Postoperative vomiting in children

Learning objectives.

By reading this article, you should be able to:

• •Identify children at risk of postoperative vomiting (POV).
• •Apply a risk stratification strategy to distinguish ‘high-risk’ children.
• •Communicate risk and treatment strategies appropriately.
• •Provide effective prophylaxis and treatment of POV in children.

Key points.

• •Children must be risk stratified for postoperative vomiting (POV) in order to identify those at risk and avoid unwanted adverse drug reactions.
• •The four main independent risk factors in children are age >3 yrs; previous history or direct family member history of postoperative nausea and vomiting; motion sickness; and strabismus surgery.
• •Regional anaesthesia, where appropriate, is beneficial, either in isolation or in combination with general anaesthesia.
• •Using appropriate prophylactic treatment can reduce the incidence of POV.

Nausea is defined as the unpleasant sensation associated with awareness of the urge to vomit. Vomiting is defined as the involuntary forceful expulsion of gastric contents through the mouth. Nausea can be difficult to report in small children, and therefore, this article will focus on postoperative vomiting (POV). POV can be defined as early (<6 h) and late (6–24 h).

For reasons that are unclear, the reported incidence of POV in children varies between 13% and 42%, which is twice the frequency in adults.1, 2

Physiology

Vomiting is integrated by the vomiting centre, located in the dorsal part of the lateral reticular formation in the medulla oblongata. The vomiting centre contains predominately acetylcholine (ACh) receptors, with some histamine (H₁) receptors. It receives input from the cerebral cortex, chemoreceptor trigger zone (CTZ), and the nucleus tractus solitarius (NTS). The cerebral cortex receives afferent CNS input from pathways mediating pain, smell, sight, emotion (fear/anxiety), and organic disturbances.

The CTZ is functionally located outside the blood–brain barrier on the floor of the fourth ventricle in the area postrema. It contains serotonin (5HT₃) and dopamine receptors. It receives input from the vestibular apparatus, which contains H₁ and ACh receptors. The NTS contains 5HT₃ receptors, and receives input from the gastrointestinal tract stretch and chemoreceptors.

The efferent pathways are carried by the vagus, hypoglossal, glossopharyngeal, trigeminal, and facial nerves to the stomach and via the spinal nerves to the diaphragm and abdominal muscles.

The causes of POV are wide ranging. They include emotional, pharmacological and chemical stimuli, and also direct physical causes, such as the presence of blood in the gastrointestinal tract. The integrated pathway is complex, dynamic and multidimensional. Consequently, management must address several factors and there are both pharmacological and non-pharmacological options.

Vomiting has two phases: ‘pre-ejection’ and ‘ejection’. In the pre-ejection phase, the sympathetic nervous system causes the unpleasant symptoms of tachycardia, tachypnoea and sweating, whilst the parasympathetic system causes salivation, upper and lower oesophageal sphincter relaxation and retrograde contraction. This prepares the body for the ejection phase. In older children, the pre-ejection phase is experienced as nausea. The ejection phase starts with cessation of respiration in midinspiration; the hyoid and larynx elevate to open the crico-oesophageal sphincter, the glottis closes, the soft palate elevates to close off the nasopharynx, and contraction of the diaphragm occurs to increase intra-abdominal pressure. Finally, the gastro-oesophageal sphincter opens and the gastric contents are ejected.

Consequences of POV

The consequences of POV are costly to the patient and family, the anaesthetist, and the institution. They also have the potential to impede recovery from surgery.

The patient with POV suffers from adverse psychological, metabolic, and physiological events, including discomfort, hunger, dehydration, electrolyte imbalance and pulmonary aspiration. Pulmonary aspiration under anaesthesia has an incidence of 9.3 per 10,000 children.³ There may be adverse surgical effects, such as wound dehiscence and reduced or delayed mobilisation. The cost to the institution results from delayed discharges or unanticipated hospital admissions of which 2% result from POV.4, 5 The institution must provide funding for the acquisition costs of drugs and strategies to prevent or treat POV, but with a cost-effective strategy, this may represent an investment. Recently, the collection of data on patients' satisfaction, such as the ‘NHS Friends and Family Test’, is becoming more widespread and is likely to be influenced by events, such as POV.

