FRCA Notes


Post-Operative Nausea and Vomiting

The curriculum asks for "appropriate regimes for prevention and treatment of PONV".

This topic appeared as an SAQ in March 2017, with the bulk of the marks on offer for knowing the risk factors for, and unwanted effects of, PONV.

An abysmal 36% pass rate in a CRQ on PONV from September 2020 left examiners "surprised at the lack of knowledge in what is a very common anaesthetic topic".

A repeat CRQ on the topic in 2024 was "well answered" except for the side effects of ondansetron.

A September 2024 CRQ on dystonic reactions to anti-emetics saw candidates drop marks on anti-emetic strategies, and a middle ear surgery question in the same sitting found candidates unable to explain why these patients are prone to PONV.

Relevant Primary FRCA topics

Resources


  • Post-operative nausea and vomiting is any nausea, vomiting or retching that occurs within 48hrs of surgery

  • Related terms include:
    • POV: post-operative vomiting (the outcome usually measured in children)
    • PDNV: post-discharge nausea and vomiting (relevant to day-case surgery)

  • 30-33% of patients undergoing elective surgery will experience N&V if no prophylaxis is given
  • Incidence may be as high as 70-80% if multiple risk factors present

  • NB PONV in paediatric patients is covered separately


Patient Organisational
Distress and reduced satisfaction Delayed return to baseline mobility
Increased pain Prolonged recovery and PACU stay
Raised ICP/IOP from retching Delayed day-case discharge
Dehydration from reduced oral intake Unplanned admission following day surgery
Secondary haemorrhage due to raised venous pressure Increased healthcare costs
Wound dehiscence
Aspiration
Oesophageal trauma
Electrolyte abnormalities


  • See the physiology of nausea and vomiting page for more detail

  • The vomiting centre is a collection of effector neurones in the medulla, projecting to the vagus and phrenic nerves and the spinal motor neurones supplying the abdominal muscles
  • It receives input from:
    • The chemoreceptor trigger zone (CTZ) in the area postrema, which lies outside the blood-brain barrier so can 'sample' circulating emetogenic drugs and toxins
    • The vestibular apparatus
    • GI tract via vagal afferents, including 5-HT released from enterochromaffin cells in the gut
    • Cardiovascular afferents, peripheral pain pathways and the higher cortical/limbic centres (anxiety, fear)

  • The large number of receptors and inputs involved is the rationale for a multimodal approach to prophylaxis and treatment:
Site Receptors Anti-emetics acting here
CTZ/area postrema D2, 5-HT3, NK1, MOP Metoclopramide, droperidol, prochlorperazine, ondansetron, aprepitant
Vomiting centre and vestibular system H1, muscarinic (M1) Cyclizine, hyoscine
GI tract (vagal afferents) 5-HT3, D2, NK1, MOP Ondansetron, metoclopramide
Nucleus tractus solitarius 5-HT3, NK1, MOP Ondansetron, aprepitant



Patient factors Anaesthetic factors Surgical factors
Female gender (OR ~3) Volatile anaesthetic agents (OR ~2, dose-dependent) Long duration of surgery
Non-smoker (OR ~2) Use of nitrous oxide (OR ~1.4) Laparoscopic surgery
History of motion sickness or PONV (OR ~2) Use of opioids intra-/post-operatively (dose-dependent) Gynaecological and urological surgery
Age <50yrs General (vs. regional) anaesthesia ENT surgery, particularly middle ear
Paediatric patient Duration of anaesthetic exposure (>45mins) Neurosurgery, particularly posterior fossa
Dehydration Hypotension (e.g. after neuraxial block or orthostatic in PACU) Strabismus (squint) surgery
Delayed gastric emptying Neostigmine reversal of NMBA Breast and thyroid surgery
Upper GI surgery
Surgery causing bleeding into the airway or GI tract e.g. dental, plastics

Why do these factors change risk of PONV?

  • Smoking:
    • Down-regulation of nicotinic acetylcholine receptors occurs
    • Smokers become habituated to the emetogenic compounds in tobacco smoke
    • Hepatic enzyme (cytochrome P450) induction speeds metabolism of emetogenic anaesthetic drugs

  • Motion sickness reflects vestibular sensitivity, with changes in position and movement post-op stimulating H1 and muscarinic receptors in the vestibular system

  • Nitrous oxide diffuses into closed air spaces
    • Middle ear: pressure changes stimulate the vestibular system
    • Bowel: distension stimulates visceral afferents

  • Volatile agents: stimulate serotonin release in the GI tract, increase vestibular sensitivity and stimulate the CTZ
    • They appear to be the main cause of early (0-2hrs) PONV

