Nearly half of the children still report moderate to severe pain during postoperative hospitalization (1). The results of a survey conducted in China showed that 42 out of 66 hospitals (63.64%) implemented pain management in children, but only eight (12.12%) achieved full coverage of pain assessment for children, including all outpatients and hospitalized children (2). Pain in early childhood that is inadequately treated, unrecognized, or poorly managed can persist into adulthood and lead to severe and persistent negative consequences (3). On the one hand, it can turn into chronic postoperative pain, which can cause functional impairment, disability, or depression (4). On the other hand, it may affect children's future healthcare experience, deepen unpleasant memories and psychological burdens, and generate anxiety and fear (5). Therefore, perioperative analgesia in children is essential.

There are limited options for analgesics for children. Ethical constraints in pediatric clinical trials have led to inadequate clinical evidence for many analgesics in children compared with adults, and the labels of most analgesics clearly state that they are not recommended for children (6). Opioids are still the cornerstone of multimodal analgesic regimens for children with moderate to severe acute pain, but their adverse events have been a frequent problem (7). The most common opioid complications, such as nausea, dizziness, and vomiting, occurred at the rates of 18.9, 18.4, and 13.4%, respectively (8). The incidence of pruritus caused by intravenous morphine injection ranges 2–10% (9). Therefore, a multimodal analgesic regimen based on non-steroidal anti-inflammatory drugs is recommended for effective perioperative pain control and the reduction of opioid consumption. Non-steroidal anti-inflammatory drugs are recommended as part of multimodal analgesia in the perioperative period in children by consensus and guidelines in several countries (10–12).

Flurbiprofen axetil is an injectable, non-selective cyclooxygenase (COX) inhibitor. Owing to the novel drug carrier system of lipid microspheres, flurbiprofen axetil can be targeted to inflammatory sites and surgical incisions and control the release of the encapsulated drug for a longer-lasting effect (13). Flurbiprofen axetil also readily penetrates into the cerebrospinal fluid, where it can reach concentrations of up to seven times higher than unconjugated plasma concentrations, contributing to the analgesic effect in the central nervous system (14). Parecoxib is the first injectable selective COX-2 inhibitor that is rapidly converted to its active metabolite, valdecoxib, after hepatic enzymatic hydrolysis in vivo (15). Parecoxib has a highly selective inhibitory effect on COX-2 and reduces peripheral and central sensitization by blocking the arachidonic acid cascade and inhibiting prostaglandin synthesis, thus providing an analgesic effect. It has no inhibitory effect on COX-1 at the recommended dose and has little impact on the normal physiological function of the gastric mucosa and platelets (16).

Clinical evidence for the use of parecoxib and flurbiprofen ester for perioperative analgesia in children has emerged as multimodal analgesia regimens and is increasingly advocated. However, only one meta-analysis published in 2015 systematically evaluated the effect of perioperative parecoxib on postoperative pain in children, demonstrating that, compared with placebo, parecoxib reduced pain scores and the incidence of adverse events (17). The efficacy and safety of flurbiprofen axetil in children remain unclear, and few studies have directly compared the drugs. Therefore, we performed a Bayesian network meta-analysis that integrated as much data as possible from direct and indirect comparative evidence to systematically evaluate the efficacy and safety of parecoxib and flurbiprofen axetil in pediatric perioperative analgesia.

Our analysis included 49 RCTs with 3,657 children aged 6 months to 14 years, and the aim of this study was to systematically evaluate the efficacy and safety of parecoxib and flurbiprofen axteil in perioperative analgesia in children using a Bayesian network meta-analysis. The results demonstrated that the pain scores of the flurbiprofen axetil group were significantly lower than those in the placebo group at all time points in the 24-h postoperative period, and the pain scores at all time points during the 12-h postoperative period of the parecoxib group were significantly lower than those in the placebo group. From the potential rankings, flurbiprofen axetil was superior to other drugs in reducing pain scores at 2, 4, and 12 h postoperatively, while parecoxib had some advantages in reducing pain scores at 0, 0.5, 1, 6, 8, and 24 h postoperatively.

Few meta-analyses have been conducted to investigate the efficacy and safety of parecoxib or flurbiprofen axetil for perioperative analgesia in children, but evidence in adults is constantly being updated. A previous meta-analysis by Li et al. (68) reported that low to moderate evidence indicated that parecoxib relieved postoperative orthopedic pain while sparing opioid analgesic consumption without increasing the incidence of adverse events (68). Sun et al. (69) reported that the perioperative administration of flurbiprofen was effective in reducing postoperative pain, nausea, and vomiting in Chinese surgical patients (69). Another meta-analysis comparing flurbiprofen axetil vs. parecoxib found that parecoxib had an advantage over flurbiprofen axetil for patients with postoperative pain beyond 6 h (70). It has been reported that intravenous administration of parecoxib (1.0 mg/kg to a maximum of 40 mg) in children resulted in free valdecoxib concentrations above the IC₅₀ for COX-2 for at least 12 h, and the area under the valdecoxib concentration-time curve was similar to that in adults (40 mg) (71). A population pharmacokinetic study of intravenous flurbiprofen axetil in Chinese patients with postoperative pain discovered that between-subject variability for flurbiprofen was associated with body height and weight (72). In children aged 3 months to 13 years, the estimated clearance rate was 0.96 L/h/70 kg and the volume of distribution at steady state was 8.1 L/70 kg, which was significantly lower than that in adults (14). Therefore, the efficacy and safety of parecoxib and flurbiprofen axetil for perioperative analgesia in children need to be confirmed, as the results of the adult studies cannot be directly extrapolated to children because of the limitations of individual differences between adults and children.

