Efficacy and safety of application of Sivelestat in pediatric intensive care unit: a systematic review and meta‑analysis

Abstract

Background

The use of Sivelestat in pediatrics remains controversial. Given the limited available evidence, drug labeling does not recommend its administration in children, yet it is increasingly used off-label for pediatric patients with cardiopulmonary bypass (CPB)-related injury or acute respiratory distress syndrome (ARDS). This meta-analysis synthesized the existing data to evaluate the efficacy and safety of Sivelestat in pediatric populations, aiming to provide evidence for rational clinical medication.

Methods

A computerized search was conducted to identify reports on the application of Sivelestat in critically ill pediatric patients in databases PubMed, Embase, Cochrane Library, Web of Science, China national knowledge infrastructure (CNKI), Wanfang Data and VIP from inception to May 2026. The meta-analysis was performed using Review Manager 5.3 version. The risk of bias was assessed using the Cochrane Handbook Risk of Bias Assessment tool.

Results

Finally, 8 controlled studies and 5 case reports were included, involving a total of 608 and 10 children respectively. The results of the meta-analysis showed that the overall P/F ratio was significantly higher in the SE group than in the control group [mean difference (MD) =27.17, 95% confidence interval (CI): 11.84, 42.50, P<0.001]. However, subgroup analysis revealed no significant difference in P/F ratio between the two groups among children with CPB. In children with ARDS, a marked increase in the P/F ratio was observed in the SE group [MD =24.65, 95% CI: 18.12, 31.19, P<0.001). A low risk of bias was identified for this outcome, and further research is warranted to confirm the results. The overall ICU stay time was significantly shorter in the SE group [MD =−2.57, 95% CI: −4.26, −0.87, P=0.003). Similarly, subgroup analysis showed no significant difference in children with CPB, while a marked reduction in ICU stay was observed in the SE group among children with ARDS [MD =−4.03, 95% CI: −7.12, −0.94, P=0.01). The hospital stay time was significantly shorter in the SE group [MD =−2.96, 95% CI: −3.68, −2.24, P<0.001). The overall levels of inflammatory factors were significantly lower in the SE group, including WBC count, CRP, IL−6, IL-8, and TNF-α. Among them, the subgroup findings for WBC and TNF-α were inconsistent. Only a marked reduction was found in children with ARDS. Furthermore, its incidence of complications and adverse events was significantly lower [OR =0.38, 95% CI: 0.22, 0.65, P<0.001). However, existing data are insufficient for a full safety assessment.

Conclusions

In pediatric patients with severe conditions such as ALI/ARDS, Sivelestat may improve the oxygenation index and inhibit the release of inflammatory mediators, but safety assessment remains limited. However, given the limited number of included studies and inter-study heterogeneity, large-scale trials are still needed to verify the results.

Highlight box.

Key findings

• For pediatric patients with severe condition such as acute lung injury (ALI)/acute respiratory distress syndrome (ARDS), Sivelestat shows promising effects on oxygenation improvement and inflammatory response suppression.

What is known and what is new?

• The application of Sivelestat in pediatric patients remains relatively scarce, and its clinical efficacy remains controversial.

• This is the first meta-analysis conducted to assess the efficacy and safety of Sivelestat for pediatric critically ill patients. Sivelestat significantly improves oxygenation index and inhibits the release of inflammatory mediators, and it is safe in pediatric patients.

What is the implication, and what should change now?

• Our findings preliminarily support the efficacy and safety of Sivelestat for critically ill children with ALI/ARDS. Nevertheless, large-scale prospective multicenter trials are still needed to consolidate and validate the current evidence.