Giving antiemetic prophylaxis injudiciously to all patients can lead to unnecessary and unjustified drug reactions, as demonstrated in the RCoA's Sixth National Audit Project, which examined perioperative anaphylaxis.⁶ Furthermore, it has not been shown to improve outcome.⁷ It is thereby imperative that all patients are risk stratified objectively and treated appropriately.

Who should we treat?

Determining each patient's baseline risk allows the clinician to inform the patient and family, and to instigate measures to reduce preventable factors and administer prophylaxis or treatment most appropriately (Table 1).

Table 1.

Baseline assessment of the risk of POV

Steps for the clinician	Considerations and options
Assess baseline risk factors	Age >3 yrs
	Surgery >30 min
	Strabismus and tonsillectomy surgery
Inform patient and parent	Good communication
Decrease baseline risk	Regional anaesthesia
	Propofol
	Hydration
Treat	Prophylaxis
	Treatment

Risk factors

Children may have independent features, which can be classified as low or high risk for developing POV, and should receive appropriate treatment (Table 2). The first step in determining who requires treatment is to assess a patient's baseline risk in terms of patient, anaesthesia or surgery-related factors.

Table 2.

Risk assessment for postoperative vomiting.

Risk category	Factors	Risk factors	Treatment
No risk	Patient	<3 yrs of age, no history of POV or motion sickness	No prophylaxis
	Anaesthetic	None
	Surgical	<30 min duration
Low risk	Patient	>3 yrs of age, postpubertal female	Single prophylaxis
	Anaesthetic	Volatiles, opioids, and anticholinesterases
	Surgical	>30 min duration
High risk	Patient	Previous history of POV or motion sickness	Double prophylaxis; consider TIVA using propofol as a single agent
	Anaesthetic	Volatiles, N2O, opioids, and anticholinesterases
	Surgical	Strabismus, tonsillectomy with or without adenoidectomy, and middle ear surgery

Patient related

Nausea is difficult to recognise in very young children. This differs from the adult population, where the presence of symptoms of nausea allows for treatment before vomiting occurs. Children aged 3 yrs and above are at greater risk than those aged <3 yrs. In infancy, the rate of POV is around 5% compared to 20% in those of preschool age.⁸ A large study demonstrated an incremental increase in incidence of 0.2–0.8% per year from the age of 3 yrs until adolescence.⁹ Children with a prior history or direct family member history of POV, or those with a history of motion sickness are at an increased risk. Adult female patients are more at risk than males. In prepubertal children, there are no differences between boys and girls, but adolescent girls have two to four times the risk of boys.9, 10 Management of anxiety in children is key to providing positive experience for patients in many areas of anaesthesia. There is low-level scientific but strong anecdotal evidence that anxiety increases the risk of POV, especially with air swallowing during a traumatic induction of anaesthesia.¹¹ The incidence of POV may be decreased by passive smoking; but this is not recommended as a preventative strategy.¹² There is weak evidence that higher BMI contributes to POV.

Anaesthesia related

The use of all inhalational anaesthetic agents, including nitrous oxide (N₂O), contributes to the risk of POV. This is amplified by the use of opioids and anticholinesterases. TIVA is as effective as a single-dose antiemetic, and therefore, its use may be considered.¹³ Maintaining adequate hydration decreases the risk of POV. The current recommendation for clear fluid fasting in children is 1 h, but institutional guidelines should be followed.¹⁴ Where possible, the use of opioid-sparing regional anaesthesia should be considered.

