This single-center cross-sectional study was conducted in the Pediatric Infectious Disease Ward and COVID-19 pademic ward at the University of Health Sciences Dr Behcet Uz Children’s Hospital with a 360-bed tertiary care hospital in Izmir, Turkey, from 11, 2020, to 11, 2022. This hospital is a pediatric referral center in the Aegean Region of Turkey with annual approximately 600,000 outpatients and 24,000 hospitalizations. The study included all children who required HFNC upon admission or during follow-up. Patients were split into two groups: Group 1 included the patients who tested positive for SARS-CoV-2, group 2 included the patients who tested and found negative for SARS-CoV-2
COVID-19 infection was diagnosed using quantitative real-time reverse transcriptase-PCR positivity with detection of double targets, N-gene and ORF ab1 region at cycling threshold value under 35 cycles [SARS-CoV-2 (2019-nCoV) qPCR Detection Kit, Bio-Speedy, Turkey][11]. The respiratory viruses were detected using a multiplex real-time PCR test (Bosphore Respiratory Pathogens Panel Kit V4,Anato lia Geneworks, Turkey) that is capable of identifying viral pathogens including influenza viruses (influenza A, pandemic H1N1 influenza A, seasonal H1N1 influenza A, and influenza B), parainfluenza viruses (PIVs; PIV-1, PIV-2, PIV-3, and PIV-4), human coronaviruses (CoV OC43, CoV NL63, CoV HKU1, and CoV 229E), RSV A/B, rhinovirus, hMPV, enterovirus, bocavirus, adenovirus, and parechovirus. For detection of respiratory viruses’ specific master mix reagents which include targeted genomic regions of microorganisms were used and cycling threshold values under 35 cycles was considered as positive[12]
Data collection:
Data of the patients was collected from medical records including information on demographic characteristics (age, gender, and medical history); underlying diseases or co-morbidities, indications for HFNC and physical examination findings. If present laboratory examinations taken from the submissions were recorded, including complete blood count (total lymphocyte count, absolute lymphocyte count, hemoglobin, and platelet count), levels of serum coagulation parameters (PT, aPTT, INR, fibrinogen, and D-dimer), C-reactive protein (CRP), procalcitonin, and troponin I/T. The arterial venous parameters such as pH (mmHg), PCO2 (mmHg), PaO2 (mmHg), HCO3 (mmol/L), Lactat (mmol/L), So2(%), and FiO2(%)were also recorded from medical records.
Respiratory distress is typically characterized by signs of increased work of breathing, such as tachypnea, use of accessory muscles, nasal flaring, and/or retractions. The diagnosis of respiratory failure requires at least two clinical signs of respiratory distress and one laboratory criterion (arterial PaCO2 > 50 mmHg and PaO2 < 50 mm Hg in room air; PaCO2 >50 mm Hg and pH 60 mmHg and PaO2 < 60 mm Hg when FiO2 0.60 in patients without cyanotic heart disease; oxygen saturation. Oxygenation of the patients was monitored by pulse oximetry and lower than 94% was accepted as hypoxemia [13]
The indication and duration of HFNC, the length of hospital stay, admission or transfer to the PICU each patient’s stay in the hospital and if present the mortality rates by group were recorded. On high-flow, children received high-flow at weight specific flows (Table-1) delivered via age-appropriate OptiflowTM Junior 274 Nasal Interfaces, OptiflowTM Junior 2+ Nasal Interfaces or Adult cannula and a high-flow delivery system, AirvoTM 2 System (Fisher&Paykel Healthcare, Auckland, New Zealand) [14]. Inspired oxygen fraction (FiO2) was adjusted to obtain oxygen saturation between 92 and 98%.
Statistical analysis
The descriptive properties (mean, median, number, and percentage) of the variables were determined. The numeric variables were checked for fit with normal distribution. While comparing the two groups, the Student’s t-test was used for numeric variables with normal distribution. The Mann-Whitney U test was performed for numeric variables not normally distributed. The chi-square test was performed to compare categorical variables between group I and group II. A p-value < 0.05 was considered statistically significant. Statistical Package for the Social Sciences (SPSS) version 17 (Chicago, Illinois, USA) software was used to analyze the results.
The study protocol was approved by local ethical committee.