35 N
The elliptical boundary captures the interaction -- at higher magnitudes, even smaller angular deviations push the vector outside the success zone.
Why an ellipse?

The zone boundary follows an equation of the form (A/Aw)2 + (M/Mh)2 = 1. This means the tolerable angle shrinks as magnitude increases, and vice versa -- matching the clinical intuition that high force at any angle, or any force at extreme angles, indicates risk.

Calibrating the Zone Boundaries

The zone boundaries shown above are illustrative. In practice, the angle and magnitude tolerances would be calibrated independently from outcome data -- there is no reason to expect symmetric sensitivity along both axes. Adjust each knob to see how the ellipse reshapes and how classification performance changes.

Controls how wide the ellipse is horizontally -- how much angular deviation is tolerated before crossing the boundary.

Controls how tall the ellipse is vertically -- how much shear magnitude is tolerated before crossing the boundary.

Classification Performance
97%
Sensitivity
67%
Specificity
86%
Accuracy

Notice how widening angle tolerance alone may improve sensitivity without affecting specificity -- and vice versa for magnitude. The optimal ellipse is almost certainly asymmetric.

1. Outcome data collection

Prospective cohort measuring postoperative S-GSVS1-a and S-GSVS1-mpaired with HRQOL scores (ODI, SF-36, VAS back pain) at minimum 2-year follow-up. Target sample: 200+ patients across the severity spectrum.

2. Boundary optimization

Two-dimensional ROC analysis: vary the elliptical boundary parameters and compute sensitivity and specificity against MCID achievement. The optimal boundary maximizes the Youden index (sensitivity + specificity - 1).

3. Subgroup refinement

Zone boundaries may differ by age, BMI, diagnosis, and surgical approach. Bayesian hierarchical models can estimate group-specific ellipses while sharing statistical strength across subgroups.

Part IV

Interactive Tools

Apply the GSV to real patient data with 2D and 3D calculators

Interactive Calculator

Explore how spinal alignment affects shear forces. Select a preset, upload your own X-ray image, or edit coordinates directly.

Well-balanced spine with cervical lordosis, thoracic kyphosis, and lumbar lordosis.

Spine View

C2C3C4C5C6C7T1T2T3T4T5T6T7T8T9T10T11T12L1L2L3L4L5S1

Global Shear Vector

Ant0Post-41
S-GSVS1-a (degrees)
18
S-GSVS1-m (Newtons)

Regional Vectors

Cervical (RSV-C)
4 N-81 deg
Thoracic (RSV-T)
14 N-58 deg
Lumbar (RSV-L)
7 N36 deg

Curvature Estimates

Cervical Lordosis (C2-C7)1 deg
Thoracic Kyphosis (T4-T12)-26 deg
Lumbar Lordosis (L1-S1)28 deg
Upper Lumbar (L1-L4)17 deg
Lower Lumbar (L4-S1)11 deg
Per-Level Data
LevelAngleShear (N)Normal (N)Cum GSV (N)Cum Angle
C2-96-5.752.30.5-84
C3-100-9.551.81.3-81
C4-102-10.851.52.3-80
C5-100-9.351.83.1-80
C6-97-6.452.23.6-80
C7-95-4.252.54-81
T1-90-0.687.74.1-81
T2-8410.599.32.8-74
T3-8020.9113.41.31
T4-7826.41273.457
T5-8026.1141.75.967
T6-8416.5157.67.470
T7-90-0.31747.370
T8-96-18.4189.75.963
T9-102-42.9203.83.827
T10-105-58.22236.1-37
T11-104-60.6246.310.9-56
T12-104-69.4272.617.5-63
L1-103-70.7299.723.5-67
L2-99-50.3330.627.4-69
L3-93-19.8359.328.7-70
L4-8628.3387.626.8-68
L5-7887.2408.222.1-59
S1-75117.5430.517.7-41

Coordinates

LevelXY
C23530
C33250
C42870
C52490
C621110
C719130
T118150
T219170
T322190
T426210
T530230
T633250
T734270
T833290
T930310
T1025330
T1120350
T1215370
L110390
L26410
L34430
L44450
L57470
S112490
Generate Report:Landscape report with measurements for patient chart

From 2D to 3D: The Total GSV

Everything above describes the GSV in the sagittal plane -- but real spinal deformity is three-dimensional. By repeating the same calculation on a coronal (AP) radiograph, we obtain a coronal GSV. The T-GSVS1 combines both into a single 3D resultant vector.

RtLtCaudalPostAntGSVS,S1GSVC,hGSVT,S1GSVT,S1 = sqrt(GSVS,S1² + GSVC,h²)Drag to rotateReset view

S-GSVS1 -- Sagittal Global Spine Vector

Calculated from a lateral (sagittal) radiograph. Captures shear forces in the anterior-posterior plane (forward/backward bending). This is the traditional GSV described in the earlier sections.

