Effect of Prophylactic Tropisetron on Post-Operative Nausea and Vomiting in Patients Undergoing General Anesthesia: Systematic Review and Meta-Analysis with Trial Sequential Analysis

This systematic review and meta-analysis of randomized controlled trials (RCTs) with trial sequential analysis (TSA) aimed to comprehensively evaluate and compare the efficacy of the prophylactic administration of tropisetron in the prevention of the incidence of post-operative nausea and vomiting (PONV) in patients undergoing surgery under general anesthesia. This study was registered with PROSPERO (CRD42024372692). RCTs comparing the efficacy of the perioperative administration of tropisetron with that of a placebo, other anti-emetic agents, or a combination of anti-emetic injections were retrieved from the databases of Ovid-MEDLINE, Ovid-EMBASE, the Cochrane Central Register of Controlled Trials, and Google Scholar. The frequency of rescue anti-emetic use (RA) and the incidence of PON, POV, and PONV (relative risk [RR]: 0.718; 95% confidence interval [CI] 0.652–0.790; I2 = 0.0, RR: 0.587; 95% CI 0.455–0.757; I2 = 63.32, RR: 0.655; 95% CI 0.532–0.806; I2 = 49.09, and RR: 0.622; 95% CI 0.552–0.700; I2 = 0.00, respectively) in the tropisetron group were lower than those in the control group; however, the incidence of complete response (CR) was higher in the tropisetron group (RR: 1.517;95% CI 1.222–1.885; I2 = 44.14). TSA showed the cumulative Z-curve exceeded both the conventional test and trial sequential monitoring boundaries for RA, PON, POV, and PONV between the tropisetron group and the control group. Thus, the prophylactic administration of tropisetron exhibited superior efficacy in the prevention of PON, POV, and PONV. Furthermore, a lower incidence of RA and a higher incidence of CR were observed with its use.


Introduction
Post-operative nausea and vomiting (PONV) are frequent and uncomfortable side effects of surgery that affect patient well-being, recovery processes, and healthcare expenses [1].PONV continues to affect 20-30% of the general surgical population and up to 80% of high-risk individuals despite the advances in the field of anesthesia and perioperative care [2,3].
Selective serotonin 5-HT 3 receptor antagonists are the cornerstone of pharmacological interventions for the management of PONV [4,5].Tropisetron, a selective serotonin 5-HT 3 antagonist with minimal affinity for 5-HT 4 receptors, is a promising anti-emetic agent owing to its efficacy and favorable safety profile [6,7].Tropisetron exhibits a prolonged duration of action, with an elimination half-life of 7 to 8 h and 30 to over 40 h in rapid and slow metabolizers, respectively.This prolonged duration of action distinguishes it from other agents in its class and renders it suitable for use in single-dose PONV prophylaxis [8,9].Furthermore, tropisetron exhibits partial agonistic activity at the α 7 -nicotinic receptors, suggesting potential modulatory effects other than serotonin receptor blockade [10][11][12].
Various clinical guidelines for the prevention and management of PONV, such as the First Consensus Guidelines for the Management of PONV [13] and the subsequent revisions (including the most recent fourth consensus guidelines issued in 2020), recommend the administration of tropisetron [1].Although evidence has been accumulated to support the inclusion of tropisetron, systematic reviews and meta-analyses exploring its efficacy have reported conflicting and variable results [14][15][16].Furthermore, the absence of updated references within these guidelines, particularly in those of systematic reviews and metaanalyses, underscores the necessity for a thorough re-evaluation of the available data to ascertain the efficacy of tropisetron.
This meta-analysis aimed to systematically examine and evaluate randomized controlled trials (RCTs) that evaluated the prophylactic efficacy of tropisetron in the prevention of PONV in patients who underwent surgery under general anesthesia.Trial sequential analysis (TSA) was used in this study to enhance statistical robustness.

Protocol and Registration
The methodological framework of this systematic review and meta-analysis with TSA was designed following recommendations from the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P).The study protocol was registered in the PROSPERO network (registration number: CRD42024372692; www.crd.york.ac.uk/Prospero (accessed on 26 April 2024)).This study was performed following the protocol recommended by the Cochrane Collaboration [17,18] and adhering to the PRISMA statement guidelines [19].

Inclusion and Exclusion Criteria
The inclusion and exclusion criteria were determined before commencing the study.RCTs that compared the efficacy profiles of tropisetron with those of controls, other antiemetic agents, or a combination of anti-emetic agents were eligible for inclusion.The PICO criteria of this study were as follows: 1.
Intervention (I): Tropisetron injection to prevent the incidence of PONV.

3.
Comparison (C): A placebo or other anti-emetic agents, such as HT3 receptor blockers, NK receptor blockers, antihistamines, anticholinergics, and steroids, or a combination of anti-emetic injections applied equally to both groups.4.

5.
Outcomes recorded during all post-operative phases were included [20].The data from the first time point were selected as the outcome of interest if data were collected at multiple time points.For instance, the data collected at 0 h were selected if the study reported data collected at 0, 2, 4, 6, and 24 h post-operatively.Data from studies that did not mention a specific time point were included to maximize the number of studies.6.
Study design (SD): RCTs.Studies that satisfied the following criteria were excluded: (1) studies that included pediatric patients; (2) studies that included patients who did not receive general anesthesia (for instance, studies that included patients who received spinal or epidural anesthesia); (3) studies that investigated the treatment effect of tropisetron; and (4) case reports, case series, editorials or letters to the editor, reviews, and animal or laboratory studies.

