National Institutes of Health Stroke Scale (NIHSS)

Purpose of the measure

The NIHSS is a 15-item impairment scale, intended to evaluate neurologic outcome and degree of recovery for patients with stroke. The scale assesses level of consciousness, extraocular movements, visual fields, facial muscle function, extremity strength, sensory function, coordination (ataxia), language (aphasia), speech (dysarthria), and hemi-inattention (neglect) (Lyden, Lu, & Jackson, 1999; Lyden, Lu, & Levine, 2001). The NIHSS was designed to assess differences in interventions in clinical trials, although its use is increasing in patient care as an initial assessment tool and in planning postacute care disposition (Schlegel et al., 2003; Schlegel, Tanne, Demchuk, Levine, & Kasner, 2004).

Available versions

Original version. Brott, Adams, Olinger, Marler, Barsan, Biller, Spilker, Holleran, Eberle, Hertzberg, Rorick, Moomaw, and Walker (1989).

Features of the measure

Items:
Items of the NIHSS are based on three previously used scales, the Toronto Stroke Scale, the Oxbury Initial Severity Scale and the Cincinnati Stroke Scale (Brott et al., 1989).

The scale has 15 items in total which assess the following:

  1. Level of consciousness
  • Responsiveness of the patient (rated from 0 – 3).
  • Questions: Patients are asked to state the month and their age (rated from 0 – 2).
  • Commands: The patient is asked to open and close the eyes and then to grip and release the non-paretic hand (hand not affected by partial motor paralysis) (rated from 0 – 2).
  1. Best gaze
  • Horizontal eye movements of patient (rated from 0 – 2).
  1. Visual
  • To assess the presence of hemianopia (rated from 0 – 3).
  1. Facial palsy
  • Patients are asked to show their teeth or raise their eyebrows and close their eyes. Look for symmetry (rated from 0 – 3).
  1. Motor arm
  • Left arm: Arm is extended (palms down) 90 degrees (if sitting) or 45 degrees (if supine). Drift is scored if the arm falls before 10 seconds (rated from 0 – 4, or UN if amputation or joint fusion).
  • Right arm: Same as in a.
  1. Motor leg
  • Left leg: Leg is raised at 30 degrees (supine). Drift is scored if the leg falls before 5 seconds (rated from 0 – 4, or UN if amputation or joint fusion).
  • Right leg: Same as in a.
  1. Limb ataxia
  • Finger-to-nose and heel-to-shin test (rated from 0 – 2, or UN if amputation or joint fusion).
  1. Sensory function
  • If level of consciousness is impaired, score if a grimace or an asymmetric withdrawal is observed (rated from 0 – 2).
  1. Best language (aphasia)
  • Standard pictures are named (rated from 0 – 3).
  1. Dysarthria
  • Patient is asked to read or repeat words from a list (rated from 0 – 2, or UN if intubated or other physical barrier).
  1. Extinction and inattention (formerly called neglect)
  • Sufficient information to detect neglect may be obtained from prior testing (rated from 0 – 2).

An additional item that measures distal motor function has been used in a few drug trials, but is not widely used in ongoing research or in clinical practice.

Time:
The examination requires less than 10 minutes to complete.

Scoring:
Each item is scored from 0 – 2, 0 – 3, or 0 – 4, and untestable items are scored as “UN”. A score of 0 indicates normal performance. Total scores on the NIHSS range from 0 – 42, with higher values reflecting more severe cerebral infarcts. Stroke severity is further stratified in the following way:

(Source: Brott et al., 1989)

≥ 25

Very severe neurological impairment

5-14

Mild to adequately severe neurological impairment

< 5

Mild impairment

The predictive value of the scale can also aid in planning a patient’s rehabilitation or long-term care needs, even as early as the day of admission. NIHSS scores can be interpreted in the following way:

(Source: Schlegel et al., 2003; Rundek et al., 2000; Goldstein & Samsa, 1997; DeGraba, Hallenbeck, Pettigrew, Dutha, & Kelly, 1999)

≥ 14

Severe: Long-term care in nursing facility required

6-13

adequate: Acute inpatient rehabilitation required

≤ 5

Mild: 80% with this score are discharged home

The NIHSS can be completed and scored automatically at the following link:
http://sitemaker.umich.edu/chant/yale_nihss_calculator

Equipment:
None typically reported.