0 deg = straight down (optimal). Positive = anterior (left on X-ray).

C-GSVh -- Coronal Global Spine Vector

Calculated from a frontal (coronal) radiograph. Captures shear forces in the left-right plane, quantifying lateral deviation as seen in scoliosis.

0 deg = straight down (optimal). Positive = right deviation.

T-GSVS1 -- Total Global Spine Vector

The 3D resultant combining both views. Provides a single metric for overall spinal deformity severity.

T-GSVS1 = sqrt(S-GSVS1² + C-GSVh²)

A patient with both sagittal and coronal deformity will have a T-GSVS1 larger than either component alone.

In Development

3D GSV Calculator

Upload sagittal (lateral) and coronal (AP) radiographs to compute S-GSV, C-GSV, and the total (T-GSV) from real patient images.

Upload at least one radiograph to begin.

Use a lateral (sagittal) view for S-GSV and/or an AP (coronal) view for C-GSV. Both are needed for T-GSV.

Part V

Advanced Topics

Comparative evolution, surgical simulation, kinematics, and segment-specific modeling

Comparative Vertebrate Analysis

The calculators above work for any lateral radiograph -- but why does the human spine have an S-curve at all? Not all bipeds have one. By running different spinal geometries through the same GSV engine, we can see exactly how curvature pattern determines shear burden -- and why the S-curve is uniquely optimized for sustained upright walking and running.

How the S-Curve Cancels Shear

The diagram below shows the same gravitational decomposition on two spines standing upright. Each arrow is the shear force at one vertebral level -- colored by direction. Red arrows point anteriorly (forward); blue arrows point posteriorly (backward). The purple arrow is the net GSV: the vector sum of all per-level shear forces.

Human S-CurveUpright postureGorilla C-CurveForward-pitched trunkGSV 18 NGSV 89 NCervicallordosisThoracickyphosisLumbarlordosisContinuouskyphosisAnterior shearPosterior shearNet GSVOpposing forces cancelAll forces compound

The Cancellation Principle

When a vertebral endplate tilts forward (kyphosis), gravity pulls the body into anterior shear -- the red arrows. When it tilts backward (lordosis), gravity creates posterior shear -- the blue arrows.

The human S-curve alternates between kyphosis and lordosis, so each region generates shear in the opposite direction from its neighbors. When these opposing forces are summed, they partially cancel. The net GSV is much smaller than any individual region's contribution.

A C-curve (like the gorilla) has no alternation. Every endplate tilts the same way, so every per-level shear arrow points in the same direction. Instead of canceling, they compound. The net GSV is the sum of all forces -- dramatically larger.

Why This Matters for Muscles

The GSV represents the net shear that muscles, facets, and ligaments must actively resist to keep you upright. A low GSV means the spine is self-balancing -- minimal muscular effort needed. A high GSV means continuous paraspinal contraction, leading to fatigue, energy expenditure, and eventually pain and degeneration.

GSV Across Bipedal Vertebrates

Select any configuration below to run it through the GSV calculator. All models use the same 24-level weight distribution and gravitational decomposition -- only the curvature changes.

Modern Human

Homo sapiens
CervicalThoracicLumbarC2C5T1T4T7T10L1L4FH
70 kg
S-curve (3 curves)Obligate bipedBipedalism: 100%

Three alternating curves -- cervical lordosis, thoracic kyphosis, lumbar lordosis -- evolved over ~6 million years to minimize shear under sustained upright posture.

  • --Cervical lordosis positions the head over the pelvis
  • --Thoracic kyphosis accommodates rib cage and viscera
  • --Lumbar lordosis is the key human adaptation -- it reverses shear direction, creating cancellation with the thoracic region
  • --Opposing regional shear vectors partially cancel, yielding a low net GSV

GSV Analysis at 70 kg

18
S-GSVS1-m (N)
Net shear
-41°
S-GSVS1-a
Direction
431
Normal (N)
Compression
Regional Curvature
CL
Cervical lordosis
26°
TK
Thoracic kyphosis
28°
LL
Lumbar lordosis
4
Cervical GSV
14
Thoracic GSV
7
Lumbar GSV
Pelvic Parameters (literature)
53°
PI
Pelvic incidence
13°
PT
Pelvic tilt
40°
SS
Sacral slope
Legaye et al. (1998), Boulay et al. (2006) -- Asymptomatic adult mean values
Literature
  • Schwab et al. (2006): normal TK 20-50 deg, LL 40-60 deg
  • Roussouly & Pinheiro-Franco (2011): sagittal balance classification
  • Legaye et al. (1998): PI 53 +/- 10 deg in asymptomatic adults
  • Boulay et al. (2006): PI-SS-PT relationships in normal population
  • Selby et al. (2019): humans have longer lumbar spines, greater lumbosacral angle, dorsally wedged bodies
  • Hasegawa et al. (2025): evolutionary signals in sagittal alignment across 5 countries

