It is well recognised that regurgitation and aspiration of gastric contents during anaesthesia causes morbidity and mortality . Risk of this is increased by a large gastric volume and by low gastric pH . Trauma is thought to delay gastric emptying , further complicating identification of patients who require additional methods to prevent aspiration during general anaesthesia. Several techniques have been recommended to reduce the risk of regurgitation and aspiration during anaesthesia, e.g. provoking vomiting with apomorphine , induction of anaesthesia in the foot-down position , cricoid pressure  and rapid induction of anaesthesia . These techniques have various disadvantages including increased risk of awareness, side effects of the necessary drugs and patient discomfort. These techniques are therefore avoided in those patients who are considered not to be at risk of aspiration. The aim of this study was to compare the current practice of anaesthetists in predicting 'gastric safety' with the available evidence.
Anaesthetists in Nottingham, Mansfield, Lincoln, Derby and Sheffield hospitals (in the Trent region of England) were surveyed by postal questionnaire. The questionnaire was developed following a local pilot study and is provided in the appendix. Anaesthetists were asked to rate (on a 0-5 scale) the reliability of the criteria given in the appendix for identifying those patients at-risk of pulmonary aspiration of gastric contents following trauma. They were also asked to state whether they currently used each criterion during their clinical practice. Data were examined as 'all grades' and separately as junior trainees (senior house officers), specialist trainees (specialist registrars and senior registrars) and specialists (consultants). Each criterion was expressed as the proportion of anaesthetists using it and as the median score for 'reliability' allocated to it. Data recording and analyses were performed using Microsoft® Excel (version: Office 97).
Two hundred and forty-seven postal questionnaires were sent out and 196 completed replies were received, giving a 79% response rate. Presence or absence of bowel sounds was the least used and lowest rated criterion (median score 0; used by 7%). Interval of eating to injury was the most used and highest rated criterion (median score 4; used by 88%). The use of opioids was the next most frequently used criterion (median score 3; used by 75%), and nausea, pain and vomiting were all highly rated. Specialists (consultants) used most criteria less frequently than trainees and rated most criteria lower. Junior trainees [SHO (senior house officer) grade] rated most criteria higher and used most criteria more frequently than senior trainees [SpR (specialist registrar and SR (senior registrar) grades]. Data for each criterion are given in Table 1.
The results of the survey suggest that most criteria are considered to be valuable in identifying patients at risk of aspiration following trauma. Only the presence or absence of hunger or bowel sounds were perceived as not useful. Most anaesthetists identified the interval from eating to injury as the most useful criterion and this was widely used by all grades of anaesthetist, implying that precautions are taken if this period is considered inadequate.
Factors predisposing to the regurgitation of gastric contents include lower oesophageal sphincter dysfunction and delayed emptying of stomach contents [8,9]. Pulmonary aspiration of gastric contents was first reported as a complication of general anaesthesia in 1848, but Mendelson is credited with describing the pathophysiology of the acid aspiration syndrome nearly 100 years later . Attempts have been made to define the volume and composition of fluid necessary to produce severe or fatal aspiration pneumonia. In 1974, Roberts and Shiley, studying acid aspiration during Caesarean section, suggested that the at-risk patient had a gastric pH of less than 2.5 and volume greater than 0.4 mL kg−1. These data, although derived from unpublished work in rhesus monkeys, rapidly gained widespread acceptance. However, more recent animal studies suggest that approximately 0.8 mL kg−1 instilled directly into the trachea is necessary to reproduce aspiration pneumonitis , and Schreiner, in a recent editorial, proposed that gastric fluid volume may be a poor surrogate for assessing risk of aspiration pneumonitis .