Introduction

Sivelestat (SE) is a selective, competitive, and low-molecular-weight synthetic inhibitor of neutrophil elastase which generally treats some serious disease, such as systemic inflammatory response syndrome (SIRS), acute lung injury (ALI), and acute respiratory distress syndrome (ARDS) (1,2). Recently, studies have suggested that the use of SE may be a beneficial strategy for the treatment of coronavirus disease 2019, pulmonary fibrosis, chronic obstructive pulmonary disease, chronic liver disease and cirrhosis, cardiopulmonary bypass (CPB), and the perioperative period in patients with esophageal cancer (3). It can be seen that the clinical application study of SE is constantly expanding and has achieved favorable results in the treatment of ARDS since its launch (4). However, these studies primarily focus on adult patients, with limited clinical applications in pediatric patients, in which it is only being utilized for the treatment of conditions such as ARDS, CPB, and pneumonia, while the clinical efficacy remains controversial (1,2,4). On the one hand, findings from different clinical studies were inconsistent (1). On the other hand, its safety and efficacy in pediatric patient have not been fully verified, so it is not recommended for patients under 18 years of age according to the drug label. Nevertheless, given the critical role of neutrophil elastase in the inflammatory cascade of CPB and ARDS, multiple clinical studies (5,6) and expert consensus (7) support its off-label use in critically ill pediatric patient, particularly those after CPB surgery or diagnosed with ARDS. However, due to the unique characteristics of the pediatric population, for pediatric critically ill patients with exceptionally rapid disease progression, the efficacy and safety of SE use undoubtedly warrant close attention. Therefore, it is essential to conduct a systematic review on the application of SE in critically ill pediatric patients. This is the first systematic review and meta-analysis performed to evaluate the efficacy and safety of SE in critically ill pediatric patients, providing evidence for the safe and rational clinical application of SE. We present this article in accordance with the PRISMA reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0119/rc).

Methods

Data sources and search strategy

All data were gathered from publicly available sources. Databases including PubMed, Embase, Cochrane Library, Web of Science, China National Knowledge Infrastructure (CNKI), Wanfang Data and VIP were systematically searched for articles reporting the application of SE in critically ill pediatric patients from inception to May 2025, with an updated search conducted in May 2026. The following key words were used: (“Sivelestat” OR “elaspol”) AND (“child” OR “children” OR “pediatrics” OR “infant” OR “adolescent”). Language or region was not restricted. Taking PubMed as an example, the search strategy is presented (Figure 1).

Figure 1.

Flowchart of search strategy (PubMed as an example).

The retrieved literature was imported into EndNote software, and duplicate records were removed. All unique titles and abstracts of identified studies were assessed. The full texts of potentially relevant articles were independently examined by two researchers to determine their eligibility for inclusion. Disagreements were resolved through consultation with a third researcher.

Inclusion and exclusion criteria

The population, intervention, comparator, outcome and study design (PICOS) approach was used to establish the selection criteria for our systematic review and meta-analysis. We excluded incomplete or missing data. Studies meeting the following criteria were included:

• ❖ P: patients aged under 18 years.

• ❖ I: use SE.

• ❖ C: the SE group versus control group. The control group underwent a standard procedure, while the SE group received SE in addition to the control group.

• ❖ O: (I) the ratio of partial arterial pressure of oxygen (PaO₂) to fractional concentration of inspired oxygen (FiO₂) (P/F ratio); (II) mechanical ventilation (MV) time; (III) intensive care unit (ICU) stay time; (IV) hospital stay time; (V) white blood cell (WBC) count; (VI) C-reactive protein (CRP); (VII) interleukin-6 (IL-6); (VIII) interleukin-8 (IL-8); (IX) tumor necrosis factor-α (TNF-α); (X) neutrophil count (NC); (XI) the incidence of adverse events (AEs).

• ❖ S: systematic review and meta-analysis. We selected randomized controlled trials (RCTs), prospective cohort studies, and due to the scarcity of studies, we also included retrospective cohort studies and matched case-control series.

Data extraction

The collected data including the first author’s name, publication year, country, sample size, mean age, sex, disease status, and intervention, were independently extracted by two researchers. Reported endpoints were also extracted by them. Disagreements were solved by consulting a third researcher.

