162

Jean-Francois Hardy MY FRCeC

Review Article

Large volume gastro- esophageal reflux: a rationale for risk reduction in the perioperative period

Contents

introduction Incidence and clinical significance Prevention of reflux in the normal individual Predisposing conditions leading to aspiration pneumonitis The pregnant patient The effect of anaesthetics on the lower oesophageal

sphincter Definition of the patient at risk Pharmacologic prophylaxis of aspiration pneumonitis Anaesthetic technique for the patient at risk Conclusions

Gastroesophageal reflux or regurgitation "is the process whereby gastric contents passively flow from the stomach across the gastroesophageal sphincter into the oesophagus and on into the pharynx. ''t

If laryngeal protective reflexes are impaired, the regurgitated material may eventually reach the lungs and produce pulmonary damage. Thus, knowledge of the pathophysiology involved and of the influence of anaes- thetic drugs on the upper digestive tract is essential to anaesthetists for the prevention of aspiration pneumonitis.

Incidence and clinical significance Few studies have examined the incidence of gastro- esophageal reflux associated with anaesthesia. "Silent" (small volume) regurgitation occurs in 4 to 26 per cent of all cases of general anaesthesia, the higher incidence reported in older studies, z-4 Fifty to 76 per cent of

From the Department of Anaesthesia, Maisonneuve-Rosemont Hospital at the Umvemity of Montreal, 5415 L' Assomptioa Boulevard, Montreal, Quebec, H I T 2M,:l.

regurgitations led to aspiration (i.e., some gaslric contents entered the lungs), but: "The postoperative course of patients in whom aspiration occurred was surprisingly uneventful. ,,4

More recently, Blitt vt al. reviewed 734 general anaesthetics and found 68 (nine per cent) episodes of regurgitation, six of which led to aspiration. Only one of these patients developed pulmonary complications post- operatively, but it was not possible to"definitely relate the pulmonary complication to the aspiration.'5 Of the 283 patients given general anaesthesia by face mask alone or face mask and an oral airway, 12 regurgitated and two aspirated. This incidence is similar to that associated with the use of an endotracheal tube. Tumdorf et al. 6 in 1974 reported similar figures. Regurgitation occurred in 14.5 per cent of their 152 patients which were all intubated. Tracheal aspiration of regurgitated material did not occur. Carlsson et al. 7 found a 20 per cent incidence of regurgitation of acidic material in the pharynx of patients undergoing emergency laparoscopy. No correlation of this finding with either pulmonary aspiration or pneumon- itis is given.

Thus, the clinical significance of silent regurgitation associated with general anaesthesia remains to be estab- lished.

The incidence of large volume gastroesophageal re- flux, defined in this text as regurgitation into the pharynx of a sufficient volume of gastric contents capable of causing pulmonary damage should it be aspirated, is unknown. It can be catastrophic if it occurs in the interval between the loss of protective airway reflexes and tracheal intubation with a cuffed tube. It is epidemiologically difficult to differentiate between passive gastroesophageal reflux and active vomiting, since these events are often associated with difficult, prolonged and repeated attempts at intubation (during which the effect of short-acting neuromuscular blocking drugs may wear off). Large

C A N J A N A E S T H 1988 ! 3 2 : 2 ! pp 162-73

Hardy: LARGE VOLUME GASTROESOPHAGEAL REFLUX

volume gastroesophageal reflux may lead to clinically significant aspiration pneumonitis.

The importance of aspiration pncumonitis as a major cause of anaesthesia-related deaths has led to attempts to measure the incidence of this event. Surprisingly, reliable figures are difficult to obtain. This probably arises from difficulties in establishing a firm diagnosis of aspiration pneumonia and evaluating the contributing role of non- anaesthesia-related factors (e.g., impaired consciousness prior to anaesthesia). Failure to report clearcut cases of anaesthesia-related aspiration also occurs in the current medico-legal climate.

According to Hutchinson, s the incidence of severe pulmonary acid aspiration in Auckland, New Zealand, decreased from 1 in 1 t,000 obstetrical and surgical patients in 1967-69, to 1 in 40,000 in 1970-72. The decrease coincided with the introduction of routine preoperative administration of particulate antacids, but mortality re- mained unchanged (1 / ! 10,1300 versus 1 / 120,000).

Kaunitz et al.9 studied the causes of maternal mortality in the United States by reviewing death certificates (from 1974 to 1978). Aspiration of gastric contents accounted for 28 deaths, or 1.35 per cent of the 2,067 deaths in pregnancies not associated with an abort ive outcome.

l'he "Report on Confidential Enquiries into Maternal Deaths in England and Wales" (1976-78), ~~ indicates thai 11 patients died as a direct result of inhalation of regurgitated gastric contents. In three other patients, aspiration was most probably the event leading to death. These l I to 14 deaths represent approximately one third to one half of the 30 deaths directly attributable to anaesthe- sia for the three-year period. Despite an overall reduction in death rate (per million pregnancies), inhalation of stomach contents appears to remain a major cause (eight patients) of anaesthesia-related mortality (29 patients) according to the latest Report (1979-81). 1~ Comparisons with other studies are difficult since the number of anaesthetics delivered is not known.

Three recent studies tend to confirm that the incidence of aspiration pneumonitis, and related mortality, is low. Tiret et al. t2 reported 27 cases of aspiration of gastric contents in a review of 198,103 anaesthetics given in Fral~ce, an incidence nf 1.4/10,000. These 27 cases of aspiration, of which 13 occurred during the postanaesthetic period, led to four deaths and left two patients comatose. In Canada, Cohen et al.13 reported 34 and 38 cases of aspiration of gastric contents in a study of 112,721 anaesthetics during the periods extending from 1975 to 1978 and from 1979 to 1983 respectively, a mean stable incidence of 6.4/10,000. In Sweden, Olsson et ai. 14

reviewed 185,358 computer-based anaesthetic records (where aspiration could be specifically recorded) and found an incidence of 4.7 aspirations in 10,000 anaesthetics.

163

Aspiration led to radiologically confirmed pneumonitis in 47 per cent of cases and to death in 0.2 cases per 10,000 anaesthetics.

Despite the alleged underreporting of this complication for fear of medicolegal involvment and despite other limitations pertaining to the study of this type of event, the incidence of aspiration pneumonitis related mortality appears to be low.

The significance to anaesthetists is that this complica- tion (1) can be the result ofjudgemental or technical errors, (2) is often preventable, and (3) may affect perfectly healthy patients coming for minor surgery.

Prevention of reflux in the normal individual Various mechanisms have been proposed in the prevention of gastroesophageal reflux. These include mechanical factors and modification of lower <)esophageal sphincter (LES) tone.

The intra-abdominal portion of the oesophagus, which is usually submitted to a mean pressure of approximately 10 cm H~O, while the inlra-thoracic portion of the oesophagus is submitted to a mean pressure of - 5 cm H20, may act as a flutter valve. This mechanism may become especially important during stress of the cardio- oesophageal junction. ~s

A mucosal valve mechanism, fashioned by folds of gastric mucosa, has been demonstrated in animals and human subjects, is

The angle with which the oesophagus meets the fundus of the stomach, acting as a flap valve, has also received great attention. Distortion of this angle by pulling down on the stomach lessens competency, while pushing up the fundus into the left hemi-diaphragm increases competency, t7 The diaphragmatic crura may increase competency by maintaining a normal angle and also act as a pinchcock mechanism.

