American Spinal Injury Association (ASIA) Impairment Scale

Original Editor - Naomi O'Reilly.

Top Contributors - Naomi O'Reilly, Stacy Schiurring, Arnold Fredrick D'Souza, Kim Jackson, Jess Bell, Admin, Abbey Wright, Tarina van der Stockt, Rucha Gadgil and Olajumoke Ogunleye  

Disclaimer: digital rectal examination involving deep anal pressure and voluntary anal contraction is discussed on this page. This is an invasive and advanced assessment technique and should only be performed after demonstrating competency after in-person mentorship.

Introduction

This Physiopedia Page has been updated to reflect the most recent 2026 version of the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) or ASIA exam.

Spinal cord injury (SCI) is a major cause of long-term disability worldwide. The Global Burden of Disease Study estimated that 20.6 million people were living with SCI in 2019, with 0.9 million new cases that year. Falls and road injuries were the two leading causes, and SCI at cervical levels were more common than SCI below the level of the neck.

For a review of spinal cord injury, please read this article.

Spinal cord injury affects the conduction of sensory and motor signals across the site of the lesion, as well as the autonomic nervous system. Systematic examination of the dermatomes and myotomes allows the clinician to determine which cord segments are affected.

The International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI), often called the "ASIA exam", is the most widely used method of documenting the level and severity of SCI. The original Standards were introduced by the American Spinal Injury Association (ASIA) in 1982. They are now maintained jointly by the International Standards Committee of ASIA and the International Spinal Cord Society (ISCoS). The 8th edition was published in 2019 and introduced a taxonomy for documenting impairments not caused by the SCI and a revised definition of the zone of partial preservation. The 9th edition was released in April 2026. It does not substantially change the examination or the classification schema, but it clarifies a number of definitions and rules that caused confusion after 2019 (see What's new in the 9th edition below). Because definitions change between editions, the version used should always be stated in clinical records and research.

The Standards describe two examination components, sensory and motor, each of which includes part of the anorectal examination. For teaching purposes it is helpful to think of three parts:

1. a dermatome-based sensory examination (light touch and pin prick at 28 key sensory points on each side)
2. a myotome-based motor examination (10 key muscle functions on each side)
3. an anorectal examination (deep anal pressure, the sensory component, and voluntary anal contraction, the motor component)

The examination findings are then used in six steps of classification, which follow the order printed on the worksheet: (1) sensory levels, (2) motor levels, (3) the neurological level of injury (NLI), (4) complete or incomplete injury, (5) the ASIA Impairment Scale (AIS) grade and (6) the zones of partial preservation (ZPP). The NLI and AIS alone do not fully describe the extent and pattern of impairment, so the Committee recommends a full examination that also reports the sensory and motor levels, ZPPs, upper and lower extremity motor scores, and total sensory scores.

ISNCSCI is a classification, not a comprehensive neurological examination. It does not include reflexes, tone or proprioception, and it does not capture autonomic function. The International Standards to document Autonomic Function after SCI (ISAFSCI) are recommended as an adjunct. The skeletal (radiological) level of injury is not part of ISNCSCI, because not all SCIs involve bony injury, bony and neurological injury do not consistently correspond, and the skeletal level cannot be revised to document neurological change.

Terminology

Caudal/rostral anatomical positioning

Tetraplegia (preferred to "quadriplegia") is impairment or loss of motor and/or sensory function due to SCI with an NLI at or rostral to T1. It

typically affects the upper limbs and usually also the trunk, lower limbs and pelvic organs. It does not include brachial plexus lesions or injury to peripheral nerves outside the spinal canal.

Paraplegia is impairment or loss of motor and/or sensory function due to SCI with an NLI caudal to T1. Upper limb function as tested by ISNCSCI is normal, while trunk, lower limb and pelvic organ function may be impaired depending on the level and severity of injury. The term includes conus medullaris and cauda equina injuries, but not lumbosacral plexus lesions or injury to peripheral nerves outside the spinal canal.

Since 2026 both terms are defined by the NLI and apply only to people with SCI. One consequence is that a person with an NLI of T1 is classified as having tetraplegia even if the T1 key muscle is graded 5. The Committee accepted this for simplicity, because fewer than 2% of people with acute SCI have an initial NLI of T1 with normal T1 key muscle strength.

The terms tetraparesis and paraparesis are discouraged, because they describe incomplete injuries imprecisely and wrongly imply that tetraplegia and paraplegia apply only to complete injuries. Severity is described with the AIS instead.

What's new in the 9th edition (2026)

The 2026 edition contains no substantial changes to the examination or classification schema. It adds the following specifications and clarifications:

1. Tetraplegia and paraplegia are now defined by the NLI: tetraplegia at or rostral to T1, paraplegia caudal to T1.
2. Muscle strength grading: grades 4 and 5 are tested in a joint-specific test position, not through the full range. A muscle that moves actively with gravity eliminated but does not achieve full range is graded 1.
3. NT versus NT*: NT is used when testing is prevented by a condition caused by the SCI (e.g. spasticity, neuropathic pain); NT* when it is prevented by a non-SCI condition (eg. amputation, burn, cast, contracture). ND* is used when a level, grade or ZPP cannot be determined because of a non-SCI condition.
4. Motor level in regions without testable myotomes: non-testable myotomes at or above the sensory level are assumed normal; those below it are assumed to be less than grade 3. Ten worked examples were added.
5. AIS E is renamed "Normal ISNCSCI". Impairments not tested by ISNCSCI, such as autonomic dysfunction, may still be present.
6. Asterisk tagging: the rule that the NLI must always be tagged when any level is tagged has been removed. Sum scores are never tagged. A step-by-step algorithm decides which classification variables need an asterisk.
7. Patterns of Spinal Cord Injury: the section is renamed (previously "Patterns of Incomplete Spinal Cord Injuries") and now includes posterior cord syndrome.
8. Deep anal pressure should be tested routinely, even when light touch or pin prick is already present at S4–5.
9. Minor wording changes in the booklet and worksheet.

