Written by: Dr T Green, CT3 Anaesthetics, Morriston Hospital, Swansea

SmartTots Update Regarding Anesthetic Neurotoxicity in the Developing Brain. Order BA, Suresh S, Evers AS.

Anesthesia & Analgesia 2018 doi: 10.1213/ANE.0000000000002833

In December 2016, the US Food and Drug Administration (FDA) issued a Drug Safety Communication about the potential for neurotoxicity from anaesthetic agents on the developing brain of young children and foetuses.  SmartTots is a public-private partnership between the International Anaesthesia Research Society and the US FDA.  SmartTots work in collaboration with researchers to investigate such potential neurotoxicity from anaesthetic drugs.

This special article provides an update to the current understanding with a brief discussion of two clinical trials partially sponsored by the SmartTots collaboration, one completed and one ongoing.  There is then a brief discussion of recent epidemiological studies, paving the way for the upcoming conclusion of the TREX trial.

The first SmartTots sponsored trial is the General Anaesthesia Compared to Spinal Anaesthesia trial.  This aims to compare scores on an Intelligence Quotient (IQ) test measured at five years of age, comparing children undergoing surgery with general anaesthesia to those receiving a regional anaesthetic technique.  The secondary outcome of performance differences at the age of two years is available and encouragingly shows performance deficit when comparing the two groups.  The primary outcome will be available later in 2018.

The second SmartTots sponsored trial was the Pediatric Anesthesia and Neurodevelopment Assessment Study, an observational cohort design with the primary outcome to assess neurocognitive performance and behavioural outcomes in children aged eight to fifteen years, who underwent hernia repair before three years of age compared to the performance of their unexposed siblings.  The results did not support an adverse effect of anaesthetic agents on the assessed parameters.  Given the short duration of anaesthesia for a hernia repair operation, these results are felt to be in keeping with preclinical data indicating the safety of short (less than two hours) exposure to anaesthetic drugs.

An epidemiological study independent of SmartTots examined the performance of children on a primary school entrance exam, finding no performance deficit in children undergoing general anaesthesia when aged less than two years compared to age matched controls.  Interestingly in contradiction to other preclinical and clinical studies, there was a small performance deficit in children aged greater than two years when undergoing general anaesthesia.

A second epidemiological study compared overall school grades and military conscription IQ tests between those undergoing general anaesthesia on at least one occasion before the age of four years.  The results were similar for one exposure to general anaesthesia and for multiple exposures, with a small reduction in school grades and military conscription IQ tests scores although no details of duration of general anaesthesia or agent used is available.

A Study to Compare the Long-term Outcomes After Two Different Anaesthetics (TREX) Trial is an ongoing trial comparing neurodevelopmental outcomes after anaesthesia for surgery with low dose sevoflurane in combination with remifentanil and dexmedetomidine versus standard sevoflurane-based anaesthesia with anaesthetic duration of at least two and a half hours.

A description of preclinical evidence for the neurotoxicity of anaesthetic agents on the developing brain is given but is covered in the summary of “Long-term neurocognitive outcomes following surgery and anaesthesia in early life.”

This article is intended as a brief update to the current evidence in the debate over the safety of general anaesthesia in young children with immature brain physiology.  It is encouraging that clinical trials have not demonstrated a significant deficit in neurocognitive or neurodevelopmental parameters following exposure of the developing brain to anaesthetic agents.  The mainstream media and tabloids were avidly reporting the potential for harm from general anaesthesia to the developing brain when the FDA Drug Safety Communication was released and the studies summarised in this update are helpful in counselling parents of young children who need to undergo anaesthesia for surgery.

Long-term neurocognitive outcomes following surgery and anaesthesia in early life. Hansen TG, Engelhard T.

Current Opinion in Anaesthesiology 2018 doi:10.1097/ACO.0000000000000577

Controversial and alarming warnings about the potential for harm to the developing brain from exposure to anaesthetic agents have been given, such as the US Food and Drug Administration Drug Safety Communication.  These have been largely based on preclinical animal studies.  This brief review assesses these preclinical studies alongside the results of some human studies.

