Abstract
Ketamine is an N-methyl-
Introduction
Ketamine is a derivative of the hallucinogen phencyclidine, and was developed in the 1960s as a fast-acting, non-barbiturate anaesthetic agent.
1
Its molecular mechanisms of anaesthetic action centre on its potent non-competitive inhibition of the N-methyl-
Ketamine is also recognised widely to have analgesic properties. Its analgesic action is complex, probably involving opioid receptor activity as well as NMDA receptor antagonism in the brain and spinal cord, and it may inhibit the development of opioid-induced hyperalgesia.5,6 In the perioperative setting, there is mixed evidence from small clinical trials that ketamine reduces acute postoperative pain scores and opioid requirements, but with an unclear dose–response relationship. 7 Whether perioperative ketamine prophylaxis can also reduce the risk of progression from acute to chronic postsurgical pain (CPSP) after surgery remains undetermined from small trial data,8,9 and is currently the subject of a large multi-centre trial, the ROCKet (Reduction Of Chronic postsurgical pain with Ketamine) Trial, which is supported by the Australian and New Zealand College of Anaesthetists Clinical Trials Network (ANZCA CTN). The ROCKet trial will also provide a large amount of data on the effectiveness of ketamine on acute postoperative pain and associated drug side-effects and safety data. 10
However, the recommended indications for use of intravenous ketamine by the FDA and the Australian Therapeutic Goods Administration (TGA) remain solely as an anaesthetic or sedative adjunct to anaesthesia. Despite widespread clinical practice, use of ketamine for postoperative analgesia remains ‘off-label’, which calls into question the appropriateness and relevance of current longstanding regulatory approvals. We therefore conducted a survey of ANZCA Fellows regarding their customary practice regarding administration of ketamine for perioperative analgesia. Specifically, we aimed to (a) define patient factors and surgical categories that influence the anaesthetist’s decision-making regarding perioperative ketamine analgesia administration; and (b) determine their standard prescribing practices and dosage of intravenous (IV) ketamine analgesia, both intraoperatively and postoperatively, for a range of major surgical categories. We hypothesised that perioperative practice of prescribing of ketamine in Australia and New Zealand is highly variable, with variations in dosing, regimens (bolus and infusion) and patient and surgery selection. As exploratory analyses, we looked for potential associations with ketamine prescribing including respondent seniority, formal pain medicine qualifications and predominant practice environment (public versus private practice). In addition, we took the opportunity to survey the opinion of anaesthetists about whether it is likely that perioperative ketamine reduces CPSP incidence in major surgery.
Methods
Formal ethical approval was obtained prospectively under the low/negligible risk pathway from the Austin Health Human Research Ethics Committee (HREC/79152/Austin-2022), in addition to endorsement and approval following review by the ANZCA CTN in January 2023.
Following HREC approval, the authors distributed a pilot questionnaire among consultants in the Department of Anaesthesia at Austin Health, Melbourne, Victoria, in order to (a) examine the robustness and useability of the electronic audit-data collection format, and (b) receive local, peer-reviewed feedback regarding the question format and content. The questionnaire design underwent final review by the ANZCA CTN for brevity and wording, to improve question clarity and increase response rate before widespread distribution.
The survey was conducted as a closed-questioned, online (electronic) questionnaire, created using Survey Monkey, with a multiple-choice answer format with an additional free text option for many questions. All information collected was de-identified, with collected data stored via the Survey Monkey interface, under password protection, without internet protocol (IP) addresses or responder details being obtained. All responses were stored on the Survey Monkey website database per third-party security. When data collection was completed, pooled data were stored on a secured database under password protection.
Response data collected consisted of a combination of categorical data for demographics, quantitative, categorical data for drug dosing regimens and a modified Likert scale to assess responder likelihood for different prescribing regimens, to decrease bias and ambiguity. The questions included practitioner demographic data (seniority, geographical location, clinical involvement in both acute and chronic pain and possession of formal pain qualifications) expressed as percentages of total responses, and patient factors influencing ketamine prescribing likelihood, expressed as percentages of total responses. Likely ketamine dosing regimens (loading dose and infusion rates) in multiple defined ranges were sought, with responses recorded as percentages of total responses. Likelihood of ketamine prescription for various major surgical categories was sought, asked on a five-point Likert scaled from very unlikely to very likely, expressed as percentages of total responses. Hierarchy of ketamine use within any analgesic regimen to treat acute postoperative pain for various operations was also sought, on a quantitative scale from first line to third/rescue line, expressed as percentages of total responses. Finally, responder impression was sought of likelihood that ketamine (any regimen) would reduce the risk of development of CPSP after each of the major surgical categories, asked on a five-point Likert scale from very unlikely to very likely, expressed as percentages of total responses.