Systematic Search
Two researchers (KH and HHJ) performed an independent systematic literature search in April 2024.The databases of Ovid-MEDLINE, Ovid-EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL), and Google Scholar were searched to retrieve relevant articles.The Supplementary Materials provides a detailed overview of the search strategy, which incorporates free text, Medical Subject Headings (MESH), and EMTREE terms.In addition, clinical trial registries were searched to identify completed RCTs that have not been published yet.Open SIGLE was used to retrieve gray literature.The reference lists were imported into Endnote software 9.3 (Thompson Reuters, Los Angeles, CA, USA), and duplicate articles were excluded.The reference lists of the selected original articles were thoroughly reviewed to retrieve additional articles until no further relevant references could be identified.Language-or publication date-related constraints were not applied.The Supplementary File lists the search terms.

Study Selection
Two investigators (KIJ and CGJ) independently reviewed the titles and abstracts identified using the search strategy.The full text of the article was obtained and screened if the title or abstract was deemed suitable for inclusion.The full text of relevant articles selected by at least one author was procured and assessed.Papers published by the same authors, organizations, or countries were compared to minimize data duplication.Two investigators independently evaluated the articles that met the inclusion criteria.Any disagreements between the investigators were resolved by reaching a consensus via discussion.A third investigator (KH) was consulted if a consensus could not be reached.

Data Extraction
Two researchers (KIJ and HHJ) independently retrieved all inter-related data from the included studies using a pre-defined, standardized data collection form and cross-checked the findings.Any disagreements between the investigators were resolved by reaching a consensus via discussion.A third investigator (KH) was consulted if a consensus could not be reached.The spreadsheet for data extraction included the following items: (1) title; (2) name of the first author; (3) name of the journal; (4) year of publication; (5) study design; (6) clinical trial registration number; (7) country; (8) language; (9) risk of bias; (10) conflict of interests; (11) number of patients included; (12) sex of the patients; (13) age of the patients; (14) height of the patients; (15) weight of the patients; (16) duration of anesthesia or surgery; (17) physical status according to the American Society of Anesthesiologists (ASA) classification; (18) inclusion criteria; (19) exclusion criteria; (20) type of anesthesia; (21) type of surgery; (22) the agent used for anesthetic induction; (23) the agent used for the maintenance of anesthesia; (24) the type and dose of experimental drug (tropisetron or other anti-emetic agents or combination of anti-emetic agents); (28) rescue analgesic agents; (29) definitions of nausea, vomiting, and retching; (30) number of cases with PON, POV, and PONV; (31) the frequency of administering RA; and (32) CR.
The data were first extracted from tables or text.The corresponding authors of studies with incomplete or missing data were contacted to obtain the necessary information.Plot Digitizer (version 2.6.8;http://plotdigitizer.sourceforge.net(accessed on 29 March 2024)), an open-source software program, was used to extract the missing data from the available data if attempts to contact the corresponding author were unsuccessful.

Risk of Bias Assessment
Two independent investigators (CGJ and HHJ) critically appraised the quality of each study using the revised Cochrane risk of bias tool for randomized trials (RoB 2.0 version) [21].Each domain of the included studies was evaluated and rated as follows: D1, bias resulting from the randomization process; D2, bias resulting from deviations from the intended interventions; D3, bias resulting from missing outcome data; D4, bias resulting from the measurement of the outcome; and D5, bias resulting from the selection of reported results.The overall risk of bias was classified as follows: (1) low risk, the risk of bias for all domains is low; (2) high risk, the risk of bias for at least one domain is high or the risk of bias for multiple domains is of some concern; and (3) some concern, the overall judgment is neither low nor high.Any disagreements between the investigators were resolved by reaching a consensus via discussion.A third investigator (KH) was consulted if a consensus could not be reached.

Conventional Meta-Analysis
Comprehensive Meta-Analysis software (version 2.0; Englewood, NJ, USA, 2008) was used to conduct all meta-analyses.Two authors (KIJ and HHJ) independently entered all the data into the software and calculated the pooled risk ratios (RRs) and 95% confidence intervals (Cis) for each outcome.Cochran's Q test, Higgins' I 2 , τ using the DerSimonian-Laird estimator, and the prediction interval (PI) method were used to assess heterogeneity.The PI was not calculated if τ = 0.0 [22].A significance level of 10% (p < 0.1) in Cochran's Q test or an I 2 value of >50% indicated considerable heterogeneity.A fixed effects model was selected if the significance level in Cochran's Q test was >0.10 and the I 2 value was <50%.A random effects model was selected if the I 2 value was >50% [23].A sensitivity analysis was performed in the case of heterogeneous outcomes by removing one study at a time to detect changes in the results.The mean and standard deviation were calculated if the data were reported as median (P 25 -P 75 ), median (range), or mean (standard error of the mean) [17,24].The number needed to treat (NNT) was calculated based on the absolute risk reduction to estimate the overall clinical impact of the intervention.
Begg's funnel plot and Egger's linear regression test were used to assess publication bias.The asymmetry of the funnel plot or a p-value of <0.1 in Egger's test indicated the presence of publication bias.A trim-and-fill analysis was performed in such cases.The presence of publication bias was not assessed if <10 studies were included [25].

