White Blood Cell Count Improves Prediction of Delayed Cerebral Ischemia Following Aneurysmal Subarachnoid Hemorrhage

2019 
BACKGROUND: Immune dysregulation has long been implicated in the development of delayed cerebral ischemia (DCI) following aneurysmal subarachnoid hemorrhage (aSAH). OBJECTIVE: To determine the relationship of inflammatory cell biomarkers with DCI. METHODS: We evaluated 849 aSAH patients who were enrolled into a prospective observational cohort study and had a white blood cell (WBC) differential obtained within 72 h of bleed onset. RESULTS: WBC count > 12.1 × 109/L (odds ratio 4.6; 95% confidence interval [CI]: 1.9-11, P < 0.001) was the strongest Complete Blood Count (CBC) predictor of DCI after controlling for clinical grade (P < .001), thickness of SAH blood on admission computed tomography (P = .002), and clipping aneurysm repair (P < .001). A significant interaction between clinical grade and WBC count (odds ratio 0.8, 95% CI: 0.6-1.0, P = .02) revealed that good-grade patients with elevated WBC counts (49%: 273/558) had increased odds for DCI indistinguishable from poor-grade patients. Multivariable Cox regression also showed that elevated WBC counts in good-grade patients increased the hazard for DCI to that of poor-grade patients (hazard ratio 2.1, 95% CI 1.3-3.2, P < .001). Receiver operating characteristic curve analysis of good-grade patients revealed that WBC count (area under the curve [AUC]: 0.63) is a stronger DCI predictor than the modified Fisher score (AUC: 0.57) and significantly improves multivariable DCI prediction models (Z = 2.0, P = .02, AUC: 0.73; PPV: 34%; NPV: 92%). CONCLUSION: Good-grade patients with early elevations in WBC count have a similar risk and hazard for DCI as poor-grade patients. Good-grade patients without elevated WBC may be candidates to be safely downgraded from the intensive care unit, leading to cost savings for both patient families and hospitals.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    51
    References
    31
    Citations
    NaN
    KQI
    []