Timing of the examination

In the acute phase, findings may be affected by spinal shock, sedation, pain or other injuries, so serial examinations are needed to establish the true extent of injury. Early examinations still matter for prognosis. For example, the widely used clinical prediction rule for walking one year after traumatic SCI is based on age and ISNCSCI motor and light touch scores at L3 and S1, recorded within 15 days of injury.

Patterns of spinal cord injury

These clinical syndromes are not part of the ISNCSCI examination or the AIS classification. They are qualitative descriptions of anatomical patterns of injury. As of 2026, this section is called "Patterns of Spinal Cord Injury" rather than "Patterns of Incomplete Spinal Cord Injuries" because some of these syndromes can be complete.

Syndrome	Typical presentation
Central cord syndrome	The most common syndrome. Often seen in older people with cervical spondylosis after a hyperextension injury, usually a fall, with or without fracture. Presents as a motor incomplete injury with greater weakness in the upper limbs than the lower limbs.
Brown-Séquard syndrome	Hemisection of the cord. Ipsilateral loss of proprioception, vibration and motor control at and below the lesion; contralateral loss of pain and temperature sensation. The pure form is rare; mixed presentations with features of central cord syndrome are sometimes called Brown-Séquard-plus syndrome.
Anterior cord syndrome	Relatively rare; classically due to reduced blood supply to the anterior two-thirds of the cord. Loss of motor function, pain and temperature sensation at and below the level, with light touch and joint position sense preserved.
Posterior cord syndrome	The rarest syndrome (added in 2026). Mainly affects the dorsal columns: impaired vibration and joint position sense, and to a lesser extent light touch, below the level, with motor function, pain and temperature sensation generally preserved.
Conus medullaris syndrome	Usually follows a thoracolumbar bony injury. A mixed picture of upper and lower motor neuron signs. Sacral reflexes (bulbocavernosus, anal wink) are occasionally preserved with higher conus lesions. May be hard to distinguish clinically from cauda equina injury.
Cauda equina syndrome	Injury to the lumbosacral nerve roots, which may spare the cord itself. Lower motor neuron picture: flaccid paralysis of the affected lower limb muscles, areflexic bowel and bladder, and absent sacral reflexes.

Adapted from Rupp et al. 2026.

Anatomy of the lumbosacral spinal cord

The ISNCSCI examination

International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) or "ASIA" exam, 2026 version, page 1

International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) ASIA Impairment Scale (AIS), 2026 version, page 2

Positioning and spinal precautions

The examination is performed in supine, except for the anorectal examination, which can be performed side-lying. Supine testing allows valid comparison of scores across all phases of care. If the spine is unstable and not stabilised, the patient should be log-rolled for the anorectal examination, or an abbreviated examination performed in supine.

Special Topic: Spinal Precautions

• In a potentially unstable spine, take care with all manual muscle testing.
• With a suspected acute traumatic injury below T8, do not allow the hip to flex beyond 90° actively or passively, because of the kyphotic stress on the lumbar spine. Test hip flexion isometrically and unilaterally, keeping the opposite hip extended to stabilise the pelvis.
• Follow local spinal precautions until spinal instability has been excluded or managed.

Sensory examination

The required sensory examination tests one key point in each of the 28 dermatomes from C2 to S4–5, on the right and left sides. Each key point is tested for light touch (LT) and pin prick (PP; sharp/dull discrimination), compared with sensation on the face (preferably the cheek) as the normal reference.

Sensory scoring

Score	Meaning
0	Absent
1	Altered: either decreased/impaired sensation or hypersensitivity
2	Normal or intact (similar to the face)
NT	Not testable because of a condition directly related to the SCI (e.g. neuropathic pain including allodynia)
NT*	Not testable because of a non-SCI condition (e.g. cast, burn, amputation, or inability to perceive sensation on the face)
0*, 1*	Abnormal score affected by a non-SCI condition

Adapted from Rupp et al. 2026. See Recording not testable scores and non-SCI conditions below.

Sensory testing techniques

• Light touch: a tapered wisp of cotton stroked once across no more than 1 cm of skin, with the patient's eyes closed or vision blocked. Score 2 if the touch feels the same as on the face, 1 if it is felt but different, and 0 if it is not felt.
• Pin prick: a disposable safety pin, stretched apart so both ends can be used; the pointed end tests sharp and the rounded end dull.
  • The patient must reliably distinguish sharp from dull. If in doubt, 8 out of 10 correct answers is the suggested standard, which reduces the probability of correct guessing to less than 5%.
  • 0 = unable to distinguish sharp from dull, or no feeling
  • 1 = distinguishes sharp from dull reliably, but the intensity of sharpness differs from the face (greater or lesser)
  • 2 = distinguishes sharp from dull reliably, and the sharpness feels the same as on the face

Key sensory points

Please see the Additional Resources section at the end of this article for video examples of testing all key sensory points.

