National Institutes of Health Stroke Scale — 15-item bedside neurological exam quantifying stroke severity (0–42) · оцінка тяжкості інсульту за шкалою NIHSS
The NIH Stroke Scale opens with the single most prognostically important item on the entire instrument: consciousness. A patient who cannot be aroused, cannot answer, or cannot follow commands is signaling a large, often devastating, cerebral or brainstem insult. Items 1a–1c probe arousal, orientation, and command-following in three graded sub-tests, each contributing points that compound quickly for severely affected patients.
The examiner scores the patient's responsiveness even before any verbal exchange:
0 = Alert, keenly responsive 1 = Not alert, arousable by minor stimulation to obey, answer, or respond 2 = Not alert, requires repeated or strong/painful stimulation to attend 3 = Responds only with reflex motor or autonomic effects, or is totally unresponsive, flaccid, areflexic
This item is scored even if a full evaluation is precluded by factors such as an endotracheal tube, language barrier, or orotracheal trauma. A score of 3 is only given for total unresponsiveness — it does not require the patient to be unarousable to pain, but rather that no purposeful response is elicited.
A patient scoring 2–3 on Item 1a alone frequently carries a total NIHSS in the moderate-to-severe range even before any focal deficit is tested — depressed consciousness is one of the strongest single predictors of large territory or brainstem stroke.
1b. LOC Questions — the patient is asked the current month and their age: 0 = Answers both correctly 1 = Answers one correctly 2 = Answers neither correctly (includes dysarthric, aphasic, or intubated patients unable to respond) Only the initial answer counts; the examiner does not coach.
1c. LOC Commands — the patient is asked to open/close the eyes and grip/release the non-paretic hand: 0 = Performs both tasks correctly 1 = Performs one task correctly 2 = Performs neither task correctly If hands cannot be used, a substitute one-step command (e.g., tongue protrusion) may be used. Only the first attempt is scored — the examiner gives credit for an unequivocal, though incomplete, effort.
Together, items 1a–1c create an early gate: a comatose patient (1a=3) typically cannot meaningfully attempt 1b/1c and will score the maximum on those as well, compounding to 7 points before any lateralized sign is examined.
Items 2 through 4 probe the cranial nerve and cortical pathways governing eye movement, visual perception, and facial musculature. Because the corticobulbar and optic pathways are highly organized and somatotopically mapped, these three items are powerful localizers — a conjugate gaze preference toward one side, a homonymous visual field cut, or a lower-face droop each point toward a specific hemisphere and vascular territory.
Only horizontal eye movements are tested, voluntarily or with the oculocephalic (doll's eyes) reflex if needed: 0 = Normal 1 = Partial gaze palsy — gaze is abnormal in one or both eyes but forced deviation or total gaze paresis is not present 2 = Forced deviation, or total gaze paresis not overcome by the oculocephalic maneuver Isolated peripheral nerve palsy (CN III, IV, VI) scores 1. Frontal eye field lesions classically produce ipsiversive gaze preference — the eyes look toward the lesion, away from the hemiparetic side ("wrong-way eyes").
Visual fields (upper and lower quadrants) are tested by confrontation, using finger counting or visual threat if needed: 0 = No visual loss 1 = Partial hemianopia (including quadrantanopia) 2 = Complete hemianopia 3 = Bilateral hemianopia (blind from any cause, including cortical blindness) A homonymous hemianopia strongly implicates the contralateral optic radiation or occipital cortex — commonly a posterior cerebral artery (PCA) territory infarct. If extinction is present on double simultaneous stimulation, score 1 and report separately under Item 11.
The patient shows teeth, raises eyebrows, and closes eyes tightly: 0 = Normal symmetrical movement 1 = Minor paralysis (flattened nasolabial fold, asymmetry on smiling) 2 = Partial paralysis (total or near-total paralysis of lower face) 3 = Complete paralysis of one or both sides (absence of facial movement in the upper and lower face) A central (upper motor neuron) facial palsy from stroke classically spares the forehead — the patient can still wrinkle the brow on the affected side because the frontalis muscle receives bilateral cortical innervation — distinguishing it from a peripheral (CN VII nucleus/nerve) lesion, which paralyzes the entire hemiface including the forehead.
