Article Figures & Data
Figures
- FIGURE.
NIHSS and LVO distribution in the tested cohort. A, Relative frequency of NIHSS scores as stratified by the presence-versus-absence of large-artery occlusion. B, Distribution of NIHSS scores according to large-artery occlusion status. There was no difference in the distribution of NIHSS scores (bins) between subjects with absent-to-mild versus moderate-to-severe leukoaraiosis (P = .319, χ2 test).
Tables
- Table 1:
Baseline characteristics (unadjusted) of the studied patient population as stratified by the presence-versus-absence of proximal LVOa
Characteristics All Patients (n = 274) LVO (n = 46) No LVO (n = 228) P Value Age (yr) 69 (60–80) 68 (58–77) 70 (60–81) .347 Female sex 126 (46%) 24 (52%) 102 (45%) .418 Admission NIHSS score 5 (2–10) 14 (4–17) 4 (2–9) <.001 Prestroke mRS 0 (0–1) 0 (0–1) 0 (0–1) .774 LVO stroke scale scores 3I/SS 1 (0–2) 2 (1–3) 1 (0–2) <.001 VAN 0 (1–2) 3 (0–4) 0 (0–2) <.001 CPSSS 0 (0–1) 1 (0–3) 0 (0–1) .001 FAST-ED 2 (0–4) 4 (2–6) 1 (0–3) <.001 RACE 2 (0–4) 5 (1–7) 1 (0–3) <.001 Laboratory data Admission glucose level (mg/dL) 119 (100–142) 120 (105–143) 119 (99–142) .515 Admission creatinine level (mg/dL) 0.95 (0.78–1.20) 0.90 (0.73–1.0) 0.97 (0.79–1.20) .090 HbA1c level (%) (n = 161) 5.9 (5.7–6.5) 6.0 (5.6–6.3) 5.9 (5.7–6.6) .794 LDLc level (mg/dL) (n = 171) 93 (72–123) 89 (65–111) 94 (74–123) .274 Neuroimaging Fazekas scale score (0–2) 129 (48%) 30 (65%) 99 (43%) .009 Pre-existing risk factors Hypertension 210 (77%) 34 (74%) 176 (77%) .703 Hyperlipidemia 152 (56%) 26 (57%) 132 (58%) .076 Diabetes 75 (27%) 11 (24%) 64 (28%) .594 Prior stroke or TIA 59 (22%) 6 (13%) 53 (23%) .168 Atrial fibrillation 42 (15%) 8 (17%) 34 (15%) .823 Coronary artery disease 58 (21%) 4 (9%) 54 (24%) .028 Congestive heart failure 20 (7%) 1 (2%) 19 (8%) .215 Peripheral arterial disease 18 (7%) 1 (2%) 17 (8%) .221 Preadmission medications Statin 124 (45%) 19 (41%) 105 (46%) .689 Antihypertensive 185 (67%) 32 (70%) 153 (67%) .863 Antiglycemic 53 (19%) 9 (20%) 44 (19%) 1.000 Antiplatelets 145 (53%) 24 (52%) 121 (53%) 1.000 Oral anticoagulant 19 (7%) 4 (9%) 15 (7%) .749 Note:—HbA1c indicates glycated hemoglobin A1c; LDLc, low-density lipoprotein cholesterol.
↵a Proximal LVO is defined as an occlusion of the internal carotid artery or middle cerebral artery M1 segment. Data are No. (%) or median (25th–75th percentile).
- Table 2:
Diagnostic accuracy of prehospital stroke scales in predicting LVO stratified by leukoaraiosis severitya
Scale Sensitivity (%) Specificity (%) PPV (%) NPV (%) κ Total cohort (n = 274) 3I/SS ≥4 22 (11–36) 90 (86–94) 31 (19–47) 85 (83–87) 0.14 (0–0.28) VAN ≥2 70 (54–82) 63 (57–69) 28 (23–33) 91 (87–94) 0.20 (0.10–0.30) CPSSS ≥2 41 (27–57) 79 (74–84) 29 (19–47) 87 (84–90) 0.18 (0.05–0.31) RACE ≥5 50 (35–65) 85 (80–89) 40 (31–51) 89 (86–92) 0.32 (0.18–0.46) FAST-ED ≥4 57 (41–71) 79 (73–84) 35 (27–43) 90 (83–87) 0.28 (0.15–0.41) Absent-to-mild leukoaraiosis (Fazekas scale score, 0–2; n = 129) 3I/SS ≥4 50 (21–79) 88 (80–94) 33 (18–52) 94 (89–96) 0.09 (0–0.27) VAN ≥2 73 (54–88) 68 (58–77) 41 (32–50) 89 (82–94) 0.32 (0.16–0.48) CPSSS ≥2 37 (20–56) 87 (79–93) 46 (30–63) 82 (77–86) 0.25 (0.06–0.45) RACE ≥5 53 (34–72) 91 (83–96) 64 (46–78) 87 (81–90) 0.47 (0.29–0.65) FAST-ED ≥4 57 (37–75) 84 (75–90) 52 (38–65) 86 (81–91) 0.39 (0.09–0.57) Moderate-to-severe leukoaraiosis (Fazekas scale score, 3–6; n = 145) 3I/SS ≥4 25 (7–52) 92 (86–96) 29 (12–53) 91 (88–93) 0.18 (0–0.41) VAN ≥2 63 (35–85) 60 (51–68) 16 (11–23) 93 (87–96) 0.10 (0–0.22) CPSSS ≥2 50 (25–75) 74 (65–81) 19 (12–29) 92 (88–95) 0.14 (0–0.29) RACE ≥5 44 (20–70) 81 (73–87) 22 (13–35) 92 (88–94) 0.17 (0–0.35) FAST-ED ≥4 56 (30–80) 74 (66–82) 21 (14–32) 93 (89–96) 0.11 (0–0.27) Note:—NPV indicates negative predictive value; PPV, positive predictive value.
↵a Data in parentheses are 95% CI.
- Table 3:
Multivariable logistic regression analysis for the ability of previously defined clinical stroke scale cutoffs to predict large-vessel occlusion in the examined cohort
Stroke Scale Crude OR (95% CI) P Value Adjusted OR (95% CI)a P Value 3I/SS ≥4 4.736 (1.513–14.826) .008 5.389 (0.856–33.908) .073 VAN ≥2 3.918 (1.979–7.760) <.001 2.468 (0.845–7.206) .098 CPSSS ≥2 2.710 (1.388–5.290) .003 2.794 (0.973–8.028) .056 RACE ≥5 5.706 (2.881–11.299) <.001 3.236 (1.099–9.527) .033 FAST-ED ≥4 4.749 (2.447–9.217) <.001 3.740 (1.291–10.839) .015 ↵a Each model was adjusted for leukoaraiosis burden (as determined by the Fazekas scale score) as well as Leukoaraiosis × Stroke Scale Interaction. When additionally entered into the model, the prestroke modified Rankin Scale, age, sex, hypertension, history of stroke/transient ischemic attack, and atrial fibrillation were not retained in the final step of the analysis. In all analyses, leukoaraiosis and stroke scales were entered as dichotomized variables. When all analyses were repeated by entering the Fazekas scale score and stroke scale scores as ordinal variables, the VAN (P = .029), CPSSS (P = .024), FAST-ED (P = .003), RACE (P = .002), and 3I/SS (P = .006) scores predicted LVO independent of the leukoaraiosis burden (not shown). There was no Leukoaraiosis × Stroke Scale Interaction in any of the examined models (P > .05, each).