Level	Key point for sensory testing
C2	At least 1 cm lateral to the occipital protuberance (alternatively 3 cm behind the ear)
C3	Supraclavicular fossa (posterior to the clavicle) at the midclavicular line
C4	Over the acromioclavicular joint
C5	Lateral (radial) side of the antecubital fossa, just proximal to the elbow crease
C6	Thumb, dorsal surface of the proximal phalanx
C7	Middle finger, dorsal surface of the proximal phalanx
C8	Little finger, dorsal surface of the proximal phalanx
T1	Medial (ulnar) side of the antecubital fossa, just proximal to the medial epicondyle of the humerus
T2	Apex of the axilla
T3	Midclavicular line and 3rd intercostal space. Alternatively, palpate the manubriosternal joint (level of the 2nd rib), move slightly lateral and count down to the 3rd rib; the point is in the intercostal space below it
T4	Midclavicular line and 4th intercostal space (nipple line)
T5	Midclavicular line and 5th intercostal space (midway between T4 and T6)
T6	Midclavicular line and 6th intercostal space (level of the xiphisternum)
T7	Midclavicular line and 7th intercostal space (midway between T6 and T8)
T8	Midclavicular line and 8th intercostal space (midway between T6 and T10)
T9	Midclavicular line and 9th intercostal space (midway between T8 and T10)
T10	Midclavicular line and 10th intercostal space (umbilicus)
T11	Midclavicular line and 11th intercostal space (midway between T10 and T12)
T12	Midclavicular line over the midpoint of the inguinal ligament
L1	Midway between the key sensory points for T12 and L2
L2	Anterior-medial thigh, midway on a line from the midpoint of the inguinal ligament (T12) to the medial femoral condyle
L3	Medial femoral condyle above the knee
L4	Medial malleolus
L5	Dorsum of the foot at the 3rd metatarsophalangeal joint
S1	Lateral heel (calcaneus)
S2	Midpoint of the popliteal fossa
S3	Ischial tuberosity or infragluteal fold
S4–5	Perianal area less than 1 cm lateral to the mucocutaneous junction (taken as one level)

Adapted from Rupp et al. 2026.

ISNCSCI key sensory testing points, 2026 version

Optional sensory elements

Joint movement appreciation, position sense and deep pressure appreciation may be tested. They are not used for classification, and there is no field for them on the worksheet, so results are recorded in the Comments box.

• Joint movement appreciation and position sense are graded 0 (absent), 1 (impaired) or 2 (normal). Grade 1 means the patient reports large joint movements correctly in at least 8 of 10 trials but cannot reliably report small movements. Grade 2 means at least 8 of 10 correct on both small (about 10°) and large movements. Joints that can be tested include the thumb interphalangeal joint, little finger proximal interphalangeal joint, wrist, great toe interphalangeal joint, ankle and knee.
• Deep pressure appreciation is tested only where light touch and pin prick are both 0. Firm pressure is applied to the skin for 3–5 seconds at the wrist, fingers, ankles and toes, and graded as present or absent, using firm pressure on the chin as the reference.

Motor examination

The required motor examination tests 10 key muscle functions on each side, representing the C5–T1 and L2–S1 myotomes. They are tested in a rostral-to-caudal sequence in supine, with standard positioning and stabilisation. Poor positioning or stabilisation allows substitution by other muscles and gives an inaccurate grade.

Motor grading

Grade	Meaning
0	Total paralysis
1	Palpable or visible contraction
2	Active movement, full range of motion with gravity eliminated
3	Active movement, full range of motion against gravity
4	Active movement, full range of motion against gravity and moderate resistance in a muscle specific position
5	(Normal) active movement, full range of motion against gravity and full resistance in a muscle specific position expected from an otherwise unimpaired person
NT	Not testable because of a condition directly related to the SCI (e.g. spasticity, neuropathic pain)
NT*	Not testable because of a non-SCI condition (e.g. immobilisation, amputation, contracture, heterotopic ossification, burns, musculoskeletal pain)
0*, 1*, 2*, 3*, 4*	Abnormal grade affected by a non-SCI condition

Adapted from Rupp et al. 2026.

Clarifications in the 2026 edition:

• Partial range with gravity eliminated = grade 1. A weak muscle that produces active movement with gravity eliminated but does not reach full range is graded 1, not 2. Correct positioning at each end of the range is needed for accurate grading.
• Grades 2 and 3 are tested through range; grades 4 and 5 are not. For grades 4 and 5, the muscle is tested in its joint-specific test position (see Key muscle functions). If the joint cannot be placed in that position, record NT or NT*, depending on the cause.
• Limited range of motion. If range is limited by an SCI-related condition (spasticity, neuropathic pain) or a non-SCI condition (contracture, heterotopic ossification, burns, musculoskeletal pain) and more than 50% of normal range remains, grade through the available range. If less than 50% remains, record NT for an SCI-related cause and NT* for a non-SCI cause.
• Non-SCI weakness. The "5*" grade has not been used since 2019. If pain, disuse or another non-SCI factor reduces strength but still allows a forceful contraction, record the grade actually achieved, tag it with an asterisk, and explain in the Comments box how it should be treated for classification (for example "treat as normal").
• If pain, positioning, hypertonicity or other factors prevent standardised testing altogether, record NT (or NT* for a non-SCI cause).

ISNCSCI motor scoring, 2026 version

Key muscle functions

Please see the Additional Resources section at the end of this article for video examples of testing all 10 key muscle functions.