Explanation for the repetition of some literature

Since some studies measured the values of the same indicator at multiple time points, but our study was unable to specify which time point to use, the authors considered whether the results obtained by averaging multiple data would be less biased than the subjective selection. Therefore, all of them were included.

Quality assessment

We used the Cochrane Handbook Risk of Bias Assessment Tool to investigate the quality of included RCT. The tool is composed of seven domains, and the domains guide a rating of unclear, low, or high risk of bias (8). We used the Methodological Index for Non-Randomized Studies (MINORS) to investigate the quality of included non-RCT, which is composed of 12 items. The items are scored 0 (not reported), 1 (reported but inadequate) or 2 (reported and adequate). The global ideal score is 16 for non-comparative studies and 24 for comparative studies (9). Two investigators independently evaluated the evidence, and disagreements were solved by discussion.

Statistical analysis

All statistical analyses were performed using Review Manager 5.3 version (Cochrane Collaboration, Oxford, UK). The effect measures included continuous and dichotomous variables. For outcome analyses, mean difference (MD) was presented for continuous data and odds ratio (OR) was reported for dichotomous measures. We converted median (interquartile range) data to mean ± standard deviation using the method proposed by Luo (10) for subsequent pooled analysis. All metrics were reported with 95% confidential interval (CI). The Chi-squared (χ²) test and inconsistency index (I²) were applied to quantify the heterogeneity among different studies. If I²>50%, heterogeneity was deemed considerable. Statistical analyses were performed using a random-effect model. For outcomes with substantial heterogeneity, we executed a leave-one-out sensitivity analysis to examine how the included studies impact the pooled measures. Publication bias was investigated by funnel plots. An asymmetrical spread of the points indicates a publication bias. P values <0.05 were considered statistically significant.

Results

Search results and study characteristics

A total of 167 references were obtained initially, of which 65 were duplicates. One hundred and two studies were left for the final review. After reviewing the titles, abstracts, and full texts, 8 controlled studies (5,6,11-16) and 5 case reports (17-21) were ultimately included (Figure 2). The 8 controlled studies comprise 608 pediatrics patients. The baseline characteristic is presented in Table 1. The 5 case reports cannot be subjected to meta-analysis, and only a descriptive analysis was conducted.

Figure 2.

Flowchart of study selection. CNKI, China National Knowledge Infrastructure.

Table 1. Baseline characteristic of controlled studies included.

Study	Country	Type	Group	Cases	Age (months)	Sex (M:F)	Disease status	Intervention [mg/(kg·h)]	Outcome indicator
Toyama [2008] (11)	Japan	Retrospective	S	10	2.9±2.2	NR	CPB	0.2	(I)(II)(III)(XI)
			C	15	3.7±2.8	NR		Standard procedure†
Inoue [2013] (5)	Japan	RCTs	S	13	10.8±5.9	9:4	CPB	0.2	(I)(II)(III)(V)(VI)(VII)(VIII)(IX)(X)
			C	13	13.4±7.8	4:9		0.9% NaCl
Kohira [2013] (12)	Japan	RCTs	S	15	9.5±6.7	NR	CPB	0.2	(I)(V)(VI)(VIII)(X)
			C	15	12.2±6.3	NR		0.9% NaCl
Nomura [2013] (13)	Japan	RCTs	S	7	2.7±2.1	2:5	CPB	0.2	(II)(III)(IV)(VIII)
			C	6	2.5±2.3	2:5		Saline
Liu [2026] (14)	China	RCTs	S	50	19.62±22.36	22:28	CPB	0.2	(I)(II)(III)(V)(VI)(VII)(IX)(X)(XI)
			C	50	15.01±15.65	25:25		0.9% NaCl
Wang [2022] (6)	China	Retrospective	S	125	61.7±76.6	67:58	ARDS	NR	(I)(III)(V)(VI)(XI)
			C	87	60.4±79.6	42:45		NR
Wang [2024] (15)	China	RCTs	S	51	39.1±11.8	26:25	ARDS	0.2	(II)(III)(IV)(VI)(VII)(IX)(XI)
			C	51	37.1±10.4	23:28		Symptomatic treatment
Nadire [2025] (16)	China	Retrospective	S	40	116.9±18.1	25:15	ARDS	0.2	(I)(II)(IV)(V)(VI)(VII)(IX)(XI)
			C	60	116.0±17.4	38:22		Symptomatic treatment

