King-Moe Classification of Thoracic AIS
- The King (King-Moe) classification, described in 1983, is a system of FIVE types of THORACIC adolescent idiopathic scoliosis based on the relationship and relative size/flexibility of the thoracic and lumbar curves; it was devised specifically to guide SELECTIVE THORACIC FUSION - that is, which curves needed to be included in a fusion - in the era of HARRINGTON rod instrumentation.
- King I is an S-shaped curve in which BOTH the thoracic and lumbar curves cross the midline and the LUMBAR curve is the larger or more rigid of the two; King II is the S-shaped 'false double major' in which the THORACIC curve is larger and the lumbar curve, although it crosses the midline, is more FLEXIBLE - King II is the classic indication discussed for selective thoracic fusion.
- King III is a thoracic curve in which the lumbar spine does NOT cross the midline (essentially an isolated thoracic curve); King IV is a LONG thoracic curve in which L4 tilts into the curve; and King V is a DOUBLE THORACIC curve with a STRUCTURAL upper thoracic component, recognised by the first rib or T1 being tilted into the upper curve (important because the upper curve must be included in the fusion to avoid postoperative shoulder imbalance).
- The KING II CONTROVERSY is the key clinical learning point: selectively fusing only the thoracic curve in a King II, as the classification suggested, was associated in some patients with postoperative coronal DECOMPENSATION and progression of the unfused lumbar curve - so 'selective thoracic fusion' must be applied with judgement (curve flexibility, magnitude and the lumbar modifier), and this limitation drove the move to better classifications.
- The King classification has well-documented LIMITATIONS: it describes only THORACIC curves (it does not categorise thoracolumbar, lumbar, double-major or triple curves comprehensively), it was developed for two-dimensional Harrington instrumentation and does not address the sagittal plane, and it has POOR inter- and intra-observer RELIABILITY - a major weakness for a system meant to guide surgery.
- Because of these shortcomings, the King classification was SUPERSEDED by the LENKE classification (2001), which is more comprehensive (six curve types covering all AIS patterns, with a lumbar spine modifier and a sagittal thoracic modifier), more reliable, and explicitly treatment-based; King remains important to KNOW for exams and as the conceptual origin of selective thoracic fusion, but Lenke is the modern system in practice.
- “King (King-Moe, 1983) = 5 types of THORACIC AIS, devised to guide SELECTIVE THORACIC FUSION in the Harrington era.
- “King II = false double major (thoracic larger than lumbar, lumbar flexible) -> selective thoracic fusion, but watch for DECOMPENSATION; King V = double thoracic (structural upper curve - include it to avoid shoulder imbalance).
- “Limitations: thoracic-only, 2D/Harrington, no sagittal plane, POOR reliability -> superseded by the Lenke classification.
Five types of thoracic AIS (King-Moe, 1983) devised to guide selective thoracic fusion in the Harrington era. King II = false double major; King V = double thoracic.
Thoracic-only, 2D/no sagittal plane, poor reliability, and the King II decompensation problem - superseded by Lenke.
The Five Types & the King II Controversy
The King-Moe classification (1983) sorts thoracic adolescent idiopathic scoliosis into five types based on the relationship and relative size/flexibility of the thoracic and lumbar curves, and its purpose was to guide selective thoracic fusion in the Harrington rod era. The clinically pivotal type is King II - the S-shaped 'false double major' with a larger thoracic curve and a more flexible lumbar curve - because selectively fusing only the thoracic curve in a King II was, in some patients, followed by coronal decompensation and progression of the unfused lumbar curve. King V, the double thoracic curve with a structural upper thoracic component, matters because the upper curve must be included in the construct to avoid postoperative shoulder imbalance.
| Type | Pattern | Key point |
|---|---|---|
| King I | S-shaped; both thoracic and lumbar cross midline; LUMBAR larger/more rigid | Usually fuse both curves |
| King II | S-shaped 'false double major'; THORACIC larger; lumbar crosses midline but FLEXIBLE | Classic selective thoracic fusion - beware decompensation |
| King III | Thoracic curve only; lumbar does NOT cross the midline | Fuse the thoracic curve |
| King IV | LONG thoracic curve; L4 tilts into the curve | Long thoracic pattern |
| King V | DOUBLE THORACIC; structural upper curve (T1/first rib tilted into it) | Include the upper curve - avoid shoulder imbalance |
Limitations & Why Lenke Replaced It
- Thoracic-only. It categorises thoracic curves but not thoracolumbar, lumbar, double-major or triple curves comprehensively.