Items 5 and 6 are the workhorse motor exam of the NIHSS, each limb tested independently and scored 0–4. Because each of the four limbs can contribute up to 4 points, motor testing alone accounts for 16 of the scale's 42 points — making hemiparesis severity the single largest contributor to total NIHSS and a direct proxy for corticospinal tract involvement.
The arm is extended (palms down) to 90° if sitting, 45° if supine, and held for 10 seconds: 0 = No drift; limb holds 90°/45° for full 10 seconds 1 = Drift; limb holds initially but drifts down before 10 seconds, without hitting the bed/support 2 = Some effort against gravity; limb cannot get to or maintain (if placed) the position, drifts down to bed, but has some effort against gravity 3 = No effort against gravity; limb falls immediately 4 = No movement UN = Amputation or joint fusion (explain, does not count toward total). Each side (5a right, 5b left) is scored separately, and the examiner counts aloud to standardize timing.
With the patient supine, the leg is held at 30° for 5 seconds: 0 = No drift; leg holds 30° position for full 5 seconds 1 = Drift; leg falls by the end of the 5-second period but does not hit the bed 2 = Some effort against gravity; leg falls to bed by 5 seconds, but has some effort against gravity 3 = No effort against gravity; leg falls immediately 4 = No movement UN = Amputation or joint fusion. Each side (6a right, 6b left) is scored separately. Testing the weaker-appearing side first is a common bedside strategy, but formally the non-paretic side is tested for baseline comparison.
Motor drift is exquisitely sensitive to even mild corticospinal tract dysfunction — a subtle downward pronator drift of the outstretched arm can be the only objective sign of an otherwise clinically silent small subcortical infarct, making Items 5–6 indispensable even in "mild" presentations.
The final block of individual items captures cerebellar coordination, somatosensory perception, higher-order language function, speech articulation, and hemispatial attention. These five items are less about raw strength and more about how the brain integrates and interprets information — deficits here often reflect cortical, thalamic, or cerebellar involvement rather than pure corticospinal tract injury.
Finger-nose-finger and heel-shin testing is performed on both sides; ataxia is scored only if disproportionate to any weakness present: 0 = Absent 1 = Present in one limb 2 = Present in two limbs UN = Amputation, joint fusion, or the patient cannot understand. If the patient is hemiplegic and cannot perform the test, ataxia is scored 0 (it cannot be assessed if there is no voluntary movement to observe) — ataxia must be out of proportion to weakness, so a fully paralyzed limb, by definition, cannot demonstrate ataxia.
8. Sensation: tested with pinprick (or withdrawal from noxious stimulus in obtunded patients) in as many body areas as needed — arm, leg, trunk, face: 0 = Normal, no sensory loss 1 = Mild-to-moderate loss; patient feels pinprick less sharp or is dull, or there is a loss of superficial pain with pressure sense preserved 2 = Severe to total sensory loss; patient is unaware of being touched in the face, arm, and leg
9. Best Language: assessed via picture description, naming, and reading. This is the item that captures aphasia: 0 = No aphasia, normal 1 = Mild-to-moderate aphasia; some obvious loss of fluency or comprehension without significant limitation on ideas expressed 2 = Severe aphasia; fragmentary expression, examiner carries burden of communication, listener has to make guesses 3 = Mute, global aphasia; no usable speech or auditory comprehension Broca's (expressive), Wernicke's (receptive), and global aphasia all map to this single graded item, weighted by functional communication rather than by lesion location.
10. Dysarthria: the patient reads or repeats listed words, assessing articulation rather than content: 0 = Normal 1 = Mild-to-moderate; slurring of some words, can be understood with some difficulty 2 = Severe; speech so slurred as to be unintelligible, or mute/anarthric UN = Intubated or physical barrier
11. Extinction and Inattention (formerly "Neglect"): tested with double simultaneous visual and tactile stimulation: 0 = No abnormality 1 = Visual, tactile, auditory, spatial, or personal inattention, or extinction to bilateral simultaneous stimulation in one sensory modality 2 = Profound hemi-inattention or extinction in more than one modality; does not recognize own hand, orients to only one side of space Neglect is classically associated with non-dominant (usually right) parietal lesions and, when severe, is a strong marker of poor functional recovery independent of raw motor strength.
