FT20 Perioperative pain management guidelines: why don’t they work?

Today, despite all the analgesic drugs and techniques available, adequate management of postoperative pain remains challenging, particularly for some patients e.g. patients presenting with preoperative chronic pain conditions or longlasting opioids intake. 1 There is evidence that the administration of analgesic treatments is sometimes inconsistent and does not reflect the best available evidence. In example, a recent analysis of perioperative pain management in 10415 patients (data extracted from 2017 until 2019) reported the use of one non-opioid analgesic in 57% of the patients and the use of two non-opioid analgesics in only 38% of the patients, while multimodal analgesia is strongly recommended from evidence-based medicine to enhance recovery.2 Clearly, there is a gap between published recommendations and their clinical application. The principal causes of the problem are a lack of familiarity with the medical literature from practicing physicians and an inability to effectively synthetize a large body of existing literature into meaningfully beneficial clinical practice changes.3 Perioperative guidelines have been developed to overcome the problem i.e. to provide a synthesis of existing literature and recommendations. By definition, clinical guidelines are ‘systematically developed statements to assist practitioner and patient decisions about appropriate healthcare for specific clinical circumstances’.4 The implementation of practice guidelines is expected to reduce variations in practice, to allow for standards for measurement of clinical performance and to improve the efficacy of healthcare delivery.4 Further, medical practice guidelines have also medicolegal implications.5 Consequently, the field of perioperative medicine has seen a dramatic rise in clinical guidelines and consensus statements published to improve patient care.6 Postoperative pain guidelines and their observance in practiceTwo very recent surveys point out the poor observance as well as the ignorance of postoperative pain guidelines. Following a Danish survey that highlighted that 66% of women reported severe pain at any time during the first 24 h after cesarean section, some authors aimed to review the standard practice for postoperative pain management in 22 Danish delivery centers.7 All centers provided answers. Most centers relied on oral opioids in addition of paracetamol and NSAIDs. Only one center used intrathecal morphine (very low dose of 40 µg). Truncal nerve blocks were used for rescue analgesia in 73% of centers and as prophylactic analgesia in only 9% of the centers.7 Recent guidelines on perioperative pain management for cesarean section are available from PROSPECT working group8 and from the National Institute for Clinical Excellence (NICE guidelines released in 2021, updated in 2024). A second interesting survey piloted by the members of PROSPECT working group and colleagues was sent to ESRA and ASRA members to question the current utilization of postoperative pain guidelines. Anonymous responses from 1340 persons (56.9% from Europe) were analysed.9 Not surprisingly the survey showed a relative underutilization of postoperative pain guidelines: only 43.8% of respondents followed available national guidelines and 38.9% used international PROSPECT guidelines in their daily practice. Some important suggestions were also made by the respondents to improve guidance: increased visibility, need to make surgeons aware of recommendations, regular update, promoting flexible guidelines to be applicable in various environments and countries, focusing guidelines on complex and challenging scenarios. The two previous examples and moreover the suggestions made by the respondents to the last survey certainly question the actual barriers to the implementation of guidelines in clinical practice and the need of adequate strategies to overcome these barriers.Barriers to clinical guidelines implementationSeveral studies have analyzed the principal barriers involved in guidelines implementation. These barriers could be classified in 3 categories which are clearly detailed and summarized in a recent review paper from De Hert.4 Personal factors relate to the physician knowledges and attitudes. As pointed previously, ignorance about existing guidelines, national and others, and how to find them is not uncommon (prospect survey). Further, a negative attitude towards recommendations and evidence-based medicine may also exist. Some physicians think they are ‘cookbook medicine’, they suppress medical freedom, they are a way to control costs, they ignore