Bespoke intra‐operative anaesthesia – the end of the formulaic approach?

A generation ago, anaesthesia was practised using a largely formulaic approach. The use of anaesthetic agents (both intravenous (i.v.) and inhalational), neuromuscular blocking drugs, opioids, i.v. fluids and lung ventilatory patterns were all delivered according to simple algorithms: administer a fixed aliquot, calculated usually on patient body weight (actual body weight or maybe ideal body weight), patient age or some other parameter, and that particular aspect of the anaesthetic was considered completed. Gradually, we have undergone a paradigm shift towards titrating our interventions to achieve an individualised, appropriate physiological response. Fundamentally, this is logical, as for many interventions, there is a therapeutic window or sweet spot, with too little or too much often leading to adverse effects and even patient harm. Nowhere is this illustrated more clearly than in i.v. fluid management. In the 1970s, the concept of optimisation of cardiac output (and thus oxygen delivery) together with oxygen consumption in the intensive care population led to increased survival rates. This concept was later successfully applied to high-risk patients undergoing major surgery to maximise oxygen delivery and was the harbinger of modern goal-directed fluid therapy and stroke volume optimisation 1. Although the concept is not accepted by all 2 and is still practised in a variable fashion, without doubt there has been a fundamental change in i.v. fluid management, pivotal within modern enhanced recovery after surgery programmes (ERAS), consigning both older fixed regimes to the past (e.g. x ml.kg.h−1 reducing to y ml.kg.h−1 thereafter). What is less controversial is that both excess and inadequate fluid management lead predictably to serious patient harm; either fluid overload or inadequate cardiac output both ultimately leading to cellular hypoxia and dysfunction, with an appreciation that patients with documented high-risk status and/or those undergoing high-risk surgery were most likely to benefit. There are a number of cardiovascular monitoring devices described and these have recently been compared 3. In a similar way, we can also individualise or goal-direct blood transfusion management with point of care coagulation tests such as thromboelastometry (ROTEM®) or thromboelastography (TEG®), aiming for a normal coagulation profile. This minimises the risk of hypercoagulopathy, and thus thrombotic complications, as well as potential cost savings by not overtreating. In cardiac surgery, these have been shown to reduce transfusion requirements, reduce complications, reduce hospital length of stay and improve survival 4. The same logic applies to other therapies. The use of neuromuscular blocking drugs has rightly been viewed as a major advance in anaesthesia, described 75 years ago 5. However, serious problems are attached to their misuse, including accidental awareness under general anaesthesia (AAGA) reported both intra-operatively and immediately postoperatively 6, as well as major problems with inadequate reversal, such as postoperative hypoventilation and pulmonary dysfunction 7. This is not a new problem, and was first described with the older style neuromuscular blocking drugs nearly 40 years ago, but has also been described with the newer agents and indeed is still commonplace 8. Again, a purely formulaic approach to dosing (i.e. calculated as mg.kg−1), allowing a minimum time after a dose before reversal and administering a fixed dose of anticholinesterase to reverse these drugs, is flawed for a number of reasons. There is inherent variability in response of individuals to these drugs and their reversal agents – both for pharmacodynamic reasons (e.g. variability in response within individuals), and pharmacokinetic reasons (e.g. drug interactions or impaired metabolism of these drugs due to organ dysfunction or hypothermia) contributing to inadequate reversal at the end of anaesthesia, emphasising the need for neuromuscular monitors. Moreover, the current interest in deep neuromuscular blockade, which is usually maintained until just before the end of surgery, adds a new dimension to this problem 9. Although various neuromuscular monitors have been around for many years, the gold standard of objective quantitative adequate neuromuscular reversal is a train-of-four ratio of greater than 0.9, commonly measured using acceleromyography 8, and is recommended over qualitative methods 10. The ability to usefully monitor the depth of anaesthesia was, for many years, considered almost impossible, but the advent of bispectral index monitoring (BIS) to assess the degree of electroencephalographic (EEG) cortical suppression has permitted this to be estimated with some degree of confidence 11. Bispectral index allows titration of anaesthetic drugs, which has been shown to reduce postoperative delirium and also cognitive dysfunction at three months (but not at one week) in patients over 60 years of age, while permitting a reduction in both i.v. and volatile anaesthetic dosage 12. Although its definitive place in anaesthetic practice remains to be confirmed, it undoubtedly has a role in the patient who is receiving both total i.v. anaesthesia and neuromuscular blockade 10, as even with sophisticated algorithms, such as target-controlled administration, plasma concentration of propofol may still vary and lead to concerns about depth of anaesthesia. In contrast, the monitoring