Ketamine and esketamine for postoperative analgesia in gynaecological surgery

We will follow MECIR (Methodological Expectations for Cochrane Intervention Reviews) guidance when conducting the review [ 37 ], and PRISMA 2020 for reporting the review process and our findings [ 38 ]. We will include randomised controlled trials (RCTs) comparing the efficacy and safety of a subanaesthetic dosage of ketamine/esketamine, administered as an adjunct to perioperative analgesia, with the same analgesic strategy not including ketamine/esketamine, or administering a placebo, to reduce postoperative pain in participants undergoing gynaecological surgical procedures. We will consider and analyse placebo study arms and study arms not including ketamine/esketamine administration (without placebo) together. Although placebo and no‐treatment controls may estimate different effects in some clinical settings, all included studies will administer ketamine/esketamine in participants under general anaesthesia. Consequently, participants will be unlikely to be aware of treatment allocation at the time of administration, and expectancy of placebo effects related to knowledge of treatment is expected to be minimal. Furthermore, in the perioperative setting, administration of intravenous fluids and infusions is routine, which further reduces the likelihood of unblinding. We will include cluster‐randomised trials but will exclude quasi‐RCTs, in which participants are allocated to intervention groups using a non‐random method (i.e. a method that is predictable), such as allocation by date of birth, day of the week, medical record number, or order of enrolment. For cross‐over trials, we will include data from the first period only, in order to avoid potential bias arising from carry‐over effects, which may invalidate comparisons between intervention conditions. We will include studies reported as full text, studies published as abstracts only, and unpublished data. This review will include studies involving participants assigned female at birth who are undergoing any type of gynaecological surgery. Gynaecological surgery comprises a heterogeneous group of procedures that diagnose, treat, or manage conditions of the female reproductive system (uterus, ovaries, fallopian tubes, cervix, vagina, and vulva) and related pelvic organs, performed for benign or malignant conditions (see Table 1 in Description of the condition for more details). We will only include adult women over 18 years of age. We will exclude studies including only paediatric participants. For studies including paediatric and adult participants, we will only include data pertaining to participants who are 18 years of age or older. We will contact study authors to obtain necessary adult participants' data if they are not provided. For trials including a mixed population, we will preferentially extract data for the subgroup of participants meeting our inclusion criteria. We will contact study authors where necessary to obtain such data. If subgroup data are unavailable, we will consider including the study if the majority of participants meet the inclusion criteria, while assessing the potential impact of indirectness. We will explore the influence of these studies in sensitivity analyses. We will not consider the following as exclusion criteria: age (when higher than 18 years), comorbidities, diagnosis and ethnicity. We will include studies that evaluated ketamine/esketamine administered intravenously (IV) at a subanaesthetic dose as an adjunct to perioperative analgesia compared to the same analgesia strategy not including ketamine/esketamine or administering placebo. We will include studies administering ketamine/esketamine as a single or repeated IV bolus, as a continuous IV infusion, or as a bolus followed by a continuous infusion. We will include any dose considered to be subanaesthetic: for ketamine, < 1 mg/kg if single bolus, < 1 mg/kg per bolus if repeated bolus, and < 1 mg/kg/h if continuous infusion; for esketamine, < 0.5 mg/kg if single bolus, < 0.5 mg/kg per bolus if repeated bolus, and < 0.5 mg/kg/h if continuous infusion. We will exclude studies involving non‐surgical applications of ketamine/esketamine or those administering an anaesthetic dose. We will include studies where other pain drugs, like opioids, paracetamol, non‐steroid anti‐inflammatory drugs (NSAIDs) and clonidine, are co‐administered in the intervention group together with ketamine/esketamine, so long as the same therapy is also administered to the control group. We will exclude studies that compare an analgesia protocol including ketamine/esketamine with another analgesia protocol with a different intervention drug (for example, lidocaine infusion and specific opioids). We will exclude studies that compare an analgesia protocol including IV ketamine/esketamine with a different analgesia method, such as spinal or epidural analgesia. In the case that a study has more than two arms and compares different interventions with a control group (e.g. ketamine/esketamine in one group, lidocaine infusion in a second group and a control group), we will extract data from the ketamine/esketamine group and from the control group. Our critical outcomes include both benefits and potential harms associated with the intervention. The benefit‐related outcomes for the reduction of postoperative pain have been selected in accordance with the systematic review by Pushpanathan and colleagues [ 39 ], found through the Core Outcome Measures in Effectiveness Trials (COMET) Initiative website [ 40 ]. Peak postoperative pain intensity during the first 24 hours after surgery Measured as a numeric rating scale (NRS) for pain, or Measured as a visual analogue scale (VAS) for pain Opioid consumption during the first 24 hours after surgery Any oral and parenteral administered opioids Proportion of serious adverse events Number of participants with at least one serious adverse event within 24 hours Peak postoperative pain intensity during the first 24 hours after surgery Measured as a numeric rating scale (NRS) for pain, or Measured as a visual analogue scale (VAS) for pain Opioid consumption during the first 24 hours after surgery Any oral and parenteral administered opioids Proportion of serious adverse events Number of participants with at least one serious adverse event within 24 hours We will use the definition of serious adverse events (SAEs) as reported by the study authors. If no definition is provided, we will define a serious adverse event as any