What stroke severity tool helps differentiate between large vessel occlusion and non-large vessel occlusion?

Stroke Severity Tools for Differentiating Large Vessel Occlusion from Non-Large Vessel Occlusion

The National Institutes of Health Stroke Scale (NIHSS) is the most validated and reliable stroke severity tool for differentiating between large vessel occlusion (LVO) and non-large vessel occlusion, with an optimal cutoff score of ≥10 balancing sensitivity (73%) and specificity (74%) in hospital settings. 1

Key Validated Stroke Severity Tools for LVO Detection

The 2018 American Heart Association/American Stroke Association guidelines identified four most frequently validated LVO prediction instruments:

1. National Institutes of Health Stroke Scale (NIHSS)

   • Most extensively validated tool
   • Hospital setting: Optimal cutoff ≥10 (73% sensitivity, 74% specificity)
   • Higher sensitivity threshold: ≥6 (87% sensitivity, 52% specificity)
   • Area under curve (AUC): 0.70-0.85

2. Cincinnati Prehospital Stroke Severity Scale (CPSSS)

   • Prehospital setting: Cutoff ≥2
   • Sensitivity: 56-65%, Specificity: 84-85%
   • AUC: 0.75-0.80

3. Los Angeles Motor Scale (LAMS)

   • Prehospital setting: Cutoff ≥4
   • Sensitivity: 47%, Specificity: 90-96%
   • AUC: 0.70-0.85

4. Rapid Arterial Occlusion Evaluation (RACE)

   • Prehospital setting: Cutoff ≥5
   • Sensitivity: 67%, Specificity: 85%
   • AUC: 0.83

Setting-Specific Recommendations

Hospital Emergency Department Setting

In hospital settings, the NIHSS is the optimal LVO prediction instrument because:

• It has the largest validation dataset
• It's already recommended for stroke severity assessment by AHA/ASA guidelines
• It's a quality performance measure for stroke care
• It has acceptable reliability when administered by physicians or nurses
• It offers flexibility in threshold selection based on clinical priorities 1

Prehospital Setting

For emergency medical services (EMS) in the field:

• CPSSS, LAMS, and RACE are all acceptable options
• No clear evidence for superiority of one scale over others
• The shortened NIHSS for EMS (sNIHSS-EMS) shows promise with LVO prediction at cutoff ≥6 (sensitivity 70%, specificity 81%) 2

Clinical Implementation Algorithm

1. Determine clinical setting:

   • Hospital ED with neurologist available → Use full NIHSS
   • Hospital ED without neurologist → Use NIHSS or simplified scales
   • Prehospital/EMS → Use CPSSS, LAMS, or RACE

2. Apply appropriate threshold based on priorities:

   • If prioritizing sensitivity (fewer missed LVOs):
     • NIHSS ≥6 (87% sensitivity, 52% specificity)
   • If balancing sensitivity/specificity:
     • NIHSS ≥10 (73% sensitivity, 74% specificity)
     • CPSSS ≥2 (56-65% sensitivity, 84-85% specificity)
     • RACE ≥5 (67% sensitivity, 85% specificity)

3. Interpret results with caution:

   • Positive test: LVO probability 35-50% (false positive rate 50-65%)
   • Negative test: LVO still possible in ~10% of cases

Important Caveats and Pitfalls

1. No perfect tool exists - No scale can determine LVO with both high sensitivity and specificity 1

2. False negatives occur - Even with optimal thresholds, some LVO cases will be missed. For example, 13% of patients with proximal occlusions had NIHSS scores of 0 in one study 3

3. Clinical scales cannot replace vessel imaging - While useful for triage, these scales should not be used as substitutes for CT angiography or MR angiography 4

4. Prevalence affects predictive value - The underlying prevalence of LVO significantly impacts the positive and negative predictive values of these tools 1

5. Limited prehospital validation - Most studies were conducted in hospital settings rather than by EMS in the field, limiting confidence in prehospital application 1

By understanding these tools' strengths and limitations, clinicians can make more informed decisions about patient triage and imaging, potentially improving outcomes for patients with large vessel occlusions who may benefit from endovascular therapy.