Preoperative fasting in children. The evolution of recommendations and guidelines, and the underlying evidence

Abstract

This review discusses the evolution of preoperative fasting guidelines and examines the incidence of pulmonary aspiration of gastric contents and suggested treatments.

Nine guidelines developed by professional societies and published in peer-reviewed journals since 1994 were identified. The recommendations on preoperative fasting for various categories have undergone only small adaptations in the following three decades in pediatric anesthesia.

We found twelve published studies of the incidence of pulmonary aspiration, which ranges from 0.6 to 12 in 10,000 anesthetics in children. However, this variation reflects differences in the definition of aspiration as well as differences in study design. The main risk factors identified are emergency surgery, ASA physical status, and patient age. Several additional risk factors have been suggested, including non-compliance to fasting guidelines.

The duration of clear fluid fasting is not associated with an increased risk of pulmonary aspiration which may be reflected in future guideline updates in pediatric anesthesia.

1 Preoperative fasting guidelines from past to present

The notion of preoperative fasting is as old as anesthesia itself. Case reports of asphyxiation due to vomiting were available during the 19th century with Sir Joseph Lister publishing the first recommendation for preoperative fasting in 1883. Lister identified remains of solid food in the stomach as a major risk for aspiration but also intuitively acknowledged the negative effects of prolonged fasting. A cup of beef tea about 2 h before the administration of chloroform anesthesia was subsequently recommended. More elaborate regimens for limited intake of food and drink (rather than outright fasting) before anesthesia were described in the following decades. Maltby compiled a detailed history of the evolution of preoperative fasting recommendations in this journal, an article well worth re-reading [ ]. After the Second World War, the recommended strict nil-per-os from midnight rule was gradually enforced and included both solids and liquids. It took, however, another half a century before formal guidelines from national and international societies were developed and published ( Table 1 ).

Table 1

Preoperative fasting guidelines published by national or international societies. Professional societies: Norwegian Society of Anaesthesiology, ASA – American Society of Anaesthesiology, SSAI – Scandinavian Society of Anaesthesiology and Intensive Care Medicine, CAS – Canadian Society of Anaesthesiology, ESAIC – European Society of Anaesthesiology and Intensive Care. Fasting rules in hours before induction; 0+ – fluids allowed until patient called to the OR).

First author	Society	Year	Country/Region	Solids heavy meal	Solids light meal	Infant formula	Breast milk	Clear Fluids
Raeder JC [ ]	Norwegian S An	1995	Norway	None	6	NA	4	2
ASA TF [ ]	ASA	1999	USA	8	6	NA	2	2
Soreide [ ]	SSAI	2005	Scandinavia	None	6	4	4	2
Smith [ ]	ESA	2009	Europe	None	6	4	4	2
ASA TF [ ]	ASA	2011	USA	8	6	ND	4	2
ASA TF [ ]	ASA	2017	USA	8	6	NR	4	2
Dobson [ ]	CAS	2021	Canada	8	6		4	1
Frykholm [ ]	ESAIC	2022	Europe	None	6/4	4	3	1/0+
ASA TF [ ]	ASA	2023	USA	8	6		4	2

The first consensus-based guideline on perioperative fasting guidelines was published by the Norwegian Society of Anaesthesiology [ ]. This guideline included a recommended mandatory fasting period of 6 h after a light meal. Clear fluids were allowed until 2 h before anesthesia but also 150 mL of water 1 h before induction to enable oral premedication. The authors correctly predicted that an adoption by the American Society of Anesthesiologists (ASA) task force would enable international consensus. Indeed, the 1999 ‘ Practice Guidelines for Preoperative Fasting and the Use of Pharmacologic Agents to Reduce the Risk of Pulmonary Aspiration’ not only led to the adoption of the recommendations of the task force in the United States but also worldwide [ ]. The guidelines included recommendations for preoperative fasting after a large meal of solids (8 h), a light meal (6 h), breast milk (4 h), and clear fluids (2 h), commonly abbreviated to the 6-4-2 rule.

For clinicians and their patients alike, this guideline was a significant improvement from the ‘Fast from Midnight’ as it addressed the problem of starving breastfeeding infants and offered a solution to dehydration and thirst for children and adults. Very few significant changes to these core recommendations have been made since [ , ].

The Scandinavian Society of Anaesthesiology and Intensive Care Medicine (SSAI) also appointed a task force on pre-operative fasting guidelines and published almost identical recommendations in 2005 [ ]. One important exception was the inclusion of infant formula in infants undergoing elective procedures under anesthesia. In 2009, the European Society of Anaesthesiology (ESAIC) published its first guideline on preoperative fasting in adults and children [ ]. In addition to the basic 6-4-2 rule, the ESAIC guideline introduced the concept of encouraging intake of clear fluids at 2 h before anesthesia, and breast milk or formula at 4 h before anesthesia in infants <6 months of age. The recently published ESAIC guideline for preoperative fasting in children was the result of two years of work by a newly appointed ESAIC task force consisting of pediatric anesthesiologists from 11 countries [ ]. It is the first fasting guideline that used current evidence-based medicine methodology, including the definition of a set of Population Intervention Comparison Outcome queries (PICO), a systematic literature review, evaluation of the quality of the evidence according to GRADE, followed by a consensus process including several Delphi meetings, with subsequent publication in a peer-reviewed journal. The major changes compared to previous guidelines were the recommendations to encourage children to drink clear fluids until 1 h before anesthesia induction, for infants to be breastfed until 3 h before induction, or fed infant formula until 4 h before induction. Furthermore, the guideline did not advise against centers continuing to implement the concept of a well-defined light breakfast 4 h before induction or clear fluids until the patient is called to the operating theater (0+). In the current shorthand for preoperative fasting regimens, the evidence-based recommendations were summarized as 6-4-3-1, including solid food fasting for at least 6 h as per other guidelines.