Trial Sequential Analysis
TSA was performed to determine the required information size (RIS) and assess whether the results were conclusive [26].The cumulative Z-curve was constructed using a fixed or DerSimonian and Laird approach (DL) random effects model.TSA was conducted such that the overall risk of type I errors was maintained at 5%.
A sufficient level of evidence to accept or reject the anticipated intervention effect was considered to have been reached if the cumulative Z-curve crossed the trial sequential monitoring boundary or entered the futility area; no further studies were required thereafter.The evidence to conclude was considered to be insufficient if the Z-curve did not cross any boundaries and RIS was not reached, indicating the requirement for further studies.
RIS was estimated based on the proportion of patients with an outcome in the control or other anti-emetic group (the cumulative proportion of patients with an event relative to all patients in the control or other anti-emetic group), a relative risk reduction of 20% in the tropisetron group, an alpha of 5% for all outcomes, a beta of 20%, and the observed diversity, as suggested by the trials in TSA.

Quality of Evidence
The quality of the evidence was evaluated using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) system.The quality of the evidence was assessed sequentially, the risk-benefit balance was evaluated, and the strength of the recommendations was appraised subsequently [27].The quality of the evidence was classified into the following categories: (1) high, the confidence in the effect estimate is unlikely to change with further research; (2) moderate, additional research is likely to result in significant changes in the confidence in the effect estimate and alter the estimate; (3) low, additional study is likely to result in significant changes in the confidence in the effect estimate and alter the estimate; and (4) very low, no effect estimate is certain.

Literature Search and Study Selection
A total of 478 articles published until 15 April 2024 were retrieved by searching the databases of Ovid-MEDLINE, OVID-EMBASE, CENTRAL, and Google Scholar.A manual search yielded an additional 18 articles.Among the 487 articles retained after the exclusion of duplicate articles (n = 9), 441 articles that were judged to be unsuitable after reviewing the titles and abstracts were excluded.The kappa value between the two investigators for literature selection was 0.876 at this stage.A detailed review of the full texts of the 44 remaining articles resulted in the exclusion of 16 articles.The Supplementary File lists the reasons for the exclusion of these articles.The kappa value between the two investigators for article selection was 0.851 at this stage.Thirty studies involving a total of 5175 patients were included in the systematic review and meta-analysis with TSA (Figure 1). in significant changes in the confidence in the effect estimate and alter the estimate; (3) low, additional study is likely to result in significant changes in the confidence in the effect estimate and alter the estimate; and (4) very low, no effect estimate is certain.

Literature Search and Study Selection
A total of 478 articles published until 15 April 2024 were retrieved by searching the databases of Ovid-MEDLINE, OVID-EMBASE, CENTRAL, and Google Scholar.A manual search yielded an additional 18 articles.Among the 487 articles retained after the exclusion of duplicate articles (n = 9), 441 articles that were judged to be unsuitable after reviewing the titles and abstracts were excluded.The kappa value between the two investigators for literature selection was 0.876 at this stage.A detailed review of the full texts of the 44 remaining articles resulted in the exclusion of 16 articles.The Supplementary File lists the reasons for the exclusion of these articles.The kappa value between the two investigators for article selection was 0.851 at this stage.Thirty studies involving a total of 5175 patients were included in the systematic review and meta-analysis with TSA (Figure 1).

Study Characteristics
Table 1 describes the characteristics of the 30 studies that satisfied the inclusion criteria.

Study Characteristics
Table 1 describes the characteristics of the 30 studies that satisfied the inclusion criteria.
However, when comparing between tropisetron and droperidol, the cumulative Zcurve for the incidence of POV (represented as the complete blue curve) exceeded the conventional test boundary (represented by the dotted red line) and met the trial sequential monitoring boundary (represented by the complete red curve) (Supplementary Figure S25, Table 3) [31,42,[47][48][49][50].
The cumulative Z-curve (represented by the complete blue curve) exceeded the conventional test boundary (represented by the dotted red line) but not the trial sequential monitoring boundary (represented by the complete red curve) for comparing the incidence of PONV between tropisetron and metoclopramide (Supplementary Figure S36, Table 3) [34,36,38,43,44].
TSA indicated that only 21.9% (239 of 1093 patients), 29.1% (452 of 1551 patients), 42.0% (408 of 971 patients), 17.9% (257 of 1436 patients), 24.7% (355 of 1140 patients), and 39.3% (286 of 728 patients) of the RIS was accrued for comparing the incidence of CR between tropisetron and metoclopramide (Supplementary Figure S65, Table 3) [29, 36,38], ondansetron (Supplementary Figure S66, Table 3) [36,38,44,47], dexamethasone (Supplementary Figure S67, Table 3) [35,50,51,54], droperidol (Supplementary Figure S68, Table 3) [36,37,46], droperidol when including studies comparing tropisetron combined with dexamethasone to droperidol combined with dexamethasone [53] (Supplementary Figure S69, Table 3) [36,37,46,53], and granisetron (Supplementary Figure S70, Table 3) [44,47].The figure depicts individual trials as filled squares, with the relative sample size and 95% confidence interval (CI) of the difference as a solid line.The diamond shape represents the pooled estimate and uncertainty for the combined effect.The pooled estimate indicates that the incidence of complete response in the tropisetron group is lower than that in the control group [29, 30,33,35,36,46,50,51,56].(B) Trial sequential analysis plot.The uppermost and lowermost complete red curves represent the trial sequential monitoring boundary lines for benefit and harm, respectively.The horizontal dotted red line represents the conventional boundaries for statistical significance.The triangular red lines on the right side represent the futility boundaries.The blue solid line represents the cumulative z-curve.The number on the x-axis indicates the required information size (n = 2838).The cumulative Z-curve exceeded the conventional boundary but not the trial sequential monitoring boundary, favoring the use of tropisetron over that of the control for complete response.