Although pulmonary aspiration of gastric contents is a rare event, it is one of the commonest causes of death related to anaesthesia . Estimates of the incidence of aspiration pneumonitis range from one to six per 10 000 anaesthetics with a mortality of 5% [15-19]. It is therefore difficult to experimentally identify factors that increase the incidence of aspiration. Most researchers investigating the risk of aspiration have examined the volume and pH of the stomach contents at induction of anaesthesia. Three studies of children who had suffered trauma demonstrated gastric volumes greater than 0.4 mL kg−1 and a pH less than 2.5 at induction of anaesthesia in 33-62% of patients and concluded that all children are at significant risk of aspiration following trauma [20-22]. However, the presence of adequate gastric contents to produce aspiration pneumonitis does not necessarily imply that there is a significant risk of aspiration. Using gastric intubation, which underestimates the volume of gastric contents , several studies have demonstrated that the stomach contents of a significant proportion of patients presenting electively exceed these critical values [24-27], implying that patients presenting for elective surgery are at significant risk of aspiration pneumonitis. In view of the rarity of the problem, this is not the case . The identification of gastric contents greater than 0.4 mL kg−1 with a pH less than 2.5 lacks the specificity to identify patients at risk of sustaining pulmonary injury due to aspiration of gastric contents. Other factors such as the protective role of the lower oesophageal sphincter are also important in the prevention of this complication. Plourde et al. determined that the gastric volume needed to produce regurgitation in cats was at least 20 times greater than the volume required to produce pulmonary damage by direct instillation into the trachea .
Another approach to quantifying the risk of pulmonary aspiration is measurement of the rate of gastric emptying. This can be performed using several different techniques, e.g. radiological methods, scintigraphy, ultrasound and epigastric electrical impedance techniques although most of these techniques are unsuitable for use in patients awaiting surgery . Paracetamol absorption is a simple non-invasive test that relies upon the minimal absorption of paracetamol from the stomach . It has been used in several studies of gastric emptying in relation to anaesthesia [31-33].
The evidence supporting the use of the criteria listed in Table 1 is discussed below:
Intervals of ingestion to injury to surgery
It seems likely that trauma causes some slowing or retardation of gastric emptying  but the evidence is conflicting. The survey showed a strong reliance by anaesthetists on the predictive accuracy of the interval from ingestion to injury. In the study by Bricker et al., 62% of children had a gastric aspirate greater than 0.4 mL kg−1 and they therefore concluded that following trauma all children potentially have a 'full' and 'at-risk' stomach . Because similar volumes are found in the stomachs of patients presenting for elective surgery, the significance of these findings is questionable. Using paracetamol absorption, Steedman et al. found no delay in gastric emptying in patients presenting within 4 h of a Colles' fracture compared with repeat testing at the patients' first outpatient appointment . In a small study using paracetamol absorption, Power et al. found a trend towards delayed emptying in recently head-injured patients , although they failed to find statistically significant changes. It appears that trauma does not invariably increase the volume of gastric contents nor increase the likelihood of aspiration.
Ingestion of food slows gastric emptying with solids being emptied significantly slower than fluids . The volume, pH, temperature and osmotic pressure of the ingested solution all influence the rate of emptying . It may be that the type of solid or fluid ingested thus has some bearing upon the likely volume of gastric contents at the time of operation.
Presence of pain or distress
The effect of pain on gastrointestinal function is difficult to quantify as the perception of pain varies considerably between individuals. Ischaemic pain in healthy volunteers delayed gastric emptying of semisolids measured using a scintigraphic technique  and Zaricznyj et al. found that gastric emptying was retarded by the development of severe pain, swelling and shock in dogs with forelimb fractures . However, intermittent immersion of the feet in iced water did not influence the absorption of paracetamol in healthy volunteers . Additionally Marsh, Spencer and Nimmo, using paracetamol absorption, found normal gastric emptying in patients with moderately severe pain . Similar contradictions can be found with the presence of emotional stress and anxiety. Lydon et al. found that gastric emptying was not delayed by anxiety  in contrast to Simpson and Stakes who reported that gastric emptying was delayed in patients with a low anxiety trait who developed a high anxiety state before surgery . Bricker et al. and Olsson and Hallen  noted that the volume of gastric aspirates following injury was related to the severity of trauma. Therefore, it appears that distress is associated with increased risk, although objective assessment may be difficult.
Administration of opioids
The effect of opioids on gastric emptying is well documented . Gastric emptying may be prevented for 2 h after an intramuscular (i.m.) dose of diamorphine and this effect is reversed by naloxone . Opioids are also known to reduce lower oesophageal sphincter tone, potentially increasing the likelihood of regurgitation . The use of opioids following trauma is potentially useful in identifying which patients are at risk of regurgitation. Alcohol is frequently associated with trauma and is thought to retard gastric emptying . However, in humans there is much anecdotal evidence with little experimental evidence. It has been suggested that wine and beer delay gastric emptying, while whiskey and brandy promote it .
