The relationship of neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio with bone mineral density in adolescent girls suffering from anorexia nervosa

Abstract

Introduction: Inflammation is supposed to be one of the factors contributing to decreased bone mineral density (BMD) in anorexia nervosa (AN). The neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) are simple and cost-effective inflammatory markers, well documented as indicators of postmenopausal osteoporosis. This study aimed to assess the relationships between these ratios and BMD in girls with AN. Material and methods: The electronic records of 73 girls hospitalized for AN were analysed retrospectively. The age range of the study group was 12.56–17.67 years. BMD was assessed by dual-energy X-ray absorptiometry (DXA) and expressed as Z-scores according to lumbar spine (s-BMD) and total body (TB-BMD) sites. NLR and PLR were calculated according to complete blood count results. Patients were divided into 2 subgroups with parallel analyses — according to the TB-BMD criterion and the s-BMD criterion: normal (Z-score > –2.0, n = 63) and low s-BMD subgroup (Z-score ≤ –2.0, n = 10), and normal (Z-score > –2.0, n = 45) and low TB-BMD subgroup (Z-score ≤ –2.0, n = 28). Results: In the low s-BMD subgroup a tendency to an increase of mean NLR, PLR, and WBC values was observed. Respective BMD Z-score values correlated significantly and negatively with PLR in the low s-BMD (R = –0.892, p < 0.001) and normal TB-BMD (R = –0.451, p = 0.002) subgroups, while with NLR only in the normal TB-BMD subgroup (R = –0.685, p < 0.001). In the low s-BMD subgroup the PLR was shown to be a significant and independent predictor of s-BMD (β = –0.881, p < 0.001). The PLR contributed to 77.6% of the s-BMD Z-score variability (R 2 = 0.776, p < 0.001). In the normal TB-BMD subgroup, the PLR and NLR levels were significant and independent predictors of TB-BMD (β = –0.352, p = 0.004; β = –0.450, p = 0.001; β = –0.339, p = 0.005, respectively) and explained 44.4% of TB-BMD Z-score variability (R 2 = 0.444, p < 0.001). Conclusions: These results indicate that there might be a relationship between bone mass loss and inflammation expressed as NLR and PLR in adolescent girls suffering from AN. These connections seem to be dependent on the examined skeletal area. NLR and PLR, which are common indicators of morbidity and mortality in many malignancies and inflammatory chronic diseases, can also be useful in the evaluation of bone condition in adolescent females with AN. However, there is a need for further investigation in this field.