of end-tidal concentrations of a volatile anaesthetic (which are equated to brain concentrations) is regarded with more confidence to help judge depth of anaesthesia. Thus, BIS may ultimately help to ascertain the depth of an anaesthetic sweet spot so that both inadequate and excessive levels of anaesthesia may be avoided, although it is acknowledged that BIS values alone may be too simplistic and anaesthetists should probably learn to interpret the EEG 13. There has been considerable interest in patients with low mean arterial pressure (MAP), low BIS readings while only receiving low minimum alveolar concentration (MAC) of volatile anaesthetic intra-operatively (the so-called ‘triple low’), as increased postoperative mortality was described in such patients 14, although this is now disputed 15. A recent study demonstrated that a prolonged double low of MAP and BIS may be associated with an increased morbidity and mortality 16. Bispectral index has also been used to guide TIVA techniques using closed-loop delivery systems. A recent meta-analysis demonstrated that BIS-guided delivery of TIVA may reduce propofol dose during induction, maintain a better target depth of anaesthesia and reduce recovery time compared with manual targets 17. This is an evolving area, and although it is acknowledged that low recorded BIS values may reflect increased patient frailty or sensitivity to anaesthetic agents (rather than excessive anaesthetic administration), correctly monitored and tailored anaesthetic depth seems fundamental for our patients. The pattern of intra-operative ventilation of the lungs has been a matter of conjecture for many years, with many anaesthetists either choosing default ventilator settings, or those which appeared to give adequate oxygenation, capnography trace and airway pressures, with a tidal volume based on ideal body weight. Once again, the impetus for change has come from intensive care patients, where low tidal volumes have been used for many years to protect patients with poor lung compliance from barotrauma. The results of various studies suggested that intra-operative lung-protective ventilation with low tidal volumes (6–8 mls.kg−1) with positive end-expiratory pressure (PEEP) of 6–8 cmH2O (with or without recruitment manoeuvres) may reduce postoperative complications 18, 19, although another trial recommended both low tidal volume and low PEEP (< 2 cmH20) without recruitment manoeuvres 20. The superiority of volume-controlled ventilation over pressure-controlled ventilation intra-operatively was confirmed in a recent large study, where there was a reduction in postoperative pulmonary complications 21. However, more recently, a different approach has been described using an individualised strategy. By plotting total dynamic compliance-volume curves it is possible to look at intra-tidal compliance, by which three basic profiles can be identified: recruitment/derecruitment; constant respiratory system mechanics; and overdistension. This then allows the optimal level of PEEP to be titrated for an individual patient to prevent either recruitment/derecruitment or overdistension 22. This permits the individualised application of best PEEP throughout, which may change intra-operatively (e.g. conversion from laparoscopic to open surgery or adopting the prone position). Conspicuously absent in the foregoing list is the objective use of intra-operative analgesia. This is a fundamental area for modern anaesthesia. Although the mainstay of intra-operative analgesia may be opioids, for more than 20 years the multi-modal use of other analgesics has led to opioid-sparing techniques. This has been fundamental in addressing not only the classical issues associated with opioids, such as nausea, vomiting, constipation, respiratory and cough suppression, pruritus, urinary retention; but also with more recent concerns such as opioid-induced hyperalgesia and tolerance. Moreover, opioid misuse has led to the so-called ‘opioid epidemic’ in the US with over 16,000 deaths annually attributable to prescription opioids 23, some of which commences following elective surgery. Postoperative patients are the second highest in frequency to receive opioid prescribing (after chronic pain patients). This has several consequences, including one in 16 surgical patients going on to become chronic opioid users 24. Finally, the importance of opioid use and long-term survival from patients undergoing major cancer surgery is currently topical 25-27. These drugs not only inhibit the function of natural killer cells (which are known to be pivotal in immunosurveillance, via the production of cytotoxic effects on tumour cells) but also have other actions, such as causing cancer cell proliferation via effects on angiogenesis and tumour cell signalling pathways. Although much of the work is in vitro or in animals, this has generated huge interest and prospective human studies are eagerly awaited. So, if we accept that opioid use needs to be minimised where possible, what options do we have? Firstly, the liberal use of remifentanil has to be addressed and its dangers appreciated 28. In addition, the concept of opioid-free anaesthesia (OFA) needs to be revisited. Although not new, many drugs may be co-administered to provide analgesia and suppress autonomic responses. These include alpha-2 agonists, lidocaine, magnesium, esmolol, ketamine, dexamethasone and gabapentinoids. These have a number of potential benefits for day surgery, ERAS programmes, obesity with obstructive