Peripheral blood parameters for predicting PICU admission and mechanical ventilation in pediatric inpatients with human parainfluenza virus-induced pneumonia.

Abstract

Human parainfluenza viruses (hPIVs)-induced pneumonia is an important cause of pediatric hospitalization, and some develop severe pneumonias requiring pediatric intensive care unit (PICU) admission and mechanical ventilation (MV). The aim of this study is to investigate the value of peripheral blood (PB) parameters available on admission in predicting the need for PICU admission and MV due to pneumonia caused by hPIVs. A total of 331 cases including 277 (83.69%) on the general ward (GW) and 54 (16.31%) on the PICU were enrolled between January 2016 and June 2021. Of 54 patients admitted to the PICU, 24 patients (7.25%) received MV, whereas 30 (9.06%) did not. For both the PICU and GW groups, infants accounted for the highest proportion while school children had the lowest. Compared with the GW group, the PICU group had significantly higher rates of premature birth, fatigue, sore throat, headache, chest pain, tachypnea, dyspnea, and underlying diseases including congenital tracheal stenosis, congenital heart disease (CHD), metabolic disorder, and neurological disorder (ND), but significant lower proportion of exclusive breastfeeding and Z-scores for weight-for-height, weight-for-age, height-for-age, and body-mass-index (BMI)-for-age (BMIZ). Higher levels of some leukocyte differential counts (LDC)-related parameters including counts of neutrophil (N), ratios of neutrophil-to-lymphocyte ratio (NLR), derived neutrophils/(leukocytes minus neutrophils) ratio (dNLR), and platelet-to-lymphocyte ratio (PLR), lower levels of some other LDC-related parameters including lymphocyte (L) and monocyte (M) counts, ratios of lymphocyte-to-monocyte ratio (LMR), lymphocyte-to-C-reactive protein ratio, and prognostic nutritional index (PNI), and lower levels of PB protein (PBP)-related parameters including red blood cell (RBC), hemoglobin, total protein (TP), and serum albumin were observed in the PB of patients in the PICU compared with those in the GW. Notably, higher PLR level and two comorbidities including CHD and ND were identified as independent risk factors for PICU admission, while lower PNI level as well as smaller numbers of RBC and L as good predictors. Low levels of TP might be a useful predictor of the need for MV. Overall, the relative contributions of LDC- and PBP-related factors for accurate identification of patients required PICU admission accounted for 53.69% and 46.31%, respectively. Thus, determination of whether a patient with hPIVs-induced pneumonia is admitted to PICU involves consideration of both the LDC- and PBP-related parameters.