Head-to-Head: All Configurations at 70 kg

Modern Human
1.0x(18N)
Australopithecus
1.4x(26N)
Chimpanzee
3.4x(62N)
Gorilla
4.9x(89N)
Penguin
0.1x(2N)

GSV Magnitude -- net shear burden the muscles, facets, and ligaments must resist. Lower is better for sustained bipedalism.

Why the S-Curve is Optimal for Bipedal Locomotion

Standing: Passive Equilibrium

When you stand still, gravity pulls downward through every vertebra. At each endplate, this force decomposes into a compressive (normal) component and a sliding (shear) component. The shear must be actively resisted by muscles and ligaments.

The S-curve arranges the spine so that the cervical and lumbar regions generate shear in one direction, while the thoracic region generates shear in the opposite direction. These opposing forces cancel across regions, leaving only a small net residual -- the GSV. This means standing upright requires minimal muscular effort, which is why humans can stand for hours without fatigue.

A chimpanzee or gorilla standing upright has no such cancellation. Their paraspinal muscles must work continuously against the full, uncanceled shear -- which is why they fatigue within minutes and return to quadrupedal posture.

Walking and Running: Dynamic Amplification

During walking, ground reaction forces oscillate between 1.0x and 1.3x body weight. During running, peak forces reach 2-3x body weight. These dynamic loads multiply every shear component proportionally.

For the human S-curve, this amplification applies to forces that are already mostly canceled. A GSV of 30 N at rest becomes ~90 N during running -- manageable for the paraspinal muscles.

For a C-curve spine, the same amplification applies to forces that are compounding. A GSV of 150 N at rest becomes ~450 N during running. This exceeds what the paraspinal muscles can sustain, which is why no living great ape can run bipedally.

The S-curve also acts as a spring during gait. The alternating curves absorb and return energy with each stride, reducing the metabolic cost of locomotion. A straight or C-curved spine transmits shock rigidly, wasting energy and accelerating disc degeneration.

No Lordosis

Great apes (gorilla, chimpanzee)

  • --All shear vectors point the same direction
  • --Forces compound from C2 to S1 with no reversal
  • --Muscles must resist the full cumulative shear
  • --Standing is costly; running is impossible
  • --Highest GSV of any bipedal configuration

Partial Lordosis

Australopithecus, early Homo

  • --Developing lumbar lordosis begins to reverse shear direction
  • --Partial cancellation reduces GSV below the ape level
  • --Sustained walking becomes possible -- the Laetoli footprints
  • --Insufficient lordosis limits running efficiency
  • --Evolutionary pressure continues to deepen the S-curve

Full S-Curve

Modern Homo sapiens

  • --Three alternating curves create maximum shear cancellation
  • --Net GSV is a small fraction of any single region's shear
  • --Standing requires minimal muscular effort
  • --Running amplifies already-canceled forces -- still manageable
  • --The S-curve doubles as a spring, storing and returning gait energy

Clinical Takeaway: Deformity as Lost Cancellation

Degenerative sagittal imbalance -- disc degeneration, vertebral fractures, loss of lumbar lordosis -- destroys the shear cancellation that the S-curve provides. Inoue and Espinoza Orias (2011) documented how nucleus desiccation and annular fissuring progressively alter disc mechanics, while Sun et al. (2019) showed concurrent multifidus atrophy that further degrades the active stabilizing system. As lordosis is lost, the spine reverts toward the ancestral C-curve pattern: high anterior shear with no opposing forces to cancel it. The result mirrors what the GSV predicts for the great ape spine -- chronic muscular fatigue, pain, and an inability to maintain upright posture without compensation.

Surgical correction through osteotomies and lordosis restoration is, in biomechanical terms, rebuilding the opposing shear vectors that cancel each other. The GSV quantifies exactly how well the correction restores that cancellation -- making it a direct measure of how effectively surgery returns the spine to its evolved equilibrium.

Evolutionary Context

The comparative analysis above quantifies shear forces, but the evolutionary and allometric data explain why these patterns exist -- and what happens when the S-curve is compromised. Whitcome et al. (2007, Nature) showed that lumbar lordosis evolved in hominins partly to counterbalance fetal load during bipedal pregnancy, linking reproductive biomechanics to the S-curve that minimizes shear.