sleep apnoea, postoperative cognitive dysfunction (POCD) in the elderly and progression to chronic pain syndromes. Secondly, providing, where possible, effective pre-emptive regional anaesthesia. For example pre-emptive neural blockade for radial fractures provided superior analgesia compared with identically administrated postoperative blockade 29. The major area, however, is the objective assessment of nociception under general anaesthesia. This would allow the correct amount of analgesics to be delivered to the patient, avoiding the risks of excessive opioid administration on the one hand and inadequate analgesia on the other. Up until a few years ago, the assessment of analgesia was somewhat crude, dependent on methods such as assessment of autonomic responses (e.g. sweating) and sympathetic overactivity. In the last few years, the ability to objectively quantify pain under general anaesthesia has sparked interest in the logical management of our patients' analgesic requirements. Several methods have been described based on autonomic responses to pain, including videopupillometry monitoring and the surgical plethsmographic index (SPI). However, the most evidence has come from the analgesia nociception index (ANI) monitor, which uses heart rate variability to assess pain. By analysing the ECG and performing frequency domain-based analysis of the high-frequency component (0.15–0.5 Hz), it is able to assess sympathetic/parasympathetic balance in the autonomic control of the heart rate. It derives a score of 0–100, the numbers indicating the magnitude of parasympathetic predominance in heart rate control. Thus, higher values indicate less nociception and generally a target of over 50 (e.g. 50–70) is used as an indicator of adequate balance between nociception and anti-nociception. Is this a reliable device? There have been several papers published and, although results vary somewhat, it is encouraging to see that this provides a relatively good nociception indicator across a range of surgeries, including paediatrics. This has sparked a number of studies aiming to explore and titrate analgesic administration. If OFA is used, it provides a logical approach to the management of intra-operative nociception. For example, in the commonly encountered clinical scenario of intra-operative tachycardia and/or hypertension, the conundrum is whether or not the patient should receive opioids or sympatholytic drugs. As the ANI score usually falls before the onset of sympathetic activation, it has been used to predict such haemodynamic changes 30. This may permit a logical approach to managing the tachycardia and/or hypertension: if the ANI score is low, then analgesia is required, whereas if high then other drugs (such as beta-blockers) are a logical treatment. It may also permit the anaesthetist to predict postoperative pain permitting a logical approach to the administration of intra-operative analgesia. In particular, it may allow logical administration of non-opioids 31, confirming/quantifying the benefit of the analgesic technique including OFA. Another area is that it may enable the quality of epidural or regional blockade to be assessed before the end of surgery, and whether or not the analgesic regimen used is likely to be providing adequate analgesia, avoiding over- or underdosing of analgesics and the problems associated with both. A major area of interest would be for targeting postoperative analgesia on arrival in the recovery ward. A study by Boselli et al. showed that an ANI score of < 50 was predictive of moderate to severe pain as assessed by a score of > 3 on the 0–10 numerical rating scale (NRS), with a high specificity and sensitivity and high negative predictive value, corresponding with an area under the ROC curve of 0.89 32. This should allow optimal titration of analgesia before the end of surgery and in the immediate postoperative period. It has also recently been described for intensive care patients, where, in spite of low specificity and sensitivity, it had a good negative predictive value for pain (> 90%) when the score was > 43 33. In the future, it may be entirely possible to administer the correct amount of analgesic drugs before the patient arrived in the recovery ward, avoiding over- or underdosing. Combined with the above, although the individual gains may seem small, as a profession undertaking millions of anaesthetics each year, these aggregation of marginal gains may be substantial, with consistently delivered meticulous anaesthesia individually tailored to our patients' needs, undoubtedly producing significant improvements. Some may bemoan the loss of clinical freedom, but the more enlightened will welcome the opportunity to reduce the variation in delivered clinical care, which very often goes pari passu with increased quality of care. Many obvious parallels have been drawn between anaesthesia and the aviation industry and this is yet one more: the future of neither belongs to, nor will be taken over by, technology. Yet both professions work closely with technology to produce the optimum in both safety and outcome, while at the same time retaining and honing our inherent skills. The technology is there to help provide quality anaesthesia every time, yet for the foreseeable future, the anaesthetist (and pilot) remain firmly at the controls, albeit guiding our specialities inexorably from art to science. The authors plan to undertake research with the ANI monitor later in 2018. No other competing interests declared.