^†, the control group underwent a standard procedure, and the sivelestat group received a continuous infusion of 0.2 mg/(kg·h) sivelestat besides standard procedure. S: Sivelestat group; C: control group. Outcome indicator: (I) oxygenation index, PaO₂/FiO₂; (II) MV time; (III) ICU stays time; (IV) hospital stay time; (V) WBC count; (VI) CRP; (VII) interleukin-6 (IL-6); (VIII) IL-8; (IX) TNF-α; (X) NC; (XI) the incidence of adverse events. ARDS, acute respiratory distress syndrome; CPB, cardiopulmonary bypass; CRP, C-reactive protein; F, female; FiO₂, fractional concentration of inspired oxygen; ICU, intensive care unit; IL-8, interleukin-8; M, male; MV, mechanical ventilation; NC, neutrophil count; NR, not reported; PaO₂, partial arterial pressure of oxygen; RCTs, randomized controlled trials; TNF-α, tumor necrosis factor-α; WBC, white blood cell.

Quality assessment

Of the eight controlled studies, the quality evaluation results of 5 RCTs were as follows: 3 studies did not specify the method of random sequence generation; only one mentioned allocation concealment; one study did not clarify whether blinding was implemented; all outcome data were complete; there was no selective reporting; other biases remained unclear (Figure 3). The MINORS scores of the three non-RCT studies were 17–20 points.

Figure 3.

Risk of bias graph and summary.

Meta-analysis outcomes

P/F ratio

A total of 6 studies (5,6,11,12,14,16) were included, among which 4 (5,6,11,14) had measurements at multiple time points. There is a statistical heterogeneity among the studies (P=0.008, I²=57%), so we use a random-effects model. Meta-analysis of the overall P/F ratio demonstrated that the SE group exhibited a significantly higher value than the control group (MD =27.17, 95% CI: 11.84, 42.50, P<0.001). The subgroup analysis revealed no significant difference in the P/F ratio between the two groups among children with CPB (MD =21.29, 95% CI: −28.50, 71.08, P=0.40). In children with ARDS, a marked increase in the P/F ratio was observed in the SE group (MD =24.65, 95% CI: 18.12, 31.19, P<0.001) (Figure 4).

Figure 4.

Forest plot of the P/F ratio in the Sivelestat and control groups. ARDS, acute respiratory distress syndrome; CI, confidence interval; CPB, cardiopulmonary bypass; IV, inverse variance; P/F, ratio of partial arterial pressure of oxygen to fractional concentration of inspired oxygen; SD, standard deviation.

MV time

A total of 6 studies (5,11,13-16) were included. There is a statistical heterogeneity among the studies (P<0.001, I²=94%), so we use a random-effects model. Meta-analysis of the overall MV time showed that there is no statistical difference between the two groups (MD =−0.87, 95% CI: −1.82, 0.08, P=0.07). The subgroup analysis showed no significant difference in children with CPB (MD =0.00, 95% CI: −0.69, 0.69, P>0.99), while a marked reduction in MV time was observed in the SE group among children with ARDS (MD =−2.21, 95% CI: −2.72, −1.70, P=0.01) (Figure 5).

Figure 5.

Forest plot of the MV time in the Sivelestat and control groups. ARDS, acute respiratory distress syndrome; CI, confidence interval; CPB, cardiopulmonary bypass; IV, inverse variance; MV, mechanical ventilation; SD, standard deviation.

