BJA
Correspondence
4 Neuman DN, et al. Trust, but verify. Examining the role of observational data in perioperative decision-making. Anesthesiology 2013; 118: 1008– 9
doi:10.1093/bja/aeu073
Choice of anaesthesia for elective total knee arthroplasty Reply from the authors to Dr Chincholkar
Declaration of interest None declared. A. Harsten Hasssleholm, Sweden E-mail: [email protected] 1 Harsten A, Kehlet H, Toksvig-Larsen S. Recovery after total intravenous general anaesthesia or spinal anaesthesia for total knee arthroplasty. Br J Anaesth 2013; 111: 391– 9 2 McCarteny CJL, Choi S. Does anaesthetic technique really matter for total knee arthroplasty? Br J Anaesth 2013; 111: 331–3
doi:10.1093/bja/aeu074
Editor—We read with interest the study by Harsten and colleagues1 ‘Recovery after total intravenous general anaesthesia or spinal anaesthesia for total knee arthroplasty (TKA): a randomized trial’ and the accompanying editorial.2 Interestingly, the study’s inclination to entertain qualitative parameters (patient’s preference for anaesthesia technique in the future) along with quantitative outcome measures reflects ‘mixedmethods’ study design,3 an exciting and relatively new methodology appropriate for clinical anaesthesia research. However, although the editorial’s complete new-take on the study, that is, comparative evaluation of the recovery outcome after ‘state-of-the-art’ general anaesthesia (SOTA-GA) vs ‘standardof-care’ spinal anaesthesia (SOC-SA) in patients undergoing elective TKA, was insightful and broadened the scope of comprehension, it also invoked some ethical issues. First, not uncommonly in clinical anaesthesiology research, in order to analyse patient outcome responses, one requires to follow both quantitative objective parameters of the anaesthetized patient and the qualitative postoperative subjective patient responses. Therefore, an appropriate ‘mixed-methods’ study design has a tacit relevance in anaesthesia outcome research that justifies study objectives and facilitates valid creation of clinical evidence.3 Secondly, ethically, given that width of the line-of-transition between any SOTA and SOC medical intervention techniques (e.g. spinal/general anaesthesia) may vary as per state of available healthcare infrastructure and the backup governance/ policy, in the current clinical anaesthesia practice/research context, irrespective of the technique applied, one must ensure that the patients in receipt of anaesthesia do not get harmed (do-no-harm)4 and/or denied the best available care (principle of justice). Further, in the absence of clarity and/or nonavailability of consistent scientific evidence as to whether anaesthesiologist’s skill-set, technical gadgetry advantage, or the type of technical intervention impacts SOC-SOTA turnover,5 a SOTA vs SOC comparative effectiveness research (CER)6 in anaesthesia may have an awkward ethical stance whenever an ‘intertechnique’ analysis (SA vs GA) is undertaken. While one may go for ‘intra-technique’, it may be ethically problematic for ‘intertechnique’ evaluation, particularly when one of the techniques is operator-dependent and has a potential for failure (e.g. SA). In the same regard, probably, an inter-technique ‘SOTA-GA’ vs ‘SOC-SA’ evaluation as undertaken in the study1 is likely to incur an ethical imbalance wider than is perceived otherwise, an effect with pronounced ramifications for a randomized study. Collectively, for an authentic outcome analysis research, the IRB investigators have an ethical obligation to closely consider the philosophical gap between two different techniques (SA vs GA) to secure participants’ safety. Also, such a CER study should be accorded appropriate methodology to analyse the outcome
765
Downloaded from http://bja.oxfordjournals.org/ at University of Sydney Library on June 9, 2015
Editor—We thank Dr Chincholkar for his letter on our clinical study.1 2 Using an i.v. opioid towards the end of a targetcontrolled infusion (TCI) anaesthetic is almost to be considered as the modus operandi of this type of anaesthesia. This is due to the fact that remifentanil has a very short-lasting analgesic effect. Hence, this could almost be considered as a part of the TCI technique. It is possible that this could have influenced the postoperative pain scores. This is more of an academic issue since one could argue that TCI and oxycodone could result in a more favourable recovery profile in a fast-track set-up. We agree that the difference in length of hospital stay (6 h) may not be clinically significant. However, sample size was obviously adequate since we were able to detect a statistically significant difference between the groups. Nausea and dizziness was monitored at fixed times twice daily (at 08:00 and again at 14:00 h). However, the number of patients having vomited represents the number of vomiting occasions from the previous measuring point. In conclusion, our study does not show that general anaesthesia (of any kind) is better than regional anaesthesia for patients undergoing total knee arthroplasty (TKA). Our results should be seen in the light of a modern general anaesthetic technique together with a fast-track set-up for patients undergoing elective TKA and it urges for further large randomized trials.