It appears that the main factor preventing gastro- esophageal reflux is the LES, a physiologic barrier consisting of a high pressure zone in the distal oesopha- gus. No anatomic sphincter has been demonstrated, but an exceptionally large number of intramural neurons in the area lends support to this proposition. 18 The sphincter characteristics are thought to lie in the circular muscular layer which lies between the longitudinal muscular layer and the muscularis mucosae.

The resting tone of the sphincter is an intrinsic property of the muscle but it can be modified by a host of neural or hormonal influences (see review by Goyal and Rattanlg). The influence of anaesthetic drugs on the LES will be reviewed later.

It is the difference between the LES pressure and intra-gastric pressure, termed the barrier pressure (BrP), which determines whether regurgitation will occur. A low

164 CANADIAN JOURNAL OF ANAESTHESIA

FIGURE 1 Aspiration pncumonitis associated with anaesthesia is usually secondary to inhalation of gastric contents and all the following conditions must be present for the syndrome to develop: gastric contents (I) must be capable of causing lung damage should they be aspirated, intragastric pressure (2) must overoume LES pressure (3) and finally, enough regurgitated material (5) must pass the larynx (4) to cause pulmonary damage.

BrP, caused by a decrease in LES or an increased gastric pressure will promote reflux. Conversely, an increase in LES pressure or the prevention of increased gastric pres- sure should reduce the incidence of reflux.

The normal individual will respond to increased gastric pressure by a reflex increase in LES pressure, thus preventing reflux. This mechanism remains competent with intra-gastric pressures reaching approximately 20 cm H20, maybe more. Unfortunately, it is not possible to separate normal individuals from those presenting reflux on the basis of resting LES tone or BrP alone, since it is the capacity to increase pressure during stress which is responsible for competency, not baseline pressure per se. 15

Predisposing conditions leading to aspiration pneumonitis The concept discussed in this section is illustrated in Figure 1. While depressed consciousness and impaired broncho-oesophageal clearing mechanisms are highly significant risk factors for pulmonary aspiration should vomiting or regurgitation occur, they do not, per se,

predispose to gastroesophageal reflux. Therefore, they will not be considered further in this text.

Function of the LES is affected by many factors. Hiatus hernia has often been equated with gastroesophageal reflux but, in a study by Cohen and Harris, hiatus hernia per se apparently had no effect on gastrnesophageal sphincter competence.2~

Gi'~en the historical importance of aspiration pneu- monitis in the obstetrical context, whether or not it is still justified, pregnancy will be considered in a separate section.

Smoking, the ingestion of caffeine, alcohol and choco- late have been shown to decrease LES tone, but the significance of this finding in the perioperative period is unknown.

The influence of gaatric intubation on gastroesophageal reflux is controversial. It is generally believed that the presence of a gastric tube will prevent sphincter closure and promote gastroesophageal reflux. In cats, whose LES closely approximates that of humans, the presence of a mannmetry ealheter (comparable in size to a gastric tube) was not associated with an increase in gastroesophageal reflux. :t While there is one human study 22 in the anaesthetic literature demonstrating a reduction in the volume of stomach contents necessary to produce regurgi- tation in the presence of a gastric tube, analysis of the published data shows that this volume is very much in excess of that reported by most authors studying residual stomach contents on induction of anaesthesia. On the other hand, a study of regurgitation in the presence or absence of a nasogastric tube in trauma patients has shown an actual tendency towards reduction of gastroesophageal reflux (6 versus 12 per cent) 23 with the tube in place. This may be explained by a reduction of the volume of intra-gastric contents and a venting mechanism whereby pressure cannot build up in the stomach when the gastric tube is patent.

Intra-gastric pressure can increase by two mechanisms. When the normal capacity of the stomach (approximately 1,000-1,500 ml in the adult) is exceeded, the pressure inside it increases. The normal capacity is determined by the compliance of the gastric pouch and by the capability of the abdominal cavity to accommodate to the increased volume. The second mechanism is an independent in- crease in intra-abdominal pressure which is directly transmitted to the stomach. The pressure inside the stomach and in the abdominal cavity are normally the same.

Internal (gastric secretion) or external (food, blood, air, x-ray conUast medium) sources may lead to increased gastric contents.

The stomach secretes only a few milliliters per hour of gastric 3uice during the interdigestive period. However, strong emotional stimuli can increase secretion to 50 ml per hour. Furthermore, the cephalic phase of gastric secretion, associated with hunger, can be associated with a rate of secretion of gastric juice as high as 500 ml per hour. 24 Patients will often experience stress and hanger while waiting for anaesthesia and surgery. This central stimulation of gastric fluid production could explain the increase in gastric fluid volume prior to induction of anaesthesia that has been demonstrated in several studies, especially in out-patients, 2s since gastric emptying is thought to remain normal (v ide infra).

Hard),: LARGE VOLUME GASTROESOPI-iAGEAL REFLUX 165

A recent meal or x-ray study of the upper digestive tract, swallowing of blood or active stomach haemorrhage are all evident causes of increased intra-gastric volume. However, relaxation of the proximal stomach will accom- modate volumes up to one liter after eating with little changes in iutra-gastric pressures. This same mechanism explains that the human stomach can accept 1600 ml of air with an increase in pressure of less than 10 mmHg. 26 It is only when these volumes are exceeded that intra-gastric pressure will increase enough to cause regurgitation, unless extra-gastric (intra-abdominal pressure) is in- creased in the presence of a smaller volume of stomach contents.

Normal emptying of gastric contents depends on the type of food ingested. Fats delay emptying most, carbo- hydrates least and proteins moderately. 24 Most patients will have a small gastric residue within the usual preopera- tive fasting period. For example, in a study by Suthefland et at. ,27 the control group had a gastric residue of approxi- mately 30 ml (25 patients; fasting interval: ! 1 hours).

The effect of waiting for anaesthesia on gastric empty- ing is another controversial topic. It is a widely held belief that stress and pain delay gastric emptying before surgery. On one hand, there are studies of gastric emptying (based on the rate of paracctamol absorption) that could not show any difference in gastric emptying in elective patients immediately prior to induction or in healthy volunteers. 2s In a small pilot study of patients with pain awaiting emergency surgery, similar results were found. 26 On the ether, a recent study concluded that gastric stasis occurred in patients with a low predispositon to anxiety who became apprehensive while waiting for surgery. 29 In children presenting for emergency surgery the severity of the trauma appears to be more important in determining an increased gastric volume than the time from last meal to induction of anaesthesia.~~

A light breakfast 2-4 h before surgery is not associated with an increased gastric residue in adults at the induction of anaesthesia.31 In elective surgical patients, who did not receive narcotic premed[cation, a drink of water ( 150 ml) between two and three hours preoperatively produced a significantly smaller residual gastric volume than a prolonged overnight fast. 27 Furthermore, using bromo- sulphthalein as a marker dye, Sutherland et al. 27 demon- strated that all the oral fluid taken more than two hours preoperatively bad passed through the stomach prior to induction. However, it appears that patients presenting for elective Caesarean section 32 and children 33 will present increased gastric residues if not fasted for a minimum of four hours.

