Canadian Neurological Scale (CNS)

Overview


For the purposes of this review, we conducted a literature search to identify all relevant publications on the psychometric properties of the CNS.

Floor/Ceiling Effects

Floor/ceiling effects have not yet been examined.

Reliability

Internal Consistency:
Cote et al. (1986) examined the internal consistency of the CNS in 34 patients with acute stroke. Four raters (one neurologist, one resident in neurology and two nurses) evaluated the patients. Internal consistency, as calculated using Cronbach’s alpha, was excellent for all domains (leg weakness 0.896; facial weakness 0.934; distal arm weakness 0.969; orientation 0.979; proximal arm weakness 0.98; and speech 1.00). No differences between professionals were found.

Intra-rater:
Intra-rater reliability has not been examined.

Inter-rater:
Cote et al. (1986) examined the inter-rater reliability of the CNS in 34 patients with acute stroke. Four raters (one neurologist, one resident in neurology and two nurses) evaluated the patients within two to four hours of each other using the CNS. Inter-rater reliability, as calculated using kappa statistics, was adequate to excellent for all domains (leg weakness 0.722-0.842; facial weakness 0.535-1.00; distal arm weakness 0.758-0.974; orientation 0.744-1.00; proximal arm weakness 0.788-1.00; and speech 0.934-1.00).

Brushnell, Johnston and Goldstein (2001) looked at the retrospective inter-rater reliability scoring of both the CNS and the National Institute of Health Stroke Scale (NIHSS). They compared data from academic medical centers to data from community hospitals with neurologists and community hospitals without neurologists. Inter-rater reliability for the CNS, as calculated using Intraclass correlation coefficient (ICC), was found to be excellent for all charts reviewed (academic medical center ICC=0.97; community hospital with neurologists 0.88; community hospital without neurologists 0.78). The inter-rater reliability for the NIHSS was excellent for the charts reviewed from the academic medical centre and the community hospital with a neurologist (ICC=0.93; 0.89 respectively), however only adequate agreement was found for charts reviewed from the community hospital without a neurologist (ICC=0.48). More data was missing for the NIHSS in comparison to that missing for the CNS likely due to the fact that the NIHSS requires a more detailed neurological examination. These results suggest that scoring the CNS retrospectively is reliable regardless of whether the medical record contains evaluation material from a neurologist.

Test-retest:
Test-retest reliability has not been reviewed.

Validity

Content:
Content validity has not been reviewed.

Criterion:
Concurrent:
Cote et al. (1989) evaluated the concurrent validity of the CNS in the original validation study involving 157 patients with acute stroke. Patients were evaluated by staff neurologists or neurology residents upon admission to the hospital and were classified as either having no, mild, moderate or severe deficit resulting from acute stroke. Nurses then evaluated the patients using the Glascow Coma Scale (GCS) and the CNS. An average interval of 3.71 hours occurred between assessments. Concurrent validity was evaluated by correlating CNS item with the appropriate components of the neurological examination, and the total score of the CNS with the global assessment on the neurological examination (no, mild, moderate or severe). Concurrent validity, as measured by Spearman rank correlation, was found to be excellent between the global neurological examination and the total CNS score (0.775). The concurrent validity between the neurological components and CNS was found to be excellent for orientation (0.716), speech (0.691) and weakness (0.767); and adequate for level of consciousness (0.574).

Predictive:
Cote et al. (1989) evaluated the predictive validity of the CNS in 157 patients with acute stroke. Three outcomes were evaluated: 1) death at 6 months; 2) any vascular event within 6 months (for example, MI, CVA or vascular death); and 3) independence in ADL at 5 months or beyond. Initial CNS scores were found to significantly predict death within 6 months, morbidity, and recovery of ADL within 5-months. For patients with scores of ≥ 11, only 2.1% had died at 6 months, 2.1% experienced another vascular event, approximately 90% were independent in ADLS at 5 months or beyond; compared to those that scored <9 initially where 13.2% had died at 6 months, 20.6% experienced another vascular event and <50% were independent in ADLS at 5 months or beyond.