Surgery related

Surgery lasting greater than 30 min in duration is an independent risk factor for POV, with an odds ratio of 3.25.9, 15 Procedures that are painful and have a high opioid requirement increase the risk. Tonsillectomy carries a high risk of POV.¹⁶ This may be compounded by the administration of opioids. Certain procedures, such as middle ear and squint surgery, carry a higher risk, probably because of direct surgical stimulation of the vagus nerve and vestibular apparatus.

Risk scores for predicting POV

Two widely used risk scoring systems in children are the postoperative vomiting in children score (POVOC) and the vomiting in the postoperative period score (VPOP). POVOC, developed by Eberhart and colleagues, identifies four independent variables, each scoring 1 point: strabismus surgery, age ≥3 yrs, duration of surgery >30 min and a history of PONV.⁹ The VPOP scoring system provides the addition of multiple doses of opioids.¹⁷

Two other studies have identified four similar independent risk factors for POV in children: duration of surgery greater than 30 min, age >3 yrs, past or direct family member history of POV and strabismus surgery. Each study demonstrated an incremental increase in POV with each risk factor in an additive manner.9, 18 Children with independent risk factors should be considered for prophylactic antiemetic medication.

Treatment: prevention and therapeutic strategies

Several guidelines on the management of POV in children have been published, including the most recent Association of Paediatric Anaesthetists of Great Britain and Ireland guidelines published in 2016.10, 15, 19 The key features are to reduce modifiable risk factors and institute appropriate treatment. Children should be assessed using the risk scores to determine the pharmacological treatment.

Decrease baseline risk

The use of regional anaesthesia to completely avoid general anaesthesia in children is largely not feasible. However, it could be considered in older children undergoing minor procedures, the aim being to reduce exposure to N₂O, volatile agents, and opioids. The combination of regional anaesthesia and general anaesthesia can be opioid-sparing, and hence reduce risk. This technique has been demonstrated to decrease POV in strabismus surgery.²⁰ Prolonged starvation should be avoided, and encouragement of good hydration decreases POV.

More recently, genetic profiling of one single-nucleotide polymorphism located in proximity to the promoter region (rs2165870) for the M₃ muscarinic ACh receptor gene, CMRM3, has been investigated, and this may have future potential for screening.²¹

Drug therapy

Multiple drugs have been studied to prevent POV (Table 3). Some are used to prevent and reduce the risk of POV, and some are used for treatment. Treatment in children differs from that in adults, where certain medications are not licensed for paediatric use or where dosages have been standardised.

Table 3.

Drug therapy for postoperative vomiting.

Class	Drugs	Dose	Prophylaxis/treatment	Special considerations
5HT3	Ondansetron	0.15 mg kg−1 (maximum: 4 mg)	Both	Avoid in children where prolonged QT is known or suspected
	Tropisetron	0.1–0.2 mg kg−1 (maximum: 2 mg)		Unlicensed for children in the UK
	Granisetron	10–80 μg kg−1 (maximum: 3 mg)		More evidence required
	Dolasetron	350 μg kg−1 (maximum: 12.5 mg)		Contraindicated in children because of cardiac arrhythmias
	Ramosetron	Undetermined		Further evaluation needed before recommendation; previous falsely reported results by a Japanese researcher
Steroids	Dexamethasone	0.0625–0.15 mg kg−1 (maximum: 8 mg)	Prophylaxis	Should not be used in patients at risk of tumour lysis syndrome
Antidopaminergics	Metoclopramide	0.25 mg kg−1 (maximum: 10 mg)		Does not reliably reduce POV and is not recommended; contraindicated in children <1 yr of age
	Prochlorperazine			Not recommended
	Droperidol	25 μg kg−1 (maximum: 1.25 mg)	Both	Second line where dexamethasone is contraindicated; contraindicated in known long QT syndrome
Antihistamine	Cyclizine			Not recommended
Antimuscarinic	Dimenhydrinate	0.5 mg kg−1 (maximum: 25 mg)	Both	May be used except in children undergoing tonsillectomy caused by sedative adverse effects