It has been clearly shown that opiate analgesics delay absorption of paracetamol and diazepam, presunmbly by delaying gastric emptying, s4,3s Despite this, a study of

125 elective adult outpatients failed to demonstrate either an increase or a decrease in gastric acidity and volume as a result of narcotic or oral premed[cation. 3'5 The role of opiates in elective premed[cation needs further evaluation in this context, but these drugs are likely to constitute a significant risk factor when they delay emptying of previously increased stomach contents, such as in the emergency patient.

Mechanical obstruction by compression of the gastric outlet by abdominal fat is a mechanism proposed by Vaughan et el . 37 to explain the small but statistically significant increase in gastric fluid volume in 56 obese patients when compared to 50 non-obese subjects (mean gastric volume after induction was 42.3 -+ 3.3 ml versus 14.7 - 1.7 ml). Gastroesophageal reflux was not studied in these patients. More significant causes of delayed gastric emptying are pylofic stenosis and bowel obstruction.

There has been no report on the rate of gastric emptying per se during an,~esthesia. However, gastric fluid volume after two hours of anaesthesia (elective surgery, excluding upper abdominal procedures) was significantly less than immediately after induction when patients were premed[- cared with morphine and the anaesthetic consisted of nitrous oxide 60 per cent with either fentanyl, enflurane or halo- thane, as These Findings suggest that gastric fluid production is minimal during anaesthesia for this type of surgery and/or that gastric emptying is normal or increased.

After operation, as in the preoperative period, opiate analgesics are an important cause of delayed gastric emptying. 26

The pressure inside the abdominal cavity is directly transmitted to the contents of the gastric pouch, thus increasing the probability of gastroesophageal reflux. The increased gastric pressure reflexly induces an increase of the LES tone, thus maintaining a normal barrier pressure. This response is however limited and gastric pressures exceeding 26 cm H20 are likely to induce regurgitation in normal anaesthetized subjects paralysed with tube- curarine. ~r

Conditions such as pregnancy, obesity, bowel obstruc- tion, ascites, large intra-abdominal tumours, and unusual surgical position will increase intra-gastric pressure and promote reflux. These pressure increases are relatively small when compared to the major increases in intra- abdominal pressure associated with coughing and active vomiting (100 mmHg or more).

Conversely, gastric pressure may be decreased in the adult patient by the use of the semi-sitting position which also impedes the progress of regurgitated nmterial towards the pharynx.

The pregnant patient Since Mendelson's first proper[ 39 in 1946, the pregnant

166 CANADIAN I O U R N A L OF ANAESTHESIA

patient has received much attention with regards to aspiration pneumonitis. In his original report, he analysed 66 cases of aspiration in a series of 44,016 deliveries at the New York Lying-In Hospital (an incidence of 15/10,000 deliveries). A mixture of N20, oxygen and ether was used in all cases where aspiration occurred. Two patients died of suffocation secondary to tracheal obstruction by solid material.

While this incidence may appear surprisingly low, particularly when the anaesthetic technique is taken into consideration, aspiration pneumonitis remains a significant cause of anaesthesia-associated maternal mortality and thus requires thorough consideration. Numerous physio- logical changes have been demons~ted during pregnancy and are thought to predispose to gastroesophageal reflux and subsequent aspiration. Interpretation of all the available data is difficult and often contradictory.

High progesterone blood levels diminish nonstrialcd muscle tone and have been held responsible for decreased gastric emptying during pregnancy. Nevertheless, using an epigastric impedance technique, O'Sullivan et al.

could not demonstrate any difference in the rate of gastric emptying between pregnant and nonpregnant females. 4~ The influence of labor on gastric emptying has also been studied and found to produce only small delays in paraeetamol absorption, while narcotic analgesics pro- duce a marked delay that can be reversed by naloxone but not by metoclopramide. 4t However, not all authors agree on this and some, like Davison et aL 42 believe that labor per se prolongs gastric emptying time. The production of gastrin, a hormone that increases both acidity and volume of gastric secretion is increased during pregnancy and that of motilin, a hormone which speeds gastric emptying, is depressed. All these hormonal changes should contribule to increase the volume of gastric contents.

Despite these hormonal changes, only 27 per cent of patients were found to have both gastric content volume >25 ml and pH <2.5 in a study of 100 patients about to undergo Caesarean section, whether in labour or not.e3 The greatest volumes were found when time to last meal was less than eight hours. In another study, '~ 5 of 16 parturients (31 percent) exceeded the same criteria. These figures are not very different from the 17 to 60 per cent of elective surgical patients reported to be at risk using the same criteria. 3g'45 During early pregnancy (less than 20 weeks), gastric volume is the same as in nonpregnant controls, despite ongoing hormonal changes. ~ Further- more, James e t a l . 45 were not able to demonstrate significant differences in gastric volume or pH between early postpartum patients and controls.

In pregnancy, the LES pressures are normal but women with symptoms of reflux have decreased barrier pres- snres. 47m This symptomatic decrease in barrier pressure

may be a sign deserving attention, even if lowered LES tone has not been associated with increased gastroeso- phageal reflux during anaesthesia.

During late pregnancy, the uterus encroaches on the stomach but the pnsition of the cardia does not change. Anatomic distortion of the gastroesophageal junction does not appear to contribute to ref~a.K. 49 However, gastric pressure is increased during pregnancy, early by a hormonal effect, 4s and later by compression due to the enlarging uterus.

Several iatrogenic factors contribute to the risk of gastroesophageal reflux and aspiration in obstetrical patients. The use of narcotics retards gastric emptying. The use of intravenous alcohol may increase gastric volume and acidity. Both narcotics and alcohol depress protective laryngeal reflexes. Oversedatlon wi~h anticonvulsants or overdosage with magnesium sulfate for the treatment of pre-cclampsia and eclampsia will have the same effect. The lithotomy position increases intra-gastrie pressure, as does application of uterine supra-fundal pressure to aid deli- very. Extradural analgesia does not affect gastric empty- ing, 26 but secondary hypotension, accidental subarachnoid block or intravascular injection may be associated with nausea, vomiting and loss of consciousness. Tlae adminis- tration of ergometrine intravenously has led to aspiration, l0 General swelling of the respiratory tract, particularly of the larynx, may render laryngoscopy and intubation difficult. Hypoxia, light anaesthesia and the waning effect of short acting neuro-muscular blocking drags, leading to cough- ing and vomiting, set the stage for disaster. Of the 14 deaths secondary to aspiration of stomach contents reported by Tomkinson etal. , nine were also associated with difficult or ocsophagcal intubation. ~o

Thus, many specific factors appear to promote gastrn- esophageal reflux in the parturient about to receive an anaesthetic. Consequently, all pregnant patients are indis- criminately considered at increased risk, even for elective surgery, despite a gastric fluid voluale and pH silnilar Io non-pregnant controls. The incidence of aspiration during Caesarean section of 15/10,000 recently reported by Olsson et al.,14 when compared to an overall incidence of 4.7/10,000, tends to support this contention. However, more information is required to determine if pregnancy p e r se puts the elective, healthy, asymptomatic parturient at an increased risk of aspiration. Conditions common to the general surgical population (emergency surgery,tO symptomatic reflux, and other factors that remain to be determined), may also explain the findings of this study.