Sensitivity Analysis
Sensitivity analysis, performed by removing one study at a time, revealed no statistically significant difference between tropisetron and control groups in terms of the incidence of PON, POV, PONV, use of RA, and CR.However, the exclusion of the studies conducted by Capouet et al. [32], Chan et al. [33], and Ebehart et al. [35] resulted in a statistically significant difference between the tropisetron and control groups in terms of the incidence of PON when including studies [35,52,54] comparing dexamethasone to tropisetron combined with dexamethasone [29, 30,32,33,35,40,42,[48][49][50][51][52]54,56,57] (Table 3).Furthermore, there was no statistically significant difference in terms of the incidence of PON, POV, PONV, use of RA, and CR between tropisetron and other anti-emetic agents.However, statistically significant difference in terms of the incidence of PONV was observed between tropisetron and metoclopramide when excluding studies conducted by Naguib et al. [44] and Jokela et al. [34,36,38,43,44] (Table 3).

Publication Bias
Begg's funnel plot and Egger's linear regression test revealed no evidence of publication bias for the following outcomes: the incidence of PON compared with that in the control group (p = 0.21605, Supplementary Figure S71) and with that in the control group when including studies comparing dexamethasone to tropisetron combined with dexamethasone (p = 0.48744, Supplementary Figure S72); the incidence of POV compared with that in the control group (p = 0.62222, Supplementary Figure S73) and with that in the control group when including studies comparing droperidol to tropisetron combined with droperidol and those comparing dexamethasone to tropisetron combined with dexamethasone (p = 0.27420, Supplementary Figure S74); and the incidence of CR compared with that in the control group when including studies comparing dexamethasone to tropisetron combined with dexamethasone (p = 0.27420, Supplementary Figure S75).
Begg's funnel plot and Egger's linear regression test indicated the possibility of the inclusion of publication bias for the following outcomes: (1) the incidence of PONV compared with that in the control group (p = 0.00312, Supplementary Figure S76) and with that in the control group when including studies comparing dexamethasone to tropisetron combined with dexamethasone and those comparing metoclopramide to tropisetron combined with metoclopramide (p = 0.00002, Supplementary Figure S77); (2) the incidence of use of RA compared with that in the control group (p = 0.06786, Supplementary Figure S78) and with that in the control group when including studies comparing metoclopramide to tropisetron combined with metoclopramide, those comparing dexamethasone to tropisetron combined with dexamethasone, and those comparing droperidol to tropisetron combined with droperidol (p = 0.02875, Supplementary Figure S79).A subsequent trim-and-fill analysis to assess publication bias, however, did not reveal any significant changes in the findings.

Quality of the Evidence
Twenty-eight outcomes were evaluated using the GRADE system (Table 4).The quality of the pooled analysis was high for the following outcomes: the incidence of PON compared with that in the control group, the incidence of PONV compared with that in the control group, the incidence of use of RA compared with that in the control group and with that in the control group when including studies comparing other anti-emetic agents to tropisetron combined with other anti-emetic agents, and the incidence of CR compared with that in the control group and with that in the control group when including studies comparing other anti-emetic agents to tropisetron combined with other anti-emetic agents.However, the quality of the pooled analysis was low for comparing the incidence of PONV and CR between tropisetron and dexamethasone.For the rest of the outcomes, the quality of the pooled analysis was of moderate quality.ROB; Risk of bias, post-operative nausea; control1; compared with control when including studies comparing other anti-emetic agents to tropisetron combined with other anti-emetic agents, ondansetron1; compared with ondansetron when including studies comparing tropisetron combined with other anti-emetic agents to ondansetron combined with other anti-emetic agents, POV; post-operative vomiting, granisetron1; compared with granisetron when including studies comparing tropisetron combined with other anti-emetic agents to granisetron combined with other anti-emetic agents, PONV; post-operative nausea and vomiting, droperidol1; compared with droperidol when including studies comparing tropisetron combined with other anti-emetics to droperidol combined with other anti-emetics, RA; rescue anti-emetics, CR; complete response.a : considered serious as the I 2 value was >50% or P chi 2 value was <0.010; b : regarded as not serious as the trim-and-fill analysis did not reveal any statistically significant changes.

