Neutrophil-to-lymphocyte ratio in children and adolescents presenting to a paediatric emergency department with non-suicidal self-injury: a retrospective case-control study with healthy comparison participants

Non-suicidal self-injury (NSSI) is a common and clinically serious behaviour in children and adolescents and is among the strongest predictors of later suicide attempt. Complete blood count (CBC) derived indices, namely the neutrophil-to-lymphocyte ratio (NLR), the platelet-to-lymphocyte ratio (PLR) and the monocyte-to-lymphocyte ratio (MLR), are inexpensive surrogates of systemic inflammation that require no laboratory testing beyond a routinely obtained CBC. Data from paediatric emergency settings, and from Türkiye, are lacking. Retrospective case-control study of 89 participants aged 9 to 18 years: 36 patients with a DSM-5 diagnosis of NSSI who presented to a paediatric emergency department and 53 healthy children attending routine outpatient wellness visits at the same hospital. All indices were recomputed from source counts and compared using the Mann-Whitney U test with rank-biserial effect sizes. The primary multivariable analysis was binary logistic regression with NSSI status as the outcome and NLR, age and sex as covariates, with Firth penalised estimation as a sensitivity analysis. Discrimination was assessed by receiver operating characteristic (ROC) analysis with DeLong confidence intervals, bootstrap internal validation and repeated cross-validation. NLR was higher in the NSSI group (median 2.03, interquartile range 1.51 to 3.33) than in controls (1.42, 1.17 to 1.74; p < 0.001, rank-biserial correlation 0.47), whereas PLR and MLR did not differ significantly. After adjustment for age and sex, each one-unit increase in NLR was associated with higher odds of NSSI status (adjusted odds ratio 2.24, 95% CI 1.29 to 3.91, p = 0.005); female sex was the strongest correlate (adjusted odds ratio 8.58, 95% CI 2.79 to 26.36). NLR showed moderate discrimination (area under the curve 0.736, 95% CI 0.630 to 0.843) that was not superior to the absolute neutrophil count alone (0.724; DeLong p = 0.80). The Youden-optimal threshold was 1.72, but its bootstrap 95% percentile interval was wide (1.22 to 2.19) and cross-validated discrimination fell to 0.717. The association persisted after excluding patients with major depressive disorder, patients with any psychiatric diagnosis and patients who presented after self-poisoning, but was attenuated and no longer significant among unmedicated patients. Presentation to a paediatric emergency department with NSSI was associated with a higher NLR relative to healthy peers, but discrimination was moderate, unstable on internal validation and no better than the neutrophil count alone. Because cases were sampled during an acute emergency presentation and carried substantial psychiatric morbidity, these findings are hypothesis-generating and do not establish NLR as an NSSI-specific or clinically deployable marker. Studies incorporating psychiatric and emergency comparison groups are required.