The questionnaire was distributed to 2000 ANZCA Fellows. All participants invited to participate in this survey held a Fellowship of ANZCA. Participants were selected randomly from the ANZCA Fellows secure email database. Invitations to complete the questionnaire were then distributed by email by ANZCA administrators. Participants were invited from every state and territory in Australia and New Zealand. Participation was voluntary, and individual consent was implied upon a participant voluntarily completing the questionnaire. One reminder email invitation was sent to any invitee who had failed to respond after four weeks. Following distribution (March 2023), the questionnaire remained available for responses for a period of three months (until July 2023), before data collation and analysis.
We used definitions [including ‘pain’, and oral morphine equivalents (oMEDD)] from the latest ANZCA Faculty of Pain Medicine (FPM) college guidelines. 11 Given the simple nature of the questions within this clinical practice audit, formal testing of the validity and reliability of individual questions was not attempted. 12 The authors aimed for a ‘adequate’ response rate of >40% in order to minimise responder bias and margin of error. 12 This was facilitated by limiting the number of questions and question length, in addition to pilot testing. The final survey that was distributed is available (Supplementary Material 1).
We used simple, descriptive statistics to report the demographic data of respondents, and prescribing practices of ketamine, including dosing regimens and procedures and patient groups more or less likely to result in ketamine administration, as this allowed us to answer our research question both clearly and directly without examining outcomes (which goes beyond the scope of this survey and report). Exploratory analyses of potential associations between respondent seniority, formal pain medicine qualifications and predominant practice environment (public versus private practice) and the likelihood of ketamine prescribing across a range of procedures and patient groups were done by constructing contingency tables from Likert scale results and using Fisher’s exact test in Stata 12 (Stata Corp, College Station, TX). Missing data (i.e. questions that were not answered by respondents) were handled as ‘missing completely at random’ and not included in the final analysis, so that reported results describe percentage of actual responses received to each question.
Results
Of the 2000 ANZCA Fellows surveyed, there were 806 successful responders (40.3% overall response rate). Respondents were of mixed seniority between junior, middle-career and senior anaesthetists, with the commonest (31.1%) range being 11–20 years following conferring of ANZCA Fellowship (FANZCA). They were from a diverse geographical spread, with the majority located in Victoria, New South Wales and New Zealand. Most respondents worked primarily within the public sector, with the majority not holding an FPM qualification (95.4%), and the majority declaring some acute pain service (APS) involvement in their regular practice (98.4%) (Table 1).
Demographic data of respondents: seniority (years since conferring of FANZCA), location, and clinical practice details. n = number of responses.
FANZCA, Fellowship of Australian and New Zealand College of Anaesthetists.
Perioperative ketamine prescribing
The most common patient factors influencing respondents towards any perioperative ketamine administration included pre-existing chronic pain (90.1%), heavy (>60 mg oMEDD equivalents per day) opioid use (85.1%) and/or multiple opiates (76.9%) preoperatively (Figure 1). Table 2 lists responses to survey Question 7, on likely ketamine dose ranges chosen for loading bolus and for starting rates of intra- and postoperative infusions in an American Society of Anesthesiologists Physical Classification 1 patient undergoing non-emergent surgery. The commonest choices of loading dose were in the range of either 0.20–0.39 mg kg–1 or 0.40–0.59 mg kg–1. The commonest choice of infusion rate was in the range of 0.10–0.19 mg kg–1 h–1.

Patient factors influencing responder prescribing. Responses to survey Question 6: ‘For the treatment and/or prevention of pain, in which of the following patient groups are you likely to prescribe ketamine intra- or postoperatively?’. EtOH, Ethanol; oMEDD, oral morphine equivalent.
Ketamine prescribing practices: responses to survey question 7. “If using a ketamine LOADING BOLUS for analgesia INTRA-OPERATIVELY, what dose would you most likely use? * Assuming ASA 1, non-emergent surgery, no chronic or acute pain issues, substance misuse and normal renal/hepatic function”.
n = number of responses.