Level	Key muscle function and muscles	Testing position for grade 4 or 5
C5	Elbow flexion Biceps brachii, brachialis	Elbow flexed at 90°, arm at the patient's side, forearm supinated
C6	Wrist extension Extensor carpi radialis longus and brevis	Wrist in full extension
C7	Elbow extension Triceps brachii	Shoulder in neutral rotation, adducted and in 90° of flexion, with the elbow in 45° of flexion
C8	Flexion of the middle finger Flexor digitorum profundus	Distal phalanx fully flexed, with the proximal finger joints stabilised in extension
T1	Abduction of the little finger Abductor digiti minimi	Fingers fully abducted
L2	Hip flexion Iliopsoas	Hip flexed to 90°
L3	Knee extension Quadriceps	Knee flexed to 15°
L4	Ankle dorsiflexion Tibialis anterior	Ankle in full dorsiflexion
L5	Long toe extension Extensor hallucis longus	First toe fully extended
S1	Ankle plantarflexion Gastrocnemius, soleus	Hip in neutral rotation, flexion/extension and abduction/adduction; knee fully extended; ankle in full plantarflexion

Adapted from Rupp et al. 2026. See the ASIA Motor Exam Guide and InSTeP (fee required) for testing positions for grades 0–3.

Non-key muscle functions

Non-key muscle functions are not used to determine motor levels or motor scores. They are used only to decide whether an injury is motor incomplete, that is, AIS B versus AIS C or D (see Step 5). In a patient with an apparent AIS B classification and absent VAC, non-key muscle functions more than three levels below the motor level on each side should be tested. The Comments box should record that non-key muscle functions were tested, together with the most caudal voluntary muscle function found.

The root levels reflect the most rostral innervation of the muscles that usually perform each movement. Movements are listed rather than individual muscles, so the examiner does not need to identify which of several possible muscles is producing the movement.

Root level	Movement
C5	Shoulder: flexion, extension, abduction, adduction, internal and external rotation. Elbow: supination
C6	Elbow: pronation. Wrist: flexion
C7	Finger: flexion at proximal joint, extension. Thumb: flexion, extension and abduction in plane of thumb
C8	Finger: flexion at MCP joint. Thumb: opposition, adduction and abduction perpendicular to palm
T1	Finger: abduction of the index finger
L2	Hip: adduction
L3	Hip: external rotation
L4	Hip: extension, abduction, internal rotation. Knee: flexion. Ankle: inversion and eversion. Toe: MP and IP extension
L5	Hallux and toe: DIP and PIP flexion and abduction
S1	Hallux: adduction

From the 2026 ISNCSCI worksheet.

Anorectal examination

Disclaimer: the anorectal examination is an intimate, invasive procedure. It requires valid consent, a chaperone in line with local policy, and competence demonstrated under in-person supervision.

The anorectal examination provides the information for Step 4 (complete or incomplete injury). It has a sensory component (DAP) and a motor component (VAC).

Deep anal pressure (DAP)

DAP is tested by inserting a gloved, lubricated index finger and applying gentle pressure to the anorectal wall, which is innervated by the somatosensory components of the pudendal nerve (S4–5). Alternatively, the thumb can gently squeeze the anus against the inserted index finger. Consistently perceived pressure is recorded as present or absent (Yes/No). Any reproducible pressure sensation in the anal area means the injury is at least sensory incomplete.

In earlier editions, DAP did not need to be tested when light touch or pin prick was present at S4–5, because the injury was already sensory incomplete. The 2026 edition recommends that DAP is assessed routinely in every examination, including in people who are already sensory incomplete. A complete anorectal examination allows neurological change to be tracked over time and helps to correct misclassification. DAP also determines whether the sensory level can be recorded as "INT" (see Step 1 below).

Voluntary anal contraction (VAC)

The external anal sphincter is innervated by the somatic motor components of the pudendal nerve (S2–4). With a gloved, lubricated finger in the rectum, the examiner asks the patient to "squeeze my finger as if holding in a bowel movement". Reproducible voluntary contraction is recorded as present or absent (Yes/No). Present VAC means the injury is motor incomplete.

• VAC must be distinguished from reflex contraction. A contraction produced only with a Valsalva manoeuvre may be reflex and should be scored as absent.
• If DAP or VAC cannot be tested, record NT or NT* and give the reason in the Comments box.

Recording not testable scores and non-SCI conditions

Not testable (NT or NT*)

NT is used when a sensory point, key muscle, DAP or VAC cannot be tested reliably because of a condition directly related to the SCI, such as spasticity or neuropathic pain. NT* is used when the cause is a non-SCI condition, such as a cast, burn, amputation or contracture, or inability to perceive sensation on the face. The reason must always be given in the Comments box. The examination should still be completed as fully as possible. Where associated injuries (for example traumatic brain injury, brachial plexus injury or limb fracture) interfere with the examination, the levels should still be determined as accurately as possible, but the sum scores, AIS grade and ZPPs may be deferred to a later examination.

The asterisk (*)

Introduced in 2019, the asterisk applies to both sensory and motor scores. Pre-existing or coexisting conditions often affect ISNCSCI scores, for example peripheral nerve injury, previous stroke, amputation, fracture, burns, pain, limb swelling and age-related weakness. When this happens:

1. Record the score as examined. Do not upgrade it.
2. Tag the abnormal score with an asterisk (sensory 0*, 1*; motor 0*–4*), or record NT* if testing is impossible.
3. In the Comments box, name the condition and state how the score should be treated for classification (for example "treat as normal" or "treat as not normal").

How these tagged scores are used during classification is described in Classifying when non-SCI conditions are present below.