  • Opioids:
    • Direct stimulation of μ receptors in the CTZ and vestibular system
    • Reduced GI motility and delayed gastric emptying cause bowel distension and ileus
    • Serotonin release acting on 5-HT3 receptors of vagal afferents

  • Hypotension: reduced cerebral blood flow stimulates the vomiting centre

  • Surgical factors:
    • Laparoscopy: raised intra-abdominal pressure from CO2 insufflation, compounded by head-down positioning
    • Strabismus surgery: traction on the extraocular muscles
    • ENT, airway and neurosurgery can cause direct stimulation of vagal and glossopharyngeal afferents and thus central vomiting centre receptors
    • Swallowed blood (tonsillectomy, dental, oral, upper GI surgery) is a potent emetogenic stimulus

Factors of limited or no relevance

  • Factors which have been shown to have little or no clinical relevance include:
    • Obesity/BMI
    • Anxiety
    • Pre-operative fasting
    • Intra-operative nasogastric tube
    • Supplemental oxygen
    • Pregnancy (in women having non-obstetric surgery)
  • ASA grade and phase of the menstrual cycle remain controversial

  • There are a few risk-scoring systems available e.g.  Apfel, Koivuranta or Palazzo
  • The Apfel and Koivuranta scores are perhaps better than the Palazzo score in predicting PONV (Anaesthesiologica Scandinavica, 2001)
  • They are at best moderately accurate and don't provide a risk threshold above which anti-emetics should be administered, or below which they should not (Anaesthesia, 2005)
  • The Apfel is probably the most widely used, as it is marginally simpler than the Koivuranta score
  • A polygenic (genetic) risk score has been developed, but adds nothing clinically meaningful to the traditional risk factors

Apfel score

  • Score 1 point for each of:
    • Female gender
    • Non-smoker
    • Previous history of motion sickness or PONV
    • Planned post-operative use of opioids

  • Each additional risk factor increases the risk of PONV by ~20%:
Score Risk of PONV in 24hrs post-op.
0 10%
1 21%
2 39%
3 61%
4 79%

  • This is reasonably similar to the outcome of the Koivuranta score, which adds an extra possible point (for surgery duration >1hr)
  • Scoring full marks on the Koivuranta score nets you a PONV risk of 87%

Apfel PDNV score

  • Used to predict nausea and vomiting after discharge from day-case surgery
  • Score 1 point for each of:
    • Female gender
    • Age <50yrs
    • History of PONV
    • Use of opioids in PACU
    • Nausea in PACU
Score Risk of PDNV
0 7%
1 20%
2 28%
3 53%
4 60%
5 89%

Using risk scores in practice

  • Two broad strategies exist:
    • Risk-adapted: the number of prophylactic anti-emetics given is tailored to the patient's risk score
    • Universal multimodal: (almost) everyone gets combination prophylaxis, with more for those at higher risk

  • The argument for risk-adapted prophylaxis:
    • In low-risk patients the NNT for prophylaxis approaches the number needed to harm, so routine prophylaxis exposes many to side-effects for little benefit
    • Prophylaxis is most cost-effective in those with >2 Apfel risk factors
    • Low-risk patients can simply be prescribed rescue therapy

  • The argument for a universal approach:
    • Risk scores are imperfect, and in practice moderate- and high-risk patients are often under-treated
    • A simple, standardised protocol is easier to follow and may reduce disparities in care
    • E.g. a simplified algorithm of triple prophylaxis (dexamethasone, ondansetron plus TIVA or droperidol) for women and dual prophylaxis (dexamethasone and ondansetron) for men significantly reduced PONV

  • Whichever strategy is used, institutional PONV rates fall with a written protocol, electronic prescribing reminders and ongoing individual performance feedback to anaesthetists

Drug (class) Mechanism NNT (0-24hrs) Notable side-effects
Ondansetron (5-HT3 antagonist) 5-HT3 antagonism centrally (CTZ) and peripherally (gut vagal afferents) 6-7 Headache, constipation, flushing, bradycardia, QTc prolongation
Dexamethasone (corticosteroid) Unclear; possibly reduced arachidonic acid release, reduced 5-HT turnover or reduced BBB permeability 4 Hyperglycaemia; caution in diabetics
Droperidol (butyrophenone) D2 antagonism at CTZ 5 Sedation, QTc prolongation, extrapyramidal effects, hypotension (α-blockade)
Metoclopramide (benzamide) D2 antagonism at CTZ, weak 5-HT3 antagonism; prokinetic 30 (10mg)
16 (25mg)
Extrapyramidal effects (up to 72hrs post-dose, more common in young women); arrhythmias and hypotension with rapid IV injection
Prochlorperazine (phenothiazine) Central D2 antagonism Extrapyramidal effects, sedation, hypotension
Cyclizine (antihistamine) H1 antagonism, with anticholinergic properties Tachycardia, drowsiness, blurred vision, pain on injection
Hyoscine (anticholinergic) Muscarinic antagonism (vestibular and vomiting centre) Sedation, dry mouth, blurred vision, central anticholinergic syndrome
Aprepitant (NK1 antagonist) NK1 (substance P) antagonism centrally Generally well tolerated; reduces efficacy of hormonal contraception
Propofol (TIVA) Enhances GABA activity; inhibits 5-HT3 receptors; reduces dopaminergic activity 5