ASA, American Society of Anaesthetists Physical status.
Prescribing by surgical categories
Responder administration of ketamine varied intraoperatively depending on the procedure being performed and is listed in Table 3. For open pelvic/abdominal surgery, thoracic surgery and major spinal surgery, the majority of respondents were either likely or very likely to administer ketamine. For laparoscopic abdominal surgery and mastectomy surgery, the majority of respondents were unlikely to do so. For major limb surgery, there was a heterogeneous spread or responses.
Responder likelihood of administering ketamine intraoperatively for different surgical categories.
n = number of responses.
Non-emergent.
Non-cardiac, non-emergent, either open or video-assisted.
Orthopaedic or vascular, non-emergent.
Responder propensity to prescribe ketamine for the treatment of acute postoperative pain also varied depending on the procedure being performed (Table 4). Respondents prescribed ketamine ‘routinely’ most commonly for major spinal procedures (13.1%) and as either a routine or second-line option for major spinal (41.7%), thoracic (41.3%) and open abdominal (41.3%) surgeries. The majority of respondents reserved ketamine as a ‘third-line or rescue option’ for all other procedure types listed.
Responder prescribing practice of ketamine postoperatively for treatment of acute pain, for different surgical categories.
n = number of responses.
Non-emergent.
Non-cardiac, non-emergent, either open or video-assisted.
Orthopaedic or vascular, non-emergent.
Exploratory analyses
Likelihood of administering ketamine intraoperatively was inversely associated with years since conferring of FANZCA (Pearson chi-square between 46.4 and 102.5, P < 0.001 for all surgical categories; Supplementary Material 2). Likelihood was also greater among respondents whose practice was predominantly in public hospitals (Pearson chi-square between 14.4 and 29.7, P < 0.01 for all surgical categories with the exception of laparoscopic surgery (Pearson chi-square 5.3, P = 0.259; Supplementary Material 3). No significant associations were found with FPM qualification, although the small proportion of respondents with an FPM qualification made this analysis statistically weak (Supplementary Material 4).
The majority of respondents thought it either ‘likely’ or ‘very likely’ that ketamine would reduce chronic postsurgical pain after thoracic surgery. Respondents most commonly selected ketamine as being ‘neither likely or unlikely’ to influence chronic postsurgical pain following laparoscopic, major limb and mastectomy surgery. A substantial minority (over 40%) of respondents believed ketamine was likely to significantly reduce postoperative chronic pain after open abdominal and major spinal and limb surgery (Table 5).
Responder perception of ketamine on influencing the development of chronic postsurgical pain, for different operations.
n = number of responses.
Non-emergent.
Non-cardiac, non-emergent, either open or video-assisted.
Orthopaedic or vascular, non-emergent.
Discussion
This survey found that, for typical patients undergoing thoracic, spinal or open abdominal surgery, administration of IV ketamine intraoperatively was likely practice by most ANZCA Fellow respondents. We found the patient factors mostly likely to lead to ketamine administration included pre-existing chronic pain, and heavy oMEDD and/or multiple opioid use. The majority of respondents were either likely or very likely to administer ketamine for most surgery types, with the exception of laparoscopic pelvic-abdominal or mastectomy surgery. Routine administration of ketamine postoperatively remained an uncommon practice, but more than two-thirds of respondents used postoperative ketamine as second-line analgesia, or as third-line/rescue analgesia for refractory postoperative pain. Among those who administered ketamine intraoperatively, the large majority used a loading dose between 0.2 mg kg–1 and 0.6 mg kg–1, and the most common intraoperative and postoperative infusion rate was between 0.1 mg kg–1 h–1 and 0.2 mg kg–1 h–1.
Our results indicate local practice amongst respondents is largely consistent with contemporary professional society guidelines, including the 2018 American Society of Regional Anesthesia and Pain Medicine (ASRAPM) consensus guidelines on Perioperative Ketamine. 13 This guideline suggests perioperative ketamine confers most benefit for upper abdominal and thoracic surgery and spinal procedures, reflective of the surgical categories selected by respondents.