ICU stay time

A total of 6 studies (5,6,11,13-15) were included. There is a statistical heterogeneity among the studies (P<0.001, I²=78%), so we use a random-effects model. Meta-analysis of the overall ICU stay time indicated it was significantly shorter in the SE group than in the control group (MD =−2.57, 95% CI: −4.26, −0.87, P=0.003). The subgroup analysis showed no significant difference in children with CPB (MD =−1.20, 95% CI: −2.64, 0.24, P=0.10), while a marked reduction in ICU stay was observed in the SE group among children with ARDS (MD =−4.03, 95% CI: −7.12, −0.94, P=0.01) (Figure 6).

Figure 6.

Forest plot of the ICU stay time in the Sivelestat and control groups. ARDS, acute respiratory distress syndrome; CI, confidence interval; CPB, cardiopulmonary bypass; ICU, intensive care unit; IV, inverse variance; SD, standard deviation.

Hospital stay time

A total of 3 studies (11-13) were included. No statistical heterogeneity among the studies (P=0.87, I²=0%), so we use a fixed-effects model. The meta-analysis of the hospital stay time demonstrated that the SE group exhibited a significantly shorter time than the control group (MD =−2.96, 95% CI: −3.68, −2.24, P<0.001) (Figure 7).

Figure 7.

Forest plot of the hospital stay time in the Sivelestat and control groups. CI, confidence interval; IV, inverse variance; SD, standard deviation.

The levels of inflammatory factors

WBC count

A total of 5 studies (5,6,12,14,16) were included, among which 2 (6,14) had measurements at multiple time points. There is a statistical heterogeneity among the studies (P=0.06, I²=51%), so we use a random-effects model. The meta-analysis of the overall WBC count showed that the SE group had significantly lower count than the control group (MD =−1.77, 95% CI: −2.49, −1.04, P<0.001). The subgroup analysis showed no significant difference in WBC count in children with CPB (MD =−1.75, 95% CI: −3.83, 0.33, P=0.10), while a marked reduction in WBC count was observed in the SE group among children with ARDS (MD =−1.99, 95% CI: −2.55, −1.43, P<0.001) (Figure 8).

Figure 8.

Forest plot of the WBC count in the Sivelestat and control groups. ARDS, acute respiratory distress syndrome; CI, confidence interval; CPB, cardiopulmonary bypass; IV, inverse variance; SD, standard deviation; WBC, white blood cell.

CRP

A total of 6 studies (5,6,12,14-16) were included, among which 3 (6,14,15) had measurements at multiple time points. These is no statistical heterogeneity among the studies (P=0.05, I²=49%), so we use a fixed-effects model. The meta-analysis of the overall CRP levels showed that the SE group had significantly lower values than the control group (MD =−6.02 95% CI: −7.25, −4.97, P<0.001). The subgroup analyses also yielded consistent results. The levels of CRP were lower in the SE group than in the control group in children with CPB (MD =−7.57 95% CI: −12.28, −2.86, P=0.002) and ARDS (MD =−5.91 95% CI: −7.18, −4.63, P<0.001) (Figure 9).

Figure 9.

Forest plot of the CRP in the Sivelestat and control groups. ARDS, acute respiratory distress syndrome; CI, confidence interval; CPB, cardiopulmonary bypass; CRP, C-reactive protein; IV, inverse variance; SD, standard deviation.

IL-6

A total of 4 studies (5,14-16) were included, among which 2 (14,15) had measurements at multiple time points. There is a statistical heterogeneity among the studies (P<0.001, I²=95%), so we use a random-effects model. The meta-analysis of the overall IL-6 levels showed that the SE group had significantly lower values than the control group (MD =−16.16, 95% CI: −28.62, −3.71, P<0.001). The subgroup analyses also yielded consistent results. The levels of IL-6 were lower in the SE group than in the control group in children with CPB (MD =−1.76 95% CI: −2.98, −0.53, P=0.005) and ARDS (MD =−26.68 95% CI: −41.80, −11.56, P<0.001) (Figure 10).

