Admission neutrophil-to-lymphocyte ratio is associated with earlier observed peak midline shift after large anterior-circulation infarction: a retrospective neurocritical care cohort study

Large anterior-circulation infarction can produce space-occupying edema, but current risk tools mainly address whether severe edema will occur rather than when maximum mass effect will be observed. We examined whether admission neutrophil-to-lymphocyte ratio (NLR) was associated with the timing of maximum observed midline shift (MLS). We performed a retrospective secondary analysis of de-identified registry data from consecutive neurocritical care patients with large anterior-circulation infarction. Serial imaging followed routine care, with noncontrast head computed tomography (CT) generally repeated at approximately 24-hour intervals early after admission and additional CT and/or magnetic resonance imaging obtained as clinically indicated. The primary outcome was the interval from last-known-well/onset-time approximation to the scan showing the maximum observed MLS. Because time was right-skewed, ln(hours + 1) was modeled by multivariable linear regression and back-transformed as the percentage difference per NLR doubling. Heteroskedasticity-consistent type 3 (HC3) robust standard errors were used, with adjustment for age, sex, infarct volume, admission National Institutes of Health Stroke Scale score, endovascular therapy, and intravenous thrombolysis. The primary cohort included 115 patients. Each doubling of admission NLR was associated with a 21.8% shorter interval to the maximum observed MLS (95% confidence interval, 8.2%-33.3% shorter; P = 0.003). The median interval was 66.4 h in the highest NLR tertile and 124.8 h in the lowest tertile ( P = 0.007). For an early observed peak (≤ 72 h), NLR alone had an area under the receiver operating characteristic curve of 0.715; the cohort-derived Youden cutoff of 7.96 had 77.3% sensitivity and 59.2% specificity. Associations with maximum MLS and severe MLS did not reach the conventional P < 0.05 threshold. Higher admission NLR was independently associated with an earlier maximum observed MLS, suggesting that systemic inflammatory status may carry information about edema tempo. Because imaging was clinically scheduled and the cutoff was internally derived, the findings are hypothesis-generating. NLR should not be used alone to determine imaging frequency or surgical decisions; prospective external validation with standardized imaging is required.