Transcript C22_Davidson

Anaesthesia and Neurotoxicity
Andrew Davidson
Royal Children’s Hospital
Melbourne AUSTRALIA
http://www.smarttots.org
Rodent data up to 2010
• Neuronal apoptosis in rodent models
– Ketamine, isoflurane, midazolam, propofol, sevoflurane
– Dose effect
– Combination worse
– Window of vulnerability day 7 in a rat
– Some evidence for long term neurobehavioural effect
Mechanism
• May be related to inactivity
• May be related to changing ontogeny of receptors
• May be due to upregulation of NMDR receptor
Slikker et al. Ketamine-Induced Neuronal Cell Death in the
Perinatal Rhesus Monkey. Toxicological Sciences 2007; 98:
145-158
• Ketamine in monkeys
• Apoptosis
– 24 hours ketamine, 5 day old monkey
• No apoptosis
– 3 hours ketamine, 5 day old monkey
– 24 hours ketamine, 35 day old monkey
• Big doses
• Need big doses in monkeys to have an effect
• Day 6 monkeys
• 5hrs isoflurane 0.7-1.5%
• Increased apoptosis
Paule et al. Ketamine anesthesia during the first week
of life can cause long-lasting cognitive deficits in rhesus
monkeys. Neurotoxicol Teratol 2011
• Monkeys exposed to 24 hrs ketamine as day 5 infants
• Now 3½ years old: cognitive impairments
– poorer performance in learning and colour and position
discrimination tasks
– deficits in accuracy of task performance & response speed
– differences in motivation
• Day 15 rat pups
• 5hrs anaesthesia: propofol,
ketamine, midazolam
• Increased dendritic spine density
Control
Propofol
Ketamine
• Day 16 rat pups
• Isoflurane, desflurane,
sevoflurane
• 30, 60, 120 minutes
• No cell death
• Increased spine density
Control
30 min
60 min
120 min
Which agents are bad?
• Isoflurane, desflurane, sevoflurane
Which agents are bad?
• Isoflurane, desflurane, sevoflurane
• Midazolam, diazepam, clonazepam
• Phenobarbital, pentobarbital
• Chloral hydrate
• Propofol
Which agents are good?
• Dexmedetomidine, xenon
– no apoptosis
– “protective”
• Opioids
– no evidence for apoptosis
Problems with animal studies
• Duration of exposure
• Dose of agent
• Monitoring
• Length of neurodevelopment
• Plasticity
• Effect of surgery
• Lumbar intrathecal morphine
• Rats – P3, P10, P21
• Therapeutic dose
• Toxicity
• Therapeutic index
Therapeutic index
• Toxic dose/effective dose
• P3:
• P21:
>3/0.01
>3/0.15
>300
>20
• Rats; P3, P7, P21
• Ketamine; 3-10 mg/kg
• Effective dose
• Toxicity
• Therapeutic index
Therapeutic index
• Toxic dose/effective dose
• P3
• P21
3/3
15/15
1
1
Human studies
2008 Mayo Clinic study
• 5357 children in a population based retrospective birth cohort
– “Rochester epidemiology project”
– Register of all children born 1976-82 in five townships in
Olmsted county Minnesota who stayed local for 5 years
• 593 surgery before age of 4
• Adjusted for gender, birth weight, gestational age
• 932 had learning disability
Unadjusted hazard
ratios
Adjusted hazard ratios
Any anaesthetic (593)
1.27 (1.05- 1.53)
1.20 (0.99-1.46)
1 (449)
1.05 (0.84- 1.32)
1.00 (0.79- 1.27)
2 (100)
1.78 (1.22- 2.59)
1.59 (1.06- 2.37)
3 or more (44)
2.50 (1.55- 4.04)
2.60 (1.60- 4.24)
Dose effect – increased risk of disability with duration and number of
anaesthetics
J Neurosurg Anesth
• 383 children born in NY state cared for by Medicaid that had a
hernia repair < 3yrs of age
• 5050 randomly selected controls matched on age
• Adjusted for age, gender, race and presence of complicating
diagnoses at birth
• Behavioral or developmental disorder
– 17 in hernia group (4.4%)
– 59 in non-hernia group (1.2%)
• Adjusted Hazard Ratio 2.3 (1.3 - 4.1)
• Danish birth cohort 1986-1990
• 2689 inguinal hernia repair
• 14,575 Controls (5% of all children in Denmark)
• Outcome school test at 9th grade (age 15-16 years)
• Hernia group do worse
• No evidence for an association when adjusted for confounding
factors
Twin Research and Human Genetics 2009
• Twin study: monozygotic concordant-discordant design
• 1143 monozygotic twin pairs born 1986-95
• Any anaesthesia
– Prior to 3
– Prior to 12
• Educational achievement at age 12
Problems with human studies
• Little idea which age is most at risk & many studies have
older children
• No idea how long an exposure is bad
• Bias is difficult to eradicate in cohort studies that compare to
population norms
• Little idea which outcome to look at & many studies have
multiple outcomes and very course outcomes
• Confounding
– Many known strong confounding factors
– Probably many unknown confounding factors
– Adjustments are not perfect & BIG doesn’t really help
Anaesthesia is associated with surgery
Surgery is associated with pathology
Surgery  poor outcome
•
•
•
•
•
•
Hormonal “Stress”
Inflammatory response
Circulatory instability
Respiratory compromise
Extra lines & handling
Temperature instability
Pathology  poor outcome
•
•
•
•
Genetic abnormality
Malformations
Prematurity
Sepsis
Surgery or anaesthesia?
• Not able to disentangle effect of surgery and anaesthesia
• Surgery may be the harm
• Anaesthesia may have benefits to reduce surgical harm
Anaesthesia
Good
Reduces stress
Reduces pain
Neuro protection
Bad
Apoptosis
Dendritic development
Effects may be disproportionate in different situations
Summary
• Animal evidence
– Strong for histological change
– Some evidence for change in function
• Human evidence
– Some evidence for an association between
surgery/anaesthesia and poor outcome
– Role of anaesthesia very unclear
Recommendations
• “Avoid elective surgery in infants”
• Don’t withhold analgesia and anaesthesia for
necessary surgery and procedures
• Is one drug better ?
– Avoid prolonged use of high dose ketamine in
infants
– Dexmedetomidine, opioids may be preferable
• Be very careful changing safe established practices
due to theoretical risks
Future studies
• GAS study
– RCT hernia GA versus RA
• Raine cohort
– Western Australian birth cohort
• PANDA study
– Hernia repair and matched