untoward medical occurrence that resulted in death, was life‐threatening, required hospitalisation or prolongation of existing hospitalisation, or resulted in persistent or significant disability, in accordance with the International Committee of Harmonization of Good Clinical Practice (ICH‐GCP) definition [ 41 ]. Our important outcomes include additional pain and recovery quality measurements, and the proportion of participants with adverse events. Postoperative pain at one to two hours (or in the Post‐Anesthesia Care Unit – PACU), and at 6, 12 and 24 hours after surgery Measured as NRS, or Measured as VAS Postoperative pain measured beyond 24 hours (up to six months after surgery) Measured as NRS, or Measured as VAS Measured as the McGill Pain Questionnaire for chronic pain score [ 42 ] Measured with other questionnaires and scales for chronic pain The proportion of participants with one or more of the predefined adverse events of interest (see below) The proportion of participants with anxiety in the PACU or during the first 24 hours postoperatively Sleep quality and sleep structure during hospital admission Quality of Recovery score (QoR) assessed within a week from surgery [ 43 ] Postoperative pain at one to two hours (or in the Post‐Anesthesia Care Unit – PACU), and at 6, 12 and 24 hours after surgery Measured as NRS, or Measured as VAS Postoperative pain measured beyond 24 hours (up to six months after surgery) Measured as NRS, or Measured as VAS Measured as the McGill Pain Questionnaire for chronic pain score [ 42 ] Measured with other questionnaires and scales for chronic pain The proportion of participants with one or more of the predefined adverse events of interest (see below) The proportion of participants with anxiety in the PACU or during the first 24 hours postoperatively Sleep quality and sleep structure during hospital admission Quality of Recovery score (QoR) assessed within a week from surgery [ 43 ] Predefined adverse events of interest include hypertension, tachycardia, sedation, hallucinations or psychomimetic effects, nausea and vomiting, and hypersalivation. We will extract adverse events as defined by the study authors. Where definitions are not reported, we will use the following criteria: hypertension (systolic arterial pressure > 180 mmHg or diastolic > 100 mmHg), tachycardia (heart rate > 120 beats per minute), sedation, defined as a Richmond Agitation–Sedation Scale (RASS) < −1, with other events defined according to clinical reporting. Predefined adverse events that meet the criteria for SAEs will be reported as SAEs only and not counted separately among predefined adverse events. We will search for studies using the guidelines described in Chapter 4 of the Cochrane Handbook for Systematic Reviews of Interventions [ 44 ]. No restrictions on language, publication year, or publication status will apply. We will search the following sources from inception to the present. Cochrane Library via Cochrane Library Online (CENTRAL, inception to present) PubMed (National Library of Medicine, 1996 to present) Embase (Elsevier, 1947 to preset) CINAHLComplete (Cumulative Index to Nursing and Allied Health Literature; EbscoHost, 1961 to present) Web of Science Core collection (Clarivate Analytics, inception to present) Cochrane Library via Cochrane Library Online (CENTRAL, inception to present) PubMed (National Library of Medicine, 1996 to present) Embase (Elsevier, 1947 to preset) CINAHLComplete (Cumulative Index to Nursing and Allied Health Literature; EbscoHost, 1961 to present) Web of Science Core collection (Clarivate Analytics, inception to present) We will also search databases of ongoing trials on ClinicalTrials.gov (www.clinicaltrials.gov/) and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (apps.who.int/trialsearch). We will provide key information about potentially relevant ongoing studies in a table of 'Characteristics of ongoing studies'. We developed a search strategy for MEDLINE, which we will adapt for the other electronic databases ( Supplementary material 1 ). This search strategy was reviewed by a medical librarian. We will perform a complete updated search on all specified databases four weeks before we submit the final review draft for editorial approval. This search will allow us to identify possible postpublication amendments (such as errata, corrigenda and retractions). In addition, we will perform a targeted search on the articles included in the review to identify possible expressions of concerns and retractions. We will try to identify other potentially eligible trials or ancillary publications by searching the reference lists of included trials, related systematic or other reviews, and clinical study reports (CSRs) and other regulatory documents, such as the US Food and Drug Administration (FDA) Drug Approval Packages. We will also search for potentially eligible trials on the websites of the American College of Obstetricians and Gynaecologists (www.acog.org) and the American Society of Anaesthesiologists (www.asahq.org). We will search OpenGrey (opengrey.eu/) and Google Scholar for grey literature. We will also search for abstracts presented at relevant conference proceedings. All identified citations will be uploaded to EndNote in order to automatically eliminate duplicates [ 45 ], then directly uploaded to Covidence for selection of studies [ 46 ]. The designated authors (two main review authors and the authors assigned to solve possible disagreement) will have access to Covidence. Two review authors will assess the abstract and title of every retrieved citation for potential eligibility, independently and in duplicate. We will retrieve the full texts for all citations judged by both authors to be potentially eligible, and will exclude studies judged not eligible by both authors. When there is a disagreement between the two assessing authors, the citation will be referred to a third author for assessment. If the third author is not able to rule on inclusion, the citation will be retrieved in full‐text and assessed by the three authors, who will then evaluate its eligibility. The two authors may also express doubt about the eligibility of a citation. If one or both of the review authors express a doubt, a third author will assess the citation. If the third author is not able to rule on inclusion, the citation will be retrieved in full‐text and assessed by the three authors, who will then evaluate its eligibility. Before starting the selection process, the two review authors responsible