Muir et al. (1996) compared the CNS, National Institute of Health Stroke Scale (NIHSS) and the Middle Cerebral Artery Stroke Score (MCANS) to see which scale best predicted good (alive at home) or poor (alive and requiring in care or dead) outcome at 3-months in 373 patients with acute stroke. Predictive accuracy of the variables was compared by ROC curves and stepwise logistic regression. Logistic regression showed that the NIHSS added significantly to the predictive value of all other scores. The overall accuracy for the CNS, NIHSS and MCANS as stand alone measures was adequate (0.79, 0.79 and 0.83 respectively).

Sensitivity/Specificity:
The sensitivity/specificity has not been examined.

Construct :
Convergent/Discriminant:
Cote et al. (1989) evaluated the discriminant validity of the CNS and the Glascow Coma Scale (GCS) by comparing results with a standard neurological examination. Results from the GCS and the CNS evaluation of 157 patients with acute stroke were compared with a standard neurological examination. Excellent correlation was found between the total CNS score and the standard neurological examination (r2=0.769), however only adequate correlation was found between the GCS and the standard neurological examination (r2=0.563). These results suggest that the CNS may better discriminate neurological deficit.

Known Groups:
The known groups validity has not been examined.

Responsiveness

Cote et al. (1989) evaluated the responsiveness of the CNS in 79 patients with acute stroke. The CNS was administered on admission and throughout the first 48 hours. Patients were classified as either 1) remaining stable over first 48 hours or 2) status changed over first 48 hours. A change in score  ≥ 1 yielded the highest negative predictive value (0.969), with a sensitivity of 0.933 and specificity of 0.508). The results of this study suggest that the CNS can be used to monitor clinically significant differences in neurological status.

Hagen, Bugge, and Alexander (2003) examined the responsiveness of the CNS and other commonly used outcome measures in 136 patients in the early post-stroke period. The outcomes measures were administered at 1, 3 and 6 months after stroke onset. The sensitivity of the CNS to detect change from 1 to 3 months and 3 to 6 months, as calculated using Standardized Response Mean, was small (SRM=0.2860 and 0.2849 respectively). These results suggest that the CNS has some ability to detect change in patients with stroke in the subacute phase of recovery.

References
  • Bushnell, C.D., Johnston, D.C.C. & Goldstein, L. B. (2001). Retrospective assessment of the initial stroke severity: Comparison of the NIH Stroke Scale and the Canadian Neurological Scale. Stroke, 32, 656-660.
  • Cote, R., Battista, R.N., Wolfson, C., Boucher, J., Adam, J., Hachinski, V. (1989). The Canadian Neurological Scale: Validation and reliability assessment. Neurology, 39, 638-643.
  • Cote, R., Hachinski, V., Shurvell, B., Norris, J. & Wolfson, C. (1986). The Canadian Neurological Scale: A preliminary study in acute stroke. Stroke, 17(4), 731-737.
  • Cuspineda, E., Machado, C., Aubert, E., Galan, L, Liopis, F, Avila, Y. (2003). Predicting outcome in acute stroke: A comparison between QEEG and the Canadian Neurological Scale. Clinical Electroencephalography, 34(1), 1-4.
  • Muir, K.W., Weir, C.J., Murray, G.D., Povey, C., Lees, K.R. (1996). Comparison of neurological scales and scoring systems for acute stroke prognosis. Stroke, 27, 1817-1820.
  • Nilanont, Y., Komoitri, C., Saposnik, G., Cote, R., Di Legge, S., Jin, Y. et al. (2010). The Canadian Neurological Scale and the NIHSS: Development and validation of a simple conversion model. Cerebrovascular Disease, 30(2), 120-126.
  • Shinar, D., Gross, C.R., Mohr, J.P., Caplan, L.R., Price, T.R., Wolf, P.A. et al. (1985). Interobserver variability in the assessment of neurologic history and examination in the stroke data bank. Archives of Neurology, 42, 557-565.