The effects of anaesthetics on the lower oesophageal sphincter The effects of anaesthesia on the LES have been reviewed by Cotton and Smith. s~ They state that greater attention

Hardy: LARGE V O L U M E G A S T R O E S O P H A G E A L R E F L U X 167

should be paid to the effects of drugs and technique on the LES, the major physiological mechanism impeding gastro- esophageal reflux.

All premedicanl drugs used customarily, with the exception of metoclopramide and domperidone, have a depressant effect on the LES tone, thus possibly facili- tating regurgitation.

Thiopentone 3 mg.kg - t produces a slight but significant decrease in BrP. Nitrous oxide 66 per cent, halothane, enflurane and narcotics also decrease LES tone. Data on which to base rational preferences in patients in whom the trachea will not be intubated is not available at this time.

Fasciculations associated with the use of succinylcholine produce an increase in gastric pressure, and this has been thought by some to facilitate regurgitation. 5~'52 This increase, however, is accompanied by an even greater increase in LES pressure and BrP actually increases in normal subjects.S3 It has been postulated that this increase in LES pressure could be caused by contraction of the diaphragmatic crura, by a cholinergic mimetic effect of succinylcholine or by a reflex adaptive response of the LES. Atropine or glycopyrrolate consistently decrease BrP 54,s5 and prior administration could prevent the reflex increase in BrP associated with succinylcholine. Fascicu- lations can be prevented by prior administration of a non-depolarizing muscle relaxant, but the effect of pre-treatment on BrP is unknown. Another important consideration is that there is no evidence that maintenance of normal barrier function can be assumed to be operative in patients with gastric or intra-abdominal pathology. 56

In a study comparing the effect of non-depolarizing muscle relaxants on LES tone, 57 it was shown that atra- curium 0.6 mg.kg-t had no effect and that pancuronium 0.1 mg.kg - t produced a significant increase in BrP one minute after injection which was sustained for five minutes. Pancuronium thus appears to be a good choice when a sparing effect on the LES is desirable. Reversal of neuromuscular blockade with neostigmine 2.5 mg and atropine 1.2 mg (or glycopyrrolate 0.6 mg) does not alter BrP significantly. Increasing the dose of neostigmine to 5 mg increases BrP significantly.

It is difficult to apply this knowledge to everyday practice because the net effect of different combinations of drugs administered in different sequences is not known. Most drugs used for anaesthesia, with the exception of succinylcholine, pancuronium and neostig- mine, decrease BrP and may result in an increased tendency to gastroesophageal reflux.

Definition of the patient at risk In 1974, Roberts and Shirley, in an article on the pevention of acid aspiration during Caesarean section, "arbitrarily defined the patient at risk as that patient with

�9 La~nx

oesophageal ~ ~ ~ sphincter . Lung

FIGURE 2 In contrast with Figure 1, the usual definition of the patient "at risk" deals with the consequences of the direct tracheal injection of gastric material, ignoring all the other conditions necessary for aspiration penumonitis to develop.

at least 25 ml of gastric juice of pH below 2.5 in the stomach at delivery. ''43 While it was clearly stated that this was a statement based on preliminary unpublished work in the rhesus monkey, this arbitrary definition of risk rapidly gained widespread recognition and has only recently been disputed, l'21 According to this definition, from 17 to 60 per cent of elective adult patients, 3s'45 76 per cent or more of paediatric patients, ss 75 per cent of obese patients 32 and 27 to 73 per cent of pregnant or post-partum patients 43,45 are said to be "at risk." Estimat- ing the incidence of aspiration pneumonitis between one to six per 10,000 surgical c ases,12-~4's9 we can hardly say that the above "risk criteria" are of any help in identifying those patients effectively at risk of sustaining pulmonary injury from regurgitated gastric material. Should one really believe that these criteria are correct, mask anaes- thesia in paediatrics for example or any type of induction of general anaesthesia other than a rapid-sequence induc- tion should be condemned. Similarly, oral premedications should be abandoned. One rapidly realizes that while these criteria give rise to much literary work, they have not been generally accepted as rational basis for the safe conduct of anaesthesia in relation to the avoidance of aspiration pneumonitis in otherwise normal patients. We believe this is because these criteria are mistakenly used to predict the incidence of aspiration pneumonia while they should only be used to predict the incidence of pulmonary damage following direct intra-tracheal injection (Figure 2).

Considering the pathophysiological mechanism of re- flux, it appears logical, but again arbitrary, to propose that those patients at risk are those with decreased LES tone or conditions leading to increased intra-gastric pressure, in the presence of enough gastric juice to cause pulmonary lesions should aspiration occur. Those patients with symptomatic gastroesophageal reflux could therefore be

168 CANADIAN t O U R N A L OF ANAESTHESIA

considered at risk, but this remains to be demonstrated. Similarly, patients with an increased intra-abdominal pressure may be considered at risk if the volume of stomach contents is sign~eantly elevated (e.g., emergency surgery).

Preoperative fasting is a simple measure which should theoretically ensure minimal gastric contents in elective surgical patients. Even though controversy exists as to the exact definition of what minimal contents should be, the gastric contents found in these patients are definitely not sufficient to cause increased intra-gastric pressure. The intm-gastric volume necessary to passively overcome patency of the LES is 20.8 ml.kg-1 in the catfl t In view of the wide range of the data observed in this study (8-41 ml.kg-~), prudence dictates that the lower limit (8 ml.kg- ') be used as the criterion for risk of regurgitation. Even so, the animal data agrees with the figures presented by Tryba et al. 2z in a human study of passive gastroeso- phageal reflux under anaesthesia in the presence or absence of a gastric tube. Thus, the small residues found in stomachs of elective surgical or obstetrical patients are not, per se, a significant risk factor.

Yet another problem is that the volume, pH and composition of aspirated gastric fluid necessary to pro- duce pulmonary damage in humans has not been deter- mined (v ide infra). Furthermore, should this "critical" volume of the "ideal" fluid be present in the stomach, all of it will have to reach the lungs to cause any significant damage. The direct intra-tracheat injection of various fluids in the trachea of any animal model is far from the everyday clinical situation. Accordingly, the risk criteria derived from these studies have a low predictive value when used to predict the incidence of aspiration pneu- monitis in humans.

Finally, another definition of the patient at risk may be inferred from the retrospective study of cases in which aspiration effectively occurred, such as the review by Olsson et al.14 Major predisposing factors in this study were: c~sophageal disease, the exlremes of age (especially the age group 0-9 years), emergency surgery, induction of anaesthesia outside daytime working hours, neurological disorders with decreased consc iousness or elevated intra- cranial pressure, obesity and a history of peptic ulcer or gastritis. Inexperience of the anaesthetist was felt to be an aggravating factor by these authors.