Preclinical animal studies including nonhuman primates have indicated an association between exposure to anaesthetic agents and neuroapoptosis and abnormal synaptic development, leading to neurocognitive impairment.  The studies seem to implicate all classes of general anaesthetic agent.

In the preclinical animal studies, the anaesthetic technique is markedly different to that employed in clinical practice.  Multimodal monitoring techniques and airway management are technically difficult or impossible to implement on neonatal small animals and therefore such monitoring and management as would be standard in clinical practice was not implemented, leading to likely hypoventilation, hypoxia and hypotension.  This is reflected in the (unreported) excessively high mortality of the animal subjects.  It is widely accepted that physiological derangements such as hypoxia and hypotension can be detrimental to the central nervous system and the studies have no mechanism for excluding this as the cause of observed neurocognitive deficits.

Studies to date are unable to pick apart the likely complex interplay of the exposure to anaesthetic agents, the stress response to surgery and the effects of the underlying condition necessitating surgery.

Also reviewed are human trials examining the neurocognitive and behavioural effects of anaesthetic exposure to the immature brain.  These are discussed further in “SmartTots Update Regarding Anesthetic Neurotoxicity in the Developing Brain”.  In essence, there is no convincing evidence of a clinically significant detrimental impact of anaesthetic agent exposure on neurocognitive and behavioural outcome in later life.

Providing surgery without anaesthesia would be ethically unacceptable and therefore the risks of delaying surgery (and therefore anaesthetic exposure) needs to be balanced against the potential risks from such anaesthetic exposure.  The evidence provided so does not justify a change to current clinical practice.  We should however be prepared to be asked about this topic by concerned parents of children requiring surgery.

*These two articles follow on from articles reviewed in the journal watch of January 2017 and April 2017. This is a topic that is unlikely to disappear soon given the coverage in mainstream and tabloid media.

Cognitive decline in the middle-aged after surgery and anaesthesia: results from the Wisconsin Registry for Alzheimer’s Prevention cohort

Bratzke LC, Koscik RL, Scheming KJ et al. Anaesthesia 2018 doi:10.1111/anae.14216

Recent interest has been seen in the neurocognitive effects of surgery and anaesthesia on children and the elderly.  The middle-aged however have been assumed to be non-susceptible to such effects and Patel et al (Anaesthesia 2016 71: 1144-52) when assessing cognitive decline in the elderly following anaesthesia and surgery state “A younger cohort would be unlikely to show any effect, since very few would be cognitively impaired before surgery”.

In this study, the authors conduct a prospective longitudinal cohort study to assess the effect anaesthesia and surgery may have on neurocognitive parameters in the middle-aged with no neurocognitive deficit at baseline.  Participants were sought from the Wisconsin Registry for Alzheimer’s Prevention (WRAP).  Serial psychometric analysis was performed at baseline, five years and nine years.  Patient-reported procedures in the intervening periods between psychometric testing were explored in further detail by analysis of healthcare records (anaesthetic charts, operation notes and discharge summaries).

964 candidates were included in this study with 212 having surgery before the first cognitive assessment and 130 having surgery before the second cognitive assessment.  The mean age at baseline was 54 years.

The findings were a decline in immediate memory in participants undergoing one or more surgeries during the study with an association noted between number of surgeries, cumulative duration of anaesthesia and surgery and the American Society of Anesthesiologists (ASA) physical grade.  No difference in working memory or executive function were identified.  The decline in immediate memory, while statistically significant, is of questionable clinical importance, being sufficiently small so as to not qualify as mild cognitive impairment or dementia.

This study does exhibit limitations, some of which are recognised by the authors.  Reliance on patient reporting of relevant surgical procedures may lead to omission of some relevant procedures with subsequent skewing of the results.  The surgery, although heterogenous in nature was controlled in that intracranial procedures and those requiring cardiopulmonary bypass (both directly linked to neurocognitive changes) were excluded.  Importantly however the type of anaesthesia was heterogenous, with general anaesthesia and neuroaxial or regional techniques combined together with no subgroup analysis reported.  Retrospective analysis of the study power showed 80% power and this combined with the small effect noted, leaves the question of cognitive decline in the middle-aged after surgery and anaesthesia unanswered.