patient preferences and they exclude individualized medical cares.4 Guideline-related factors represent to date a significant barrier to their common use. These factors are well known and have been pointed out in several publications.3 4 6 Guidelines reliability can be questioned as many have competing recommendations due to a lack of unified methodological strategy,6 they focus on expert opinions or they content recommendations that have since been contradicted by more recent clinical evidence.3 Besides the criticism of being based on expert opinion that limits transparency, guidelines mainly rely on results from RCTs and systematic reviews usually considered as high level of evidence. The problem is that level of evidence in the majority of systematic reviews is generally low to moderate, in agreement with the quality of the included clinical studies. In example, the lack of ‘basic analgesic interventions’ i.e. paracetamol, anti-inflammatory drugs in most of the trials aiming to assess locoregional techniques for mastectomy and for TKA requires critical appraisal and mitigates useful clinical implications.10 11 Systematic reviews and meta-analysis also do not apply to specific populations of patients (e.g. chronic pain patients, drugs dependent patients, frailty patients with several comorbidities).6 Those patients who are usually excluded from RCTs on perioperative pain treatments are the most difficult ones to manage. Finally, the access to guidelines recommendations may not be easy for various reasons (limited visibility), and the guidelines presentation may be too complex (unclear and unfriendly layout, no summary provided). External factors like organizational constraints, lack of resources, lack of time for application in busy clinical practice, and misalignment with patient expectations also preclude a current use of published guidelines.4 Here the weight of guidelines medicolegal impact on current practice might help to better support their application at least by hospital administration and public health services.5 Future of practice guidelines: the example of perioperative pain management guidelinesAs previously stated, the overabundance of perioperative guidelines identified in the literature does not facilitate their application by practicing clinicians.3 Consequently, there is actually a real need to improve perioperative guidelines from utility to presentation and clinical application.3 6 It is also worth noting that in a near future, artifitial intelligence (AI) might become a real competitor to existing guidelines.12 In 2022, an online AI chatbot (i.e. ChatGPT) was released and rapidely attracting attention. The use of AI might allow to shift from static internet information searching to dynamic knowledge gathering.12 Currently ChatGPT seems to provide medical information of comparable quality to available static internet information. A study assessing the reliability of medical information provided by ChatGPT found a 60% agreement between guideline recommendations and AI answers.12 Recently, the PROSPECT group (which methodology is rigorous based on critical assessment of published randomised controlled trials) has examined ChatGPT-generated recommendations for perioperative pain management across five common surgical procedures.13 The authors compared their agreement with the PROcedure-SPECific postoperative pain managemenT (PROSPECT) recommendations. Results found persistent inaccuracies between ChatGPT versions and PROSPECT recommendations.13 These differences might be explained by the insufficient precision of ChatGPT to identify relevant literature to extract their recommendations as well as limited capabilities to assess the relevance and quality of the information contained in the sources and references used to generate responses. However, ChatGPT is still involving.14 and in the future, it might become a provider of medical information, and even more, an adviser for health care professionnals and for the patients.12 Therefore, it becomes mandatory to adapt the novel guidelines to the demand of physicians, focusing on easy access, readability and clinical application.3 As an exemple, the PROSPECT working group which provides procedure-specific pain management recommendations regularly updates its methodology to further meet clinicians needs (see table 1).15 In conclusion, several perioperative pain guidelines are available to help clinicians to improve patient recovery and outcomes. However, current reports and observations show a lack of application of existing recommendations, that strongly questions the potential barriers to guidelines implementation in