Presence of bowel sounds and the presence of hunger
The survey showed that hunger and bowel sounds were not considered useful criteria and indeed the presence of hunger does not necessarily imply that the stomach is empty of solids. This is therefore an unreliable sign for predicting gastric safety although hunger may imply the absence of severe pain or anxiety. Bricker et al. commented that hunger appeared related to low gastric volume however, the experimental sample was small and a statistically significant difference was not found .
Presence of nausea and of vomiting
No evidence exists to support or deny the predictive capacity of these criteria, although the presence of vomiting implies some residual gastric volume and may prompt the assumption of ongoing risk of aspiration of gastric contents.
Presence of a nasogastric tube
A nasogastric tube may interfere with the integrity of the gastro-oesophageal sphincter, but is presence may also allow the stomach to be emptied of its liquid contents. No evidence currently supports or denies the presence of a gastric tube as a predictive factor for 'gastric safety'.
Pathological causes of delayed gastric emptying include gastrointestinal disease, ketoacidosis, hypocalcaemia and electrolyte imbalance . Studies looking at the effects of age, sex and body weight on gastric emptying have been inconsistent and although these factors may affect the pharmacokinetics of orally administered drugs it is unlikely that they influence gastric contents . Another physiological factor found to affect gastric emptying is posture, with delayed emptying occurring in the supine or left lateral position  and trauma may be associated with immobility in a position not conducive to gastric emptying.
In summary, the currently available evidence does not suggest that trauma per se delays gastric emptying or increases the risk of aspiration. However, factors associated with trauma such as distress, opioids, alcohol, and severe pain may delay gastric emptying or affect lower oesophageal sphincter tone and may thus predispose to regurgitation and pulmonary aspiration. Current practice in predicting the 'at-risk' stomach following trauma seems to include these factors and is evidence-based. However, the popular use of several criteria lacks the support of scientific evidence. The presence of the traditional 'at-risk' volumes of acidic gastric aspirate (0.4 mL kg−1 and pH less than 2.5) at induction of anaesthesia are of limited predictive value in identifying patients at risk of aspiration and further studies are required to assess gastric emptying and gastro-oesophageal reflux following trauma.
Data collection by Dr T. F. Cobby and Dr S. K. Hussain are greatly appreciated. We are also grateful for Professor A. R. Aitkenhead's critique of the manuscript during its development.
1 Stark DC. Aspiration in the surgical patient. Int Anesthesiol Clinics
2 Awe WC, Fletcher WS, Jacob SW. The pathophysiology of aspiration pneumonitis. Surgery
3 Rose EF. Factors influencing gastric emptying. J Forensic Sciences
4 Holmes JM. The prevention of inhaled vomit during obstetric anaesthesia. J Obstetrics Gynaecology
5 Snow RG, Nunn JF. Induction of anaesthesia in the foot-down position for patients with a full stomach. Br J Anaesthesia
6 Sellick BA. Cricoid pressure to control regurgitation of stomach contents during induction of anaesthesia. Lancet
7 Stept WJ, Safar P. Rapid induction-intubation for prevention of gastric-content aspiration. Anesth Analg
8 Cotton BR, Smith G. The lower oesophageal sphincter and anaesthesia. Br J Anaesth
9 Nimmo WS. Gastric emptying and anaesthesia. Can J Anaesth
10 Mendelson CL. The aspiration of stomach contents into the lungs during obstetric anaesthesia. Am J Obstetrics Gynaecol