Pharmacologic prophylaxis of aspiration pneumonitis Anaesthetists will always have to deal with patients who have a potentially full stomach and are at increased risk of regurgitating during the anaesthetic. Several measures directed towards improving LES function, or at least minimizing LES dysfunction, and preventing increases in intra-gastric pressure may help reduce the (unknown)

incidence of large-volume gastroesophageal reflux under anaesthesia. Drugs that also reduce gastric volume and/or acidity can be administered before the anaesthetic, with the object of minimizing lung damage should aspiration Occur.

Premedication with a clear oral antacid is one of the useful available options, especially for prophylaxis prior to emergency surgery when other measures may not be effective rapidly enough. Particulate antacids are best avoided since they appear to produce lung damage comparable to hydrochloric acid in dogs. ~~ This may also be the ease in humans since despite widespread use and demonstrated success in elevating gastric pH, mortality rate has not been influenced by their introduction in clinical practice, s'l~' 11 The administration of oral antacids results in a desirable increase of LES tone 6' and in a less desirable, 'albeit small, increase in gastric volume.

Thirty ml of 0.3 ml sodium citrate, two tablets of Alka Seltzer | in 30 ml of water, or a small volume of sodium bicarbonate 8.4 per cent (equal to five per cent of the volume of gastric fluid collected in the study by Faure et al.) effectively elevate the pH of gastric contents 62-64 when given 15 to 20 minutes before surgery- 6~ This protective effect lasts from one hour ~ to four hours. 64

Cimetidine and ranitidine administered orally or by infusion do not alter LES pressures 67'68 but, when used in appropriate dosage and with proper timing, their effec- tiveness in reducing gastric volume and increasing pH cannot be denied. For example, the combination of oral cimetidine, 300 mg, administered the night before surgery plus 300 mg of intramuscular cimetidine administered 60 to 90 minutes before induction increased gastric pH to 5.0 in all the patients studied by Weber and Hirshman.69 For a detailed review of cimetidine and newer H~-receptor antagonists as they relate to anaesthesia, the reader is referred to the article by Manchikanti et al. 7~

The net effect on gastroesophageal reflux and subse- quent pulmonary aspiration should therefore be beneficial, but studies demonstrating the correctness of this statement have yet to be published. This may be because control of gastrie acidity is only part of the issue. Neutralized gastric contents containing food particles cause severe lung damage in a dog model. 7' Large food particles have re- peatedly caused suffocation in humans. 1~ Even strained gastric fluid possessing a pH of 5.9 causes significant damage to the pulmonary epithelium of mice when aspir- ated. 72 The question of which exact property of the gastric juice is responsible for pulmonary damage is not com- pletely understood.

Another possible explanation for this apparent lack of reduction of anaesthetic morbidity and mortality from aspiration by H2-receptor antagonists is their ineffective- ness when used prior to most emergency procedures.

Hardy: LARGE VOLUME GASTROESOPHAGEAL REFLUX

Cimetidine and ranitidine achieve peak plasma levels within 60 to 90 minutes 7~ but do not reliably elevate the pH of contents already present in the stomach. 73 Mechanical plugging of airways and the latter inflammatory response to particulate aspiration will still occur. Therefore, prevention of gastroesophageal reflux remains the most important consideration in emergency patients.

Definite indications for the preoperative administration of cimetidine or ranitidine await further studies demon- strating a decreased morbidity or mortality in some groups of patients or certain procedures. Until such studies become available, recommendations concerning the use of these drugs prior to elective surgery can only reflect different authors' opinions.

The preoperative administration of intravenous atro- pine is associated with a significant decrease in LES pressure and an increase in free reflux and reflux associated with straining. 74 Glycopyrrolate has the same effect on BrP 75 and thus, presumably, on reflux. Since these agents are also relatively inefficacious in lowering gastric acidity, it would appear prudent to avoid their intravenous use, if possible, in the immediate pre- induction period. The intramuscular administration of atropine produces little effect on LES pressure but prevents the normal increase in BrP associated with the subsequent administration of intravenous metoclopramide. 76

Metoclopramide and domperidone stimulate gastric emptying, increase LES tone and act as antiemetics. The effectiveness of metoclopramide in reducing gastric volume preoperatively is controversial and that of dom- peridone is unknown. Their efficacy on gastroesophageal reflux in the medical setting is not very impressive, largely unknown in the anaesthetic setting, and use of these drugs as preoperative medications still awaits official approval. Their antiemetic effect could prove useful in the postoperative period, when upper airway reflexes may be obtunded by the residual effects of the anaesthetic. Unfortunately, domperidone has been found indistinguishable from metoclopramide in the prevention of postoperative vomiting: in these studies, neither drug has been consistently superior to placebo. 77

Finally, avoiding opiates in the premedication, if possible, will contribute to normal gastric emptying. However, the net effect on gastric volume and pH is less certain.

Anaesthetic technique for the patient at risk Local anaesthesia, a controlled regional anaesthetic tech- nique, and awake intubation or the rapid-sequence induc- tion of general anaesthesia in conjunction with cricoid pressure (Figure 3) are some of the few available methods to protect the airway when possible aspiration is a concem.

169

0

I I I I

FIGURE 3 Measures directed at preventing aspiration pneumonitis associated with general anaesthesia include: (l) reduction of gastric volume and acidity, (2) prevention of sudden increases of intragastric pressure, (3) avoidance of LES dysfunction (largely theoretical), and (4) erieoid pressure.

Cricoid pressure is simple and has been demonstrated to be effective in adults 78 and in children, 79 even in the presence of a nasogastrie tube. 79,8~ A nasogastrie tube need not be withdrawn before induction of anaesthesia. The tube can act as an overflow valve if a rise in intra- gastric pressure occurs, while cricoid pressure will prevent regurgitation of gastric contents. 79 Ventilation of normally compliant lungs, provided the airway remains clear, can be carried out without fear of producing gastric distension when cricoid pressure is applied, st Conversely, failure to apply cricoid pressure, relaxation before intubation or inefficient application by unskilled assistants were related to aspiration of stomach contents and death in 12 maternal deaths reported by Tomkinson (1973-75). 82

Adequate cricoid pressure requires the application of a force of approximately 4.5 kg to the area of the cricoid cartilage, s3 As a guide, assistants should be instructed to apply as much force as is required to produce tolerable pain if the same force was applied to the bridge of their nose. Proper application of cricoid pressure is therefore painful and discomfort of the neck and red discolouration of the skin may persist for some time after the release of

170 C A N A D I A N J O U R N A L OF A N A E S T H E S I A

pressure. The use of the "cricoid yoke," a device for providing consistent and reproducible pressure, may help untrained personnel provide adequate cricoid pressure, mainly by ensuring that pressure is not relaxed prema- turely. ~

It is of the utmost importance that the onset of uncon- sciousness, the achievement of full muscular relaxation and the application of adequate cricoid pressure be timed to occur simultaneously, s5 Sellick also recommended a slight head-down lilt during induction to prevent contami- nation of the airway should regurgitation occur in spite of all precautions, s4's~ However, patients with spontaneous reflux in the horizontal deeubitus position should be anaesthetized with sufficient head-up tilt to prevent reflux while awake. Provisions for adequate suction and rapid head-down positioning should always be available for all patients.