Figure 10.

Forest plot of the IL-6 in the Sivelestat and control groups. ARDS, acute respiratory distress syndrome; CI, confidence interval; CPB, cardiopulmonary bypass; IL-6, interleukin-6; IV, inverse variance; SD, standard deviation.

IL-8

A total of 3 studies (5,12,13) were included. No statistical heterogeneity among the studies (P=0.13, I²=50%), so we use a fixed-effects model. The meta-analysis of the IL-8 showed that the SE group exhibited a significantly lower than the control group (MD =−16.54, 95% CI: −32.36, −0.72, P=0.04) (Figure 11).

Figure 11.

Forest plot of the IL-8 in the Sivelestat and control groups. CI, confidence interval; IL-8, interleukin-8; IV, inverse variance; SD, standard deviation.

TNF-α

A total of 4 studies (5,14-16) were included, among which 2 (14,15) had measurements at multiple time points. There is a statistical heterogeneity among the studies (P<0.001, I²=78%), so we use a random-effects model. The meta-analysis of the overall TNF-α levels showed that the SE group had significantly lower values than the control group (MD =−0.55, 95% CI: −0.78, −0.33, P<0.001). The subgroup analysis showed no significant difference in TNF-α levels in children with CPB (MD =−1.20, 95% CI: −2.91, 0.51, P=0.17), while a marked reduction in TNF-α levels was observed in the SE group among children with ARDS (MD =−0.58, 95% CI: −0.75, −0.40, P<0.001) (Figure 12).

Figure 12.

Forest plot of the TNF-α in the Sivelestat and control groups. ARDS, acute respiratory distress syndrome; CI, confidence interval; CPB, cardiopulmonary bypass; IV, inverse variance; SD, standard deviation; TNF-α, tumor necrosis factor-α.

NC count

A total of 3 studies (5,12,14) were included, among which 2 (12,14) had measurements at multiple time points. There is a statistical heterogeneity among the studies (P=0.007, I²=72%), so we use a random-effects model. The meta-analysis of the NC count showed that the SE group had significantly lower count than the control group (MD =−1.85, 95% CI: −3.57, −0.12, P=0.04) (Figure 13).

Figure 13.

Forest plot of the NC count in the Sivelestat and control groups. CI, confidence interval; IV, inverse variance; NC, neutrophil count; SD, standard deviation.

Incidence of AE

Five studies (6,11,14-16) examined the occurrence of complications, adverse reactions or AEs. Two of the studies (11,14) showed that none occurred in either group. A total of three studies were included. There is no statistical heterogeneity among the studies (P=0.35, I²=5%), so we use a fixed-effects model. The meta-analysis of the incidence of AE count showed that the SE group exhibited a significantly lower incidence than the control group (OR =0.38, 95% CI: 0.22, 0.65, P<0.001) (Figure 14).

Figure 14.

Forest plot of the incidence of AE in the Sivelestat and control groups. AE, adverse event; CI, confidence interval; M-H, Mantel-Haenszel.

Funnel plots

By visual check of the funnel plot, a slight publication bias was revealed with the P/F ratio because two of the studies were located further away from the central axis (Figure 15A). Publication bias was not found in other indicators (Figure 15).

Figure 15.

Funnel plots of the P/F ratio (A), MV time (B), ICU stay time (C), hospital stay time (D), WBC count (E), CRP (F), IL-6 (G), IL-8 (H), TNF-α (I), NC count (J), and incidence of AEs (K). AEs, adverse events; ARDS, acute respiratory distress syndrome; CPB, cardiopulmonary bypass; CRP, C-reactive protein; ICU, intensive care unit; IL-6, interleukin-6; IL-8, interleukin-8; MD, mean difference; MV, mechanical ventilation; NC, neutrophil count; P/F, ratio of partial arterial pressure of oxygen to fractional concentration of inspired oxygen; SE, standard error; TNF-α, tumor necrosis factor-α; WBC, white blood cell.

