Objectives: To determine the time to influenza virus nucleic acid clearance and the treatment duration of neuraminidase inhibitors (NA) in children with severe influenza admitted to the pediatric intensive care unit (PICU) following NA initiation. This study further aimed to identify factors associated with delayed viral clearance and to evaluate the appropriateness of different NA discontinuation strategies. Methods: In this retrospective cohort study, we analyzed the clinical data of 79 children with severe influenza admitted to the PICU of Beijing Children's Hospital between November 2019 and July 2024. Inclusion criteria were continuous NA administration for ≥72 h after admission, and serial nasopharyngeal swab tests for influenza virus nucleic acid performed at 3 to 5 d intervals following NA initiation until clearance or discharge. Collected data included demographic characteristics, serial nucleic acid test results, body temperature, NA treatment duration, and in-hospital mortality were collected. Children were stratified into three groups based on time from NA initiation to viral nucleic acid clearance:≤5 d, 6-10 d, and >10 d, the latter 2 of which belong to the delayed clearance group. They were also categorized according to clinical status at NA discontinuation: the fever group (fever present, nucleic acid negative), the necleic acid-positive group (afebrile, nucleic acid positive) and the double-negative group (afebrile, nucleic acid negative). Select the appropriate statistical test from among the Chi-square test, Kruskal-Wallis H test, or Mann-Whitney U test to compare between-group differences in the following variables: duration of NA use, co-infection rate, proportion of patients with fever after 5 days of NA use, pediatric risk of mortality Ⅲ score (PRISM Ⅲ) at 48 h after NA discontinuation, pediatric sequential organ failure assessment score (pSOFA) , and in-hospital mortality rate. Results: Of the 79 children with severe influenza, 43 (54%) were male and 36 (46%) female. The age was 4.6 (2.3,7.7) years. Patients were categorized into the ≤5 d (30 cases), 6-10 d (34 cases), and >10 d (15 cases) groups. The overall proportion of delayed clearance was 62% (49/79). The >10 d group demonstrated a significantly longer duration of NA therapy, higher co-infection rate, and a greater proportion of patients with persistent fever after 5 d of treatment compared to the other groups (all P<0.05). The ≤5 d group had the lowest incidence of lymphopenia detected on days 4 and 5 after PICU admission (χ²=6.08, P<0.05). No significant differences were observed among the fever (11 cases), nucleic acid-positive (26 cases), and double-negative (42 cases) groups regarding in-hospital mortality, PRISM Ⅲ, or pSOFA at 48 h after NA discontinuation (P>0.05). Conclusions: A considerable proportion of children with severe influenza in the PICU experienced delayed viral clearance after NA initiation, which was associated with a longer treatment course, particularly in those requiring >10 d for clearance. Discontinuation of NA upon achieving virological clearance or defervescence may be considered, but this strategy requires careful clinical judgment tailored to the individual patient.