for selection will conduct calibration exercises to ensure the validity of the process. Consistency checks will be made by the lead author (SM). The authors' group can read retrieved full‐text studies in English, Swedish, Italian, Spanish and French. Retrieved studies in other languages will be translated by a commissioned interpreter. We will list the articles that we exclude after full‐text assessment and the reasons for their exclusion in the 'Characteristics of excluded studies' table. For studies that require further clarification to determine their eligibility, we will contact the study authors via email to obtain this information. We will present the study selection process in an adapted PRISMA flowchart [ 38 ]. We will report the total number of articles excluded after full‐text review in the PRISMA flowchart, along with the details of key excluded studies. Two review authors will extract relevant data, independently and in duplicate, using standard data extraction forms. They will resolve any disagreements by discussion or, if required, by involving a third review author. We will seek relevant missing information on the trial from the primary author(s) of the article, if required. We will extract the following study characteristics. Lead author, date, and country in which the trial was conducted. Methodology: study design, method of randomisation, intention‐to‐treat and per‐protocol populations, and loss to follow‐up. Participant characteristics: size of participant population, age, ethnicity, participant inclusion and exclusion criteria and type of intervention. Interventions: details regarding surgical intervention, surgical approach, type of anaesthesia performed, types and doses of analgesic drugs, method of administration of analgesic drugs, method of administration and dosage of the intervention drug, administration and type of placebo in the control group. Outcomes: critical and important outcomes specified in the study, time points for outcome assessment, information on measurement tools or scales and the direction of such tools or scales. Funding source of trial and declarations of interest of trialists. Lead author, date, and country in which the trial was conducted. Methodology: study design, method of randomisation, intention‐to‐treat and per‐protocol populations, and loss to follow‐up. Participant characteristics: size of participant population, age, ethnicity, participant inclusion and exclusion criteria and type of intervention. Interventions: details regarding surgical intervention, surgical approach, type of anaesthesia performed, types and doses of analgesic drugs, method of administration of analgesic drugs, method of administration and dosage of the intervention drug, administration and type of placebo in the control group. Outcomes: critical and important outcomes specified in the study, time points for outcome assessment, information on measurement tools or scales and the direction of such tools or scales. Funding source of trial and declarations of interest of trialists. Two review authors will independently manually transfer extracted data to Review Manager (RevMan) [ 47 ]. Any discrepancies in cross‐checking will be investigated and resolved through discussion or through consultation with a third review author. Two authors will independently evaluate the risk of bias for each eligible study. Any discrepancies between the two authors will be resolved through discussion, and if necessary, by involving a third author. We will evaluate risk of bias at the outcome level using the Cochrane Risk of Bias 2 (RoB 2) tool, focusing on the effect of assignment to the intervention, consistent with an intention‐to‐treat (ITT) approach [ 48 , 49 , 50 ]. We will assess the risk of bias for all outcomes included in the summary of findings table (see Certainty of the evidence assessment ). RoB 2 assesses the following domains. Bias related to the randomisation process Bias arising from departures from the intended interventions Bias due to incomplete outcome data Bias in outcome measurement Bias in selective reporting of results Bias related to the randomisation process Bias arising from departures from the intended interventions Bias due to incomplete outcome data Bias in outcome measurement Bias in selective reporting of results Given the characteristic clinical effects of ketamine and esketamine, we will not assume that blinding was successful solely on the basis of statements that participants, personnel, or outcome assessors were blinded. During the RoB 2 assessment, we will evaluate whether the study authors reported any evidence suggesting compromised blinding or potential unblinding, including recognition of treatment allocation, differential co‐interventions, or adverse effects that may have revealed allocation status. To assess selective outcome reporting, we will compare outcomes specified in trial protocols, registry entries and methods sections with those reported in the results. Each domain will be rated as presenting a low risk of bias, high risk of bias, or some concerns. These judgements will be guided by the signalling questions outlined in the RoB 2 framework, and an overall risk of bias judgement will be determined for each outcome. For any cluster‐randomised controlled trials included in the review, we will assess the risk of bias using the RoB 2 extension for cluster‐randomised trials (RoB 2 CRT), which incorporates an additional domain addressing bias associated with the timing of participant identification and recruitment. Loss of entire clusters will be considered under the domain of missing outcome data. We will express continuous data as mean differences (MDs) with 95% confidence intervals (CIs) when all studies report the outcome using the same scale, and as standardised mean differences (SMDs) when studies report the outcome using different scales. For dichotomous outcomes, we will use risk ratios (RRs) with 95% CIs. When studies report medians and interquartile ranges (IQRs), we will assume the median approximates the mean only when the data appear to be symmetrically distributed. In cases where skewness is suspected or medians are reported due to non‐normal distributions, we will apply the Wan formula for imputing missing mean values based on the lower quartile, median and upper quartile as recommended in Section 6.5.2.9 of the Cochrane Handbook for Systematic Reviews of Interventions [ 51 ]. The unit of analysis will be the individual participant. For cluster‐randomised trials, if