Descriptive analysis outcomes

Five studies (17-21) were case reports and involved a total of 10 pediatric patients. The results showed that after treatment with SE, significant improvements were observed in oxygenation, P/F ratios, and chest X-rays in 3 cases of respiratory distress syndrome (17,18), 2 cases of aspiration (19), and 3 cases of sepsis (20), accompanied by rapid declines in hematological inflammatory markers. Two patients had an elevation of pivaloylcarnitine (21) and one patient died due to cerebral ischemia (19).

Heterogeneity and sensitivity analyses

After sequentially removing studies with substantial heterogeneity, the results are presented in Table 2. With regard to the P/F ratio, ICU stay times and WBC count, we found their heterogeneity was caused by measurements at a single time point or certain individual studies, yet no reversal was observed in the combined effect size. Their outcomes are relatively robust. Although we failed to identify the exact source of heterogeneity, sensitivity analyses revealed that the pooled effect sizes of indicators MV time and NC count were reversed, suggesting that their results are unreliable.

Table 2. Heterogeneity and sensitivity analyses.

Indicators	Heterogeneity	Sensitivity
P/F ratio (5,6,11,12,14,16)	Inoue [2013] (5) [the value at ICU 3 hours (241±182 vs. 470±272)] was the main source. After it was excluded, the heterogeneity was significantly reduced (P=0.06, I2=44%)	The combined effect size did not show significant changes. Their outcomes are relatively robust
ICU stay times (5,6,11,13-15)	Liu [2026] (14) was the main source. After it was excluded, the heterogeneity was significantly reduced (P=0.11, I2=46%)
WBC count (5,6,12,14,16)	Liu [2026] (14) [the value at 24 hours (14.31±5.34 vs. 13.22±4.58)] was the main source. After it was excluded, the heterogeneity was significantly reduced (P=0.66, I2=0%)
IL-6 (5,14-16)	Heterogeneity remains significant
TNF-α (5,14-16)	Heterogeneity remains significant
MV time (5,11,13-16)	Heterogeneity remains significant	When Inoue [2013] (5) was excluded, the combined effect size changed significantly. The statistical significance reversed.
NC count (5,12,14)	Liu [2026] (14) [the value at 24 hours (10.69±3.59 vs. 9.94±3.89)] was the main source. After it was excluded, the heterogeneity was significantly reduced (P=0.10, I2=52%)	When Inoue [2013] (5) or Kohira [2013] (12) or Liu [2026] (14) (the value at 72 hours) was excluded, the combined effect size changed significantly. The statistical significance reversed

FiO₂, fractional concentration of inspired oxygen; ICU, intensive care unit; IL-6, interleukin-6; MV, mechanical ventilation; NC, neutrophil count; P/F ratio, the ratio of PaO₂ to FiO₂; PaO₂, partial arterial pressure of oxygen; TNF-α, tumor necrosis factor-α; WBC, white blood cell.

Discussion

ALI/ARDS is a life-threatening syndrome induced by various diseases, and is a common complication at ICU admission or during ICU stay. Its mortality rate is very high (1,2). At present, the treatment of ALI/ARDS is symptomatic support, such as lung-protective ventilation, whereas specific pharmacotherapies are still lacking (1,2). In the progression of ALI/ARDS, activated neutrophils play a central role by releasing various inflammatory mediators, including elastase (2). SE is an inhibitor of neutrophil elastase. Its protective effects in attenuating ALI/ARDS have been confirmed in several models of lung injury and adult patients (2). However, its application in pediatric patients remains very limited and controversial. Therefore, this study is highly importance for improving the safety and rational use of medications in children.

