Department of Family Medicine
National Institutes of Health Stroke Survey
T. Brott, H.P. Adams Jr., C/P. Olinger, J.R. Marler, W.G. Barsan, J. Biller, J. Spilker, R. Holleran, R. Eberle, V. Hertzberg, M. Rorick, C.J. Moomaw, M. Walker (1989)
Background and Development:
The NIHSS is a 15-item neurologic examination stroke scale used to evaluate the effect of acute cerebral infarction on the levels of consciousness, language, neglect, visual-field loss, extraocular movement, motor strength, ataxia, dysarthria, and sensory loss. A trained observer rates the patent’s ability to answer questions and perform activities. Ratings for each item are scored with 3 to 5 grades with 0 as normal, and there is an allowance for untestable items. The examination requires less than 10 minutes to complete. Training is minimal and is available through instructional videos: a 45-minute training program tape, and 2 certification tapes (Lyden et al. 1994). Reliability was found to be excellent overall and moderate to excellent for most individual scales (Brott et al. 1989, Goldstein 1989, Lyden et al. 1994). The survey can also reliably be used to assess patients through a remote television link (Shafqat et al. 1999), although it requires slightly more time to complete. The NIHSS has been modified from its initial configuration by the elimination of some aspects which showed poor reliability, and the expansion of other aspects (Lyden et al. 1999). From a factor analysis they conclude that there are four factors underlying the NHISS representing combinations of left and right brain function, and cortical and motor function.
The evaluation of stroke severity depends upon the ability of the observer to accurately and consistently assess the patient. While the overall reliability of the NIHSS has been shown by several studies to be excellent, there is some disagreement as to which individual items have poor to fair reliability. The original authors evaluated the scale reliability using kappa statistic and found that, while most items had good to excellent reliability (Cronbach alpha> 0.5), two items, dysarthria and consciousness, rated fair to poor (Brott et al. 1989). Goldstein et al. (1989), in a separate analysis, found that of the 15 items making up the NIHSS, 13 showed no statistical difference between the observers. The observers had poor agreement on score in determining facial palsy and limb ataxia (alpha< 0.3). Lyden et al. (1994) reported similar findings. Reliability of observer assessment increases after viewing instructional video tape recordings made by the authors (Lyden et al. 1994). Shafqat et al. (1999) report strong correlations for total NIHSS scores between bedside assessments and remote television assessments, although the remote technique required slightly more time (9.7 vs. 6.6 minutes). Generally, two of the subscores, commands and ataxia, showed poor reliability.
Comparison of the NIHSS score with infarction volume as measured by CAT scan 1 week after event was used as a measure showing a high level of validity for the scale (r=0.68) (Brott et al. 1989). Comparison of the score with 3-month clinical outcome also shows high validity (r=0.79). Correlation coefficients between the NIHSS and the Barthel Index, the Rankin Scale, and the Glasgo Outcome Scale were significant, but modest in magnitude both at baseline and 2 hours after stroke (Lyden et al. 1999). The predictive validity of the NIHSS three months after stroke is also high.
Validation in Elderly Populations:
The nature of this survey tool makes questions regarding validation in elderly populations invalid.
Validation in Minority Populations:
The nature of this survey tool makes questions regarding validation in minority populations invalid.
The NIHSS has been used in studies to evaluate the effectiveness of clinical therapy on stroke recovery. Thrombolytic therapy for acute ischemic stroke has been shown to be an effective treatment (NINDS, 1995). The authors used the scale at baseline, 24 hours after the onset of stroke, and after three months, to determine whether the administration of t-PA within 3 hours of stroke onset had a positive clinical outcome (as indicated by an improvement of 4 points over baseline). The administration of tissue plasminogen after stroke also can improve performance on the NIHSS (Brott et al. 1992, Haley et al. 1992), as does administration of high-dose naloxone (Olinger et al. 1990).
Design Strengths and Weaknesses:
The NIHSS is a reliable tool for rapidly assessing the effects of stroke (Lyden et al. 1991). Minimal amounts of training are required to reliably administer this tool, even by non-neurologists, which takes less than 10 minutes to complete. In comparison with other tools and an objective measurement of infarction volume the tool has been shown to record a highly valid assessment of the stroke severity and later effects. However, Woo et al. (1999) indicate that for a given NIHSS score, the CAT scan-measured infarction volume is larger for right hemisphere strokes than for left hemisphere strokes. Therefore, care must be taken when predicting infarction size from NIHSS score.
This measurement tool is available from The Brain Attack Coalition:
Brott, T., H.P. Adams Jr., C.P. Olinger, J.R. Marler, W.G. Barsan, J. Biller, J. Spilker, R. Holleran, R. Eberle, V. Hertzberg, M. Rorick, C.J. Moomaw, and M. Walker (1989) Measurements of acute cerebral infarction: a clinical examination scale. Stroke, 20, 964-970.
Brott, T.G., E.C. Haley, Jr., D.E. Levy, W. Barsan, J. Broderick, G.L. Sheppard, J. Spilker, G.L. Dongable, S. Massey, and R. Reed (1992) Urgent therapy for stroke, I: Pilot study of tissue plasminogen activator administered within 90 minutes. Stroke 23, 632-640.
Goldstein, L., C. Bertels, J. Davis (1989) Interrater reliability of the NIH Stroke Scale. Arch. Neurol. 46, 660-662.
Haley, E.C., Jr., D.E. Levy, T.G. Brott, G.L. Sheppard, M.C. Wong, G.L. Kongable, J.C. Torner, and J.R. Marler (1992) Urgent therapy for stroke. II: Pilot study of tissue plasminogen activator administered 91-180 minutes from onset. Stroke 23, 641-645.
Lai. S.-M., P.W. Duncan, and J. Keighley (1998) Prediction of functional outcome after stroke. Comparison of the Orpington Prognostic Scale and the NIH Stroke Scale. Stroke 29, 1838-1842
Lyden, P. and G.T. Lau (1991) A critical appraisal of stroke evaluation and rating scales. Stroke 22, 1345-1352.
Lyden, P., T. Brott, B. Tilley, K.M. Welch, E.J. Mascha, S. Levine, E.C. Haley, J. Grotta, and J. Marler (1994) Improved reliability of the NIH Stroke Scale using video training. Stroke 25, 2220-2226.
Lyden, P., Mei Lu, C. Jackson, J. Marler, Rashmi Kothari, Thomas Brott, and J. Zivin (1999) Underlying structure of the National Institutes of Health Stroke Scale: Results of a factor analysis. Stroke 30, 2347-2354.
National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group (1995) Tissue Plasminogen activator for acute ischemic stroke. N. Eng. J. Med. 333, 1581-1587.
Olinger, C.P., H.P. Adams, T.G. Brott, J. Biller, W.G. Barsan, G.J. Toffol, R.W. Eberle, and J.R. Marler (1990) High-dose intravenous naloxone for the treatment of acute ischemic stroke. Stroke 21, 721-725.
Shafqat, S., J.C. Kvedar, M.M. Guanci, Y. Chang, and L.H. Schwamm (1999) Role for telemedicine in acute stroke. Feasibility and reliability of remote administration of the NIH Stroke Scale. Stroke 30, 2141-2145.
Woo, D., J.P. Broderick, R.U. Kothari, M. Lu, T. Brott, P.D. Lyden, J.R. Marler, J.C. Grotta (1999) Does the National Institutes of Health Stroke Scale favor left hemisphere strokes? Stroke 30, 2355-2359.