Abstract WP372: Screening For Large Vessel Occlusion - Comparing the Accuracies of the VAN Screening Tool and NIHSS

Abstract

Background: Our five-hospital Comprehensive Stroke Program utilizes a large vessel occlusion (LVO) triage that includes NIHSS ≥ 10 and last known well (LKW) < 6 hours. It was determined that a more effective triage for patients with LVO was needed. The vision, aphasia, and neglect (VAN) screening tool has been validated as an effective tool to assess for LVO. An NIHSS ≥ 6 has also been shown to have reasonable accuracy. Purpose: To compare the accuracies of the VAN screening tool and NIHSS ≥ 6 for identification of LVO in our emergency departments. Methods: A retrospective chart review of patients who arrived within 6 hours of LKW over a 4-month period was conducted. The NIHSS was completed by both the nurse (RN) and tele-stroke neurologist (physician) and was reviewed for components of the VAN screening tool to determine if patients were VAN positive or negative. The ability to identify LVO patients using the VAN screening tool was compared to NIHSS of ≥ 6. Results: Of the 111 patients that arrived within 6 hours of LKW, 40 (36%) were VAN positive, and 57 (51%) had NIHSS ≥ 6 based on RN assessment. Physician assessment identified 35 (32%) VAN positive, and 58 (52%) had NIHSS ≥ 6. A total of 18 patients were found to have LVO during the assessment period. Of these LVO patients, 15 had a documented RN assessment that identified 11 (73%) VAN positive and 13 (87%) had NIHSS ≥ 6. All 18 had physician assessments, with 17 (94%) VAN positive and 18 (100%) with NIHSS ≥ 6. NIHSS ≥ 6 had greater sensitivity than VAN (87% and 100% vs 73% and 94%) based on both RN and physician assessments, respectively. Whereas VAN had greater specificity (63.3% and 74% versus 43% and 41%). Based on RN assessment, VAN had a 92.6% negative predictive value (NPV) and a 28% positive predictive value (PPV) (vs 23% for NIHSS ≥ 6). Based on physician assessment, VAN had a 98% NPV and a 49% PPV. Discussion: VAN is a useful assessment tool for identification of LVO. However, it is associated with greater accuracy when used by the physician, rather than the RN. To optimize LVO screening, the RN may require focused training to improve assessment skills for both VAN and NIHSS.