Antiemetics for the prevention and reduction of POV in children include first-line single agents: 5HT₃ antagonists, dexamethasone, dimenhydrinate (except in patients undergoing tonsillectomy because of its sedative effects), and droperidol. Propofol infusions are effective in subhypnotic doses alongside an anti-emetic to reduce baseline risk or as a single agent in TIVA.13, 22, 23 Propofol as a single agent in TIVA is recommended for children at high risk of POV undergoing high risk for POV surgical procedures.¹⁵ The use of low-dose naloxone has been shown to reduce opioid-induced nausea.¹⁰ The use of a combination of antiemetics has proved effective in high-risk children, who should receive two drugs from different classes. One of the most frequent and effective combinations in the United Kingdom is ondansetron and dexamethasone. Where dexamethasone is contraindicated (e.g. those at risk of tumour lysis syndrome), high-risk children can receive ondansetron and droperidol. In 2001, the US Food and Drug Administration issued a controversial ‘black box’ warning regarding the potential of prolonged QTc interval and fatal arrhythmias with droperidol leading to a decrease in its use. Cyclizine and prochlorperazine are not recommended in children because of unwanted adverse effects and the lack of proven efficacy.¹⁵ Ondansetron is the current antiemetic of choice for children with established POV, the current antiemetic of choice is ondansetron.¹⁰

Novel therapies

Interest from pharmaceutical companies in POV is growing, and several new agents are currently being trialled in adults, including the long-acting 5HT₃ antagonists, palonosetron and ramosetron; newer corticosteroids; butyrophenones; and antihistamine class drugs. The neurokinin-1 receptor antagonists, aprepitant and fosaprepitant, are available for use in adults. Palonosetron and neurokinin-1 receptor antagonists have a long half-life of approximately 40 h, and have shown similar effects to ondansetron in preventing PONV.24, 25

Adverse effects

As mentioned earlier, pharmacological strategies are not without adverse effects. In addition to the risk of anaphylaxis, ondansetron and some antihistamines (dimenhydrinate) can prolong the QTc interval, leading to cardiac arrhythmias. Hence, these should be avoided or used with caution in children with a history of long QT syndrome. Steroids can cause hyperglycaemia, headaches, and tumour lysis syndrome.

Non-pharmacological strategies

Non-pharmacological strategies include acupuncture and acustimulation. The ‘pericardium (P6) acupuncture point’, which, in adults, is located 4 cm proximal to the wrist between the flexor carpi radialis and palmaris longus tendons, is stimulated by needle, pressure, or an electrical impulse. This technique has been shown to be superior to placebo and demonstrates similar results to antiemetic drugs in suitable high-risk patients without the risks assoicated with drug therapy. However, acceptance of these techniques by children may be limited.¹⁵

Communication and partnership

Parents and patients should be informed of any risk factors and the treatment plan. This promotes partnership, patient satisfaction, and communication during subsequent anaesthetics.

Conclusions

POV has detrimental effects for the patient, surgical recovery, the healthcare professional and the institution. Children should be risk stratified to receive appropriate management of risk factors and antiemetic therapy. Wherever possible, strategies to decrease exposure to emetogenic substances should be used. By decreasing baseline risk factors and correctly treating those at risk, we can provide better perioperative care to children.

Declaration of interest

The authors declare that they have no conflicts of interest.

MCQs

The associated MCQs (to support CME/CPD activity) will be accessible at www.bjaed.org/cme/home by subscribers to BJA Education.

Biographies

Christa Morrison BSc FRCA is a specialty registrar in anaesthesia at Great Ormond Street Hospital. She is the trainee representative for the Association of Paediatric Anaesthetists of Great Britain and Ireland. Her main interests are in medical education and perioperative medicine.

Sally Wilmshurst MRCP FRCA is a consultant paediatric anaesthetist at Great Ormond Street Hospital. She is an educational supervisor and lead for morbidity and mortality. Her main clinical interest is in quality improvement and patient safety.

Matrix codes: 1A02, 2D02, 3D00