clustering has not been accounted for in the analysis, we will adjust for clustering effect using the reported intracluster correlation coefficient (ICC). For studies reporting repeated measurements of outcomes at multiple time points, we will select the time points that mirror our review's outcomes and analyse them separately, one time point per analysis. We will also consider each other time point reported in the studies as a separate outcome. If more than one intervention arm from a multiple‐arm study is eligible for inclusion in the same meta‐analysis, we will combine the relevant intervention arms to create a single pairwise analysis. We will omit those groups from multiple‐arm studies that are irrelevant to our comparison of interest. For example, arms that compare the intervention drug (ketamine/esketamine) with a different intervention (e.g. spinal analgesia or lidocaine continuous infusion). We will contact study authors to request missing outcome data where feasible. For data analysis, we will use an available‐case analysis approach, including only participants with observed outcome data at the relevant time point, and will not impute missing outcome data. When studies report multiple analyses, we will extract results based on the ITT principle where available, as this most closely reflects the effect of assignment to the intervention. If ITT analyses are not available, we will extract the analysis judged closest to an ITT approach (e.g. modified ITT), and otherwise use per‐protocol analyses, noting this as a potential source of bias. We recognise that reported ITT analyses may involve different methods for handling missing data (e.g. imputation or complete‐case analysis), and we will record these methods and consider their potential impact when assessing risk of bias and interpreting results. We will assess the robustness of findings in sensitivity analyses by excluding studies at high risk of bias due to missing outcome data or with substantial proportions of missing data (e.g. > 20%). We will use several strategies to mitigate the risk of non‐reporting bias. To identify unpublished or selectively reported data, we will search trial registries, as well as clinical study reports and regulatory documents (see Searching other resources ). We will have prespecified eligible results for all meta‐analyses, including outcome domain, time points, and outcome measures. These definitions will be applied during study selection and when assessing the risk of non‐reporting bias. For studies with missing results for outcomes included in the meta‐analyses, we will compare published reports with available protocols, statistical analysis plans, and trial registry records to determine whether the outcome was prespecified but not reported. When results are not reported, we will investigate the reasons for non‐reporting, including whether technical or methodological issues prevented outcome measurement or analysis. We will use the Risk of Bias due to Missing Evidence (RoB‐ME) tool to assess the risk of bias due to missing evidence arising from non‐reporting of study results [ 52 ]. When at least 10 studies are included in a meta‐analysis for a given outcome, we will assess small‐study effects using funnel plots. Funnel plot asymmetry will be evaluated using the Harbord test for dichotomous outcomes [ 53 ], and the Egger regression test for continuous outcomes [ 54 ]. To identify unpublished or selectively reported data, we will search trial registries, as well as clinical study reports and regulatory documents (see Searching other resources ). We will have prespecified eligible results for all meta‐analyses, including outcome domain, time points, and outcome measures. These definitions will be applied during study selection and when assessing the risk of non‐reporting bias. For studies with missing results for outcomes included in the meta‐analyses, we will compare published reports with available protocols, statistical analysis plans, and trial registry records to determine whether the outcome was prespecified but not reported. When results are not reported, we will investigate the reasons for non‐reporting, including whether technical or methodological issues prevented outcome measurement or analysis. We will use the Risk of Bias due to Missing Evidence (RoB‐ME) tool to assess the risk of bias due to missing evidence arising from non‐reporting of study results [ 52 ]. When at least 10 studies are included in a meta‐analysis for a given outcome, we will assess small‐study effects using funnel plots. Funnel plot asymmetry will be evaluated using the Harbord test for dichotomous outcomes [ 53 ], and the Egger regression test for continuous outcomes [ 54 ]. We will not use funnel plots and statistical tests as sole evidence of non‐reporting bias, but will consider findings alongside information from trial registries, protocols, statistical analysis plans, and ROB‐ME assessments. We will present the characteristics of included studies, outcome data, and risk of bias assessments in structured tables. Where data permit, we will conduct meta‐analyses for all outcomes of interest. We will use a random‐effects model for meta‐analysis, as we anticipate clinical and methodological heterogeneity across studies. We will estimate between‐study variance using the Restricted Maximum Likelihood (REML) estimator. To calculate confidence intervals for the pooled effect, we will use the Hartung‐Knapp‐Sidik‐Jonkman (HKSJ) method when there are at least three studies and the estimate of heterogeneity is greater than zero [ 55 ]. In other situations (e.g. meta‐analyses including only two studies, or when the estimate of heterogeneity is zero), we will use the Wald‐type method. We will organise the narrative synthesis by outcome and, in the case of postoperative pain, by time point. The narrative synthesis will describe the direction, magnitude, and consistency of effects across studies, taking into account study design, sample size, and risk of bias. We will avoid vote counting based solely on statistical significance and will interpret findings in the context of methodological quality and clinical relevance. For the overall meta‐analysis, we will use a qualitative assessment of forest plots, the Chi 2 test for heterogeneity, and the I 2 statistic to assess the magnitude of heterogeneity. We will consider an I 2 statistic between 0% and 40% as potentially not important, 40% to 60% as moderate, 50% to 90% as substantial, and 75% to 100% as considerable heterogeneity. We