Discussion
This systematic review, which included 30 RCTs involving a total of 5175 patients, demonstrated that the prophylactic administration of tropisetron reduced the incidence of PON (GRADE: High), POV (GRADE: Moderate) and PONV (GRADE: High) and the requirement of RA (GRADE: High) and increased the incidence of CR (GRADE: High).
Conventional meta-analysis showed that the prophylactic administration of tropisetron resulted in a significant reduction in the incidence of PON, POV, and PONV, as well as the incidence of use of RA.The cumulative Z-curves for these outcomes exceeded the conventional test and trial sequential monitoring boundaries, indicating that the TSA results had achieved a sufficient level of evidence and were conclusive.Furthermore, the findings of the current meta-analysis indicated that compared with the administration of placebo agents, the administration of tropisetron resulted in a higher incidence of CR.The cumulative Z-curve for CR exceeded the conventional test boundary; however, it did not exceed the trial sequential monitoring boundary owing to the sparsity of the data.The inclusion of studies comparing other anti-emetic agents and the combination of tropisetron and other anti-emetic agents resulted in no changes (except that the cumulative Z-curve for CR exceeded the trial sequential monitoring boundary).
The effects of other anti-emetic agents did not differ from those of a combination of tropisetron and other anti-emetic agents or a combination of other anti-emetic agents.However, the incidence of POV in the tropisetron group was lower than that in the droperidol group.Furthermore, the incidence of PONV and the incidence of use of RA in the tropisetron group were lower than those in the metoclopramide group in the conventional meta-analysis.
The complex pathophysiology of PONV, which involves numerous pathways and receptors, remains unclear.However, its etiology is considered to be multifactorial and various anti-emetic agents have been used to prevent and treat PONV [1,[58][59][60][61].The area postrema, which serves as the vomiting center and orchestrates emetic responses, is located on the dorsal surface of the medulla oblongata at the caudal end of the fourth ventricle.Diverse pathways, including vagal afferent fibers in the gastrointestinal tract, input from the vestibular system, activation of the chemoreceptor trigger zone, and signaling from the forebrain, can initiate emesis [59].
The 5-HT3 receptors present in the central nervous system, peripheral nervous system, and intestinal tissues play a pivotal role in coordinating emetic processes [62,63].Therefore, 5-HT3 receptor antagonists which serve as anti-emetic agents by selectively and competitively binding to these receptors to block emetogenic signals are most frequently used anti-emetics to manage PONV.Competitive antagonists targeting serotonin 5-HT 3 receptors, which were initially used for the prevention of chemotherapy-induced emesis [64], are employed prophylactically and as an RA in the management of PONV.Not only that, but the role of 5-HT 3 receptor antagonists is expanding as recent evidence show their potential therapeutic effects on neuropsychiatric and gastrointestinal disorders [65][66][67].
Selective 5-HT3 receptor antagonists available in the United States and Europe include dolasetron, granisetron, ondansetron, palonosetron, and tropisetron [68].Among these, ondansetron was first introduced as a 5-HT3 receptor antagonist.Tropisetron, which was introduced relatively later, shares many similar mechanisms of action.Nevertheless, pharmacokinetic and pharmacodynamic variations contribute to its distinct properties and drug efficacy [69].
Tropisetron exerts additional antagonist effects on the 5-HT 4 receptors, albeit with low affinity, in addition to exhibiting 5-HT 3 antagonism; thus, it possesses theoretical advantages as an anti-emetic agent [6].Certain animal models have implicated the involvement of 5-HT 4 receptors in emetic mechanisms [70,71].Tropisetron also exhibits a longer elimination half-life of 6 to 8 h as well as an extended duration of action [72,73], which facilitates prolonged anti-emetic coverage post-operatively.This feature sets it apart from other agents in its class and renders it suitable for use in single-dose PONV prophylaxis reg-imens [9].Furthermore, tropisetron can mitigate post-operative cognitive dysfunction [74], chemotherapy-induced peripheral neuropathy [75], and fibromyalgia-associated pain [76].
Numerous systematic reviews and meta-analyses have demonstrated the efficacy of tropisetron in the prevention of the incidence of PONV [14,77,78].This has led to the inclusion of tropisentron in various clinical guidelines.However, the findings of previous studies are dated and inconsistent, indicating the requirement for updated evidence.This systematic review incorporated recent data and revealed conclusive evidence supporting the efficacy of tropisetron in the prevention of PON, POV, PONV, and RA use compared with that of a placebo.Furthermore, the effects of tropisetron were not inferior to those of other 5-HT 3 receptor antagonists, such as ondansetron and granisetron, and alternative anti-emetic agents with distinct mechanisms of action.The efficacy of tropisetron surpasses that of droperidol in the prevention of POV and that of metoclopramide in reducing the incidence of PONV and the use of RAs.
This meta-analysis has certain limitations.First, the studies included in the current meta-analysis involved patients who underwent different surgeries of varying durations.Furthermore, the use, type, and dose of anesthetics and analgesics varied across the studies.These variations may have increased the between-study heterogeneity.Second, the costeffectiveness and major side effects, including headaches, QT prolongation, and clinical outcomes beyond the observed time period (such as readmission due to PONV), could not be analyzed owing to the unavailability of data.Further studies on these outcomes are warranted.Lastly, studies that included pediatric patients were excluded to limit heterogeneity.However, in contrast to adult patients, pediatric patients possess unique characteristics that vary significantly even within the pediatric population.Therefore, further studies must be conducted to confirm the efficacy of tropisetron in the prevention of PONV in pediatric patients.Nevertheless, the findings of this systematic review and metaanalysis with TSA yielded convincing evidence indicating the superior effect of tropisetron in the prevention of episodes of vomiting.

Conclusions
This systematic review and meta-analysis with trial sequential analysis provides evidence to support that the prophylactic administration of tropisetron is effective in preventing the incidence of PON and PONV, reducing the requirement for the use of RAs, increasing the incidence of CR with high evidence certainty, and preventing POV with moderate evidence certainty, compared to controls.These findings are also supported by TSA.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jpm14080797/s1, Figure S1: Forest plot for post-operative nausea comparing tropisetron and control including studies comparing dexamethasone and tropisetron combined with dexamethasone; Figure S2

Figure 1 .
Figure 1.PRISMA flow diagram of the search for randomized controlled trials and the inclusion and exclusion criteria.

Figure 1 .
Figure 1.PRISMA flow diagram of the search for randomized controlled trials and the inclusion and exclusion criteria.