‘State-of-the-art general anaesthesia’ compared with ‘standard-of-care spinal anaesthesia’ for unilateral knee arthroplasty: ethics and philosophical considerations
BJA
Correspondence
Declaration of interest None declared. A. Dutta* N. Sethi P. Choudhary New Delhi, India * E-mail: [email protected] 1 Harsten A, Kehlet H, Toksvig-Larsen S. Recovery after total intravenous general anaesthesia or spinal anaesthesia for total knee arthroplasty. Br J Anaesth 2013; 111: 391– 9 2 McCarteny CJL, Choi S. Does anaesthetic technique really matter for total knee arthroplasty? Br J Anaesth 2013; 111: 331–3 3 Curry LE, Nembhard IM, Bradley EH. Key issues in outcome research: qualitative and mixed methods provide unique contributions to outcomes research. Circulation 2009; 119: 1442–52 4 Smith CM. Origin and uses of primum non nocere—above all, do no harm! J Clin Pharmacol 2005; 45: 371– 7 5 Strauss DC, Thomas MJ. What does medical profession mean by ‘standard of care?’ J Clin Oncol 2009; 27: e192–3 6 Armstrong K. Methods in comparative effectiveness research. J Clin Oncol 2012; 30: 4208–14 7 Marshall J. Consequentialism leads to bad ethics: an illustration. Ethics Alarms. Available from Ethicsalarms.com/2012/05/31 8 Graf C, Wager E, Bowman A, Fiack S, Scott-Lichter D, Robinson A. Best practice guidelines on publication ethics: a publisher’s perspective. Int J Clin Pract 2007; 61: 1–26
doi:10.1093/bja/aeu075
Antifibrinolytic agents in current anaesthetic practice: use of tranexamic acid in lower limb arthroplasty Editor—We read with interest the review by Ortmann and colleagues1 on the role of antifibrinolytic agents as a part of a modern blood conservation strategy for patients at risk of haemorrhage in different clinical contexts. They summarized ‘the literature of the past decade with specific relevance to
766
daily anaesthetic practice, to support clinical decision-making and to enhance further discussion of the topic’. Tranexamic acid (TXA), like 1-aminocaproic acid, prevents the premature dissolution of the haemostatic clot and reduces the volume of bleeding. Although the effectiveness of the TXA to reduce perioperative bleeding has been demonstrated in many studies, its effect on perioperative thromboembolic events and mortality has not been adequately assessed and remains uncertain.2 In fact, TXA package insert recognized that the frequency of arterial or venous embolism of any localization, and also hypersensitivity reactions including anaphylaxis, is unknown.3 This may be especially relevant for the use of TXA in lower limb arthroplasty, where postoperative hypercoagulability has been demonstrated.4 Owing to the progressive population ageing, an exponential increase in the number of interventions for elective total arthroplasty of hip (THA) and knee (TKA) is expected in the forthcoming years.5 Total lower limb arthroplasty is associated with a significant perioperative blood loss and, even if a transfusion protocol is applied, up to 20–30% of patients receive allogeneic blood transfusion (ABT), with a mean transfusion index of two units.6 The efficacy and very low cost of TXA appear as major drivers of its progressively increased use in TKA and THA, despite orthopaedic surgery is not nominally included within the specific indications of TXA in the revised product package insert.3 A review of published meta-analyses showed that TXA produced a variable reduction of perioperative blood loss and proportion of patients requiring ABT,7 but with significant heterogeneity, which requires careful interpretation of the effect of TXA in this context. Since ABT is also a risk factor of hypercoagulability in TKA and THA patients8 and TXA significantly reduces ABT, a reduced rate of thromboembolic complications could be expected. As this reduction has not been demonstrated,7 it could be hypothesized that TXA administration in TKA and THA would induce an increase in the risk of thromboembolic complications, as observed in hip fracture surgery,9 which could be offset by the benefit conferred by the reduction in exposure to ABT. Future randomized trials of sufficient power should be designed to examine the safety of TXA. To confirm its safety, it would be necessary to include appropriate numbers of transfused and non-transfused patients, treated or not with this drug. Thromboembolic complications (primary security variable) must