Background: Limited resources warrant investigating models for predicting which stroke tissue-type plasminogen activator (tPA) patients benefit from admission to neurologic intensive care unit (neuroICU).
Methods: This model classifies patients who on day 1 of their ICU admission are predicted to receive one or more of 30 subsequent active life supporting treatments. Two groups of patients were compared: low risk monitor (LRM) (patients who did not receive active treatment (AT) on the first day and whose risk of ever receiving active treatment was ≤ 10%) and AT (patients who received at least one treatment on any day of their ICU admission).
Results: Compared to LRM group (21 patients), AT group (59 patients) had similar age (75 ± 13 vs. 72 ± 17, P = 0.4), similar gender (male: 56% vs. 52%, P = 0.8), similar National Institutes of Health stroke scale (NIHSS, 16 ± 9 vs. 14 ± 8, P = 0.4), and higher Acute Physiologic and Chronic Health Evaluation (APACHE) III scores (62 ± 26 vs. 41 ± 15, P = 0.0008). Compared to LRM group, AT group had longer ICU length of stay (4.5 ± 4.4 vs. 2.5 ± 1.3, P = 0.04), higher ICU mortality (22% vs. 4.7% (one patient DNR/hospice); OR: 5.6; 95% CI: 0.7 - 46.0; P = 0.1), and higher hospital mortality (36% vs. 4.7%; OR: 11; 95% CI: 1.4 - 88.0; P = 0.02).
Conclusion: The outcome of LRM patients with stroke post-tPA suggests that they may not require admission to a formal neuroICU, improving resource use and reducing costs.
Keywords: Ischemic stroke; Neurologic intensive care unit; Stroke; Utilization review; tPA.