Published in:
01-02-2018 | Correspondence
Correspondence to “Role of neutrophil to lymphocyte and monocyte to lymphocyte ratios in the diagnosis of bacterial infection in patients with fever”
Published in: Infection | Issue 1/2018
Login to get accessExcerpt
We read with interest the paper of Naess et al. [1]. The study provides more evidence to support the use of neutrophil to lymphocyte count ratio in the discrimination between different causes of infection and different causes of fever. However, we would like to address the following issues:
(1)
9 of 14 patients with viral infection had infectious mononucleosis, i.e., a viral infection typically associated with lymphocytosis. The authors note this limitation in their first study [2] stating that the infectious mononucleosis cases contributed to the higher lymphocyte count in the viral infection group. This may bias their results and limit their applicability in discriminating bacterial from other more common viral infections. It would also be useful if the authors could provide details on the viral diagnoses of the other five patients. Could the authors also clarify how a viral infection was defined for the purposes of their study?
(2)
It is well established that corticosteroids can increase neutrophils and decrease lymphocytes [3]. Therefore, the authors should report how many (if any) patients from each group were receiving corticosteroids at presentation.
(3)
The authors used logistic regression to adjust for other confounding variables. However, their study population includes a very small number of patients in the viral group (n = 14), therefore, resulting in a very small event per variable number (much less than the minimum recommended of 10) which could lead to overfitting [4], therefore, reducing the applicability of their result to external data.
(4)
The authors found that the neutrophil to lymphocyte count ratio was more useful compared to other parameters (WBC, neutrophil count and CRP) in diagnosing septicemia, an unclear term that has long been abandoned [5]. Therefore, it is important that the authors clarify how they define septicemia. Presumably, the term is used to refer to bacteremia, in which case the authors should clarify if patients in their comparison groups (pneumonia, pyelonephritis, lower UTI, other infection Table 3 of their manuscript) had blood cultures taken.