Allometric Scaling

Majoral et al. (1997) studied 425 primates and found that vertebral column length scales with negative allometry -- larger primates have relatively shorter spines for their body mass (exponent < 0.33 in 12/16 correlations). This reduces the bending moment and shear forces in larger species.

For catarrhines (humans and great apes), centrum length scales at α = 0.25, meaning larger species have proportionally shorter vertebral bodies -- constraining the lever arm through which gravitational forces act.

Pelvic Incidence Evolution

Pelvic incidence (PI) increased progressively across the hominin lineage: from ~27-28° in great apes to ~45° in Australopithecus to ~53° in modern humans (Been et al. 2012, 2013, 2014).

Higher PI demands greater lumbar lordosis to maintain sagittal balance, establishing the biomechanical foundation for the S-curve. Tardieu et al. (2013) showed that PI increases during child development as bipedal gait is acquired -- ontogeny recapitulating phylogeny.

The Ancestral Shape Hypothesis

Plomp et al. (2015, 2019) demonstrated that humans whose vertebrae retain ancestral (ape-like) morphology are more susceptible to disc herniation. Using geometric morphometrics on human, chimpanzee, and orangutan vertebrae:

  • Vertebrae with Schmorl's nodes are closer in shape to chimpanzee vertebrae than healthy human vertebrae
  • Pathological features include smaller neural foramina, shorter/wider pedicles, and rounder vertebral bodies
  • The Procrustes distance between pathological humans and chimps is smaller than between pathological and healthy humans

Energy Cost of Bipedalism

Sockol et al. (2007, PNAS) measured that chimpanzee bipedal walking costs 75% more energy than human bipedal walking. The savings derive from the passive skeletal alignment (S-curve) that minimizes muscular effort to resist shear. This energy penalty arises directly from the uncanceled shear that paraspinal muscles must continuously resist. The human S-curve, by contrast, achieves passive shear cancellation through alternating curvatures, allowing sustained upright posture at minimal metabolic cost -- what Dubousset called remaining within the "cone of economy" (Hasegawa and Dubousset, 2022).

Pontzer et al. (2009) modeled locomotor energetics in early hominins, showing that even partial lordosis in Australopithecus would have significantly reduced energy cost. Ground reaction forces amplify to 2-3x body weight during running -- making the S-curve essential for sustained bipedal running.

Vertebral Morphology Timeline

~7 MYA
Last common ancestor
C-curve
~4.4 MYA
Ardipithecus ramidus
Transitional
~3.7 MYA
A. afarensis (Lucy)
Developing S-curve
~2 MYA
A. sediba / early Homo
Near-modern lordosis
~1.5 MYA
Homo erectus
Full pelvis- spine unit
~200 KYA
Homo sapiens
Modern S-curve
In Development

3D Interactive Visualization

Bring everything together in a fully interactive 3D environment. Select a preset or upload your own data, simulate surgical osteotomies, and see how the GSV responds in real time. Drag to rotate, scroll to zoom.

Well-balanced spine with cervical lordosis, thoracic kyphosis, and lumbar lordosis.

Surgical Interventions

No interventions added. Select a level and osteotomy type above.

Loading 3D viewer...

GSV Magnitude: 18 N
GSV Angle: -41°
Mild
CervicalThoracicLumbar
In Development

Robotics Perspective

Spine Kinematics: A Robotics Approach

The spine can be modeled as a serial kinematic chain -- a sequence of rigid links connected by revolute joints, just like a robot manipulator. Forward kinematics predicts where the head ends up given vertebral angles; inverse kinematics solves the surgical planning problem of finding the angles needed to achieve a target posture.

A robot manipulator is a set of rigid links connected by joints. The spine shares this structure: vertebral bodies are the links and intervertebral segments (disc + facet complex) are the joints.

In the sagittal plane, each spinal joint is primarily a revolute joint -- it allows rotation (flexion/extension). The joint variable is the segmental angle θi, and the link length is the vertebral body height.

We fix the base frame at S1 (the sacrum, which is fused to the pelvis). The end effector is C2, representing the head position. The kinematic question: given all joint angles, where does the head end up?

Key insight: Unlike industrial robots with 6 joints, the spine has ~24 mobile segments from S1 to C2. This makes it a hyper-redundant manipulator -- there are infinitely many joint configurations that place the head at the same position. This redundancy is what makes spinal deformity correction so challenging.