Sensory and motor scores

Scores are not one of the six classification steps, but they are recorded on the front of the worksheet and are the main way of tracking change over time.

Sensory scores

ISNCSCI 2026 worksheet, sensory scores

Light touch (yellow on the image above) and pin prick (green on the image above) scores are summed across all dermatomes and both sides to give two summary scores. With a maximum of 2 at each of 28 points per side, the maximum is 56 per side and 112 in total for each modality. Sensory scores cannot be calculated if any key point is not tested.

Motor scores

ISNCSCI 2026 worksheet, motor scores

Key muscle grades are summed into an upper extremity motor score (UEMS, maximum 50; yellow on the image above) and a lower extremity motor score (LEMS, maximum 50; green on the image above), each made up of right and left subscores of 25. The two should be reported separately, not added into a total out of 100, because they behave as separate scales. Motor scores cannot be calculated if any key muscle function is not tested.

Scores and non-SCI conditions

Where a non-SCI condition is present, sensory and motor sum scores are always calculated from the examined scores (ignoring asterisks) and are never tagged with an asterisk, because they do not include clinical assumptions.

Classification: the six steps

The worksheet sets out a recommended order for classification. Each step builds on the one before, so working through them in sequence reduces errors.

Step	Determine	Based on
1	Sensory levels (right and left)	Light touch and pin prick scores (and DAP for "INT")
2	Motor levels (right and left)	Key muscle grades and, where there is no myotome to test, the sensory level
3	Neurological level of injury (NLI)	The four levels from Steps 1 and 2
4	Complete or incomplete	Sacral sparing: S4–5 light touch and pin prick, DAP, VAC
5	ASIA Impairment Scale (AIS) grade	Steps 2–4, key and non-key muscle function
6	Zones of partial preservation (ZPP)	Sacral findings, levels from Steps 1 and 2, and function below them

ISNCSCI 2026 worksheet, six steps to classification

Where a level, grade or ZPP cannot be determined, record ND (not determinable), or ND* if this is because of a non-SCI condition. Any level, grade or ZPP that depends on an examiner's assumption about asterisk-tagged scores is tagged with an asterisk.

Step 1: Sensory levels

The sensory level on each side is the most caudal dermatome with normal (grade 2) light touch and pin prick, provided all dermatomes above it are also normal. It is the intact dermatome immediately above the first dermatome with impaired or absent light touch or pin prick.

• Determine a right and a left sensory level. Considering each modality on each side can give up to four levels (R-PP, R-LT, L-PP, L-LT); the sensory level for each side is the more rostral of its light touch and pin prick levels.
• The single sensory level is the most rostral of these levels.
• If sensation is abnormal at C2 but normal on the face, the sensory level is C1.
• If light touch and pin prick are intact in all dermatomes from C2 to S4–5 on a side and DAP is present, record "INT" (intact) rather than S5.
• If all dermatomes on a side score 2 but DAP is absent, there is still some impairment at the lowest sacral segments, and the sensory level for that side is S3.

Step 2: Motor levels

The motor level on each side is the most caudal key muscle function graded at least 3 (on supine testing), provided all key muscle functions above it are graded 5. Right and left motor levels may differ; a single motor level is the more rostral of the two.

Most muscles are innervated by two spinal segments. By convention, a key muscle graded 3 or more is taken to have intact innervation from the more rostral of its two segments, so the next most rostral key muscle must be graded 5. For example, if C8 shows no activity and C7 is graded 3, the motor level is C7, provided C6 and all key muscles above it are graded 5.

Regions with no testable myotome

There is no key muscle to test at C1–C4, T2–L1 and S2–S5. For these myotomes:

• myotomes at or above the sensory level are assumed to be normal (grade 5)
• myotomes below the sensory level are assumed to have less than antigravity strength (grade <3)

Example	Findings	Motor level	Reasoning
1	Sensory level C3	C3	C2–C3 are assumed normal; C4, below the sensory level, is assumed <3.
2	Sensory level C3; C5 key muscle graded 5	C3	C4 is assumed <3, so preserved key muscle function further down cannot move the motor level caudally.
3	Sensory level C4; C5 graded <3	C4	C2–C4 are assumed normal; C5 is below antigravity strength.
4	Sensory level C4; C5 graded ≥3	C5	C2–C4 are assumed normal and C5 is at least 3.
5	Sensory level C7; all upper limb key muscles graded 5	T1	All testable myotomes to T1 are normal; T2 and below are below the sensory level and assumed <3.
6	Sensory level T3; all upper limb key muscles graded 5	T3	T2–T3 are assumed normal; T4 and below are assumed <3.
7	As example 6, but T1 graded 3 or 4	T1	T1 is not graded 5, so the motor level cannot be caudal to T1.
8	Upper limbs graded 5; sensory level L1; L2 graded <3	L1	T2–L1 are assumed normal; L2 is below antigravity strength.
9	Upper limbs graded 5; sensory level L1; L2 graded ≥3, L3 graded <3	L2	T2–L1 are assumed normal and L2 is at least 3.
10	All key muscles graded 5; sensory level L3	S1	All testable myotomes to S1 are normal; S2 and below are assumed <3. This applies whether or not VAC is present.

Adapted from Rupp et al. 2026.

In examples 5 and 10 the motor level lies caudal to the sensory level. In such cases, check whether the sensory impairment could be caused by a non-SCI condition. The same logic also means that L2 can be the motor level only if sensation at L1 and above is intact.

