NLR and NMLR Can Be Used to Differentiate Children With Tuberculosis Disease From Other Lower Respiratory Tract Infections

To the Editors: We read with interest the recent publication by Kissling et al1 on the monocyte, lymphocyte and neutrophil ratios—easy-to-use biomarkers for the diagnosis of pediatric tuberculosis. This study provides further evidence that neutrophil-to-lymphocyte-ratio (NLR) and neutrophil-to-monocyte-plus-lymphocyte-ratio (NMLR) are promising, easy-to-obtain diagnostic biomarkers to differentiate children with tuberculosis disease from other lower respiratory tract infections.1 We support and appreciate the authors’ work and agree with their conclusions, but have some concerns about some of the details in the article. Firstly, the authors noted that children with a chronic disease that potentially impacts the complete blood count values or known infection with HIV or children without complete blood count reported were excluded.1 Importantly, it has been shown that a variety of medications affect complete blood count levels, including anti-inflammatory and antiviral agents.2 How did the authors exclude the effect of medications being taken on patients’ baseline laboratory measurements? In addition, the study period for this study was from May 2017 to August 2021. During this period, how many patients were infected with COVID-19 and how did the authors exclude the effect of COVID-19 on complete blood counts [NLR, NMLR and monocyte-to-lymphocyte-ratio (MLR)]? Should additional sensitivity analyses be performed to validate the results of this study? Secondly, the study shows that the NLR varies with patient age and body mass index and, therefore, varies from person to person.3 Li et al3 investigated 3262 healthy individuals and found that NLR was positively correlated with age (P < 0.001). The oldest age group had the highest NLR, while the youngest age group had the lowest NLR. NLR also showed a slight positive correlation with systolic blood pressure, diastolic blood pressure and body mass index (P < 0.001).3 If the authors could provide the variables of age and body mass index, the results would have been more satisfactory. Because these 2 variables may be neglected confounders. Finally, the components of NLR, NMLR and MLR may be influenced by the delay between sample collection and analysis. If the authors could provide information on the time interval between sample collection and measurement of NLR, NMLR and MLR, the results would be more satisfactory, as the waiting time before analysis could influence this laboratory parameter.4 In addition, NLR, NMLR and MLR measured at a single point in time are not accurate, if the authors could monitor NLR, NMLR and MLR dynamically (our recommendation is to average the results of multiple measurements), the results would be more satisfactory. In conclusion, before these issues are clarified, this study’s findings should be interpreted cautiously. ACKNOWLEDGMENTS We thank the Department of Tuberculosis of the Fifth People’s Hospital of Suzhou (The Affiliated Infectious Disease Hospital of Soochow University) who contributed to this manuscript.