The smooth conduct of anaesthesia, avoiding any increase in intra-gastric pressure secondary to insufflated air, coughing, straining or vomiting, is a key element of our risk reduction strategy The use of an adequate intubating dose of succinylcholine (0.8 mg'kg -I) will always cause a 100 per cent paralysis of the diaphragm and of the adductor poUicis muscle, the former being paralysed prior to the latter, ss We usually use a dose of l mg.kg - I of succinylcholine (or 1.5 mg.kg -I after the administration of a defasciculating dose of 0.05 mg.kg- 1 of tubocurarine), monitor the neuromuscular junction and wait for complete relaxation before any attempt is made at intubation, to prevent increases in intra-abdominal pres- sure secondary to muscular effort. If succinylcholine cannot be used or should ventilation be necessary to prevent hypoxia during induction, gentle inflation of the lungs may be carried out, in the presence of cricoid pressure, while waiting for full paralysis to occur.

Fourteen of 21 cases of aspiration occurred during the process of a difficult intubation in a survey by Gibbs et

a l . 87 and in a significant percentage of cases reported by Olsson et a l . la The Report on Confidential Enquiries on Maternal Deaths in England and Wales tends to support this finding and also states: "Anaesthesia to a depth classified by Guedel as HI(i) or lll(ii) prevents vomiting even without an endotracheal tube and is less harmful to the neonate than having no mother."m Careful preopera- live evaluatinn will help identify those patients at risk of a difficult intubation ss and direct the anaesthetist towards a safe anaesthetic technique. Furthermore, should the anaesthetist render a patient apnoeic and then be unable to intubate, an appropriate course of action must be taken immediately. The reader is heferred to the recent review by Mclntyre for such a "failed intubation routine."ss

The postoperative situation is very different from the induction period. Coughing and straining are difficult to

avoid and gastroesophageal reflux will presumably be more frequent in the postoperative period. Nausea and active vomiting are also frequent events but, despite the possible seepage of small quantities of gastric fluid into the trachea, the endotracheal tube in place is an effective protection against any significant aspiration. Insertion of a nasogastric tube during the anaesthetic may reduce the volume of gastric contents and lessen the risk associated with postoperative regurgitation and vomiting. Use of a vented multi-oriflced tube has recently been proven effective in emptying the liquid contents of the stomach.S9 The major difference between recovery and induction lies in the return of normal consciousness and protective airway reflexes. These are the most effective safeguards against aspiration of regurgitated material. Patients at risk should be extubated awake with reflexes intact, in the lateral position.

Conclusions The relatively small incidence of large volume regurgita- tion and subsequent aspiration resulting in significant morbidity and mortality will probably prevent anaesthe- tists from ever being able to pm-~int the patient population definitely at risk. However, it should be possible to arrive at more objective criteria to better circumscribe this popu- lation. Since a prospective study of aspiration pneumonitis is virtually impossible, given the numbers involved, the usual method of studying this event is to critically analyse all reported cases. Medico-legal considerations and in- complete reporting are a major hindrance to this type of study and have prevented researchers from reaching firm conclusions.

Despite some progress, most of the evidence on which to base a rational anaesthetic conduct for a given patient is simply not yet available. This evidence is important, since recommendations as to specific conducts aimed at pre- venting gastroesophageal reflux and aspiration entail their own share of complications. For example, a rapid- sequence induction or an awake intubation may not be the best anaesthetic for a patient with limited coronary reserve. When faced with such a dilemma, the anaesthe- tist should fn'st determine if he is dealing with a clinically significant risk of aspiration. He should then establish priorities to ensure the best possible risk/benefit ratio for that particular patient.

Routine pharmacologic prophylaxis against aspiration pneumonitis should also be closely scrutinized before being adopted. Any drug has its potential complications and side-effects, some of which may not be immediately evident (e.g. fetal effects). As mentioned previously, there is, as yet, no hard evidence that the use of any of the pharmacologic agents reduces anaesthetic morbidity and mortality from aspiration. 9~

H~ry: LARGE VOLUME GASTROESOPHAGEAL REFLUX

For the present, a smooth induction and rapid intuba- tion during application of crinoid pressure when general anaesthesia is necessary, expert anaesthetic management and skilled assistance to provide the best possible care should inlubation prove difficult, are some of the more rational options. Those patients presenting with symp-

toms of gastroesophageal reflux or a definitely full stomach, such as after a recent meal or in the presence of gastric or intestinal pathology, will probably benefit from therapy aimed at neutralizing gastric acidity and reducing stomach contents. While these recommendations appear self-evident, failure to follow these well-eslablished guidelines has resulted in significant (maternal) morbidity and mortality over the past ten years, lo,s2,s7

Acknowledgments The author wishes to thank Dis Gilles Plourde and Jacques Brochu for their kind and expert editorial assistance.

References I Gibbs CP, Modell JH. Aspiration pneumonitis, chapter

58. In: Anesthesia. Mi/ler RD (Ed). 2rid edition, New York, Churchill-Livingstone 1986, pp. 2023-50.

2 Weiss WA. Regurgitation and aspiration of gastric contents during inhalation anesthesia. Anesthesiology 1950; 11: 102-9.

3 Culver GA, Maket HP, Bencher HK. Frequency of aspira- tion of gastric contents by the lungs during anesthesia and surgery. Ann Surg 1951; 133:289-92

4 Berson W, AdrianiJ. "Silent" regurgitation and aspira- tion during anesthesia. Anesthesiology 1954; 15: 644-9.

5 Blitt CD, Gutman HL, Cohen DD, Weisnmn H~ Dillon JB. "Silent" regurgitation and aspiration daring general anesthesia, Anesth Analg 1970; 49:707 13.

6 TurndorfH, Rodis ID, Clark TS. "Silent" regurgitation during general anesthesia. Anesth Analg 1974; 53: 700-3..

7 Carlsson C, Islander G. Silent gaslropharyngeal re- gurgitation daring anesthesia. Anesth Analg 1981; 60: 655-7.

8 Hutchinson BR. Acid aspiration syndrome (letter). Br J Anaesth 1979; 51: 75.

9 Kaunitz AM, Hughes JM, Grimes DA, Smith JC, Rochat RW, Kafrissen ME. Causes of matema/mortality in the United States Obstet Gyneco] 1985; 65: 605-12.

10 Tomktnson J, Tarnbull A, Robson G e t al. Report on confidential enquiries into maternal deaths in England and Wales 1976-1978. London, Her Majesty's Stationery Office, 1982, pp. 77-87.

11 HanterAR. Towards safer anaesthesia in obstetrics (edi- torial). Br J Anaesth 1987; 59: 275-6.

12 Tiret L, Desmonts JM, Hatton F, Vourc'h G. Complica-

171

tions associated with anaesthesia - a prospective survey in France. Can Anaesth Sue J 1986; 33: 336-44.