Ketamine infusion dosing regimens reported by respondents were largely in keeping with available evidence, with the 2018 ASRAPM guidelines suggesting 0.1–0.2 mg kg–1 h–1 is most ‘common’. 13 Conversely, due to heterogeneity in international consensus, bolus dosing by survey respondents was in keeping with typical dosing for some guidelines, while exceeding others. The 2018 ASRAPM guidelines recommend that bolus dosing ‘not exceed 0.35 mg kg−1’. However, there is wide variation in practice amongst patient demographic treated and surgical category, with more liberal infusion dosing reported (e.g. 0.6 mg kg−1 h−1, for a cohort of spinal surgery patients) in some studies. 14 This variability reflects a lack of strong available evidence in this area, and may at least partially explain the range of survey responses for this question.
Our survey results are also consistent with the Australian and New Zealand FPM Background Paper for Position Statement on the use of ketamine in the management of chronic non-cancer pain, which states ketamine has an established place in the perioperative management of acute pain (at subanaesthetic doses <0.5 mg kg−1), but refrains from making formal recommendations regarding either bolus or infusion dosing or regimens. 15
Survey respondents indicated that perioperative ketamine was likely to result in the greatest reduction in CPSP for thoracic surgery, while being largely undecided on the effectiveness of ketamine for this purpose for most other procedure types. Our findings suggest ANZCA anaesthetists are most likely to prescribe ketamine in patient cohorts where evidence most strongly suggests a significant risk of CPSP, including those with chronic pain preoperatively and high preoperative opiate use.13,16,17 Conversely, however, other studies indicate that preoperative psychological factors (including anxiety and depression) are major factors determining the incidence of CPSP, 18 but we found only 11.2% of respondents likely to prescribe ketamine for those receiving pharmacological treatment for anxiety or depression, and 3.8% of respondents likely to prescribe ketamine for those receiving pharmacological treatment for any other psychiatric disorder (Figure 1). Reasons for this may include concerns among practitioners about psychotropic side-effects of ketamine in such patient groups, which limit its use and dose administered for analgesia in around 10% of surgical patients overall. 2 Similarly, patients undergoing limb amputation are a high risk group for CPSP, 19 but were not captured in the question about major limb surgery (orthopaedic or vascular, non-emergent) category, since it is likely to be considered an urgent or emergent procedure. We also observed that despite 30%–50% of respondents believing that ketamine may reduce the risk of CPSP, respondents were not likely to routinely prescribe ketamine for the mastectomy cohort, despite this cohort experiencing CPSP rates of 25%–60%. 20 Although beyond the scope of this survey, this might reflect limited current high quality evidence for the effectiveness of ketamine in reducing the incidence of CPSP, and/or concern regarding potential side-effects of ketamine administration, such as postoperative nausea and vomiting, as this cohort is at a particularly elevated risk.
This survey does have several limitations. Voluntary response designs carry the potential for inherent responder bias, where those with an interest in ketamine for perioperative pain management may be more likely to participate in the survey and influence results. However, this format allowed the questionnaire to be distributed to a large cross-section of ANZCA anaesthetists (n = 2000), and allowed prospective data collection regarding current practice, rather than retrospective analysis of records. Our respondent cohort was largely representative of the broader ANZCA community in terms of seniority, geographical location and public versus private primary practice (Supplementary Material 5). This study examined only patient factors resulting in ketamine prescription being more likely and, for reasons of economy, did not include those factors that made ketamine prescription less likely. Finally, we acknowledge that not all respondents completed all questions. However, given the relatively low proportion of missing data (less than 5% for responses to questions on ketamine use for acute pain management, and less than 10% for Question 27 on estimated effect of ketamine on CPSP risk), we anticipate incomplete responses had little statistical influence on the results of this survey.
This study also had several strengths. The survey question was novel and, to our knowledge, is the first to examine the perioperative practices and attitudes towards ketamine prescribing by anaesthetic providers internationally. An unpublished work with a low response rate (5.2%) by the American Association of Nurse Anesthesiology (AANA) surveyed 1323 certified registered nurse anesthetists (CRNAs); however, this focused largely on ketamine prescribed outside the operating theatre setting and the level of supervision and multidisciplinary team involvement required. 21
The survey underwent significant drafting and pilot-testing in the local environment, and from the ANZCA CTN. We received a strong response rate (over 40%) with a large sample size of over 800 respondents, which strengthens the validity of our results, and their generalisability to broader Australian and New Zealand anaesthesia practice. This response rate meets criteria for adequacy set out by authorities on survey methodology in anaesthesia. 12 The online format minimised logistical challenges and costs, and the random distribution among ANZCA Fellows minimised distribution bias, although we acknowledge this potentially limits international generalisability of our findings.
