Results

A total of 186 patients met the inclusion criteria, 37 of whom were excluded for the following reasons: incomplete clinical data (n = 23); admission lasting <24 h (n = 13); and age >14 years (n = 1). The screening flowchart is shown in Figure 1. Among the remaining 149 patients, 88 (59.1%) were men, the median (IQR) age was 32 (10–60) months, and the median PELOD-2 score upon admission was 9 (7–11). The in-hospital mortality rate was 64.4% (96 patients), whereas 35.6% (53 patients) survived until discharge. Overall, 82 (55.0%) and 67 (45.0%) patients had direct and indirect PARDS, respectively, while 38 (71.7%) and 15 (28.3%) survivors had direct and indirect PARDS, respectively. The baseline characteristics of the study participants are presented in Table 1.

Changes in FSS Scores from Baseline

The median (IQR) FSS scores of survivors at the time of hospital admission and discharge were 6 (6–6) and 7 (6–7), respectively. A tendency toward improved functional status was seen by the time of discharge, with a median (IQR) change in FSS score of 1 (1–2). Upon discharge, 40 (75.5%) survivors exhibited good functional status, whereas nine (17.0%) and four (7.5%) exhibited mildly and moderately abnormal function, respectively. None of the survivors demonstrated severely or very severely abnormal functional status, and the new morbidity rate of survivors at discharge was 24.5%.

Assessment of FSS Domains at Discharge

Functional impairment was present upon discharge for most domains, although respiratory, feeding, and motor domains were the most commonly affected. The rates of dysfunction were the lowest in the communication domain at discharge. Among the survivors, 33 (62.3%) exhibited respiratory dysfunction, nine (17.0%) had feeding dysfunction, five (9.4%) were affected by motor dysfunction, four (7.5%) demonstrated sensory dysfunction, four (7.5%) had mental dysfunction, and two (3.8%) showed communication dysfunction (Figure 2). Additionally, 13 (24.5%) patients exhibited multi-domain dysfunction.
Respiratory dysfunction was associated with pleural effusion (9 patients, 17.0%), lobar pneumonia (14 patients, 26.4%), pneumothorax (5 patients, 9.4%), and pulmonary fibrosis (2 patients, 3.8%), for which oxygen therapy or suction was required post-discharge. None of the patients required a tracheostomy or ventilation management. Feeding dysfunction was predominantly associated with digestive problems; for example, two patients (3.8%) remained on liquid meals due to pancreatitis, four (7.5%) required preterm milk and special formula, and three (5.7%) needed nasogastric feeding. Motor dysfunction was associated with primary disease and treatment; for example, three cases (5.7%) were caused by a primary disease (one case each of cerebral development anomalies, leg edema, and fingertip necrosis). Moreover, two patients (3.8%) underwent ECMO treatment and experienced lower extremity thrombosis.

Comparison Between Survival and Non-Survival Groups

The survival and non-survival groups differed significantly in terms of PARDS severity, P/F ratio on day 1 after diagnosis, PELOD-2 score upon admission, PARDS etiology, rate of inhaled nitric oxide use, PIP, PEEP, and FiO2 measurements, duration of hospitalization, hemopurification, immunosuppression, and comorbidity. The characteristics of survivors and non-survivors are presented in Table 2.
When assessing risk factors, severe PARDS was related to survival rate (odds ratio [OR] = 0.24; 95% confidence interval [CI]: 0.07–0.80; P = 0.02), after correcting for other factors. The results of statistical analyses are presented in Table 3. OR values for the other variables in the regression analysis (Model 2) are shown in Table 4.

Survival Curves of the Direct and Indirect PARDS Groups

A comparison of the survival curves of the direct and indirect PARDS groups using the log-rank test revealed significant separation, with the indirect group having a lower survival rate than the direct group. Additionally, the indirect PARDS group had a poorer survival rate and a worse prognosis (P < 0.05) (Figure 3).