In a previous network meta-analysis, parecoxib for analgesia after tonsillectomy in children was found to be better than fentanyl, and flurbiprofen axetil was superior to tramadol (73), but it only reported one pain score and did not specify the pain scores in children at different time points. Another meta-analysis found that the perioperative use of parecoxib in children was associated with less acute postoperative pain and adverse events compared with placebo or standard treatment (fentanyl or tramadol) (17), but that study lacked an evaluation of the analgesic effects of parecoxib beyond 12 h postoperatively. Our network meta-analysis systematically evaluated the efficacy and safety of parecoxib and flurbiprofen axetil for perioperative analgesia in children while adding more recent studies to provide a more reliable evidence-based analysis.

In terms of safety, compared with children who received placebo, those who received flurbiprofen axetil or parecoxib had a lower incidence of total adverse events and postoperative agitation. Compared with tramadol, flurbiprofen axetil and parecoxib both significantly reduced the incidence of total adverse events and postoperative nausea and vomiting. Compared with flurbiprofen axetil or fentanyl, parecoxib significantly reduced the incidence of postoperative nausea and vomiting. The ranking results showed that parecoxib was more advantageous in decreasing the incidence of total adverse events and postoperative nausea and vomiting.

Because few adverse events were reported in the included trials, such as respiratory depression, pruritus, headache, dizziness, and urinary retention, only total adverse events, postoperative nausea and vomiting, and agitation were analyzed in this study because of insufficient data. A multicenter retrospective study involving 3,542 patients found that postoperative administration of parecoxib resulted in a 0.2% incidence of adverse events, primarily hypotension, nausea, vomiting, pruritus, and rash (74). Another study discovered that flurbiprofen axetil accounted for the largest number of adverse reactions (686/5597, 12.26%), with the most common being nausea, rash, vomiting, pruritus, and dizziness (75). Several studies have revealed that, compared with tramadol, parecoxib and flurbiprofen axetil both resulted in less postoperative nausea and vomiting in adults (76–78). Our study also found in children that the incidence of postoperative nausea and vomiting was lower for both parecoxib and flurbiprofen axetil than for tramadol. Bu et al. (17) found that, when compared to placebo or standard treatment, perioperative parecoxib administration for children resulted in a lower incidence of postoperative nausea, vomiting, and agitation (17). Li et al. (67) also discovered that the incidence of postoperative nausea and vomiting was significantly higher in the placebo group (11/30, 37%) than in the parecoxib group (4/30, 13%) (67). These results suggested that parecoxib was an effective and safe analgesic for children, particularly in reducing postoperative nausea and vomiting, which is consistent with the results of our study. The included trials demonstrated that none of the children had bleeding, gastrointestinal injury, renal injury, or cardiovascular events. In a pooled analysis of 28 randomized, double-blind, placebo-controlled clinical trials, patients in the parecoxib and placebo groups had similar frequencies of gastrointestinal events, renal failure and impairment, and cardiovascular embolic and thrombotic events (79). Therefore, this study considered that both parecoxib and flurbiprofen axetil decreased the risk of total adverse events compared with placebo and tramadol and that parecoxib was more advantageous in reducing postoperative nausea and vomiting.

There were a few limitations to this study. First, the included trials were associated with high risks of bias because most of them only mentioned “random” without specifying the method of generating the random sequence or describing the allocation concealment in detail. Second, most studies only recorded pain scores within the first 24 h after surgery, paying little attention to 48 or 72 h. Furthermore, we did not analyze other outcomes, such as opioid consumption or proportions of patients requiring rescue analgesia, which may have more clinical implications. Third, data on direct head-to-head comparisons of flurbiprofen axteil and parecoxib within the network were insufficient. Finally, there were some differences between studies in the type of surgery, perioperative care protocol, study drug dose, and timing of administration, which could explain the heterogeneity in our results.

Based on the available clinical evidence, our network meta-analysis reached the following conclusions. Perioperative administration of flurbiprofen axetil and parecoxib were both effective in reducing postoperative pain scores and total adverse events compared with placebo. Flurbiprofen axetil and parecoxib also caused fewer total adverse events and postoperative nausea and vomiting compared with tramadol. Flurbiprofen axetil was most effective at reducing pain scores at 2, 4, and 12 h postoperatively. Parecoxib had an advantage in terms of reducing pain scores at 0, 0.5, 1, 6, 8, and 24 h postoperatively, as well as reducing the incidence of total adverse events and postoperative nausea and vomiting.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

XiC helped with study design, the selection of included articles, data extraction, quality assessment, data analysis and interpretation, and writing the manuscript. PC and QH helped with the study design, selection of included articles, data extraction, quality assessment, data analysis and interpretation, and the final review. XiaC, MH, and KT helped with the study design and the final review. All authors contributed to the article and approved the submitted version.