The worksheet still carries the shorter wording: "in regions where there is no myotome to test, the motor level is presumed to be the same as the sensory level, if testable motor function above that level is also normal". The booklet rule above is the complete version and should be used for examples such as 5 and 10.

Clarification for an unusual pattern

Before 2026, the Standards did not say how to determine the motor level when all key muscles of the upper limbs (or of all limbs) were graded normal, sensation was impaired in the key muscle segments, and normal sensation resumed from T2 (or S2). In an analysis of 2,660 motor levels from the European Multicenter Study about Spinal Cord Injury (EMSCI), two readings of the 2019 rule gave different motor levels in 3.6% of cases. The 2026 edition resolves this by adopting the most conservative option: no assumptions about motor function are made from sensory function below the sensory level. Non-testable myotomes below the sensory level are therefore assumed to be <3, even if sensation is normal further down.

Other points

• Because only 10 myotomes and VAC are tested, "INT" should be used for the motor level only in people with AIS E.

Step 3: Neurological level of injury (NLI)

The NLI is the most caudal segment of the cord with normal sensation and antigravity (grade 3 or more) muscle function on both sides, provided that sensory and motor function above it is normal.

Four levels are determined in Steps 1 and 2: right sensory, left sensory, right motor and left motor. The single NLI is the most rostral of these four. It is good practice to record all four as well, because a single NLI can be misleading functionally, for example when the sensory level lies rostral to the motor level.

The NLI also determines whether the injury is described as tetraplegia (NLI at or rostral to T1) or paraplegia (NLI caudal to T1), and it is the reference level for distinguishing AIS C from AIS D in Step 5.

If the NLI depends on an examiner's assumption about asterisk-tagged scores, it is tagged with an asterisk. Under the 2019 edition, the NLI had to be tagged whenever any sensory or motor level was tagged. This automatic rule was removed in 2026, because there are exceptions: for example, a tagged level that is not the most rostral of the four may not affect the NLI at all.

Step 4: Complete or incomplete?

Completeness depends only on sacral sparing: any preserved sensory or motor function in the lowest sacral segments, S4–5.

• Sensory sacral sparing: light touch or pin prick preserved (intact or altered) at S4–5 on either side, or DAP present
• Motor sacral sparing: VAC present

Finding	Classification
VAC absent and all S4–5 sensory scores 0 and DAP absent	Complete
Any of the above present	Incomplete

A person can therefore be classified as complete (AIS A) even when some function is preserved several segments below the NLI. That preserved function is documented with the ZPP in Step 6.

Step 5: ASIA Impairment Scale (AIS) grade

The AIS evolved from the Frankel Scale and uses the same letters, but it is a distinct scale based on standardised, quantitative classification rules.

Grade	Type of injury	Description
A	Complete	No sensory or motor function is preserved in the sacral segments S4–5.
B	Sensory incomplete	Sensory but not motor function is preserved below the NLI and includes S4–5 (light touch or pin prick at S4–5, or DAP) AND no motor function is preserved more than three levels below the motor level on either side of the body.
C	Motor incomplete	Motor function is preserved at the most caudal sacral segments (VAC) OR the patient is sensory incomplete and has some motor function more than three levels below the ipsilateral motor level on either side. Key or non-key muscle functions may be used to establish motor incomplete status. For AIS C, less than half of the key muscle functions below the single NLI have a grade ≥3.
D	Motor incomplete	Motor incomplete status as defined above, with at least half (half or more) of the key muscle functions below the single NLI graded ≥3.
E	Normal ISNCSCI	Sensation and motor function, as tested with ISNCSCI, are graded as normal in all segments AND the patient had prior deficits, typically documented on an earlier ISNCSCI examination. Someone without an SCI does not receive an AIS grade.

Adapted from Rupp et al. 2026 and the 2026 worksheet.

Working through Step 5

1. Is the injury complete (Step 4)? If yes, AIS A.
2. Is the injury motor complete? It is motor incomplete if VAC is present, or if a sensory incomplete patient has motor function more than three levels below the motor level on either side. Before assigning AIS B, test non-key muscle functions (see Non-key muscle functions). If motor complete, AIS B.
3. Are at least half of the key muscle functions below the single NLI graded 3 or more? No: AIS C. Yes: AIS D.
4. Are sensation and motor function normal in all segments in someone with a documented SCI? If yes, AIS E.

Key points:

• Two different reference levels are used. To distinguish AIS B from C or D, motor sparing is judged against the motor level on each side (Step 2). To distinguish AIS C from D, the proportion of key muscles graded ≥3 is judged against the single NLI (Step 3).
• AIS E was renamed "Normal ISNCSCI" in 2026. It is used only at follow-up, when someone with a documented SCI has recovered normal ISNCSCI findings. If no deficit is found at the first examination, the person is neurologically intact and the AIS does not apply. A normal ISNCSCI examination does not rule out impairments the examination does not test, such as autonomic dysfunction, other sensory modalities, untested myotomes or coordination, and these can still limit activities of daily living. About 2% of people with traumatic SCI recover to AIS E. In a large European cohort, first conversion to AIS E occurred at a median of 171 days after injury.
• If the AIS grade depends on an examiner's assumption about asterisk-tagged scores, it is tagged (for example AIS C*). Some grades are unaffected by tagged scores, for example AIS A with tagged scores rostral to S4–5, and these are not tagged. If the grade cannot be determined, record ND, or ND* if the cause is a non-SCI condition.

