Association of white blood cell count with one-year mortality after cardiac arrest.

INTRODUCTION

Post-resuscitation care of cardiac arrest patients may be complicated by systemic inflammation elicited in response to whole-body ischaemia-reperfusion injury. We assessed the association between early WBC with one-year mortality in a large, unselected population of cardiac arrest patients.

METHODS

We collected a retrospective multicentre cohort of ICU-treated CA patients from the Finnish national ICU registry. We used locally estimated scatterplot smoothing (LOESS) curve to assess the association between the most abnormal WBC of the first 24 h in the ICU with the likelihood of death within a year. Multivariable logistic regression analyses were performed to assess the independent association between WBC and one-year mortality. In nested cohort analysis, we tested the association of delay from collapse to return of spontaneous circulation (ROSC) with WBC in linear regression models.

RESULTS

The LOESS curve was U-shaped, with the lowest predicted mortality at 7.5 10/L WBC. Based on this cut-off value, patients were divided into high (≥ 7.5 10/L) and low (< 7.5 10/L) WBC groups. In 4229 patients with high WBC, higher WBC was independently associated with increased one-year mortality (adjusted odds ratio (OR) 1.03 per 10/L, 95 % confidence interval (CI) 1.02-1.04, p < 0.001). In 776 patients with low WBC, lower WBC was independently associated with increased one-year mortality (adjusted OR 0.88 per 10/L, 95 % CI 0.80-0.96, p < 0.001). In a nested cohort analysis, longer ROSC-delay was associated with higher WBC in patients with a shockable rhythm (β = 0.10, R = 0.04, p < 0.001).

CONCLUSIONS

In this large retrospective cohort, WBC was independently associated with one-year mortality after CA. Mortality was lowest in patients with WBC close to the upper limit of the normal reference range. Although WBC is not useful for outcome prognostication in individual patients, our results support the concept of excess inflammation being a harmful component of the post-cardiac arrest syndrome.