11 Roberts RB, Shiley MA. Reducing the risk of acid aspiration during Cesarean section. Anesth Analg
12 Raidoo DM, Rocke DA, Brock-Utne JG et al.
Critical, for pulmonary acid aspiration: reappraisal in a primate model. Br J Anaesth
13 Schreiner MS. Gastric fluid, is it a risk factor for pulmonary aspiration? Anesth Analg
14 Morgan M. Control of intragastric pH and. Br J Anaesth
15 Tiret L, Demonts JM, Hatton F, Vourc'h G. Complications associated with anaesthesia - a prospective survey in France. Can Anaesthetists Soc J
16 Cohen MM, Duncan PG, Pope WDB, Wolkenstein C. A survey of 112a 000 anaesthetics at one teaching hospital. Can Anaesthetists Soc J
17 Olsson GL, Hallen B, Hambraeus-Jonzon K. Aspiration during anaesthesia: a computer-aided study of 185,358 anaesthetics. Acta Anaesthesiol Scand
18 Coombs DW. Aspiration pneumonia prophylaxis. Anesth Analg
19 Bodlander FMS. Deaths associated with anaesthesia. Br J Anaesth
20 Bricker SRW, McLuckie A, Nightingale DA. Gastric aspirates after trauma
in children. Anaesthesia
21 Olsson GL, Hallen B. Pharmacological evacuation of the stomach with metoclopramide. Acta Anaesthesiol Scand
22 Schurizek BA, Rybro L, Boggild-Madsen NB, Juhl B. Gastric volume and pH in children for emergency surgery. Acta Anaesthesiol Scand
23 Petring OU, Blake DW. Gastric emptying in adults: an overview related to anaesthesia. Anaesth Intensive Care
24 Cote CJ, Goudsouzian NG, Liu LM, Dedrick DF, Szyfelein SK. Assessment of risk factors related to the acid aspiration syndrome in pediatric patients - gastric pH and residual. Anesthesiology
25 Splinter WM, Stewart JA, Muir JG. The effect of preoperative apple juice on gastric contents, thirst and hunger in children. Can J Anaesth
26 Schreiner MS, Treibwasser A, Keon TP. Ingestion of liquids compared to preoperative fasting in pediatric outpatients. Anesthesiology
27 Hardy JF, Lepage Y. Occurrence of gastroesophageal reflux on induction of anaesthesia does not correlate with the volume of gastric contents. Can J Anaesth
28 Plourde G, Hardy JF. Aspiration pneumonia: assessing the risk of regurgitation in the cat. Can Anaesthetists Soc J
29 Nimmo WS. Effect of anaesthesia on gastric motility and emptying. Br J Anaesth
30 Heading RC, Nimmo WS, Prescott LF, Tothill P. The dependence of paracetamol absorption on the rate of gastric emptying. Br J Pharmacology
31 Marsh RHK, Spencer R, Nimmo WS. Gastric emptying and drug absorption before surgery. Br J Anaesth
32 McNeill MJ, Ho ET, Kenny GNC. Effect of IV metoclopramide on gastric emptying after opioid premedication. Br J Anaesth
33 Lydon A, McGinley J, Cooke T, Duggan PF, Shorten DG. Effect of anxiety on the rate of gastric emptying of liquids. Br J Anaesth
34 Strunin L. How long should patients fast before surgery? Time for new guidelines. Br J Anaesth
35 Steedman DJ, Payne MR, McClure JH, Prescott LF. Gastric Emptying following Colles' fracture. Arch Emerg Med
36 Power I, Easton JC, Todd JG, Nimmo WS. Gastric emptying after head injury. Anaesthesia
37 Dozois RR, Kelly KA, Code CF. Effect of distal antrectomy on gastric emptying of liquids and solids. Gastroenterology
38 Petring OU, Madsen PS. Experimental ischaemic tourniquet pain delays gastric emptying of semi-solids in healthy volunteers. Acta Anaesthesiol Scand
39 Zaricznyj B, Rockwood CA Jr, O'Donoghue DH, Ridings GR. Relationship between trauma
to the extremities and stomach motility. J Trauma
40 Thoren T, Tanghoj H, Wattwil M, Jarnerot G. Epidural morphine delays gastric emptying and small intestinal transit in healthy volunteers. Acta Anaesthesiol Scand
41 Simpson KH, Stakes AF. Effect of anxiety on gastric emptying in preoperative patients. Br J Anaesth
42 Nimmo WS, Heading RC, Wilson J, Tothill P, Prescott LF. Inhibition of gastric emptying and drug absorption by narcotic analgesics. Br J Clin Pharmacol
43 Nimmo WS, Heading RC, Wilson J, Prescott LF. The reversal of narcotic induced delay in gastric emptying and drug absorption by naloxone. Br Med J
44 Hall AW, Moussa AR, Clark J, Cooleay GR, Skinner DB. The effects of premedication drugs on the lower oesophageal high pressure zone and reflux status of rhesus monkeys and man. Gut
45 Barboriak JJ, Meade RC. Effect of alcohol on gastric emptying in man. Am J Clin Nutrition
46 Nimmo WS, Prescott LF. The influence of posture on paracetamol absorption. Br J Clin Pharmacol