Robot vs. Spine Analogy

Robot ManipulatorSpinal Column
Base frameSacrum / pelvis (S1)
LinksVertebral bodies
Revolute jointsIntervertebral discs + facets
Joint angle (theta)Endplate tilt / segmental angle
Link length (a)Vertebral body height
End effectorC2 (skull base / head)
Degrees of freedom6 DOF per segment (simplified to 1R in sagittal)
Forward kinematicsPredict head position from vertebral angles
Inverse kinematicsFind angles needed for target alignment

Serial Chain Schematic

S1θ1L5θ2L3θ3L1θ4T9θ5T4θ6T1θ7C5θ8C2EEC2 (Head)
In Development

Surgical Optimization

GSV vs SVA: Optimization Compared

Upload a lateral spine X-ray or use the preset deformity to define vertebral landmarks. Mark the femoral head for pelvic parameter analysis, then compare GSV-based vs SVA-based surgical correction strategies side-by-side.

Deformed Spine (drag to adjust)
AntPostSVAC2S1

Drag landmarks to adjust

Femoral Head
Patient Parameters
Weight:kg
Height:cm
Osteotomy budget
Max:

SPO/Schwab: T1-L5. PSO: T12-L4.

Optimization Targets
N
°
cm

Active targets add penalty to both optimizers

Baseline
S-GSVS1-m49.0 N (Moderate)
S-GSVS1-a20°
C7-SVA7.2 cm (Moderate)
Interventions0
Total correction0.0° eff
In Development

Hill-Type Disc Mechanics & Local Shear Cancellation

The GSV sums shear across the entire spine, but it does not tell us what happens locally at each disc. Panjabi (1992) described three subsystems that stabilize the spine: the passive osteoligamentous column, the active musculature, and the neural control system. Every functional spinal unit (FSU) engages all three subsystems to absorb gravitational shear before it propagates to adjacent levels. Here we model that local damping using a Hill-type muscle-disc framework and define a cancellation efficiency metric that links per-level tissue health to the global shear burden.

Part 1: Local Shear Cancellation

At each level, gravitational shear is partially opposed by active paraspinal muscle contraction and passive viscoelastic disc forces. Cholewicki and McGill (1996) showed that even small contributions from individual muscles can substantially augment spinal stability, and that the lumbar spine requires continuous muscular co-contraction to remain stable under physiological loads. Only the residual shear -- the portion the FSU cannot absorb -- propagates downward through the spine.

T7T8T9

Part 2: The FSU as a Hill-Type System

Each functional spinal unit can be modeled as a Hill-type arrangement: a contractile element (paraspinal muscles) in series with an elastic element (annulus fibrosus collagen), in parallel with a viscoelastic element (nucleus pulposus + facet capsules). Haeufle et al. (2014) extended the classic Hill model with serial damping and an explicit eccentric force-velocity relation -- particularly relevant here, since paraspinal muscles predominantly operate eccentrically to resist gravitational shear. Disc degeneration degrades these elements asymmetrically: Sen et al. (2012) showed that annulus fibrosus tensile modulus increases with degeneration, while nucleus pulposus hydration and viscoelastic damping decline (Inoue and Espinoza Orias, 2011).

Superior VBInferior VBCESEPE
DegeneratedHealthy

CE -- Contractile Element

Paraspinal muscles (multifidus, erector spinae) and active ligament tension. Force generation follows the Hill force-velocity relation; eccentric capacity exceeds isometric by ~1.5x (Krylow and Sandercock, 1997). Meakin et al. (2013) showed that extensor muscle volume correlates with sagittal curvature, confirming higher force demands in lordotic spines. Degrades with fatty infiltration and atrophy.

SE -- Series Elastic Element

Annulus fibrosus collagen fibers. Nonlinear spring with a quadratic toe region. Sen et al. (2012) demonstrated that AF dynamic tensile modulus increases with degeneration -- the annulus stiffens through fibrosis, becoming less compliant and reducing elastic energy storage capacity.

PE -- Parallel Elastic Element

Nucleus pulposus hydration + facet capsule elasticity. Viscoelastic: spring for static load, dashpot for rate-dependent damping. Inoue and Espinoza Orias (2011) documented how nucleus desiccation and fissuring in the annulus progressively reduce viscoelastic properties, lowering damping capacity. Wilke et al. (1999) measured intradiscal pressures in vivo, confirming the load-bearing role of the pressurized nucleus.

Part 3: Constitutive Equations

Each element in the Hill model follows a specific constitutive law. The original Hill (1938) force-velocity relation for muscle was extended by Haeufle et al. (2014) to include serial damping and explicit eccentric behavior. These equations define how the FSU generates resistive force as a function of displacement, velocity, and tissue health.

Part 4 — Two-Level Interaction

Schmidt et al. (2013) used finite element modeling to show that lumbar motion segments resist anterior shear through a combination of disc fiber tension, facet contact forces, and ligament engagement. Here, explore how the Hill-type FSU absorbs shear between two adjacent vertebrae. Adjust the applied shear, loading rate, and disc health to see how force decomposes across CE and PE elements, and how much residual escapes to contribute to the GSV.