2 It is too confusing: the ‘pro and con’ of the 6-4-3-1

At first, the addition of 1 h of clear fluid seems hardly worth it. Does it make a difference if children drink 2 h or 1 h before anesthesia, and is it worth the risk to change a rule that has been considered ‘ safe ’ for more than two decades? [ ] Will ‘ aspiration pneumonitis become the next major focus of morbidity and mortality in pediatric anesthesia ’? [ ].

We think not. Firstly, increased gastric content volume suggested as a secondary surrogate marker for aspiration risk is irrelevant as similar gastric residual volumes in healthy children allocated to 1 or 2-h clear fluid fasting (median/IQR of 0.43 (0.21-0-84) vs 0.46 (0.19–0.79 ml kg-1) have been observed [ ]. Secondly, ingestion of clear fluids does not seem to cause aspiration, as discussed below.

What are the negative effects of prolonged fasting? Fasting overnight does very little harm to healthy older children and only leads to clinically insignificant changes in blood glucose and ketone bodies, blood pressure, or difficulties of intravenous insertion during anesthesia [ ]. However, more vulnerable populations may show a higher risk of hypoglycemia of one in four at induction of anesthesia [ ]. Most recently, a prospective Swiss multicenter audit of pulmonary aspiration following the implementation of the ESAIC guideline reported no increase in the incidence of gastric aspiration compared with previous national data [ ].

An argument against the change has been that studies comparing 1 and 2-h clear fluid fast have failed to demonstrate any benefit as most children allocated to the 1-h rule were fasted more than 2 h anyway. However, the reduction to 1 h resulted in a reduction of mean fasting times of almost 50%, from around 6 h to around 3 h [ ]. The incidence of fasting more than 6 h decreased from 33% to 6% [ ]. Most clinicians realize that the modern fasting rule aims to prevent unnecessarily prolonged fasting. There is a difference in encouraging clear fluid intake rather than forcing healthy children to have a drink exactly 1 h before anesthesia.

The change of the clear fluid fasting rule from a minimum of 2 to 1 h before induction in children is thus an effective means to reduce mean fasting times with minimal effects on the residual gastric content volume. It has the potential to reduce the risk of hypoglycemia and metabolic effects in vulnerable populations and avoid discomfort and thirst in healthy children.

3 Quantifying the risk of pulmonary aspiration – a challenging task

Several studies of the incidence of pulmonary aspiration of gastric contents in children have been published; the incidence ranges from 0.6 to 12 in 10,000 ( Table 2 ). No significant change over time can be observed. A problem with pooling data from the studies in Table 2 is the difference in design and differences in the definition of aspiration. Four of the audits were retrospective, four prospective using data from reported incidents or medical records, while three studies used case report forms (CRF) for all included procedures, i.e. both numerator and denominator data. The latter design is arguably the strongest, providing details about the severity of each incident, more accurate background data, and pre-defined risk factors. However, if available data from all the studies are analyzed in more detail we can distinguish four degrees of severity of outcome for aspiration incidents: 1) death, 2) intensive care/patient harm 3), escalation of care but not intensive care, and 4) transient symptoms such as desaturation or coughing.

Table 2

Incidence of pulmonary aspiration. Study design (P-prospective, R – retrospective, P (CRF) – prospective with a CRF for both numerator and denominator data. Fasting rule (e.g. 6-4-2 means 6 h fasting for solids, 4 h for breast milk, and 2 h for clear fluids). Events and published incidences are those reported or extracted from the articles. Outcome categories, number of reported cases and calculated incidence [Death – number of cases when aspiration contributed to death, ICU – number of cases (incidence) that required intensive care due to aspiration, Escalation of care – number of aspirations (incidence) leading to escalation of care, Transient – only transient symptoms, no treatment needed]. All incidence numbers are expressed as N:10,000).

First author/year	Design	Rule	N	Events	Published incidence	Death N	ICU N (Incidence)	Escalation of care N (Incidence)	Transient N (Incidence)
Borland [ ] 1998	P	NA	50,880	52	10	0	15 (3)	23 (5)	37 (7)
Warner [ ] 1999	P	NA	56,138	24	4	0	5 (1)	9 (2)	15 (3)
Murat [ ] 2004	P	6-4-2	24,165	10	4	0	ND	ND	ND
Neelakanta [ ] 2006	R	6-4-2	30,695	7	2	0	3 (1)	1 (0.3)	3 (1)
Walker [ ] 2013	P	6-4-2	118,371	24	2	0	5 (0.4)	9 (0.8)	10 (0.8)
Andersson [ ] 2015	R	6-4-0	10,015	3	2	0	0	3 (3)	14 (10)
Tan and Lee [ ] 2016	R	6-4-2	102,425	22	2	0	2 (0.2)	7 (0.7)	13 (1)
Habre [ ] 2017	P (CRF)	6-4-2	31,127	29	9	0	0	11 (0.4)	18 (0.6)
Eisler [ ] 2018	R	6-4-2	47,272	26	6	0	1 (0.02)	18 (3)	8 (2)
Beck [ ] 2020	P (CRF)	6-4-2/1/0	12,093	14	12	0	0	4 (3)	10 (8)
Pfaff [ ] 2020	P	6-4-2	2,440,810	135	0.6	2	53 (0.2)	110 (0.4)	ND
Schmitz [ ] 2023	P (CRF)	6-4-1	22,766	25	11	0	1 (0.4)	10 (4)	14 (6)