Figure 2 .
Figure 2. Incidence of post-operative nausea in the tropisetron and control groups.(A) Forest plot.The figure depicts individual trials as filled squares, with the relative sample size and 95% confidence interval (CI) of the difference represented as a solid line.The diamond shape represents the pooled estimate and uncertainty for the combined effect.The pooled estimate indicates that the incidence of post-operative nausea in the tropisetron group is lower than that in the control group [29,30,32,33,35,40,42,48-51,56,57].(B) Trial sequential analysis plot.The uppermost and lowermost complete red curves represent the trial sequential monitoring boundary lines for benefit and harm, respectively.The horizontal dotted red line represents the conventional boundaries for statistical significance.The triangular red lines on the right side represent the futility boundaries.The blue solid line represents the cumulative z-curve.The number on the x-axis indicates the required information size (n = 992).The cumulative Z-curve exceeds the conventional and trial sequential monitoring boundaries, favoring the use of tropisetron over the use of the control for the prevention of post-operative nausea.

Figure 2 .
Figure 2. Incidence of post-operative nausea in the tropisetron and control groups.(A) Forest plot.The figure depicts individual trials as filled squares, with the relative sample size and 95% confidence interval (CI) of the difference represented as a solid line.The diamond shape represents the pooled estimate and uncertainty for the combined effect.The pooled estimate indicates that the incidence of post-operative nausea in the tropisetron group is lower than that in the control group [29,30,32,33,35,40,42,48-51,56,57].(B) Trial sequential analysis plot.The uppermost and lowermost complete red curves represent the trial sequential monitoring boundary lines for benefit and harm, respectively.The horizontal dotted red line represents the conventional boundaries for statistical significance.The triangular red lines on the right side represent the futility boundaries.The blue solid line represents the cumulative z-curve.The number on the x-axis indicates the required information size (n = 992).The cumulative Z-curve exceeds the conventional and trial sequential monitoring boundaries, favoring the use of tropisetron over the use of the control for the prevention of post-operative nausea.

Fig 3 Figure 3 .
Fig 3 Figure 3. Incidence of post-operative vomiting in the tropisetron and control groups.(A) Forest plot.The figure depicts individual trials as filled squares, with the relative sample size and 95% confidence interval (CI) of the difference represented as a solid line.The diamond shape represents the pooled estimate and uncertainty for the combined effect.The pooled estimate indicates that the incidence of post-operative vomiting in the tropisetron group is lower than that in the control group [29,30,32,33,35,40,42,46,48-51,56,57].(B) Trial sequential analysis plot.The uppermost and lowermost complete red curves represent the trial sequential monitoring boundary lines for benefit and harm, respectively.The horizontal dotted red line represents the conventional boundaries for statistical significance.The triangular red lines on the right side represent the futility boundaries.The blue solid line represents the cumulative z-curve.The number on the x-axis indicates the required information size (n = 5314).The cumulative Z-curve exceeds the conventional and trial sequential monitoring boundaries, indicating a statistically significant difference favoring the use of tropisetron to reduce the incidence of post-operative vomiting before reaching the required information size.

Figure 4 .
Figure 4. Incidence of post-operative nausea and vomiting in the tropisetron and control groups.(A)Forest plot.The figure depicts individual trials as filled squares, with the relative sample size and 95% confidence interval (CI) of the difference represented as a solid line.The diamond shape represents the pooled estimate and uncertainty for the combined effect.The pooled estimate indicates that the incidence of post-operative nausea and vomiting in the tropisetron group is lower than that in the control group[28,30,[34][35][36]41,43,44,46,51].(B) Trial sequential analysis plot.The uppermost and lowermost complete red curves represent the trial sequential monitoring boundary lines for benefit and harm, respectively.The horizontal dotted red line represents the conventional boundaries for statistical significance.The triangular red lines on the right side represent the futility boundaries.The blue solid line represents the cumulative z-curve.The number on the x-axis indicates the required information size (n = 2380).The cumulative Z-curve exceeds the conventional and trial sequential monitoring boundaries, favoring the use of tropisetron over that of the control for the prevention of the incidence of post-operative nausea and vomiting before reaching the required information size.

Fig 5 Figure 5 .
Fig 5 Figure 5. Use of rescue anti-emetic agents in the tropisetron and control groups.(A) Forest plot.The figure depicts individual trials as filled squares, with the relative sample size and 95% confidence interval (CI) of the difference represented as a solid line.The diamond shape represents the pooled estimate and uncertainty for the combined effect.The pooled estimate indicates that the use of rescue anti-emetic agents in the tropisetron group is lower than that in the control group [29,30,32,33,35,36, 40,43,44,46,48-51,56,57].(B) Trial sequential analysis plot.The uppermost and lowermost complete red curves represent the trial sequential monitoring boundary lines for benefit and harm, respectively.The horizontal dotted red line represents the conventional boundaries for statistical significance.The triangular red lines on the right side represent the futility boundaries.The blue solid line represents the cumulative z-curve.The number on the x-axis indicates the required information size (n = 1027).The cumulative Z-curve exceeded the conventional and trial sequential monitoring boundaries, favoring the use of tropisetron over that of the control to reduce the use of rescue anti-emetic agents.