T11T12Shearη = 1.00cancellation efficiency
Applied Shear
20.0 N
CE Force
128.2 N
PE Force
7.0 N
Residual
0.0 N
Efficiency:
100%

Healthy disc. The FSU fully absorbs applied shear through muscle contraction and viscoelastic damping. No residual shear propagates to adjacent segments.

Part 5: Global Cancellation Efficiency

The global cancellation efficiency measures how much of the total per-level shear is absorbed before reaching the GSV resultant. A healthy spine has high efficiency. Polikeit et al. (2004) showed that disc degeneration alters load transfer patterns in the FSU, shifting stress from the nucleus to the annulus and endplates. Cornaz et al. (2021) further demonstrated that degeneration degrades spinal ligament biomechanics, compounding the loss of passive restraint. As these structures fail, more shear propagates and the GSV rises.

η_global = 1 - (GSV_M / Σ|F_shear_i|)
η = 0.977

Global cancellation efficiency (70 kg, normal spine)

Per-Level Cancellation Efficiency (η per level)

C21.00C31.00C41.00C51.00C61.00C71.00T11.00T21.00T31.00T41.00T51.00T61.00T71.00T81.00T91.00T101.00T111.00T121.00L11.00L21.00L31.00L41.00L51.00S11.00

Part 6: Clinical Implications

The Hill-type disc mechanics model connects local disc health to the global shear metric, offering clinical insight into three common scenarios.

Healthy Spine

All discs have high damping capacity (eta > 0.85). Per-level shear is efficiently canceled by paraspinal co-contraction and viscoelastic disc response. Meakin et al. (2013) showed that extensor muscle volume scales with lordosis, confirming that the healthy spine dedicates proportional muscle mass to shear resistance. GSV magnitude is low because opposing shear forces are locally absorbed before they accumulate.

Degenerative Cascade

One or two degenerated discs (low eta) fail to absorb their local shear. Inoue and Espinoza Orias (2011) described how nucleus desiccation, disc height loss, and annular fissuring progressively degrade viscoelastic properties. Sun et al. (2019) showed that multifidus atrophy correlates with deformity severity, indicating concurrent CE degradation. The residual propagates to adjacent discs, overloading them and accelerating further degeneration -- each failing FSU increases the shear burden on its neighbors.

Post-Surgical ASD

Fusion eliminates motion at a segment (eta = 0 for damping, though the rigid construct transmits force differently). Senteler et al. (2017) showed that fusion angle directly affects adjacent segment joint forces, and Wang et al. (2024) confirmed using patient-specific FE models that adjacent discs experience elevated stress after fusion. This concentrated overload at transition zones is a primary biomechanical mechanism of adjacent segment disease (ASD).

GSV as a Measure of Damping Failure

The GSV does not merely measure spinal alignment -- it measures the net shear that the spine's local damping mechanisms failed to cancel. A rising GSV after surgery or with aging reflects not just worsening geometry, but degrading disc mechanics. By combining the GSV with per-level eta values from the Hill model, clinicians can distinguish between geometric deformity (alignment-driven shear) and mechanical insufficiency (damping-driven shear) -- two conditions that require fundamentally different treatment strategies.

In Development

Modified GSV: Segment-Specific Parameters

The optimization framework above treats every vertebral segment as biomechanically identical. In reality, disc degeneration, facet arthropathy, and surgical fusion fundamentally alter how load distributes across the spine.

Part 1 — Load Redistribution

When one segment loses stiffness — for example, severe L4 disc degeneration — the shear load it can no longer bear transfers to adjacent levels. The mGSV accounts for this by weighting each level's contribution by its local stiffness coefficient.

Healthy Spine

C5T1T4T8T12L1L3L4L5
NormalElevatedOverloaded

Matched Healthy

C5T1T4T8T12L1L3L4L5
NormalElevatedOverloaded
L4 overloaded: The degenerated disc transfers approximately 30% of its shear load to L3 and L5. Without segment-specific weighting, the GSV underestimates the true mechanical risk at adjacent levels.

Part 2 — Segment Parameters

Six pathological factors each contribute a per-level stiffness multiplier. A composite stiffness coefficient k_i is computed for each vertebral segment and applied to the shear force at that level.

Disc Degeneration

GradesNormal / Mild / Moderate / Severe
Stiffness range1.0 - 0.25

Pfirrmann grade 1-5. Degeneration reduces disc stiffness and alters load transfer.

Disc Height

GradesNormal / Reduced / Severely Reduced
Stiffness range1.0 - 0.40

Loss of disc height reduces load-bearing surface area and axial stiffness.

Facet Joint Grade

GradesNormal / Mild / Moderate / Severe
Stiffness range1.0 - 0.30

Weishaupt 0-3. Facet degeneration reduces rotational and shear restraint.