be carefully explored, including not only the incidence of deep venous thrombosis and pulmonary embolism but also that of myocardial infarction, cerebral ischaemia, and mortality. Having in mind the low incidence of these complications (,3%), to detect a 1% difference with an 80% power (b-error) and a 95% confidence interval (a-error), data from at least 5000 patients per arm (control and TXA) would be needed. In the mean time, we should keep in mind that: (i) the available evidence seems to argue for caution, not complacency, regarding the use of TXA in TKA and THA surgery, and (ii) there are other efficacious, safe, and cost-effective alternatives for reducing bleeding, the need for ABT, or both in these
Downloaded from http://bja.oxfordjournals.org/ at University of Sydney Library on June 9, 2015
response(s) across patient-participant’s dual existential state (anaesthetized, awake) in order to facilitate research that upholds participants’ contribution and precludes studies borne out of a priori consequentialistic stance.7 Finally, although the editorial’s representation of ‘add-on’ study limitations of a study that diligently followed statutory ethical requirements (IRB-approval, informed consent) seems relevant and improves overall understanding, it might be misconstrued as ‘displaced-disclosure’ of the ‘missed’ limitations in the original article and therefore undermine publication ethics.8 Epistemologically, unless appropriate measures are taken to ensure that the readers must go through them in combination, it may sometimes lead to suboptimal gain of the propositional knowledge and inconsistent dissemination, especially when they go exclusively through either the original study or the accompanying editorial.
Correspondence
4 Neuman DN, et al. Trust, but verify. Examining the role of observational data in perioperative decision-making. Anesthesiology 2013; 118: 1008– 9
doi:10.1093/bja/aeu073
Choice of anaesthesia for elective total knee arthroplasty Reply from the authors to Dr Chincholkar
Declaration of interest None declared. A. Harsten Hasssleholm, Sweden E-mail: [email protected] 1 Harsten A, Kehlet H, Toksvig-Larsen S. Recovery after total intravenous general anaesthesia or spinal anaesthesia for total knee arthroplasty. Br J Anaesth 2013; 111: 391– 9 2 McCarteny CJL, Choi S. Does anaesthetic technique really matter for total knee arthroplasty? Br J Anaesth 2013; 111: 331–3
doi:10.1093/bja/aeu074
Editor—We read with interest the study by Harsten and colleagues1 ‘Recovery after total intravenous general anaesthesia or spinal anaesthesia for total knee arthroplasty (TKA): a randomized trial’ and the accompanying editorial.2 Interestingly, the study’s inclination to entertain qualitative parameters (patient’s preference for anaesthesia technique in the future) along with quantitative outcome measures reflects ‘mixedmethods’ study design,3 an exciting and relatively new methodology appropriate for clinical anaesthesia research. However, although the editorial’s complete new-take on the study, that is, comparative evaluation of the recovery outcome after ‘state-of-the-art’ general anaesthesia (SOTA-GA) vs ‘standardof-care’ spinal anaesthesia (SOC-SA) in patients undergoing elective TKA, was insightful and broadened the scope of comprehension, it also invoked some ethical issues. First, not uncommonly in clinical anaesthesiology research, in order to analyse patient outcome responses, one requires to follow both quantitative objective parameters of the anaesthetized patient and the qualitative postoperative subjective patient responses. Therefore, an appropriate ‘mixed-methods’ study design has a tacit relevance in anaesthesia outcome research that justifies study objectives and facilitates valid creation of clinical evidence.3 Secondly, ethically, given that width of the line-of-transition between any SOTA and SOC medical intervention techniques (e.g. spinal/general anaesthesia) may vary as per state of available healthcare infrastructure and the backup governance/ policy, in the current clinical anaesthesia practice/research context, irrespective of the technique applied, one must ensure that the patients in receipt of anaesthesia do not get harmed (do-no-harm)4 and/or denied the best available care (principle of justice). Further, in the absence of clarity and/or nonavailability of consistent scientific