Comprehensive geriatric assessment for older people admitted to a surgical service (Review). Eager G, Taheri A, Chen S et al.

Cochrane effective practice and organisation of care group 2018 doi:10.1002/14651858.CD012485.pub2

Comprehensive Geriatric Assessment (CGA) is a multidisciplinary diagnostic process utilising medical, psychosocial and functional assessment to develop an individualised management plan for older people.  Members of the multidisciplinary team include geriatric assessment trained physicians, nursing staff, physiotherapists and occupational therapists.  Previous research has shown it to be effective in both internal medicine and orthogeriatric patients.

Given the ageing population and the estimate that more than fifty percent of surgery is performed on those aged over sixty five years, mechanisms to improve overall care and outcomes from surgery in older people is of extreme importance.

This Cochrane systematic review aims to assess the use of CGA in the surgical population.  The primary outcomes were mortality and discharge to an increased level of care compared to baseline.  Secondary outcomes were length of stay, readmission rate, postoperative complications and total cost of care.  Subgroup analysis was performed to assess:

  • Orthopaedic vs other surgical specialties
  • Timing of CGA (preoperative vs postoperative)
  • Emergency vs elective surgery

Eight randomised trials were included in the systematic review, representing 1843 patients overall.  Of the eight trials, seven were specific to hip fractures and one was specific for elective oncology surgery.

Primary outcome results are reported as orthopaedic surgery alone or combined with the solid tumour surgery trial.  Mortality was reduced in hip fracture surgery but this was largely mitigated when elective oncology surgery was included.  Discharge to an increased level of care was reduced in patients undergoing surgery for hip fractures however inclusion of elective oncology surgery patients resulted in little to no effect on length of stay.

Secondary outcome results showed marked heterogeneity for length of stay but an overall small reduction was found.  No change in the rate of readmission was found but only two orthopaedic trials and one elective oncology surgery trial reported this measure, with relatively small numbers of patients overall.  Cost analysis was limited as only a single study reported this.  There appears to be a modest reduction in total cost at one year follow-up (€59,486 standard care vs €54,332 CGA).  Postoperative complications were not amenable to systematic review due to differing ways of reporting these.

Subgroup analysis is reported above for orthopaedic vs other surgical specialties.  Post-operative CGA was found to probably reduce mortality and length of stay compared to standard care but the confidence interval crosses 1.  No assessment of emergency vs elective surgery was made.

Overall, this systematic review provides evidence that CGA provides reductions in mortality and length of stay in hip fracture patients but is unable to generalise these findings to other orthopaedic procedures or non-orthopaedic surgery due to the paucity of studies.  As such, this systematic review is unable to answer the intended question – does Comprehensive Geriatric Assessment improve the outcome after admission to surgical services?  Further research is needed to extend the variety of trials in non-hip fracture orthopaedic surgery and non-orthopaedic surgery.

Preoperative versus postoperative ultrasound-guided rectus sheath block for improving pain, sleep quality and cytokine levels in patients with open midline incisions undergoing transabdominal gynaecological surgery: a randomized-controlled trial. Feng Jin, Zhe Li, Wen-fei Tan, Hong Ma, Xiao-qian Li and Huang-wei Lu.

BMC Anesthesiology 2018, 18:19 doi: 10.1186/s12871-018-0485-9

Rectus Sheath Block (RSB) is a commonly used, effective and safe analgesic technique for midline incisions for surgical procedures.  It is commonly performed preoperatively and has been shown to reduce the intraoperative anaesthetic requirements.  It has been suggested however that improved results may be evident with postoperative RSB as this will prolong the duration of action of the local anaesthetic, with improved analgesia.