Listhesis

GradesNone / Grade I / Grade II / Grade III-IV
Stiffness range1.0 - 0.25

Spondylolisthesis grade reflects segmental instability and reduced load transfer.

Fusion Status

GradesMobile / Instrumented / Solid Fusion
Stiffness range1.0 - 5.0

Fused segments have elevated stiffness; adjacent levels absorb redistributed loads.

Canal/Foraminal Stenosis

GradesNone / Mild / Moderate / Severe
Stiffness range1.0 - 0.45

Stenosis grade reflects narrowing that correlates with reduced segmental stiffness.

Part 3 — Interactive Demo

Select a pathology preset or customize individual segment parameters to see how segmental stiffness changes the GSV in real time.

All segments normal -- mGSV equals standard GSV

Modifier Map (T10 - S1)

T101.00T111.00T121.00L11.00L21.00L31.00L41.00L51.00S11.00
<0.7~1.01.0-1.3>1.3

Segment Parameters

DiscHeightFacetListhFusionStenosis
Lumbar
L1
L2
L3
L4
L5
S1

Standard GSV

Magnitude18.0 N
Angle-41.0 deg

Modified GSV (mGSV)

Magnitude18.0 N
Angle-41.0 deg

mGSV shifts S-GSVS1-m by +0.0 N, S-GSVS1-a by +0.0 deg

Stiffness coefficients are preliminary and subject to clinical validation.

Mathematical Derivation

Step-by-step formulation of the modified GSV.

References

The educational content throughout this application draws on the following peer-reviewed literature. References are grouped by topic and linked to PubMed for verification.

Sagittal Alignment & Classification

  1. Schwab F, Lafage V, Boyce R, et al. Gravity line analysis in adult volunteers: age-related correlation with spinal parameters, pelvic parameters, and foot position. Spine. 2006;31(25):E959-967. PMID: 17139212
  2. Liu Y, Liu Z, Zhu F, et al. Validation and refinement of the SRS-Schwab adult spinal deformity classification. Spine. 2013;38(12):1077-1082. PMID: 23222646
  3. Smith JS, Klineberg E, Schwab F, et al. Change in classification grade by the SRS-Schwab adult spinal deformity classification predicts impact on health-related quality of life measures. Spine. 2013;38(25):2190-2198. PMID: 23759814
  4. Lowe T, Berven SH, Schwab FJ, et al. The SRS classification for adult spinal deformity: building on the King/Moe and Lenke classification systems. Spine. 2006;31(19S):S119-S125. PMID: 16946628
  5. Abelin-Genevois K. Sagittal balance of the spine. Orthop Traumatol Surg Res. 2021;107(1S):102769. PMID: 33321235

Pelvic Parameters

  1. Boulay C, Tardieu C, Hecquet J, et al. Sagittal alignment of spine and pelvis regulated by pelvic incidence: standard values and prediction of lordosis. Eur Spine J. 2006;15(4):415-422. PMID: 16179995
  2. Tardieu C, Bonneau N, Hecquet J, et al. How is sagittal balance acquired during bipedal gait acquisition? Comparison of neonatal and adult sacropelvic morphology. Rev Chir Orthop. 2008;94(4):327-335. PMID: 18555858
  3. Ryan DJ, Protopsaltis TS, Ames CP, et al. T1 pelvic angle (TPA) effectively evaluates sagittal deformity and assesses radiographic severity. Spine. 2014;39(15):1203-1210. PMID: 25171068
  4. Protopsaltis T, Schwab F, Bronsard N, et al. The T1 pelvic angle, a novel radiographic measure of global sagittal deformity, accounts for both spinal inclination and pelvic tilt. JBJS. 2014;96(19):1631-1640. PMID: 25274788

Spinal Stability & Muscle

  1. Panjabi MM. The stabilizing system of the spine. Part I. Function, dysfunction, adaptation, and enhancement. J Spinal Disord. 1992;5(4):383-389. PMID: 1490034
  2. Panjabi MM. Clinical spinal instability and low back pain. J Electromyogr Kinesiol. 2003;13(4):371-379. PMID: 12832167
  3. Cholewicki J, Juluru K, Radebold A, et al. Lumbar spine stability can be augmented with an abdominal belt and/or increased intra-abdominal pressure. Eur Spine J. 1999;8(5):388-395. PMID: 10552322
  4. Meakin JR, Fulford J, Seymour R, et al. The relationship between sagittal curvature and extensor muscle volume in the lumbar spine. J Anat. 2013;222(6):608-614. PMID: 23600615
  5. Sun XY, Kong C, Zhang TT, et al. Correlation between multifidus muscle atrophy, spinopelvic parameters, and severity of deformity in patients with adult degenerative scoliosis. J Orthop Surg Res. 2019;14(1):276. PMID: 31455401

