BACKROUND

In children, community-acquired pneumonia (CAP) has a high mortality and morbidity rate. The purpose of our study is to determine the diagnostic value of complete blood parameters for CAP and to look into their relationship to disease severity and mortality.

METHODS

A retrospective, the cross-sectional study enrolled children aged 0-18 years who were diagnosed with CAP in the pediatric intensive care unit (PICU) between January 1, 2023 and December 31. The patients included in the study were divided into two groups according to their clinical severity: the moderate group (MG) receiving non-invasive ventilation (NIV) and the severe group (SG) receiving invasive ventilation (IMV) support. In addition to demographic analyses, a comprehensive battery of laboratory tests was conducted on all included patients, and the relationships between mortality and clinical severity were compared. Furthermore, the study also involved the comparison of various biological markers such as mean platelet volume (MPV), C-reactive protein (CRP), albumin, lactate, CRP/MPV, neutrophile/lymphocyte ratio (NLR), lactate/albumin ratio (L/A), and white blood cell (WBC)/MPV values between the two groups, along with an evaluation of their clinical implications.ü RESULTS: A total of 120 patients were included in the study, the mean age was 44.13 ± 53.03 months, and the mean PICU stay was 11.52 ± 19.75 days. Clinical severity assessment showed that 54 (45%) patients received NIV and 66 (55%) required IMV. Mortality was significantly associated with the type of infectious agent (p = 0.022), respiratory support type (p = 0.019), length of stay (LOS) (p = 0.014), lactate (p = 0.037), and L/A ratio (p = 0.022). ROC analysis revealed that the L/A ratio had a superior predictive value over lactate alone in mortality prediction. Significant differences between MG and SG were observed for LOS (p = 0.006), CRP (p = 0.019), CRP/MPV (p = 0.017), and L/A ratio (p = 0.047). Both CRP/MPV and L/A ratios were significantly higher in the IMV group (p = 0.017, p = 0.047). In the linear regression model, CRP/MPV and L/A ratio were associated with mortality (R²=0.189), with CRP/MPV being 1.1 times more effective in determining respiratory support (p = 0.037, accuracy: 56.7%). CRP/MPV showed a moderate positive correlation with L/A (r = 0.435, p < 0.001), while a weak positive correlation was found between NLR and WBC/MPV (r = 0.264, p = 0.004). No significant correlation was found between LOS and CRP/MPV, NLR, L/A, or WBC/MPV (p > 0.05).

CONCLUSIONS

Mortality was significantly associated with infectious agent type, respiratory support type, LOS, lactate, and L/A ratio. CRP, CRP/MPV, and L/A ratio differed significantly between clinical severity groups. CRP/MPV may be useful in assessing mortality, clinical severity, and respiratory support type. However, L/A ratio was found to be a more valuable predictor of mortality than lactate alone.