The subjects ultimately included in our study were mainly pediatric patients with ARDS and CPB, mainly. CPB can induce SIRS, ALI and ARDS, which are contribute significantly to the postoperative morbidity and mortality (5,11,12). Therefore, it is important to prevent SIRS and ALI/ARDS by CPB in the perioperative management. The P/F ratio is considered a critical and reliable indicator for assessing pulmonary health status. It frequently exhibits significant reductions in pediatric patients with ARDS. Our meta-analysis showed that compared with the control group, the SE group significantly increased the overall P/F value of children, suggesting it can improve lung tissue compliance and oxygenation. It is consistent with the findings in adult patients (4). It is worth noting that our outcome was assessed to have a low risk of bias. Further studies are still needed to validate our findings. However, subgroup analysis stratified by CPB and ARDS revealed that SE failed to improve the P/F ratio in children undergoing CPB, but exerted a beneficial effect only in those with ARDS. Theoretically, as lung compliance improves, the body becomes less dependent on MV, which should significantly shorten the duration of MV. However, our meta-analysis results indicate that there is no statistically significant difference in the duration of MV between the SE group and the control group. This seems contradictory. It is noteworthy that the sensitivity analysis of this outcome indicates the occurrence of statistical reversal, which means the result is not robust. The reason may be the limited number of included studies and the small sample size. Another possible factor lies in disease heterogeneity, as subgroup analyses indicated that SE exerted beneficial effects in children with ARDS. Uncontrolled neutrophil-dominant inflammation is one of the common pathophysiological features of ARDS (1). Our results indicate that SE can significantly reduce the levels of WBC, CRP, IL-6, IL-8, TNF-α, and NC in pediatric patients. It is suggested that SE exerts potent anti-inflammatory effects by inhibiting the release of inflammatory mediators, improving lung injury symptoms (6). Among them, the sensitivity analysis of IL-8 levels and NC count shows instability, more large-sample studies are still needed for verification. Moreover, subgroup analyses yielded inconsistent results for WBC count and TNF-α levels. SE failed to improve these two indicators in children with CPB, but exerted beneficial effects only in patients with ARDS. Furthermore, in our study, the ICU stay time and hospital stay time for the SE group were significantly shorter. It is consistent with the findings in adult patients (4). It is means that SE can reduce medical expenses and alleviate the economic burden on families. Interestingly, our subgroup analyses demonstrated that SE exerted beneficial effects on all measured indicators in children with ARDS. In contrast, only a small proportion of parameters were improved among children undergoing CPB, suggesting that SE may have limited application value in the CPB population. Notably, given the very limited number of included studies focusing on ARDS patients, these findings require further verification. In terms of safety, the SE group exhibited lower incidence rates of complications, adverse reactions, and AEs compared to the control group. Safety data primarily included pulmonary complications, bacteremia, allergic reactions, headaches, and mortality. The SE treatment can significantly reduce the incidence of complications in pediatric patients and demonstrate a favorable safety.

There are several limitations in this study. Firstly, the number of included studies and cases are relatively small. Currently, the application of SE in critically ill children remains relatively limited, which might be due to the particularities of the pediatric population and severe diseases, as well as the controversies about SE in clinical practice. Consequently, clinicians must use the drug with caution. However, this further highlights the value and significance of the present study. Secondly, heterogeneity was observed in the P/F ratio, MV time, ICU stay, WBC, IL-6, TNF-α and NC count indicators, but the underlying causes of this heterogeneity could not be identified. Finally, the limited studies also encompassed three retrospective studies, and therefore the quality of the literature was not high enough compared to RCT. The possibility of bias in the results of this study is greater.

Conclusions

Our results indicate that SE can significantly improve the P/F ratio, reduce ICU and hospital length of stay, decrease inflammatory markers such as WBC, CRP, IL-6, and TNF-α, and lower the incidence of complications and AEs in pediatric patients with severe conditions such as ALI/ARDS. However, due to the limitations of this study, large-scale, high-quality studies are still needed in pediatric populations to validate these results.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.