Step 6: Zones of partial preservation (ZPP)

The ZPP records how far partially preserved function extends below the sensory and motor levels. Before 2019 it was used only in complete (AIS A) injuries. Since the 2019 revision, the motor and sensory ZPPs are decided separately, based on the sacral findings rather than the AIS grade:

1. The motor ZPP is recorded whenever VAC is absent.
2. The sensory ZPP is recorded whenever there is no sacral sensation (no DAP, and no light touch or pin prick at S4–5).

As a result, a ZPP can be recorded in some AIS B, C and D injuries. It does not apply when the relevant sacral function is present; for example, most AIS D injuries have both VAC and sacral sensation, so neither ZPP applies.

Worked examples: does a ZPP apply?

Example	Sacral findings	Motor ZPP	Sensory ZPP
AIS A	No VAC, no sacral sensation	Recorded	Recorded
AIS B	No VAC, sacral sensation present	Recorded	NA
AIS C or D, motor incomplete because of muscle function more than 3 levels below the motor level	No VAC, sacral sensation present	Recorded	NA
AIS C or D with VAC but no sacral sensation (uncommon)	VAC present, no sacral sensation	NA	Recorded
AIS C or D with VAC and sacral sensation (common)	Both present	NA	NA

When the ZPP is "not applicable" (NA)

Finding at S4–5	Sensory ZPP	Motor ZPP
VAC present	Depends on sensory findings	NA on both sides
VAC absent	Depends on sensory findings	Recorded on both sides
DAP present	NA on both sides	Depends on VAC
DAP absent, light touch or pin prick present on one side only	NA on that side; recorded on the other side	Depends on VAC
DAP absent, light touch and pin prick absent on both sides	Recorded on both sides	Depends on VAC

Adapted from Rupp et al. 2026 and Schuld et al. 2024. The worksheet summarises this as "with sacral sparing of sensory function, the sensory ZPP is not applicable"; the 2026 booklet gives the side-by-side rule shown here.

Recording the ZPP

ISNCSCI 2026 worksheet, zone of partial preservation

• Four ZPPs are recorded: right sensory, left sensory, right motor and left motor.
• Each is a single segment (not a range): the most caudal segment on that side with any sensory or motor function. For example, if the right sensory level is C5 and some sensation extends from C6 to C8, record "C8".
• If there is no partially preserved function below a level, record the level itself as the ZPP.
• The motor ZPP does not follow sensory function. It is based only on voluntary muscle contraction below the motor level. For example, if the sensory and motor levels and the NLI are T4, with some sensation at left T6, the left sensory ZPP is T6 but the motor ZPP remains T4.
• Non-key muscle functions are generally not included in the motor ZPP. However, when the most caudal non-key muscle function is the only determinant of an AIS C or D grade, its root level is recorded as the motor ZPP for that side.
• ZPPs that depend on an examiner's assumption about asterisk-tagged scores are tagged with an asterisk; if a ZPP cannot be determined, record ND or ND*.

Classifying when non-SCI conditions are present

The six steps are the same when a non-SCI condition is present, but asterisk-tagged scores must first be converted into assumed scores, and the results then checked to see which classification variables need an asterisk.

Assumed scores:

• Above the sensory or motor level: tagged scores are usually treated as normal (sensory 2, motor 5).
• At or below the level: the non-SCI condition is superimposed on the SCI. Classify using the examined score and every higher score except normal (for example, for motor 2*, classify with 2, 3 and 4). If every possible score gives the same result for a classification variable, record that result with an asterisk. If the results differ, record ND*.

Which classification variables need an asterisk? If in doubt, use this algorithm:

1. Work through Steps 1–6 using the assumed scores (the scores the examiner believes would be present without the non-SCI condition).
2. Record these results in the classification boxes on the worksheet.
3. Work through Steps 1–6 again using the examined scores, ignoring the asterisks.
4. Tag with an asterisk each variable recorded in step 2 whose result differs between steps 1 and 3.

Worked example. A patient with a T10 SCI has an old right ulnar nerve injury. Sensory levels are T10 bilaterally, all upper limb key muscles are graded 5 except right T1 (little finger abduction), which is graded 2, and the injury is AIS A. The score is recorded as 2*, with the comment "right ulnar neuropathy; treat as normal for classification".

• Assumed scores (right T1 = 5): right motor level T10, NLI T10, AIS A.
• Examined scores (right T1 = 2): right motor level C8 (C8 graded 5, T1 <3), NLI C8, AIS A.
• Result: the right motor level and NLI differ, so they are recorded as T10*. The left motor level and both sensory levels are unaffected and are not tagged. The AIS grade is A either way, so it is not tagged.

The asterisk on the NLI tells the reader that classifying this injury as paraplegia rather than tetraplegia depends on the examiner's clinical judgement.

Not testable scores. When NT (an SCI-related cause) prevents a level, grade or ZPP from being determined, record ND. When NT* (a non-SCI cause) does so, or when the assumed scores do not give a consistent result, record ND*. Where NT or NT* does not affect a level or grade, that level or grade can still be documented.

Expedited and computerised classification

Expedited ISNCSCI (E-ISNCSCI)

In some situations, such as initial screening or follow-up in the chronic phase, a full examination may not be practical. The International Standards Committee therefore developed the expedited ISNCSCI, which determines the NLI and AIS grade with the fewest possible steps using standard ISNCSCI testing procedures. Version 1 (2020) was replaced by version 2 after validation work found unacceptably high misclassification rates. In one study, substituting S1 findings for the anorectal examination misclassified 45% of true AIS B injuries.