  • Ondansetron is ineffective for vomiting caused by motion sickness or dopamine agonists; reduce the dose in hepatic impairment
  • Metoclopramide appears most effective for PONV when given at the end of surgery rather than at induction
  • The use of droperidol declined owing to concerns over QTc prolongation
  • Hyoscine is well absorbed via transdermal, SC and IM routes, but poorly absorbed orally

Perioperative approach to reduce incidence of PONV


  • Identify patients at risk of PONV (and PDNV in day-case patients) at anaesthetic (pre-)assessment
  • Use of prophylaxis should be guided first by the presence of risk factors, and second by the potential morbidity PONV would cause for that patient

  • Anxiolysis with benzodiazepine premedication may reduce PONV in at-risk patients
    • Anxiety is not a robustly proven risk factor for PONV, but treating it may help
    • Benzodiazepines may modulate central inputs to the vomiting centre, or reduce anticipatory nausea, thus helping prevent PONV
  • Keeping starvation times to a minimum and use of peri-operative IV fluid hydration

Address modifiable risk factors

  • Avoid GA altogether i.e. use a regional or neuraxial technique
  • If GA is required, still use regional/neuraxial techniques in order to reduce both opioid and maintenance anaesthetic requirements

  • Use a TIVA technique
    • Reported absolute risk reduction in the region of 15 - 25%
    • Benefit comes both from avoiding volatile agents and nitrous oxide, and from propofol's own anti-emetic properties
    • A 2018 meta-analysis found propofol maintenance was associated with a relative risk of PONV of ~0.6 vs. volatile maintenance
    • Even employing a mixed propofol-volatile technique is superior to volatile agents alone with regards to incidence of PONV

    • Using sub-hypnotic doses of propofol (17mcg/kg/min) during volatile anaesthesia is mooted to reduce incidence of PONV
    • This has not been replicated and it seems sensible to just use TIVA if avoidance of PONV is the goal

  • Don't use nitrous oxide

  • Use sugammadex reversal of NMBA, which is associated with a significantly lower incidence of PONV vs. neostigmine reversal (BJA, 2022)

  • Use generous fluid hydration, especially in those undergoing day surgery
  • Avoid hypotension, including that following spinal or epidural anaesthesia
  • Minimise the duration of anaesthetic exposure where possible

Non-pharmacological prophylaxis

  • Stimulation of the P6 (pericardium 6) acupuncture point is effective in preventing PONV (Cochrane review)
    • Located on the volar forearm proximal to the distal wrist crease, between the tendons of flexor carpi radialis and palmaris longus
    • Free from side-effects

Anti-emetic prophylaxis

  • Use a multimodal anti-emetic approach; each additional antiemetic reduces risk of PONV by ∽30%
    • High-risk patients are best managed with a combination of agents rather than a single agent
    • Some brave (American) souls used a five-drug combination in their "Aim for Zero" PONV and achieved ∽90% success

  • Which drug is best? One should naturally choose the drug(s) which offer the most benefit (effectiveness, pleiotropic effects) and least harm (side-effects, cost, availability)
  • If you can only plump for a single agent, it seems a NK1 receptor antagonist is probably your best bet in terms of efficacy, though isn't readily available in most UK trusts
  • Combination therapy is, however, de rigueur and a triple-whammy of dexamethasone, ondansetron and metoclopramide is about as good as it gets

  • Dexamethasone + ondansetron is probably the modal anti-emetic combination
  • Its efficacy (risk ratio 0.3), cost (about £5 for the two), minimal side-effects and the other benefits of dexamethasone make it a suitable combination

  • Use of ye olde multimodal, opioid-sparing analgesic approach to minimise opioid-induced nausea/vomiting
  • Avoid orthostatic hypotension in PACU through adequate hydration and cautious mobilisation
  • Prescribe anti-emetics 'as required' for all patients
  • Where retching and vomiting could cause particular harm (e.g. upper GI surgery), continue prophylaxis into the early post-operative period

  • Rescue treatment should use an anti-emetic from a different class to any given as prophylaxis
  • Once established, it seems ondansetron is most efficacious at treating PONV, especially in those who haven't received any prophylaxis (Anesthesia & Analgesia, 2022)