Subscales:
The subscale items encompass level of consciousness, vision, extraocular movements, facial palsy, limb strength, ataxia, sensation, and speech and language.

Training:
A trained observer rates the patent’s ability to answer questions and perform activities. Training is minimal and is available through instructional videos: a 45-minute training program tape, and two certification tapes (Lyden et al., 1994), or alternatively one can be trained and certified online at the following website: http://www.nihstrokescale.org/. A new training DVD is now available and has established reliability (Lyden et al., 2005).

It is important to note that one must be both trained and certified in order to administer the NIHSS.

As the NIHSS was designed as an observational scale, measurement by self-report or by telephone is not possible. However, measurement by video telemedicine appears to be reliable and could offer a method for remote assessment (Meyer et al., 2005; Shafqat, Kvedar, Guanci, Chang, & Schwamm, 1999). This method of administration would require slightly more time to complete.

To see video clips of the NIHSS items being administered by telemedicine, visit the following link: https://telestroke.massgeneral.org/about.asp

Schmülling, Grond, Rudolf, and Kiencke (1998) examined whether the NIHSS could be reliably administered without any formal training program. The results of this study suggest that good inter-rater reliability of the NIHSS depends on adequate training of the raters. Inter-rater reliability among untrained raters was only poor (kappa = 0.33).

Alternative forms of NIHSS
  • 11-item modified NIHSS (mNIHSS).
    Developed by deleting poorly reproducible or redundant items (level of consciousness, face weakness, ataxia, and dysarthria) and collapsing the sensory item from 3 into 2 responses (Lyden, Lu, Levine, Brott, & Broderick, 2001). The mNIHSS consists of ten items with excellent reliability and one item with adequate reliability (Meyer, Hemmen, Jackson, & Lyden, 2002). The total score for the mNIHSS is 31.
  • 5-item NIHSS (sNIHSS-5) and 8-item NIHSS (sNIHSS-8).
    For pre-hospital assessment of stroke severity, an 8-item and a 5-item NIHSS have undergone preliminary evaluation. The 8 items that were most predictive of “good outcome” three months after stroke were: right leg, left leg, gaze, visual fields, language, level of consciousness, facial palsy, and dysarthria. The sNIHSS-8 comprises all 8 of these items and the sNIHSS-5 contains only the first 5. In the validation models, receiver operator characteristic’s (ROC) for the sNIHSS-8 and sNIHSS-5 were adequate (ROC = 0.77 and 0.76, respectively). Furthermore, no significant difference between the sNIHSS-8 and the sNIHSS-5 was observed. The sNIHSS-5 retained much of the predictive performance of the full NIHSS (Tirschwell et al., 2002).
Client suitability

Can be used with:

  • Patients with stroke.

The NIHSS is designed so that virtually any stroke will register some abnormality on the scale.

Should not be used in:

  • The NIHSS can be administered to virtually any patient with stroke, however, a potential flaw with the NIHSS is that there may be a ceiling effect below the theoretical limit in patients with very severe stroke because many scale items cannot be tested in these patients (Muir, Weir, Murray, Povey, & Lees, 1996).
  • Can be estimated retrospectively from the admission neurological examination (Bushnell, Johnston, & Goldstein, 2001; Kasner et al., 1999; Williams, Yilmaz, & Lopez-Yunez, 2000), although actual testing is preferable.
In what languages is the measure available?

The NIHSS has been translated into the following languages: (http://www.proqolid.org/)

  • Cantonese for Hong-Kong
  • Estonian
  • Hindi
  • Hungarian
  • Italian
  • Marathi
  • Portuguese
  • Telugu

The NIHSS has been translated and validated in the following languages:

  • Chinese (Sun, Chiu, Yeh, & Chang, 2006)
  • German (Berger et al., 1999)
  • Spanish (Dominguez et al., 2006)