Version 2:

• includes a full test of all key muscle functions, so motor scores can still be calculated
• uses a different sequence of sensory testing
• no longer allows an S1 sensorimotor examination to substitute for the anorectal examination, although not every part of the anorectal examination is always needed

E-ISNCSCI is published as separate guidance on the ASIA website (version 2 protocol) and is not part of the Standards booklet. Findings are recorded on the full ISNCSCI worksheet, with a note in the Comments box that the classification is based on E-ISNCSCI version 2. The full examination remains the reference standard for documenting SCI.

Computerised classification

Classification errors are common, even among experienced clinicians. Validated online algorithms can check hand classification and are recommended as a learning and quality tool. They do not replace the examination itself.

• EMSCI ISNCSCI Calculator (free access)

Psychometrics

Reliability and validity

• Trained examiners: the motor and sensory examinations show high inter-rater and intra-rater reliability when performed by trained examiners.
• Non-traumatic SCI: in people with non-traumatic spinal cord lesions, ISNCSCI showed substantial internal consistency and inter-rater agreement for AIS grade, motor and sensory scores, and the motor score correlated with the Spinal Cord Independence Measure (SCIM). Evidence in non-traumatic SCI is still limited. The 2026 Standards state that ISNCSCI should be used only in non-traumatic causes that produce a distinct level, such as spinal cord infarction, distinct spinal tumours or degenerative cervical myelopathy. It should not be used in generalised spinal cord conditions with no specific level, such as hereditary spastic paraplegia or multiple sclerosis, or where impairments caused by the SCI cannot be confidently distinguished from those caused by the underlying condition.
• Children: in a multicentre study of 236 children with chronic SCI, no child under 6 years could complete the examination, while inter-rater reliability for the motor and sensory examinations was high in children aged 6 and over. ISNCSCI is therefore recommended from 6 years of age. Examiners should complete the InSTeP and WeeSTeP training first. More research is needed for children under 6.

Training and classification accuracy

Classification is complex, and errors are common. In an international survey of 107 SCI professionals using the 2019 edition, overall classification accuracy was 74.6%, and errors were concentrated in the concepts introduced in 2019. Motor levels, AIS grade and ZPPs carry the highest error rates, and incomplete injuries are harder to classify than complete ones. Formal training in scoring and classification improves accuracy. The International Standards Committee has published case series on common classification challenges (one, two), and adds new cases to the ISNCSCI classification workbook. Discrepancies between clinicians' classifications and computerised reclassification have been documented, so algorithm checks are a useful safeguard.

Prognostic value

ISNCSCI findings are the main basis for predicting neurological recovery after traumatic SCI:

• Most AIS conversion and motor recovery occurs in the first 6–9 months, with the fastest motor recovery in the first 3 months.
• Conversion from complete to incomplete injury is more common in tetraplegia than paraplegia.
• Total motor recovery is greater after initial AIS B than AIS A, and greater after initial AIS C than after motor complete injury.
• Older age has a negative effect on neurological and functional recovery, although the threshold (over 50 or over 65 years) is unclear. Penetrating injury reduces the likelihood of conversion.
• In injuries with an initial NLI between T6 and T12, a sensory ZPP of 3 or more segments has prognostic value for AIS conversion.
• Results from very early examinations (within about 4 hours of injury) should be interpreted with caution.
• Walking: a clinical prediction rule using age (under or over 65 years), L3 and S1 motor scores, and L3 and S1 light touch scores, recorded within 15 days of injury, predicted independent walking at one year with high accuracy (area under the curve 0.96). It added prognostic value within each AIS grade, particularly B and C.

What ISNCSCI does not measure

ISNCSCI does not assess autonomic function, so it should not be used on its own to predict bladder, bowel, sexual, cardiovascular or thermoregulatory outcomes. The International Standards to document Autonomic Function following SCI (ISAFSCI), second edition (2021) cover cardiovascular, thermoregulatory and sudomotor, bronchopulmonary and sacral (bladder, bowel and sexual) function, and are recommended alongside ISNCSCI. ISNCSCI also does not measure function or participation; tools such as the Spinal Cord Independence Measure (SCIM) are needed for these.

Additional Resources

Worksheets and Standards

• ISNCSCI: 9th edition 2026 (open access): the current Standards
• ISNCSCI worksheet, 2026 (REV 04/26): free download; licensed CC BY-NC-ND 4.0
• ISNCSCI 9th edition booklet (ASIA store)
• ISNCSCI resources page (ASIA): booklets, papers, calculators and training links
• ISNCSCI: Revised 2019 (open access): previous edition, for reference
• Expedited ISNCSCI (E-ISNCSCI), version 2 protocol
• ISAFSCI, second edition

Testing guides

• ASIA Key Sensory Points guide
• ASIA Motor Exam Guide
• Dermatome testing videos
  • Sensory exam technique video
  • Sensory exam overview
• Myotome testing videos
  • C5
  • C6
  • C7
  • C8
  • T1
  • L2
  • L3
  • L4
  • L5
  • S1

Training and practice

• InSTeP: International Standards Training e-Learning Program, with five modules covering anatomy, sensory, motor and anorectal examination, and classification. It has been updated for the 2026 edition. WeeSTeP (paediatric) and ASTeP (autonomic) are also available.
• ISNCSCI Classification Workbook: practice cases
• ISNCSCI Classification Workbook: cases and answers: 26 practice cases with explanations

Classification calculators

• EMSCI ISNCSCI Calculator

For patients and families

• What is the ISNCSCI exam? Adult brochure