- Two-dimensional / Harrington-era. Developed for Harrington instrumentation; it does not address the sagittal plane.
- Poor reliability. It has documented poor inter- and intra-observer reliability - a serious flaw for a system intended to guide surgery (computer-assisted methods improved reliability, underlining the manual variability).
- The King II problem. Selective thoracic fusion guided by the classification was associated with coronal decompensation in some patients.
- Superseded by Lenke (2001): six curve types covering all AIS patterns, with a lumbar modifier and a sagittal thoracic modifier - more comprehensive, more reliable and treatment-based.
The King classification is important to know because it introduced the concept of selective thoracic fusion and because the King II decompensation experience is a recurring teaching point - but it should not be used as the sole modern basis for planning AIS surgery. Its weaknesses are real: it only addresses thoracic curves, it ignores the sagittal plane, and it has poor observer reliability, all of which the Lenke classification was designed to correct. In practice, classify and plan with Lenke (curve type plus lumbar and sagittal modifiers), while retaining the King-derived principle that a selective thoracic fusion must respect curve flexibility and magnitude to avoid coronal decompensation, and that a structural upper thoracic curve (King V / Lenke 2) must be included to prevent shoulder imbalance.
Evidence & Key Studies
Lenke classification addressing the shortcomings of the King-Moe system
- The Lenke classification was developed to address shortcomings of the King-Moe classification by providing a more comprehensive, reliable and treatment-based categorisation of ALL adolescent idiopathic scoliosis deformities.
- Even within Lenke, fine distinctions (e.g. the direction of L4 tilt subdividing Lenke 1A curves) affect selection of the lowest instrumented vertebra and surgical outcomes.
- This illustrates why curve classification matters for fusion-level selection - the very problem the King system originally tried, imperfectly, to solve.
Reliability of the King classification (improved by computer-assisted measurement)
- Evaluation of scoliosis radiographs is inherently unreliable because of technical and human judgmental errors, affecting both Cobb angle and King classification.
- A computer-assisted measurement protocol made the King classification MORE reliable than unassisted observers (intra-observer kappa about 0.85), underlining the manual variability of the classification.
- The reliability limitation of manual King classification is a recognised weakness of the system.
According to PubMed, the statement that the Lenke classification was developed to address the shortcomings of the King-Moe system (providing a more comprehensive, reliable and treatment-based categorisation) comes from the cited Miyanji study, and the inherent unreliability of manual King classification (improved by computer-assisted measurement) from the cited Stokes study. The definitions of the five King types, the King II selective-fusion controversy/decompensation, the King V structural upper curve, and the Harrington-era origin are standard, well-established teaching. (See also our Adolescent Idiopathic Scoliosis and Lenke Classification topics.)
Clinical Decision Scenarios
Practise clinical reasoning and management decisions out loud
“What is the King classification, and why has it largely been replaced?”
Mnemonics & Memory Aids
KING
Hook:KING: King-Moe 5 types, Intended for selective fusion, Now superseded, Go to Lenke.
What & why
- King-Moe (1983): 5 types of THORACIC AIS
- Devised to guide selective thoracic fusion (Harrington era)
- Based on thoracic vs lumbar curve relationship/size/flexibility
The five types
- I: both cross midline, lumbar larger/rigid; II: false double major (thoracic larger, lumbar flexible)
- III: thoracic only, lumbar doesn't cross midline; IV: long thoracic (L4 tilts in)
- V: double thoracic, structural upper curve (include it)
Limitations
- Thoracic curves only; 2D / Harrington-era; ignores sagittal plane
- Poor inter/intra-observer reliability
- King II selective fusion -> coronal decompensation risk
Modern practice
- Superseded by Lenke (6 types + lumbar + sagittal modifiers)
- Lenke: comprehensive, reliable, treatment-based
- Retain King principles (flexibility for selective fusion; include structural upper curve)