Our survey is clinically relevant and timely, providing insight into the current practice of ANZCA anaesthetists regarding ketamine administration for perioperative analgesia. More broadly, longstanding FDA and TGA approved indications for ketamine administration do not reflect current widespread clinical practice in anaesthesia in Australia and New Zealand, or current international consensus guidelines. This discrepancy may reflect a lack of high quality evidence on the effectiveness of perioperative ketamine administration on postoperative pain outcomes. An improvement in the reliability of the evidence base for perioperative ketamine effectiveness might prompt a review of regulatory approvals. Data on the effectiveness and safety of perioperative ketamine is expected from the ROCKet trial, and should help inform future guidelines and regulatory approvals. In this context, it is of note that in 2019, the United States FDA granted approval for s-ketamine delivered by the nasal route for use in treatment-resistant depression. This was the first significant change to FDA approvals for ketamine since its original introduction in 1970.
Any such review must make careful assessment of drug safety along with treatment efficacy. Although our survey did not attempt to inform these assessments, the ROCKet Trial is expected to provide a substantial amount of prospective data on both side-effect rates and effectiveness compared with placebo on acute postoperative pain up to three days postsurgery, as measured using postoperative pain scores and opioid sparing, in addition to its primary endpoint of assessing its effectiveness in primary prevention of CPSP. 10 With this in mind, it is reassuring that the perioperative dose regimen (loading dose 0.5 mg kg−1 and subsequent infusion 0.125 mg−1 kg−1 h−1) being trialled in ROCKet is similar to the most common dose range identified by our survey. This suggests that the ROCKet trial results will reliably inform the safety and effectiveness of current widespread clinical ketamine administration practice.
Conclusion
Our survey exploring current off-label prescribing practices of intravenous ketamine for perioperative analgesia received over 800 responses from ANZCA Fellows, and found administration of IV ketamine intraoperatively is likely practice by most respondents for patients undergoing thoracic, spinal or open abdominal surgery, and those with pre-existing chronic pain and heavy opioid use. Postoperative ketamine is prescribed usually as second- or third-line/rescue analgesia, typically at up to 0.2 mg−1 kg−1 h−1. Practice and dosage regimens were largely consistent with published international guidelines.
Supplemental Material
sj-pdf-1-aic-10.1177_0310057X241309655 - Supplemental material for Survey of administration of intravenous ketamine for perioperative pain management in Australia and New Zealand
Supplemental material, sj-pdf-1-aic-10.1177_0310057X241309655 for Survey of administration of intravenous ketamine for perioperative pain management in Australia and New Zealand by Patryck J Lloyd-Donald and Philip J Peyton in Anaesthesia and Intensive Care
Supplemental Material
sj-pdf-2-aic-10.1177_0310057X241309655 - Supplemental material for Survey of administration of intravenous ketamine for perioperative pain management in Australia and New Zealand
Supplemental material, sj-pdf-2-aic-10.1177_0310057X241309655 for Survey of administration of intravenous ketamine for perioperative pain management in Australia and New Zealand by Patryck J Lloyd-Donald and Philip J Peyton in Anaesthesia and Intensive Care
Supplemental Material
sj-pdf-3-aic-10.1177_0310057X241309655 - Supplemental material for Survey of administration of intravenous ketamine for perioperative pain management in Australia and New Zealand
Supplemental material, sj-pdf-3-aic-10.1177_0310057X241309655 for Survey of administration of intravenous ketamine for perioperative pain management in Australia and New Zealand by Patryck J Lloyd-Donald and Philip J Peyton in Anaesthesia and Intensive Care
Footnotes
Author Contribution(s)
Acknowledgements
The authors acknowledge Karen Goulding, and the Australian and New Zealand College of Anaesthetists Clinical Trials Network for their help in the distribution of the survey and are grateful to the respondents to the survey for their time and assistance.
Funding
Funding/support was provided from internal resources in the Austin Health Department of Anaesthesia for the cost of use of ‘Survey Monkey’ software (paid subscription).
Declaration of conflicting interests
The authors declare no conflicts of interest. P.P. is Chief Investigator of the ROCKet (Reduction of Chronic postsurgical pain with Ketamine) trial.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