evidence as to whether anaesthesiologist’s skill-set, technical gadgetry advantage, or the type of technical intervention impacts SOC-SOTA turnover,5 a SOTA vs SOC comparative effectiveness research (CER)6 in anaesthesia may have an awkward ethical stance whenever an ‘intertechnique’ analysis (SA vs GA) is undertaken. While one may go for ‘intra-technique’, it may be ethically problematic for ‘intertechnique’ evaluation, particularly when one of the techniques is operator-dependent and has a potential for failure (e.g. SA). In the same regard, probably, an inter-technique ‘SOTA-GA’ vs ‘SOC-SA’ evaluation as undertaken in the study1 is likely to incur an ethical imbalance wider than is perceived otherwise, an effect with pronounced ramifications for a randomized study. Collectively, for an authentic outcome analysis research, the IRB investigators have an ethical obligation to closely consider the philosophical gap between two different techniques (SA vs GA) to secure participants’ safety. Also, such a CER study should be accorded appropriate methodology to analyse the outcome
765
Downloaded from http://bja.oxfordjournals.org/ at University of Sydney Library on June 9, 2015
Editor—We thank Dr Chincholkar for his letter on our clinical study.1 2 Using an i.v. opioid towards the end of a targetcontrolled infusion (TCI) anaesthetic is almost to be considered as the modus operandi of this type of anaesthesia. This is due to the fact that remifentanil has a very short-lasting analgesic effect. Hence, this could almost be considered as a part of the TCI technique. It is possible that this could have influenced the postoperative pain scores. This is more of an academic issue since one could argue that TCI and oxycodone could result in a more favourable recovery profile in a fast-track set-up. We agree that the difference in length of hospital stay (6 h) may not be clinically significant. However, sample size was obviously adequate since we were able to detect a statistically significant difference between the groups. Nausea and dizziness was monitored at fixed times twice daily (at 08:00 and again at 14:00 h). However, the number of patients having vomited represents the number of vomiting occasions from the previous measuring point. In conclusion, our study does not show that general anaesthesia (of any kind) is better than regional anaesthesia for patients undergoing total knee arthroplasty (TKA). Our results should be seen in the light of a modern general anaesthetic technique together with a fast-track set-up for patients undergoing elective TKA and it urges for further large randomized trials.
‘State-of-the-art general anaesthesia’ compared with ‘standard-of-care spinal anaesthesia’ for unilateral knee arthroplasty: ethics and philosophical considerations
BJA
Correspondence
Declaration of interest None declared. A. Dutta* N. Sethi P. Choudhary New Delhi, India * E-mail: [email protected] 1 Harsten A, Kehlet H, Toksvig-Larsen S. Recovery after total intravenous general anaesthesia or spinal anaesthesia for total knee arthroplasty. Br J Anaesth 2013; 111: 391– 9 2 McCarteny CJL, Choi S. Does anaesthetic technique really matter for total knee arthroplasty? Br J Anaesth 2013; 111: 331–3 3 Curry LE, Nembhard IM, Bradley EH. Key issues in outcome research: qualitative and mixed methods provide unique contributions to outcomes research. Circulation 2009; 119: 1442–52 4 Smith CM. Origin and uses of primum non nocere—above all, do no harm! J Clin Pharmacol 2005; 45: 371– 7 5 Strauss DC, Thomas MJ. What does medical profession mean by ‘standard of care?’ J Clin Oncol 2009; 27: e192–3 6 Armstrong K. Methods in comparative effectiveness research. J Clin Oncol 2012; 30: 4208–14 7 Marshall J. Consequentialism leads to bad ethics: an illustration. Ethics Alarms. Available from Ethicsalarms.com/2012/05/31 8 Graf C, Wager E, Bowman A, Fiack S, Scott-Lichter D, Robinson A. Best practice guidelines on publication ethics: a publisher’s perspective. Int J Clin Pract 2007; 61: 1–26
doi:10.1093/bja/aeu075
Antifibrinolytic agents in current anaesthetic practice: use of tranexamic acid in lower limb arthroplasty Editor—We read with interest the review by Ortmann and colleagues1 on the role of antifibrinolytic agents as a part of a modern blood conservation strategy for patients at risk of haemorrhage in different clinical contexts. They summarized ‘the literature of the past decade with specific relevance to