plan to carry out the following subgroup analyses, where possible. Type of surgery: we will follow the authors' definitions of major and minor surgery when reported. When no definition is reported, and as definitions of ‘major’ and ‘minor’ gynaecological surgery are not standardised, we will categorise procedures based on surgical complexity and extent of tissue dissection. Major surgery will be defined as procedures involving substantial tissue manipulation or entry into the abdominal or pelvic cavity (e.g. hysterectomy, myomectomy, oophorectomy) and all procedures performed as laparotomy; whereas minor surgery will include less invasive or primarily diagnostic procedures with limited tissue disruption (e.g. hysteroscopy, dilatation and curettage, cervical procedures). We will also consider the time of surgery (less than or more than two hours) when reported, to identify more and less extensive procedures. This subgroup analysis will allow us to explore the effect of ketamine/esketamine on postoperative pain associated with different degrees of surgical invasiveness. This distinction is particularly relevant for the population under study, as previous evidence suggests that, in specific contexts, minor gynaecological procedures may be associated with greater postoperative pain than major surgery, likely due to factors such as the underlying condition and the presence of chronic preoperative pain. The presence of preoperative chronic pain or preoperative opioid use. We will perform a subgroup analysis comparing participants with preoperative chronic pain or opioid use and participants who do not have preoperative pain and do not consume opioids, as defined by the authors. When not defined, we will include participants with long‐term (over six months) or recurring opioid consumption (e.g. women consuming opioids during menstruation). This analysis may identify a subgroup of participants who derive a specific benefit from the intervention. Type of intervention drug: we will perform a subgroup analysis comparing studies using racemic ketamine with those using esketamine. This analysis aims to explore potential differences between the racemate and the S‐enantiomer in terms of both analgesic efficacy and safety profile. We will categorise studies according to low‐dose and high‐dose regimens. A low dose for ketamine is < 0.4 mg/kg/h; in studies using a single bolus or repeated boluses, a threshold of < 0.4 mg/kg per bolus will be applied. For esketamine, a low dose is < 0.2 mg/kg/h; in studies using a single bolus or repeated boluses, a threshold of < 0.2 mg/kg per bolus will be applied. A high dose of ketamine is ≥ 0.4 mg/kg/h (or ≥ 0.4 mg/kg per bolus), and for esketamine, it is ≥ 0.2 mg/kg/h (or ≥ 0.2 mg/kg per bolus). As the most effective subanaesthetic dose of ketamine/esketamine for postoperative pain control has not been established, this subgroup analysis will explore potential dose–response differences in both efficacy and safety. Type of surgery: we will follow the authors' definitions of major and minor surgery when reported. When no definition is reported, and as definitions of ‘major’ and ‘minor’ gynaecological surgery are not standardised, we will categorise procedures based on surgical complexity and extent of tissue dissection. Major surgery will be defined as procedures involving substantial tissue manipulation or entry into the abdominal or pelvic cavity (e.g. hysterectomy, myomectomy, oophorectomy) and all procedures performed as laparotomy; whereas minor surgery will include less invasive or primarily diagnostic procedures with limited tissue disruption (e.g. hysteroscopy, dilatation and curettage, cervical procedures). We will also consider the time of surgery (less than or more than two hours) when reported, to identify more and less extensive procedures. This subgroup analysis will allow us to explore the effect of ketamine/esketamine on postoperative pain associated with different degrees of surgical invasiveness. This distinction is particularly relevant for the population under study, as previous evidence suggests that, in specific contexts, minor gynaecological procedures may be associated with greater postoperative pain than major surgery, likely due to factors such as the underlying condition and the presence of chronic preoperative pain. The presence of preoperative chronic pain or preoperative opioid use. We will perform a subgroup analysis comparing participants with preoperative chronic pain or opioid use and participants who do not have preoperative pain and do not consume opioids, as defined by the authors. When not defined, we will include participants with long‐term (over six months) or recurring opioid consumption (e.g. women consuming opioids during menstruation). This analysis may identify a subgroup of participants who derive a specific benefit from the intervention. Type of intervention drug: we will perform a subgroup analysis comparing studies using racemic ketamine with those using esketamine. This analysis aims to explore potential differences between the racemate and the S‐enantiomer in terms of both analgesic efficacy and safety profile. We will categorise studies according to low‐dose and high‐dose regimens. A low dose for ketamine is < 0.4 mg/kg/h; in studies using a single bolus or repeated boluses, a threshold of < 0.4 mg/kg per bolus will be applied. For esketamine, a low dose is < 0.2 mg/kg/h; in studies using a single bolus or repeated boluses, a threshold of < 0.2 mg/kg per bolus will be applied. A high dose of ketamine is ≥ 0.4 mg/kg/h (or ≥ 0.4 mg/kg per bolus), and for esketamine, it is ≥ 0.2 mg/kg/h (or ≥ 0.2 mg/kg per bolus). As the most effective subanaesthetic dose of ketamine/esketamine for postoperative pain control has not been established, this subgroup analysis will explore potential dose–response differences in both efficacy and safety. We will use the formal test for subgroup differences in RevMan to assess the subgroups [ 47 ]. Subgroup analyses will be conducted only when a sufficient number of studies are available; we plan to proceed when three or more studies report adequate data for the relevant subgroup. For studies including participants belonging to more than one subgroup without separate reporting, we will contact the study authors to request additional information. We will conduct sensitivity analysis to determine if important methodological factors or other study features identified during the review process have an effect on our primary and secondary