Disc & FSU Biomechanics

  1. Wilke HJ, Neef P, Caimi M, et al. New in vivo measurements of pressures in the intervertebral disc in daily life. Spine. 1999;24(8):755-762. PMID: 10222525
  2. Sen S, Jacobs NT, Boxberger JI, et al. Human annulus fibrosus dynamic tensile modulus increases with degeneration. Mech Mater. 2012;44:93-98. PMID: 22247579
  3. Inoue N, Espinoza Orias AA. Biomechanics of intervertebral disk degeneration. Orthop Clin North Am. 2011;42(4):487-499. PMID: 21944586
  4. Polikeit A, Nolte LP, Ferguson SJ. Simulated influence of osteoporosis and disc degeneration on the load transfer in a lumbar functional spinal unit. J Biomech. 2004;37(7):1061-1069. PMID: 15165876
  5. Cornaz F, Widmer J, Farshad-Amacker NA, et al. Intervertebral disc degeneration relates to biomechanical changes of spinal ligaments. Spine J. 2021;21(8):1399-1407. PMID: 33901629
  6. Schmidt H, Bashkuev M, Dreischarf M, et al. Computational biomechanics of a lumbar motion segment in pure and combined shear loads. J Biomech. 2013;46(14):2513-2521. PMID: 23953504

Hill-Type Muscle Model

  1. Haeufle DF, Gunther M, Bayer A, et al. Hill-type muscle model with serial damping and eccentric force-velocity relation. J Biomech. 2014;47(6):1531-1536. PMID: 24612719
  2. Krylow AM, Sandercock TG. Dynamic force responses of muscle involving eccentric contraction. J Biomech. 1997;30(8):847-854. PMID: 8970921

Body Segment Parameters

  1. de Leva P. Adjustments to Zatsiorsky-Seluyanov's segment inertia parameters. J Biomech. 1996;29(9):1223-1230. PMID: 8872282

Osteotomy & Surgical Planning

  1. Bridwell KH. Decision making regarding Smith-Petersen vs. pedicle subtraction osteotomy vs. vertebral column resection for spinal deformity. Spine. 2006;31(19S):S171-S178. PMID: 16946635
  2. Gupta MC, Kebaish K, Engeku KA, et al. Pedicle subtraction osteotomy. JBJS Essent Surg Tech. 2020;10(1):e0028. PMID: 32368407

Adjacent Segment Disease

  1. Senteler M, Weisse B, Rothenfluh DA, et al. Fusion angle affects intervertebral adjacent spinal segment joint forces -- model-based analysis of patient specific alignment. J Orthop Res. 2017;35(1):131-139. PMID: 27364167
  2. Wang Y, Shen Q, Liang C, et al. Biomechanical analysis of adjacent segments after spinal fusion surgery using a geometrically parametric patient-specific finite element model. J Vis Exp. 2024;(203):e65876. PMID: 38314842

Comparative Evolution

  1. Whitcome KK, Shapiro LJ, Lieberman DE. Fetal load and the evolution of lumbar lordosis in bipedal hominins. Nature. 2007;450:1075-1078. PMID: 18075592
  2. Been E, Gomez-Olivencia A, Kramer PA. Lumbar lordosis of extinct hominins. Am J Phys Anthropol. 2014;153(S58):56-57. PMID: 24838427

Sagittal Balance & Morphotypes

  1. Roussouly P, Pinheiro-Franco JL. Biomechanical analysis of the spino-pelvic organization and adaptation in pathology. Eur Spine J. 2011;20(Suppl 5):609-618. PMID: 21809016
  2. Bakouny Z, Assi A, Yared F, et al. Roussouly sagittal morphotypes as determinants of gait kinematics. Gait Posture. 2017;55:1-7. PMID: 28242570
  3. Wang W, Pei B, Wu X, et al. Biomechanical responses of the lumbar spine according to Roussouly classification. PLoS One. 2022;17(7):e0271832. PMID: 35905050
  4. Hasegawa K, Dubousset JF. Cone of economy: concept and clinical significance. Spine Surg Relat Res. 2022;6(5):449-455. PMID: 36051675

Kinematics & Cervical Alignment

  1. Anderst WJ, Donaldson WF, Lee JY, et al. Six-degrees-of-freedom cervical spine range of motion during dynamic flexion-extension after single-level anterior arthrodesis. JBJS. 2013;95(6):497-506. PMID: 23515984
  2. Scheer JK, Tang JA, Smith JS, et al. Cervical spine alignment, sagittal deformity, and clinical implications: a review. J Neurosurg Spine. 2013;19(2):141-159. PMID: 23768023