Fig 6 Figure 6 .
Fig 6 Figure 6.The incidence of complete response in the tropisetron and control groups.(A) Forest plot.The figure depicts individual trials as filled squares, with the relative sample size and 95% confidence interval (CI) of the difference as a solid line.The diamond shape represents the pooled estimate and uncertainty for the combined effect.The pooled estimate indicates that the incidence of complete response in the tropisetron group is lower than that in the control group [29,30,33,35,36,46,50,51,56].(B) Trial sequential analysis plot.The uppermost and lowermost complete red curves represent the trial sequential monitoring boundary lines for benefit and harm, respectively.The horizontal dotted red line represents the conventional boundaries for statistical significance.The triangular red lines on the right side represent the futility boundaries.The blue solid line represents the cumulative z-curve.The number on the x-axis indicates the required information size (n = 2838).The cumulative Z-curve exceeded the conventional boundary but not the trial sequential monitoring boundary, favoring the use of tropisetron over that of the control for complete response.

:
Trial sequential analysis plot for post-operative nausea comparing tropisetron and control including studies comparing dexamethasone and tropisetron combined with dexamethasone; FigureS3: Forest plot for post-operative nausea comparing tropisetron and metoclopramide; FigureS4: Forest plot for post-operative nausea comparing tropisetron and ondansetron; FigureS5: Forest plot for post-operative nausea comparing tropisetron and ondansetron including studies comparing tropisetron combined with droperidol and ondansetron combined with droperidol; FigureS6: Forest plot for post-operative nausea comparing tropisetron and dexamethasone; FigureS7: Forest plot for post-operative nausea comparing tropisetron and droperidol; FigureS8: Forest plot for post-operative nausea comparing tropisetron and granisetron; FigureS9: Trial sequential analysis plot for post-operative nausea comparing tropisetron and ondansetron; FigureS10: Trial sequential analysis plot for post-operative nausea comparing tropisetron and ondansetron when including study comparing tropisetron combined with droperidol and tropisetron combined with ondansetron; FigureS11: Trial sequential analysis plot for post-operative nausea comparing tropisetron and dexamethasone; Figure S12: Trial sequential analysis plot for post-operative nausea comparing tropisetron and granisetron; Figure S13: Forest plot for post-operative vomiting comparing tropisetron and control groups including studies comparing droperidol and tropisetron combined with droperidol and comparing dexamethasone and tropisetron combined with dexamethasone; Figure S14: Trial sequential analysis plot for post-operative vomiting comparing tropisetron and control including studies comparing droperidol and tropisetron combined with droperidol and comparing dexamethasone and tropisetron combined with dexamethasone; Figure S15: Forest plot for post-operative vomiting comparing tropisetron and droperidol; Figure S16: Forest plot for post-operative vomiting comparing tropisetron and metoclopramide; Figure S17: Forest plot for post-operative vomiting comparing tropisetron and ondansetron; Figure S18: Forest plot for postoperative vomiting comparing tropisetron and ondansetron including studies comparing tropisetron combined with droperidol and ondansetron combined with droperidol; Figure S19: Forest plot for post-operative vomiting comparing tropisetron and dexamethasone; Figure S20: Forest plot for postoperative vomiting comparing tropisetron and granisetron; Figure S21: Forest plot for post-operative vomiting comparing tropisetron and granisetron including studies comparing tropisetron combined with droperidol and granisetron combined with droperidol; Figure S22: Trial sequential analysis plot for post-operative vomiting comparing tropisetron and ondansetron; Figure S23: Trial sequential analysis plot for post-operative vomiting comparing tropisetron and ondansetron when including studies comparing tropisetron combined with droperidol and ondansetron combined with droperidol; Figure S24: Trial sequential analysis plot for post-operative vomiting comparing tropisetron and dexamethasone; Figure S25: Trial sequential analysis plot for post-operative vomiting comparing tropisetron and droperidol, Figure S26: Trial sequential analysis plot for post-operative vomiting comparing tropisetron and granisetron; Figure S27: Trial sequential analysis plot for post-operative vomiting comparing tropisetron and graniseetron when including studies comparing tropisetron combined with droperidol and granisetron combined with droperidol; Figure S28: Forest plot for post-operative nausea and vomiting comparing tropisetron and control including studies comparing dexamethasone and tropisetron combined with dexamethasone and comparing metoclopramide and tropisetron combined with metoclopramide; Figure S29: Trial sequential analysis plot for postoperative nausea and vomiting comparing tropisetron and control including studies comparing dexamethasone and tropisetron combined with dexamethasone and comparing metoclopramide and tropisetron combined with metoclopramide; Figure S30: Forest plot for post-operative nausea and vomiting comparing tropisetron and metoclopramide; Figure S31: Forest plot for post-operative nausea and vomiting comparing tropisetron and ondansetron; Figure S32: Forest plot for postoperative nausea and vomiting comparing tropisetron and dexamethasone; Figure S33: Forest plot for post-operative nausea and vomiting comparing tropisetron and droperidol; Figure S34: Forest plot for post-operative nausea and vomiting comparing tropisetron and droperidol including studies comparing tropisetron combined with dexamethasone with droperidol combined with dexamethasone; Figure S35: Forest plot for post-operative nausea and vomiting comparing tropisetron and granisetron; Figure S36: Trial sequential analysis plot for post-operative nausea and vomiting comparing tropisetron and metoclopramide; Figure S37: Trial sequential analysis plot for post-operative nausea and vomiting comparing tropisetron and ondansetron; Figure S38: Trial sequential analysis plot for post-operative nausea and vomiting comparing tropisetron and dexamethasone; Figure S39: Trial sequential