daily practice. Some barriers directly rely to the guidelines themthelves (design and process) while others concern physician knowledges and attitudes. These barriers to guidelines implementation should be addressed to improve perioperative pain management.Abstract FT20 Table 1PROSPECT working group efforts to provide updated perioperative pain guidelines PROSPECT recommendations: methodology advantages Guidelines procedures in general: methodology limitations Synthesis of clinical evidence for PROcedure-SPEcific pain management (balance risk vs benefits, clinical relevance) - Results from systematic reviews/meta-analysis are not taken as granted (as they may include bias and are sometimes flawed) Bias in actual RCTs and meta-analysis/systematic reviews might limit the recommendations accuracy- Lack of use of basic analgesics (paracetamol, NSAIDs, wound infiltration…)- Patient/population-SPEcific trials are missing (usually the most difficult patients to manage) Unique and well defined methodology with - Rigorous approach to quality assessment used to assign the level of evidence- Collaborative experts consensus (anaesthesiologists and surgeons) to validate clinical application - different approaches used to synthetize the current scientific evidence (lack of unified methodological strategy )- conflictual recommendations Methodology and surgical procedures are updated periodically (5-yrs time lay out) Guidelines presentation include easy and friendly approach for clinicians- summary of analgesic procedures which are recommended or not- infographic presentation summary References Komann M, Baumbach P, Stamer UM, Weinmann C, Arnold C, Pogatzki-Zahn E, Meissner W. Desire to receive more pain treatment - a relevant patient-reported outcome measure to assess quality of post-operative pain management? Results from 79,996 patients enrolled in the pain registry QUIPS from 2016 to 2019. J Pain. 2021;22:730–738.Jena PORG, Chinese POUTn, Dutch POUTn, Mexican POUTn, Serbian POUTn, Spanish POUTn, French POUTn, Italian POUTn, Swiss POUTn, Irish POUTn, et al. Status quo of pain-related patient-reported outcomes and perioperative pain management in 10,415 patients from 10 countries: analysis of registry data. Eur J Pain. 2022;26:2120–2140.Gregory RJ, Gregory SH. Guidelines in anesthesiology: considering when, how, and why. J Cardiothorac Vasc Anesth. 2019;33:2372–2373.De Hert S, Paula-Garcia WN. Implementation of guidelines in clinical practice; barriers and strategies. Curr Opin Anaesthesiol. 2024;37:155–162.Kranke P, Afshari A, Meybohm P, Buhre W, Wiege S, Romero CS. Decoding the meaning of medical guidelines and their medicolegal implications. Eur J Anaesthesiol. 2024;41:109–114.Kehlet H, Memtsoudis SG. Perioperative care guidelines: conflicts and controversies. Br J Surg. 2020;107:1243–1244.Wikkelso AJ. Postoperative pain management for caesarean section in Denmark: a survey of current clinical practice. Acta Anaesthesiol Scand. 2025;69:e70012.Roofthooft E, Joshi GP, Rawal N, Van de Velde M, Anaesthesia PWGotESoR, Pain T. PROSPECT guideline for elective caesarean section: a reply. Anaesthesia 2023;78:1173–1174.Joshi G, Wu C, Moka E, Van de Velde M, Lobo D, group Pw. A survey on knowledge about PROcedure-SPEcific Postoperative Pain Management (PROSPECT) guidelines. Submitted.Mija D, Kehlet H, Joshi GP. Basic analgesic use in randomised trials assessing local and regional analgesic interventions for mastectomy: a critical appraisal and clinical implications. Br J Anaesth. 2023;131:921–924.Joshi GP, Stewart J, Kehlet H. Critical appraisal of randomised trials assessing regional analgesic interventions for knee arthroplasty: implications for postoperative pain guidelines development. Br J Anaesth. 2022;129:142–144.Walker HL, Ghani S, Kuemmerli C, Nebiker CA, Muller BP, Raptis DA, Staubli SM. Reliability of medical information provided by ChatGPT: assessment against clinical guidelines and patient information quality instrument. J Med Internet Res. 2023;25:e47479.Mija D, Kehlet H, Rosero EB, Joshi GP. Evaluating the role of ChatGPT in perioperative pain management versus procedure-specific postoperative pain management (PROSPECT) recommendations. Br J Anaesth. 2024;133:1318–1320.Malek MA, du Fosse N, Boon M. Evaluating the role of ChatGPT in perioperative pain management: importance of version and prompt sensitivity. Comment on Br J Anaesth 2024;133:1318–20. Br J Anaesth. 2025;134:1241–1243.Joshi GP, Albrecht E, Van de Velde M, Kehlet H, Lobo DN, Anaesthesia PWGotESoR, Pain T. PROSPECT methodology for developing procedure-specific pain management recommendations: an update. Anaesthesia 2023;78:1386–1392.