Review of A Large Clinical Series: Is the Band Count Useful in the Diagnosis of Infection? An Accuracy Study in Critically Ill Patients

Abstract

Background: The presence of immature neutrophils (bands) in the circulating blood is often used as a clinical indicator of sepsis. Indeed, a band count greater than 10% is one of the American College of Chest Physicians/Society of Critical Care Medicine’s systemic inflammatory response syndrome (SIRS) criteria used to diagnose sepsis. However, the literature regarding the diagnostic accuracy of an elevated band count for the diagnosis of infection is limited. Aim: To determine the accuracy of a band count greater than 10% and increased or decreased white blood cell (WBC) count in diagnosing infection in a heterogeneous group of intensive care unit (ICU) patients. Methods: We prospectively recorded the WBC and band count on consecutive patients admitted to our ICU. Each patient was evaluated for the presence of infection according to defined criteria. The diagnostic accuracy of band count and total WBC count were determined by standard statistical methods. Results: Overall, 145 patients were enrolled, of whom 42 (29%) had a defined infection on admission to the ICU. On multiple logistic regression, the odds ratio for infection was 8.67 (95% CI 3.36-22.39; P < .001) for patients with band count greater than 10% and 1.6 (95% CI 0.78-3.29; P = .2) for a WBC count greater than 12 × 109/L. A band count greater than 10% had sensitivity of 43% (95% CI 28%-59%), specificity of 92% (95% CI 85%-97%), positive likelihood ratio of 5.52 (95% CI 2.6-11.7), and negative likelihood ratio of 0.62 (95% CI 0.47-0.81) for the diagnosis of infection. Conclusion: Band count may provide useful information in patients whose diagnosis of infection is uncertain. An elevated band count leads to a moderate increase in the likelihood of infection. A negative test, however, leads to only a small change in the posttest probability of infection.