CPAK Classification Explained: aHKA, JLO, Nine Types

The Coronal Plane Alignment of the Knee (CPAK) classification gives surgeons a common language for describing a knee's native alignment. Instead of a single "varus" or "valgus" label, CPAK captures two independent facts: how the leg is aligned overall, and how the joint line is tilted. Those two facts combine into nine phenotypes, and knowing a knee's phenotype shapes how you plan a total knee arthroplasty (TKA).

This is a spoke of our pillar on automated CPAK alignment planning. Here we stay at the plain-language level: what the two axes mean, where the numbers come from, and how to read the nine-box grid.

The two building blocks: LDFA and MPTA

CPAK is built entirely from two coronal-plane angles measured on a long-leg image:

• LDFA (lateral distal femoral angle): the angle between the mechanical axis of the femur and the joint line at the distal femur, measured on the lateral side.
• MPTA (medial proximal tibial angle): the angle between the mechanical axis of the tibia and the joint line at the proximal tibia, measured on the medial side.

If you want a refresher on how these angles relate to the overall mechanical axis and the hip-knee-ankle (HKA) angle, our companion piece on mechanical axis alignment (LDFA, MPTA, HKA) walks through each measurement.

CPAK's insight from Samuel MacDessi and colleagues was that you can take these same two angles and combine them two ways to describe a knee independently of its arthritic deformity.

aHKA: the overall limb alignment

The arithmetic HKA (aHKA) estimates the leg's constitutional (pre-arthritic) alignment. It is the difference of the two angles:

aHKA = MPTA - LDFA

Because it is derived from bony angles rather than the weight-bearing mechanical axis, aHKA is less affected by joint-space narrowing and cartilage wear than a measured HKA. It answers: was this leg built varus, neutral, or valgus?

In the original classification, aHKA is grouped as:

JLO: how the joint line is tilted

The joint line obliquity (JLO) describes how the knee's joint line sits relative to the floor in double-leg stance. It is the sum of the same two angles:

JLO = MPTA + LDFA

A JLO near 180 degrees means the joint line is roughly parallel to the ground. A higher sum means the joint line points upward at its center (apex proximal); a lower sum means it points downward (apex distal).

A simple way to remember it: aHKA is a subtraction (overall alignment), JLO is an addition (joint-line tilt). Two numbers from the same two measurements.

The nine phenotypes

Cross the three aHKA categories with the three JLO categories and you get a 3-by-3 grid of nine phenotypes, numbered I through IX:

In the original 1,000-knee validation (healthy young adults from Belgium plus an arthritic cohort from Sydney, Australia), Type II (neutral aHKA, apex distal JLO) was the single most common phenotype. Later population studies report that as osteoarthritis severity increases, the distribution shifts toward Type I (varus, apex distal), and that phenotype prevalence varies by geography, with apex-distal types (I, II, III) dominating across most populations and apex-proximal types (VII to IX) being rare. This is why local, cohort-specific distributions matter, and why an automated tool that measures a whole scan set consistently is useful for characterizing a population.

Why phenotype matters for planning

CPAK does not by itself prescribe a surgical technique, but it frames the decision. Different alignment philosophies (mechanical, kinematic, restricted kinematic, functional) restore the joint line and limb axis differently, and a knee's starting phenotype determines how much a given target would move it. Knowing that a knee is Type I versus Type V changes what "returning to native alignment" or "correcting to neutral" actually requires, and how much soft-tissue release or bone resection asymmetry that implies.

CPAK also standardizes communication and research. Reporting outcomes by phenotype lets studies compare like with like, rather than lumping every varus knee together.

How Salnus fits in (RUO)

Measuring LDFA and MPTA by hand on every case is slow and introduces observer variability, which is exactly the interobserver reliability question the CPAK literature keeps revisiting. Salnus is building a software-only, browser-based research tool that derives coronal-plane geometry from a CT scan: it performs CT-based 3D bone segmentation and then computes the coronal angles that feed aHKA and JLO, placing a knee into its CPAK phenotype automatically.