766
daily anaesthetic practice, to support clinical decision-making and to enhance further discussion of the topic’. Tranexamic acid (TXA), like 1-aminocaproic acid, prevents the premature dissolution of the haemostatic clot and reduces the volume of bleeding. Although the effectiveness of the TXA to reduce perioperative bleeding has been demonstrated in many studies, its effect on perioperative thromboembolic events and mortality has not been adequately assessed and remains uncertain.2 In fact, TXA package insert recognized that the frequency of arterial or venous embolism of any localization, and also hypersensitivity reactions including anaphylaxis, is unknown.3 This may be especially relevant for the use of TXA in lower limb arthroplasty, where postoperative hypercoagulability has been demonstrated.4 Owing to the progressive population ageing, an exponential increase in the number of interventions for elective total arthroplasty of hip (THA) and knee (TKA) is expected in the forthcoming years.5 Total lower limb arthroplasty is associated with a significant perioperative blood loss and, even if a transfusion protocol is applied, up to 20–30% of patients receive allogeneic blood transfusion (ABT), with a mean transfusion index of two units.6 The efficacy and very low cost of TXA appear as major drivers of its progressively increased use in TKA and THA, despite orthopaedic surgery is not nominally included within the specific indications of TXA in the revised product package insert.3 A review of published meta-analyses showed that TXA produced a variable reduction of perioperative blood loss and proportion of patients requiring ABT,7 but with significant heterogeneity, which requires careful interpretation of the effect of TXA in this context. Since ABT is also a risk factor of hypercoagulability in TKA and THA patients8 and TXA significantly reduces ABT, a reduced rate of thromboembolic complications could be expected. As this reduction has not been demonstrated,7 it could be hypothesized that TXA administration in TKA and THA would induce an increase in the risk of thromboembolic complications, as observed in hip fracture surgery,9 which could be offset by the benefit conferred by the reduction in exposure to ABT. Future randomized trials of sufficient power should be designed to examine the safety of TXA. To confirm its safety, it would be necessary to include appropriate numbers of transfused and non-transfused patients, treated or not with this drug. Thromboembolic complications (primary security variable) must be carefully explored, including not only the incidence of deep venous thrombosis and pulmonary embolism but also that of myocardial infarction, cerebral ischaemia, and mortality. Having in mind the low incidence of these complications (,3%), to detect a 1% difference with an 80% power (b-error) and a 95% confidence interval (a-error), data from at least 5000 patients per arm (control and TXA) would be needed. In the mean time, we should keep in mind that: (i) the available evidence seems to argue for caution, not complacency, regarding the use of TXA in TKA and THA surgery, and (ii) there are other efficacious, safe, and cost-effective alternatives for reducing bleeding, the need for ABT, or both in these
Downloaded from http://bja.oxfordjournals.org/ at University of Sydney Library on June 9, 2015
response(s) across patient-participant’s dual existential state (anaesthetized, awake) in order to facilitate research that upholds participants’ contribution and precludes studies borne out of a priori consequentialistic stance.7 Finally, although the editorial’s representation of ‘add-on’ study limitations of a study that diligently followed statutory ethical requirements (IRB-approval, informed consent) seems relevant and improves overall understanding, it might be misconstrued as ‘displaced-disclosure’ of the ‘missed’ limitations in the original article and therefore undermine publication ethics.8 Epistemologically, unless appropriate measures are taken to ensure that the readers must go through them in combination, it may sometimes lead to suboptimal gain of the propositional knowledge and inconsistent dissemination, especially when they go exclusively through either the original study or the accompanying editorial.