outcomes. We will perform sensitivity analysis as follows. Only including studies with an overall low risk of bias (determined by risk of bias assessment) to test the effect of key methodological factors on the intervention outcomes. Excluding studies with substantial missing data to determine the impact of missing data on the outcomes. Excluding studies in which only a proportion of participants met the review eligibility criteria (i.e. studies included because the majority, but not all, participants were eligible) to assess the impact of their inclusion on the review findings. Only including studies with an overall low risk of bias (determined by risk of bias assessment) to test the effect of key methodological factors on the intervention outcomes. Excluding studies with substantial missing data to determine the impact of missing data on the outcomes. Excluding studies in which only a proportion of participants met the review eligibility criteria (i.e. studies included because the majority, but not all, participants were eligible) to assess the impact of their inclusion on the review findings. We will reassess the need for sensitivity analyses after study selection, as characteristics of the included studies may raise specific methodological concerns that warrant additional analyses. We will construct a summary of findings table using GRADEpro GDT software [ 56 ], according to the methods described in Chapter 14 of the Cochrane Handbook for Systematic Reviews of Interventions [ 57 ], for the comparison of ketamine/esketamine versus placebo or no treatment for the following outcomes (Table 2). Table 2. Template for summary of finding table Outcome Time point Measurement / analysis Peak postoperative pain intensity 0 to 24 hours (peak) NRS or VAS (continuous) Opioid consumption 0 to 24 hours Total opioid dose (continuous; morphine equivalents) Serious adverse events (SAEs) 0 to 24 hours ≥ 1 SAE per participant (dichotomous) Postoperative pain intensity 1 to 2 hours (or during admission in the PACU) NRS or VAS (continuous) Postoperative pain intensity 6 hours NRS or VAS (continuous) Postoperative pain intensity > 24 hours (up to 6 months postoperatively) NRS, or VAS (continuous), or McGill questionnaire for chronic pain, or other questionnaire and scale for pain Sleep quality / structure During hospital admission Questionnaire or objective measures (continuous) Abbreviations: NRS: numerical rating scale; PACU: Post‐Anesthesia Care Unit; SAE: serious adverse events; VAS: visual analogue scale For all the outcomes included in the summary of findings table, we will use the GRADE approach to assess the certainty of a body of evidence based on the five GRADE considerations (risk of bias – RoB 2, consistency of effect, imprecision, indirectness, and risk of publication bias) [ 58 ]. Two review authors will independently assess the level of certainty for each outcome per comparison, with any disagreements being resolved through discussion or through involvement of a third review author. We will assess the certainty of the evidence as high, moderate, low, or very low, downgrading certainty based on our evaluation of the five GRADE domains. Equity considerations will be informed by the PROGRESS‑Plus framework (these characteristics relate to Place of residence, Race/ethnicity/culture/language, Occupation, Gender/sex, Religion, Education, Socioeconomic status, Social capital, Personal characteristics associated with disability, features of relationships, and time‐dependent relationships). For this review, the most relevant characteristics are: Age: differences in age may influence both the efficacy and safety of esketamine, as older patients may have an increased susceptibility to adverse effects (e.g. cardiovascular or neurocognitive events), while younger patients may differ in pain perception and analgesic response. Gender/sex and reproductive status: although the population undergoing gynaecological surgery consists predominantly of individuals assigned female at birth, variations in hormonal status (e.g. premenopausal versus postmenopausal) may affect pain perception, pharmacodynamics, and postoperative recovery. Socioeconomic status: socioeconomic factors may influence access to esketamine, perioperative care, and follow‐up, thereby affecting the applicability of findings across different healthcare systems and patient populations. Comorbidities and disability: the presence of comorbid conditions, particularly psychiatric disorders (e.g. depression, anxiety), may modify both the efficacy and safety profile of esketamine and will be considered when interpreting results. Other PROGRESS‑Plus factors: where reported, additional characteristics (such as education level or social support) will be considered, particularly in relation to patient‐reported outcomes. Age: differences in age may influence both the efficacy and safety of esketamine, as older patients may have an increased susceptibility to adverse effects (e.g. cardiovascular or neurocognitive events), while younger patients may differ in pain perception and analgesic response. Gender/sex and reproductive status: although the population undergoing gynaecological surgery consists predominantly of individuals assigned female at birth, variations in hormonal status (e.g. premenopausal versus postmenopausal) may affect pain perception, pharmacodynamics, and postoperative recovery. Socioeconomic status: socioeconomic factors may influence access to esketamine, perioperative care, and follow‐up, thereby affecting the applicability of findings across different healthcare systems and patient populations. Comorbidities and disability: the presence of comorbid conditions, particularly psychiatric disorders (e.g. depression, anxiety), may modify both the efficacy and safety profile of esketamine and will be considered when interpreting results. Other PROGRESS‑Plus factors: where reported, additional characteristics (such as education level or social support) will be considered, particularly in relation to patient‐reported outcomes. We will extract and report these characteristics where available and consider their potential impact on the applicability and generalisability of the review findings. Where sufficient data exist, we will explore their contribution to between‐study heterogeneity. Consumer involvement will not be undertaken in this review due to limited time and resources. We recognise that involving consumers could have strengthened the relevance and applicability of the review findings.