13 Cohen MM, Duncan PG, Pope WDB, Wolkenstein C. A

survey of 112,000 anaesthetics at one teaching hospital. Can Anaesth Soc J 1986; 33: 22-31.

14 Olsson GL, HaUen B, Hambraeus-Jonzon K. Aspiration

during anaesthesia: a computer-aided study of 185,358 anaesthetics. Acta AnaesthesioI Seand 1986; 30: 84-92.

15 Clark J, CuschieriA. Evidence I-or the flutter valve mech- anism of the lower oesophageal high pressure zone. Br J Surg 1980; 67: 599-603.

16 Botha GSM. Mucosal folds at the cardia as a component of the gastro-oesophageal closing mechanism. Br J Surg 1958; 45: 569-80.

17 Oreenan J. The cardio-oesophageal junction. Br J Anaesth 1961; 33: 432--9r

18 Rash RM, Thomas MD. The intrinsic ianervation of the gastro-oesophageal and pyloro-duodenal junctions. J Anal Lond 1962; 96: 389-96.

19 GoyalRK, Rattan S. Ncurohumeral, hormonal, and drug receptors for the lower esnphageat sphincter. Gastroen- terology 1978; 74: 598-619.

20 Cohen S, Harris LD. Does hiatus hernia affect compe- tence of the gastroe.snphageal sphincter? N Engl J Med 1971; 284: 1053-6.

21 Plourde G, Hard), ,IF. Aspiration pneumonia: assessing the risk of regurgitation in the cat. Can Anaesth Soc J 1986; 33: 345-8.

22 Tryba M, Zenz M, Mlasowsky B, Huchzermeyer H. Does a stomach tube enhance regurgitation during general anes- thesia? Anaesthetist 1983; 32: 407-9.

23 Satiani B, Bonner JT, Stone H. Factors influencing intra- operative gastric regurgitation - a prospective random study of nasogastric tube drainage. Arch Surg 1978; 113: 721-3,

24 GuytonAC. Textbooknfmedica~physiology. 7fred. Phila- delphia, WB Sounders, 1986, pp. 774-8

25 Ong BY, Palahni~k R J, Cumming M. Gastric volume and pH in out-patients. Can Anaesth Son J 1978; 25: 36-9.

26 Ninuuo WS. Effect of anaesthesia on gastric motility and emptying. Br J Anaesth 1984; 56: 29-36.

27 Sutherland AD , Maltby JR, Sate j P , Reid CRG. The effect of preoperative oral fluid and ranitidine on gastric fluid volume and pH. Can J Anaesth 1987; 34:117 21.

28 Marsh RHK, Spencer R, Nimmo WS. Gastric emptying and drug absorption before and after surgery. Br J Anaesth 1983; 55:242 p.

29 Simpson KH, Stakes AF. Effezt of anxiety on gastric emptying in preoperative patients. Br J Anaesth 1987; 59: 540-4.

30 Olssan GL, Hallen tl. Pharmacological evacuation of the stomach with metoelopramide. Aeta Anaesthesiol Stand 1982; 26: 417-20.

172

31 Miller M, Wishart HY, Nimmo WSr Gastric contents at induction of anaesthesia. Is a 4 hour fast necessary? Br f Anaesth 1983; 55:1185-8.

32 LewtsM, CrawfordJ$. Can one risk fasting the obstetrical patient for less than 4 hours? Br J Anaesth 1987; 59: 312-4.

33 Meakm G, Dingwall AE, Addison GM. Effects of fasting and oral premedication on the pH and volume of gastric aspirate in children. Br J Anaesth 1987; 59: 678-82.

34 Gamble JAS, Gaston 3tl, Nair SG, Dundee JW. Some pharmacological factors influencing the absorption of diazepam following oral administration. Br J Anaesth 1976; 48: 1181-5.

35 ToddJG, Nimmo WS. Effect of premedication on drug absorption and gastric emptying. Br J Anaesth 1983; 55: 1189-93.

36 Manchikanti L, Canella MG, Hohlbein L J, Calhver JA.

Assessmen: of effect of various modes of premedication on acid aspiration risk factors in outpatient surgeD'. Anesth Analg 1987; 66: 81-4.

37 Vaughan RW, Bauer S, Wise L. Volume and pH of gastric juice in obese patients. Anesthesiology 1975; 43: 686-9.

38 Stoelting RK_ Gastric fluid volume and pl-1 after fcntanyl. enflurance or halothane-nitrous oxide anesthesia with or without atropine or glycopyrrolate. Aoesth Analg 1980; 59: 287-90.

39 Mendelson CL. The aspiration of stomach r into the lungs during obstetric anesthesia. Am J Obstet Gyn- ecol 1946; 52: 191-204,

40 O'Sullivan GM, Sutton A J, Thompson SA, Carrie LE, Bullingham RE. Noninvasive measurcment of gastric emptying in obstetric patients, Anest Analg 1987; 66: 505-11.

41 Nimmo WS, Wilson J. Prescott LF. Narcotic analgesics and delayed gastric emptying during labour. Lancet 1975; 1: 890-3.

42 Davison JS, Davison MC. Hay DM. Gastric emptying time in late pregnancy and labour. J Obstet Gynaecol Br Comrnonw 1970; 77: 37-41.

43 Roberts RB. Shirley MA. Reducing the risk of acid aspira lion during cesarean section. Anesth Analg 1974; 53: 859-68.

44 Wheatley RG, Kallus FT~ Reynolds RC. Giesecke AH.

Milk of magnesia is an effective preinduction antacid in obstetric anesthesia. Anesthesiology 1979; 50: 514-9.

45 James CF, Gibbs CP, Banner T. Postpartum perioperative risk of aspiration pneumonia. Anesthesiology 1984; 61 : 756-9.

46 WynerJ, Cohen SE. Gastric volume in early pregnancy. Anesthesiology 1982; 57: 209-12.

47 Lind JF, Smith AM, Mclver DK, Coopland AT, Crispin

JS, Heartburn in pregnancy - a manometric study. Can Med Assoc J 1968; 9g: 571-4.

C A N A D I A N JOURr~AL O F A N A E S T H E S I A

48 Brock-Ume JG, Dow GB, Dimopoulos GE, Welman S.

Downing JW, Moshal MG. Gastric and lower oesopha- geal sphincter pressures in early pregnancy. Br J Anaesth 1981; 53: 381-4.

49 Williams NH. Variable significance of hearthum. Am I Obstet Gynecol 1941; 42: 814-9.

50 Cotton BR, Smith G. The lower oesophageal ~phincter and anaesthesia. Br J Anaesth 198a: 56: 37-46.

51 Andersen N. Changes in intragastric pressure following the administration of suxamethonium. Br I Anaesth 1962; 34:363 7.

52 Muravchick S, Burkelt L, Gold Ml. Succinylcholine- induced fasciculations and intragastric pressure during induction of anesthesia. Anesthesiology 1981 ; 55:180-3.

53 Smith G, Dulling R, Williams 17R. Gastro-oesophageal pressure gradient changes produced by induction of anaesthesia and suxamethonium. Br J Anaesth 1978; 50: 1137-42.

