Table 2. Primary and Secondary End Points, by Cochran-Mantel-Haenszel Test Unless Otherwise Indicated.
| Variable | OPM Group (n = 53) | MSU Group (n = 63) | P Value | Difference (95% CI) |
|---|---|---|---|---|
| Primary End Point, No./Total No. (%) | ||||
| Triage accuracy for LVO and ICHa | 37/53 (69.8) | 63/63 (100) | <.001 | 30.2 (17.8 to 42.5) |
| Sensitivity | 6/17 (35.3) | 11/11 (100) | NA | NA |
| Specificity | 31/36 (86.1) | 52/52 (100) | NA | NA |
| Positive predictive value | 6/11 (54.5) | 11/11 (100) | NA | NA |
| Negative predictive value | 31/42 (73.8) | 52/52 (100) | NA | NA |
| Secondary End Points, No./Total No. (%) | ||||
| Triage accuracy for LVO aloneb | 41/53 (77.4) | 63/63 (100) | <.001 | 22.6 (13.2 to 35.9) |
| Sensitivity | 2/9 (22.2) | 3/3 (100) | NA | NA |
| Specificity | 39/44 (88.6) | 60/60 (100) | NA | NA |
| Positive predictive value | 2/7 (28.6) | 3/3 (100) | NA | NA |
| Negative predictive value | 39/46 (84.8) | 60/60 (100) | NA | NA |
| Secondary transfer rate | 7/17 (41.2) | 0/11 (0) | .02 | 41.2 (17.8 to 64.6) |
| Treatment rate | ||||
| Recombinant tissue-type plasminogen activator | 14/39 (35.9) | 16/32 (50.0) | .22 | 14.1 (−8.8 to 37.1) |
| Intra-arterial therapy | 6/9 (66.7) | 3/3 (100) | NA | NA |
| Stroke Management Time From Call to Various Metrics, Mean (SD), minc | ||||
| To first contact to MSU or hospital | 41.5 (12.8) | 10.3 (3.6) | <.001 | 31.2 (27.5 to 34.8) |
| To end of noncontrast imaging | 80.0 (39.6) | 39.3 (7.8) | <.001 | 40.7 (29.6 to 51.7) |
| To end of vascular imaging | 686.0 (1549.0) | 47.8 (8.8) | .009 | 638.2 (167.0 to 1109.0) |
| To imaging-based triage and therapy decisiond | 583.0 (1427.0) | 47.6 (9.0) | .009 | 535.4 (142.0 to 929.0) |
| To admission of LVO or ICH to comprehensive stroke center | 133.9 (86.2) | 85.5 (18.8) | .08 | 48.4 (−7.1 to 103.9) |
| To thrombolysis needle | 84.9 (30.2) | 50.1 (10.1) | <.001 | 34.8 (16.7 to 52.7) |
| To groin puncture | 183 (84) | 109, 114, 200e | .46 | 42.3 (84.3 to 168.9) |
| mRS Outcome at Day 90, Median (IQR)f | ||||
| All | 3 (1 to 5) | 1 (0 to 3) | .15 | NA |
| Patient with confirmed strokes | 3 (1 to 5) | 3 (1 to 4) | .12 | NA |
Abbreviations: ICH, intracranial hemorrhage; IQR, interquartile range; LVO, large-vessel occlusion; mRS, Modified Rankin Scale; MSU, mobile stroke unit; NA, not applicable; OPM, optimized prehospital management.
According to the presence (to be triaged to a comprehensive stroke center) or absence (to be triaged to a primary stroke center) of LVO or ICH. The correct triage decision to a comprehensive stroke center in 1 patient in the OPM group and to a primary stroke center in 1 patient in the MSU group was not realized (both because of patient choice).
By Fisher exact test.
By Wilcoxon rank sum test for independent groups.
Defined as the end of all required diagnostic procedures (neurologic examination and imaging, noncontrast imaging in the presence of ICH, and vascular imaging in the absence of ICH).
All 3 available values are displayed for this variable.
By logistic regression for ordinal end points (adjusted for baseline mRS score and study region).