All surgical procedures are associated with a degree of postoperative pain; however, moderate to severe pain remains a significant clinical problem, affecting approximately 20% to 40% of patients [ 1 ]. Poorly controlled postoperative pain is associated with increased morbidity, delayed recovery, and reduced quality of life. In addition, inadequate analgesia is a recognised risk factor for prolonged opioid use, opioid dependence, and the development of persistent postsurgical pain [ 2 ]. Persistent pain one year after surgery occurs in more than 10% of patients and is particularly associated with major procedures, including limb amputation, mastectomy, and thoracotomy [ 3 ]. Notably, the severity of postoperative pain does not correlate directly with the invasiveness of the procedure. For example, minor laparoscopic surgery may result in higher pain intensity than extended laparotomic procedures, possibly due to factors including the underlying pathology, the presence of pre‐existing pain, preoperative opioid use, and differences in the use of multimodal analgesic strategies. Gynaecological surgery comprises a heterogeneous group of procedures that diagnose, treat, or manage conditions of the female reproductive system (uterus, ovaries, fallopian tubes, cervix, vagina, and vulva) and related pelvic organs, performed for benign or malignant conditions. The definition includes procedures performed for gynaecological conditions, even when those procedures solely involve organs outside the reproductive system. Table 1 provides a detailed, though not exhaustive, list of the most common gynaecological procedures. Table 1. Gynaecological surgery Category Examples Uterine surgery Hysterectomy, myomectomy, endometrial ablation, dilation and curettage, hysteroscopic surgery for polyps or fibroids Ovarian and tubal surgery Oophorectomy, enucleation of ovarian cysts, salpingectomy, tubal ligation, surgery for ectopic pregnancy Pelvic floor and prolapse surgery Anterior/posterior repair, sacrocolpopexy, sling procedures for urinary incontinence Endometriosis and pelvic pain surgery Laparoscopic excision or ablation of endometriosis, adhesiolysis Gynaecologic oncology surgery Surgery for cervical, ovarian, uterine, vulvar, and vaginal cancers, including staging and debulking procedures Vaginal and vulvar surgery Vulvar lesion excision, vaginoplasty, repair of vaginal tears or fistulas Diagnostic procedures Laparoscopy, hysteroscopy, biopsy procedures These procedures include both minor, minimally invasive interventions (e.g. hysteroscopy, minor vulvar surgery) and major abdominal operations (e.g. open hysterectomy, cytoreductive surgery), reflecting varied surgical trauma and postoperative pain. Pain outcomes depend on procedure type and patient factors. Common operations like hysterectomy often cause moderate to severe pain [ 4 ]. Uterine and pelvic surgery pain is predominantly visceral, harder to manage than somatic pain and more prone to chronification [ 5 ]. Patients undergoing gynaecological surgery are at increased risk of severe and persistent pain. Many have pre‐existing conditions (e.g. endometriosis, cancer‐related pain) with central sensitisation [ 6 ], and altered pain perception [ 7 ]. Psychosocial factors are significant, including anxiety, fertility and body image concerns, and feelings of loss [ 8 , 9 ], alongside perioperative mood disturbances and sleep disruption — all contributing to worse pain control [ 10 , 11 ]. Overall, this population is heterogeneous, with interacting procedural, biological, and psychosocial factors that increase pain management complexity and adverse outcomes. Ketamine and esketamine are cyclohexanone‐derived general anaesthetics and N‐methyl‐D‐aspartate (NMDA) receptor antagonists with analgesic properties [ 12 , 13 ]. They are used clinically for the induction and maintenance of anaesthesia, for acute and chronic pain management, and more recently for the treatment of depression [ 14 ]. In addition to NMDA receptor blockade, ketamine and esketamine interact with other receptor systems, including opioid receptors — contributing to their analgesic effects — and dopamine transporters, which are thought to play a role in their antidepressant properties. Ketamine exists as two enantiomers, S‑ketamine (esketamine) and R‑ketamine (arketamine), administered as a racemate (a solution containing both enantiomers) or pure esketamine. At anaesthetic doses, they produce dissociation with preserved respiration and airway reflexes, cardiovascular stimulation (increased heart rate and blood pressure), and bronchodilation. Common adverse effects include dysphoria, hallucinations, drowsiness, nausea, hyperthermia, and cardiovascular stimulation, though these are less frequent at subanaesthetic doses [ 15 , 16 ]. Compared with racemic ketamine (containing both enantiomers), esketamine more often causes transient blood pressure increases but less nausea. Contraindications include conditions where sympathetic stimulation or hypertension is harmful (e.g. uncontrolled hypertension, severe cardiovascular or intracranial disease) and severe psychiatric disorders, especially psychosis. In recent years, increasing attention has been given to ketamine/esketamine use at subanaesthetic doses as an adjunct to the standard perioperative analgesia. They have been shown to reduce postoperative pain and opioid consumption across a range of surgical procedures [ 17 ]. However, evidence in specific or vulnerable populations — such as patients with neurodevelopmental disorders, cognitive impairment, or chronic disability — is limited and heterogeneous, with variable responses and a potential risk of neuropsychiatric adverse effects [ 18 , 19 ]. In gynaecological surgery, several studies, including