54 Cotton BR, Smith G. Comparisonoftheeffectsofatropine and glycopyrrolate on lower oesophageal sphincter pres-

sure. Br | Anaesth 1981; 53: 875-9. 55 CottonBR, Smith G. Single and combined effects of atrc.-

pine and metoclopramlde on the lower oesophageal sphincter pressure. Br. J. Anaesth 1981; 53: 869-74.

56 MuravchickS. (le:ter). Anesthesiology 1982;56: 410-1. 57 Hunt PCW, Cotton BR, Smith G. Comparison of the

effects of pancuronium and atracurium on the lower eso- phageal sphincter. Anesth Analg 1984; 63: 65-8.

58 Carp C J, Goudsouzian NG, Liu LMP , Dedrick DF, Szyfel-

bein SK~ Assessment of risk factors related to the acid aspiration syndrome in pediatric patients - gastric pH and ress volume, Anesthesiology 1982; 56: 70-2.

59 Coombs DW. Asp~-ation pneumonia prophylaxis (editorial). Anesth Analg 1983; 62: 1055-8.

60 Gibbs CP, Schwartz D J, Wynne JW, Hood CL Kuck EJ~ Antacid pulmonary aspiration in the dog. Anesthesiology 1979; 51: 380-5.

61 Higgs RH. Smyth RD . Castell DO. Gastric alkalinization - effect on lower esophageal sphincter pressure and serum gastrin. N Engl I Med 1974; 291: 486-90.

62 Gibbs CP, Banner TC. Effectiveness of Bicitra as a pre- operative antacid. Anesthesiology 1984; 61: 97-9.

63 Chen CT, Toung T.IK, Haupt HM, Hutehins GM, Cameron 3/.,. Evaluation of the efficacy of Alka Seltzer efferves- cent in gastric acid neutralization. Anesth Analg 1984; 63: 325-9.

64 Fuure EAM. Lira HS. Block BS, Tan PL. Rolzen MF. Sodium bkcatbouate buffers gastric acid during surgery in obstetric and gynecologic patients. Anesthesiology 1987; 67: 274-7.

65 Viegaz 03, Ravindran RS. Shumacker CA_ Gastric fluid pH in patients receiving sodium citrate. Anesth Analg 1981; 60: 521-3,

Hardy: LARGE VOLUME GASTROESOPHAGEAL REFLUX

66 Viegas O J, Ravindran RS, Stoops CA. Duration of effica-

cy of sodium citrate as an antacid (abstract). Aneslh

Analg 1982; 61: 220.

67 gravitz JJ, Shape WJ, Cohen S. Effect of histamine and histamine antagonists on human lower esophageal sphincter function. Gastroenterology 1978; 74: 435-40.

68 Denis P, Galmiche JP, Ducrotte P, Colin R, Pa,wuis P, Lefrancois R. Effect of ranitidine on resting pressure and pentagastrin response of human lower esophageal sphincter. Dig Dis Sci 1981; 26: 999-1002.

69 Weber L, Hirshraan CA. Cimetidine for prophylaxis of aspiration pneumonitis: comparison of intramuscular and oral dose schedules. Anesth Analg 1979; 58: 426-7.

70 Manchikanti LM, Kraus JW, Edds SP. Cimetidine and related drugs in anesthcsia. Anesth Anal 8 1982; 61: 595-608.

71 Schwartz D J, Wynne JW, Gibbs CP, Hood CI, Kuck El. The pulmonary consequences of aspiration of gastric con- tents at pH values greater than 2.5. Am Rev Respir Dis 1980; 121: 119-26.

72 WynneJW. Aspiration pneumonitis- correlation of experi- mental models with clinical disease. Clin Chest Med 1982; 3: 25-34.

72 Coomb.~DW, MooperD, PageauM. Emergencycimetidine prophylaxis against acid aspiration. Ann Emer~ Med 1982; 11: 252-4.

74 Brock-Ume JG, Rubin J, McAravey Re t aL The effect of hyoscine and atropine on the lower oesophageal sphincter. Anaesth Intensive Care 1977; 5: 223-5.

75 Brock-Utne JG, Rubin J, Welman S. Dimopoulos GE, Moshal MG, Downing JW. The effect of glyeopyrrolate (Robinul) on the lower oesophageal sphincter. Can Anaesth SocJ 1978; 25: 144-6.

76 Fell D, Cotton BR, Smith G. I.M. atropine and regurgita- tion (lelter). Br J Anaesth 1983; 55: 256-7.

77 Brogden RN, Carmine AA, Heel RC, Speight TM, Avery GS. Domperidone - a review of its pharmacological activity, pharmacokinetics and therapeutic efficacy in the symptomatic treatment of chronic dyspepsia and as an antiemetic. Drugs 1982; 24: 360-400.

78 Sellick BA. Cricoid pressure to control regurgitation of stomach contents during induction of anaesthesia. Lancet 1961; 2: 404-6.

79 Salem MR, Wong AY, F izotti GF. Efficacy of cricoid pres- sure in preventing aspiration of gastric contenls in paedia- tric padems. Br J Aoaesth 1972; 44: 401-4.

80 SalemMR,JosephNJ, HeymanHJ, BelaniB, PaalissianR, FerraraTP. Cricoid compression is effective in obliterating the esophageal lumen in the presence of a nasogastdc tube. Anesthesiology 1985; 63: 443-6.

81 Lawes EG, Campbell l, Mercer D. Inflation pressure, gas- trio insufflation and rapid sequence induction. Br J Anaesth 1987; 59: 315-8.

173

82 Tomkinson J, Turnbull A~ Robson G, Ctoake E, Adelstein

AM, Weatherall J. Report on confidential enquiries

into maternal deaths in England and Wales 1973-1975. London, Her Majesty's Stationery Office, 1979, pp.

78-86. 83 Lawes EG. Crieoid pressure with or without the "cricoid

yoke." Br J Anaesth 1986; 58: 1376-9. 84 Lawes EG, Duncan PW, Bland B, Gemmet L, Downing

JW. The cricoid yoke - a device for providing consis- tent and reproducible ericoid pressure. Br J Anaesth 1986; 58: 925-31.

85 Sellick BA. Rupture of the oesophagus following ericoid pressure? (letter), Anaesthesia 1982; 37: 213-4.

86 Pansard JL, Chauvin M, Lebrautt C, Gauneau P, Davatdestin P. Effect of an intubating dose of succinyl- choline and atracurium on the diaphragm and the adductor pollicis muscle in humans. Anesthesiology 1987; 67'.

326-30. 87 Gibbs CP, Rotbin SH, Norman P. Cause and prevention of

maternal aspiration (letter). Anesthesiology 1984; 61: 111-2.

88 MclntyreJWR. The difficult tracheal intubation. Can J Anaesth 1987; 34: 204-13.

89 Hardy JF, Plourde (7, Lebrt~n M. Cdtd C, Dab, S, Lepage Y. Determining gastric contents under general anaes- thesia: evaluation of two methods. Can J Auaesth 1987; 34: 474-7.

90 ManchikantiL. Questions concerning aspiration prophy- laxis study (reply). Anesthesiology 1987; 66: 443.