Two design choices are worth noting:

• Client-side. DICOM parsing and inference run on the surgeon's own machine (Cornerstone3D with ONNX Runtime Web); images are not uploaded to a server. In practice this means going from scan to a draft plan in minutes without a cloud round-trip, and PHI stays on-device.
• Implant-agnostic and vendor-neutral. The geometry is computed independently of any implant catalog, which is why the output is useful across systems and to manufacturers evaluating their own workflows.

Salnus is currently a Research Use Only (RUO) tool in pilot, not a cleared medical device. Its peer-reviewed footprint so far is a controlled laboratory study on 3D-printed patient-specific guides for multiligament knee reconstruction (tunnel convergence), not a clinical-outcomes trial of alignment planning. For related research directions, see our notes on the OJSM patient-specific instrumentation study and on the PCL as a research target.

FAQ

What does CPAK stand for? Coronal Plane Alignment of the Knee. It is a classification introduced by MacDessi and colleagues in 2021 that describes a knee using two independent variables, aHKA and JLO, yielding nine phenotypes.

How is aHKA different from a measured HKA? The measured HKA comes from the weight-bearing mechanical axis and is affected by cartilage wear and joint-space narrowing. The arithmetic HKA is calculated as MPTA minus LDFA from bony angles, so it better estimates the constitutional (pre-arthritic) alignment.

What is the difference between apex distal and apex proximal JLO? JLO equals MPTA plus LDFA. A sum below 177 degrees is apex distal (joint line points downward at its center); above 183 degrees is apex proximal (points upward); 177 to 183 is neutral.

Which CPAK type is most common? In the original validation, Type II (neutral aHKA, apex distal JLO) was most common overall. In arthritic populations, and as osteoarthritis severity increases, the distribution shifts toward Type I (varus, apex distal). Exact proportions vary by region.

Does CPAK tell me which alignment technique to use? No. CPAK describes the knee; it does not prescribe a philosophy. It gives a shared framework for choosing and reporting an alignment target (mechanical, kinematic, restricted, or functional) relative to the knee's native phenotype.

Can software classify CPAK automatically? Yes, in a research setting. Tools that measure LDFA and MPTA consistently can compute aHKA and JLO and assign a phenotype. Salnus does this client-side from CT as an RUO tool; results should be reviewed by a qualified clinician.

Explore the Salnus Surgeon Portal →

---

Disclaimer: This article is for educational and research purposes only. Salnus tools are designated for Research Use Only (RUO) and are not cleared medical devices. Mention of third-party products is for educational context only and does not imply endorsement or comparison of clinical equivalence. Clinical decisions should be made by qualified physicians, and regulatory status should be independently verified for your jurisdiction.

References:

• MacDessi SJ, Griffiths-Jones W, Harris IA, Bellemans J, Chen DB. Coronal Plane Alignment of the Knee (CPAK) classification: a new system for describing knee phenotypes. The Bone & Joint Journal. 2021;103-B(2):329-337.
• Griffiths-Jones W, Chen DB, Harris IA, Bellemans J, MacDessi SJ. The Coronal Plane Alignment of the Knee Classification: how it works, how to apply it, and how it can affect outcomes in knee reconstruction surgery. Journal review, 2025.
• Coronal Plane Alignment of the Knee (CPAK) Type Shifts Toward Constitutional Varus with Increasing Kellgren and Lawrence Grade, JBJS.
• A systematic review of geographic differences in knee phenotypes based on the Coronal Plane Alignment of the Knee (CPAK) classification, PMC.
• Cirdi YU, Serteser B, Mavi A, Ergun S, Akgun U. 3D-Printed Patient-Specific Guides Reduce Femoral Tunnel Convergence in Anatomic Knee Multiligament Reconstruction: Controlled Laboratory Study. Orthopaedic Journal of Sports Medicine. 2026.