randomised controlled trials, have reported beneficial effects of subanaesthetic ketamine/esketamine doses administered as an adjunct to standard perioperative analgesia on postoperative pain, opioid consumption [ 20 , 21 , 22 ], overall recovery quality [ 23 ], and postoperative nausea and vomiting [ 24 , 25 ]. These agents have been reported to provide greater relief than opioids from visceral pain, common after many gynaecological procedures [ 26 ]. In addition, the efficacy of ketamine/esketamine in reducing neuropathic pain has been shown by recent data collections and preclinical studies [ 27 , 28 ]. Beyond analgesia, as mentioned previously, ketamine and esketamine have demonstrated rapid‐acting antidepressant effects [ 29 ]. This property may be particularly relevant in gynaecological populations, where psychological distress, including depression and sleep disturbances, is common before and after surgery [ 8 , 9 ]. These drugs' potential to improve mood and sleep may contribute indirectly to better pain control and overall recovery, especially in procedures such as hysterectomy, debulking surgery and procedures for ectopic pregnancy, which are often associated with emotional and psychological burden and more severe postoperative pain [ 30 ]. The perioperative antidepressant effect of esketamine has been studied primarily in postpartum depression following caesarean section, where it was found to reduce the risk [ 31 , 32 ]. Although several clinical studies and meta‐analyses have demonstrated the efficacy of ketamine/esketamine in reducing postoperative pain across a range of surgical settings [ 17 , 25 ], a high‐quality, up‐to‐date synthesis focused on gynaecological populations is lacking. Existing systematic reviews either focus on specific procedures, such as caesarean section, or include heterogeneous surgical populations in which gynaecological procedures are not analysed separately [ 33 , 34 ]. Gynaecological surgery represents a distinct clinical context, for multiple reasons. These procedures are frequently associated with a higher incidence of visceral pain [ 4 ], and, in specific subgroups such as patients with endometriosis or gynaecological malignancies, an increased prevalence of neuropathic pain and central sensitisation, key predictors of severe postoperative pain and increased opioid consumption. Moreover, gynaecological procedures — including hysterectomy, ectopic pregnancy management, and endometriosis surgery — may carry a specific psychological burden due to their impact on reproductive organs and body integrity. These features, together with the involvement of highly innervated and hormone‐responsive tissues, suggest that findings from other surgical populations may not be directly generalisable [ 9 ]. This review will only include individuals assigned female at birth, offering a rare opportunity to evaluate the efficacy and safety of ketamine and esketamine in a sex‐specific population. Such evidence remains limited, despite the increasing recognition that pain mechanisms and responses to analgesics may differ between men and women [ 35 ]. In addition, people undergoing gynaecological surgery are often younger than the broader surgical population, with years of professional and social life ahead. This underscores the importance of preventing persistent postoperative pain and highlights the relevance of evaluating strategies, such as ketamine and esketamine, that may reduce this risk. Interest in pain related to gynaecological procedures is increasing, both in the scientific literature and in public discourse, highlighting unmet needs in the management of pain during and after interventions. Ketamine and esketamine are already used worldwide as adjunct analgesics, particularly in patients with complex pain profiles or opioid‐related concerns [ 36 ]. A rigorous synthesis of the evidence in this population may inform clinical practice, support the development or updating of guidelines, and guide decision‐making by healthcare providers and policymakers. Considering the wide spectrum of gynaecological surgery, this review will include a subgroup analysis based on the type of intervention and level of surgical invasiveness. Further subgroup analyses will be conducted according to background condition (cancer, endometriosis, or other benign conditions), type of drug used (ketamine or esketamine), dosage, and preoperative chronic pain and opioid use.

The primary objective of this review is to evaluate the benefits and harms of a subanaesthetic dosage of ketamine/esketamine as an adjunct to perioperative analgesia versus standard analgesia or placebo for reducing postoperative pain in participants undergoing gynaecological surgical procedures. A secondary objective is to assess whether these outcomes differ by type of surgery, background condition, type of drug used (ketamine or esketamine), dosage, and presence of preoperative chronic pain or preoperative long‐term or recurrent opioid use.

Supplementary materials are available with the online version of this article: 10.1002/14651858.CD016377. Supplementary materials are published alongside the article and contain additional data and information that support or enhance the article. Supplementary materials may not be subject to the same editorial scrutiny as the content of the article and Cochrane has not copyedited, typeset or proofread these materials. The material in these sections has been supplied by the author(s) for publication under a Licence for Publication and the author(s) are solely responsible for the material. Cochrane accordingly gives no representations or warranties of any kind in relation to, and accepts no liability for any reliance on or use of, such material. Supplementary material 1 Search strategies