The 15 individually scored items sum to a single number between 0 and 42, and that number drives some of the most time-critical decisions in acute medicine: whether to give intravenous thrombolysis, whether to pursue mechanical thrombectomy, which unit to triage the patient to, and what to tell the family about likely recovery. The NIHSS is not a diagnostic test — a normal score does not exclude stroke, particularly posterior circulation and lacunar strokes — but as a severity ruler it is unmatched in acute stroke care.
0 — No stroke symptoms: a fully normal 15-item exam. Does not rule out stroke (e.g., small lacunar or posterior fossa infarcts can be NIHSS-silent).
1–4 — Minor stroke: subtle, often single-domain deficits (e.g., isolated mild facial droop or dysarthria). Historically under-treated with thrombolysis due to perceived low disability risk, though current guidelines emphasize treating any *disabling* deficit regardless of numeric score.
5–15 — Moderate stroke: the classic middle-cerebral-artery territory presentation — hemiparesis, hemisensory loss, gaze preference, aphasia or neglect in combination. This band contains the majority of thrombolysis- and thrombectomy-eligible patients.
16–20 — Moderate-to-severe stroke: substantial multi-domain deficit, often depressed consciousness beginning to contribute; large-vessel occlusion (LVO) increasingly likely.
21–42 — Severe stroke: large-territory or malignant MCA infarction, brainstem catastrophe, or extensive hemorrhage. Mortality and severe disability rates rise sharply; decisions about decompressive hemicraniectomy and goals-of-care become central.
An NIHSS ≥ 6 is one of several factors that increases the pretest probability of a large-vessel occlusion (LVO) amenable to mechanical thrombectomy — but LVO screening scales (RACE, VAN, CPSSS) are typically used prehospital, while the full NIHSS remains the in-hospital gold standard once the patient reaches definitive imaging.
Score thresholds inform, but do not alone dictate, thrombolysis decisions: a very low NIHSS (1–4) with a clearly disabling deficit (e.g., isolated aphasia or hand paresis in a violinist) may still warrant treatment, while historically some clinicians hesitated to treat "too mild" a deficit. On the other end, very high NIHSS scores (>25) were an original NINDS trial relative exclusion criterion due to hemorrhage risk, though real-world practice and evolving guidelines have narrowed strict cutoffs. Mechanical thrombectomy eligibility for large-vessel occlusion is assessed largely independent of the exact NIHSS number once LVO is confirmed on imaging, though higher scores raise LVO pretest probability.
Serial NIHSS scoring (on arrival, post-treatment, 24 hours, discharge) is itself a monitoring tool: a rise of ≥4 points suggests neurological deterioration (hemorrhagic transformation, malignant edema, re-occlusion) and triggers urgent re-imaging.
The 90-day modified Rankin Scale (mRS, 0=no symptoms to 6=death) is the standard long-term functional outcome measure in stroke trials, and admission NIHSS is its single strongest independent predictor. Population-level studies (not a fixed formula) show a broad, illustrative gradient: patients presenting with NIHSS 0 overwhelmingly reach mRS 0; NIHSS 1–4 patients frequently achieve mRS 0–2 (functional independence); the moderate 5–15 band spans a wide outcome range (mRS 2–3 on average) heavily modified by treatment and age; 16–20 shifts the average toward mRS 3–4; and NIHSS >20 is associated with a markedly increased likelihood of mRS 4–6 (severe disability or death).
These mappings are population-level tendencies from large cohort and trial data, not individual guarantees — age, comorbidities, time-to-treatment, collateral circulation, and stroke mechanism all modulate the ultimate outcome for any single patient.
This simulator's mRS outlook readout is explicitly illustrative — a rough population-derived band, not a validated calculator, and should never substitute for individualized clinical prognostication.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| Consciousness, Questions, Commands | |||
| Gaze, Visual Fields, Facial Palsy | |||
| Motor Arm (L/R), Motor Leg (L/R) | |||
| Limb Ataxia | |||
| Sensory (pinprick) | |||
| Best Language (aphasia) | |||
| Dysarthria | |||
| Extinction / Inattention (Neglect) |