analysis plot for post-operative nausea and vomiting comparing tropisetron and droperidol; Figure S40: Trial sequential analysis plot for post-operative nausea and vomiting comparing tropisetron and droperidol when including study comparing tropisetron combined with dexamethasone with droperidol combined with dexamethasone; Figure S41: Trial sequential analysis plot for post-operative nausea and vomiting comparing tropisetron and granisetron; FigureS42: Forest plot for use of rescue anti-emetics comparing tropisetron and control groups including studies comparing metoclopramide and tropisetron combined with metoclopramide and comparing dexamethasone and tropisetron combined with dexamethasone and comparing droperidol and tropisetron combined with droperidol; FigureS43: Trial sequential analysis plot for use of rescue anti-emetics comparing tropisetron and control when including studies comparing metoclopramide and tropisetron combined with metoclopramide, dexamethasone, and tropisetron combined with dexamethasone, droperidol, and tropisetron combined with droperidol; FigureS44: Forest plot for use of rescue anti-emetics comparing tropisetron and metoclopramide; FigureS45: Forest plot for use of rescue anti-emetics comparing tropisetron and ondansetron; FigureS46: Forest plot for use of rescue anti-emetics comparing tropisetron and ondansetron when including studies comparing tropisetron combined with droperidol and ondansetron combined with droperidol; FigureS47: Forest plot for use of rescue anti-emetics comparing tropisetron and dexamethasone; FigureS48: Forest plot for use of rescue anti-emetics comparing tropisetron and droperidol; FigureS49: Forest plot for use of rescue anti-emetics comparing tropisetron and granisetron; FigureS50: Forest plot for use of rescue anti-emetics comparing tropisetron and granisetron including studies comparing tropisetron combined with droperidol and granisetron combined with droperidol; FigureS51: Trial sequential analysis plot for use of rescue anti-emetics comparing tropisetron and metoclopramide; FigureS52: Trial sequential analysis plot for use of rescue anti-emetics comparing tropisetron and ondansetron; FigureS53: Trial sequential analysis plot for use of rescue anti-emetics comparing tropisetron and ondansetron including studies comparing tropisetron combined with droperidol and ondansetron combined with droperidol; FigureS54: Trial sequential analysis plot for use of rescue anti-emetics comparing tropisetron and dexamethasone; FigureS55: Trial sequential analysis plot for use of rescue anti-emetics comparing tropisetron and droperidol; FigureS56: Trial sequential analysis plot for use of rescue anti-emetics comparing tropisetron and granisetron including studies comparing tropisetron combined with droperidol and granisetron combined with droperidol; FigureS57: Forest plot for complete response comparing tropisetron and control including studies comparing dexamethasone and tropisetron combined with dexamethasone; FigureS58: Trial sequential analysis plot for complete response comparing tropisetron and control including studies comparing dexamethasone and tropisetron combined with dexamethasone; FigureS59: Forest plot for complete response comparing tropisetron and metoclopramide; FigureS60: Forest plot for complete response comparing tropisetron and ondansetron; FigureS61: Forest plot for complete response comparing tropisetron and dexamethasone; FigureS62: Forest plot for complete response comparing tropisetron and droperidol; FigureS63: Forest plot for complete response comparing tropisetron and droperidol including studies comparing tropisetron combined with dexamethasone with droperidol combined with dexamethasone; FigureS64: Forest plot for complete response comparing tropisetron and granisetron; FigureS65: Trial sequential analysis plot for complete response comparing tropisetron and metoclopramide; FigureS66: Trial sequential analysis plot for complete response comparing tropisetron and ondansetron; FigureS67: Trial sequential analysis plot for complete response comparing tropisetron and dexamethasone; FigureS68: Trial sequential analysis plot for complete response comparing tropisetron and droperidol; FigureS69: Trial sequential analysis plot for complete response comparing tropisetron and droperidol when including studies comparing tropisetron combined with dexamethasone with droperidol combined with dexamethasone; FigureS70: Trial sequential analysis plot for complete response comparing tropisetron, and granisetron.
FigureS71: Funnel plot for postoperative nausea compared with control; FigureS72: Funnel plot for postoperative nausea compared with control including studies comparing dexamethasone and tropisetron combined with dexamethasone; FigureS73: Funnel plot for postoperative nausea compared with control including studies comparing dexamethasone and tropisetron combined with dexamethasone; FigureS74: Funnel plot for postoperative vomiting compared with control including studies comparing droperidol and tropisetron combined with droperidol and comparing dexamethasone and tropisetron combined with dexamethasone; FigureS75: Funnel plot for complete response compared with control including studies comparing dexamethasone and tropisetron combined with dexamethasone; FigureS76: Funnel plot for post-operative nausea and vomiting compared with control; FigureS77: Funnel plot for post-operative nausea and vomiting compared with control including studies comparing dexamethasone and tropisetron combined with dexamethasone and comparing metoclopramide and tropisetron combined with metoclopramide; FigureS78: Funnel plot for use of rescue anti-emetics compared with control; FigureS79.Funnel plot for use of rescue anti-emetics compared with control including studies comparing metoclopramide and tropisetron combined with metoclopramide, dexamethasone and tropisetron combined with dexamethasone, droperidol and tropisetron combined with droperidol.

Table 1 .
Study characteristics of included studies.

Table 2 .
Risk of bias.

, Year Bias Arising from the Randomization Process Bias Due to Deviations from Intended Intervention Bias Due to Missing Outcome Data Bias in Measurement of the Outcome Bias in Selection of the Reported Results Overall Bias
)[29,30,32,33,35,40,42,48- 52,54,56,57].

Table 3 .
Summary of the meta-analysis.

Table 4 .
GRADE evidence quality for each outcome.