This study aimed to examine preoperative versus postoperative RSB with specific regards to time to first use of patient-controlled analgesia (PCA) pump, total dose of opioid analgesic required using the PCA pump, levels of inflammatory cytokines (Interleukin-6 / IL-6, Interleukin-1β / IL-1β, Interferon-γ / IFN-γ, Tumour Necrosis Factor-α / TNF-α) measured by serial venepuncture and sleep quality assessed using Bispectral Index (BIS) monitoring (elevated IL-6 has been associated with reduced sleep quality).  Several studies have been performed validating the use of BIS monitoring on depth of natural sleep (e.g. Giménez et al 2017, J Clin Monit Comput.  Doi 10.1007/s10877-015-9805-5)

Sixty four patients undergoing elective non-cancer open abdominal gynaecological surgery were randomised to either receiving preoperative RSB with 0.5% ropivacaine followed by postoperative RSB with saline or the reverse.  The RSB was performed by a single, experienced anaesthetist and the overall anaesthetic technique was standardised.

The findings showed no significant difference between the visual analogue pain scale rating between the two groups, the time to first use of the PCA pump nor the total opioid requirement at twenty four or forty eight hours postoperatively.  Markers of surgical stress in the form of proinflammatory cytokines were consistently and significantly reduced and BIS data showed a statistically significant improvement in time spent at stage N2 and N3 sleep for those receiving preoperative RSB.

Statistical analysis found that thirty two patients would be needed in each arm for a power of 80% however due to problems with BIS monitoring in two patients, the preoperative RSB group ultimately had thirty patients (thirty two in the postoperative RSB group), leaving the study underpowered overall.

This study is a neat demonstration of the benefit to using preoperative rectus sheath block in open non-cancer gynaecological surgery.  With small numbers leaving the study underpowered and the inclusion of only one type of surgery, further work is needed.

Effective dose of dexmedetomidine as an adjuvant sedative to peripheral nerve blockade in elderly patients. Wang C, Zhang H, Fu Q.

Acta Anaesthesiologica Scandinavia 2018 doi:10.1111/aas.13087

Dexmedetomidine is a relatively new drug with an increasing repertoire for use.  This study assesses the dose required for sedation following peripheral nerve blockade for total knee arthroplasty in older patients.

Forty two patients aged 65 – 85 years were included in the study (two patients were excluded following failure of peripheral nerve blockade and one was excluded due to vomiting affecting the assessment of sedation).  The patients were stratified into two groups – “young-old” aged 65 -74 years and “middle-old” aged 75 – 85 years.  A single calculated dose of dexmedetomidine was administered as a ten minute infusion and the level of sedation assessed using the Observer’s Assessment of Alertness / Sedation Scale.  This was correlated with Bispectral Index (BIS) value alongside usual monitoring parameters.

The dose of dexmedetomidine was adjusted for serial patients based on the previous patient’s reposnse (the Dixon up-down method).  From these data, Effective Dose 50% (ED50) and Effective Dose 95% (ED95) can be determined.

The results showed:

  Young-old Middle-old
ED50 (mcg/kg) 0.57 (0.47 – 0.65) 0.38 (0.28 – 0.46)
ED95 (mcg/kg) 0.76 (0.67 – 1.08) 0.57 (0.48 – 0.89)
BIS50 86 (81 – 91) 85 (84 – 87)
BIS90 72 (53 – 78) 73 (69 – 76)


The ED50 showed increased sensitivity with advancing age, in-keeping with previous studies and the BIS values showed no statistical difference between the age groups.  No patient experienced oxygen desaturation and haemodynamic monitoring showed a modest reduction in heart rate and systolic blood pressure without adverse effects.

Limitations of the study include the use of midazolam (1mg) and fentanyl (50 micrograms) prior to nerve block as these agents will confound the sedative effect of dexmedetomidine.  The study is limited to a single surgical procedure and anaesthetic technique and the data from this study are therefore not generalisable to other operations or anaesthetic techniques.  The number of patients enrolled is quite small but this study serves an important role for setting a benchmark for single dose dexmedetomidine as a sedative adjunct to peripheral nerve blockade.