Pulmonary aspiration, an uncommon occurrence in nonemergency airway management, may lead to a spectrum of sequelae, from no discernable effects to significant morbidity and mortality. In this chapter, we will outline the known factors that increase the risks of aspiration and how airway management may be optimized to reduce the risks to the patient.

Although reported in the literature as a relatively uncommon complication of nonemergency airway management, a majority of airway practitioners will acknowledge that the risk of aspiration is a major concern to them in daily practice. Most would acknowledge that if they have not had an episode of aspiration in one of their patients, they know a colleague who has had to deal with the complication.¹ Kluger reported in 1998 that over 71% of all anesthesiologists responding to a national mail-in survey in New Zealand had at least one case of aspiration in their careers.

When Was Gastric Aspiration First Described in Anesthesia Practice?

Although aspiration has been referenced in the anesthesia literature since the mid-1800s, Mendelson² was the first to describe in a detailed manner the occurrence of aspiration in conjunction with the delivery of anesthesia. Since that time, a plethora of publications have followed, outlining the risks and ways of preventing the problem. Unfortunately, much of the information is conflicting, and conclusions have been derived from studies with surrogate endpoints that may have very little to do with actual clinical risks. For example, the often-quoted study by Roberts and Shirley³ suggested a gastric volume of greater than 25 mL and pH of less than 2.5, as a specific risk factor for aspiration. This postulation was accepted by subsequent investigators who directed their efforts for prevention of aspiration to the assumption that these specific values were critical factors in predicting the outcome of aspiration.

What Was Sellick’s Approach to Minimize the Risk of Gastric Aspiration?

In the discussion section of his 1961 paper advocating the use of cricoid pressure (CP) during the induction of anesthesia to prevent the aspiration of gastric contents, Sellick⁴ examined alternatives available at the time. He identified inhalational induction in the supine or lateral position (with head-down tilt) and rapid IV induction of anesthesia in the sitting position. He commented that, with inhalational induction, vomiting usually occurred in lighter stages of anesthesia when protective reflexes were hopefully still present and noted that any difficulty during induction predisposed to regurgitation and anoxia. Rapid IV induction in the sitting position often led to cardiovascular collapse in critically ill patients, and pulmonary aspiration was made more likely by the sitting position, if gastric reflux occurred. Sellick advocated the use of CP during induction of anesthesia as a third option.

Sellick suggested that the stomach should be emptied before induction and the nasogastric tube then be removed. He was of the opinion that the nasogastric tube would prevent esophageal occlusion with CP. The patient was positioned with the head and neck fully extended and, following denitrogenation, induction ideally occurred with an IV barbiturate-muscle relaxant combination. CP was instituted before induction, moderate pressure was applied during induction, and this was increased to firm pressure once consciousness was lost. Sellick suggested that the lungs may be ventilated without risk of gastric regurgitation. Once intubation was completed and the cuff inflated, CP could be safely released.

In his description of CP, Sellick⁴ reported its application in 23 high-risk cases. He noted no instance of pulmonary aspiration in any patient but did report that, in three cases, release of CP after intubation was followed immediately by reflux into the pharynx of gastric or esophageal contents, suggesting that CP had indeed been effective.

What Is the Incidence of Aspiration in Anesthesia Practice?

The difficulty in determining the actual incidence of aspiration relates to a number of factors. First, because it occurs rarely, studies addressing the topic need to be very large. Most, if not all, of the better studies to date have been derived from large computerized databases. Second, aspiration is not always easy to recognize and, as will be illustrated below, rarely leads to clinical findings, let alone serious sequelae. Hence, it is an event likely to be missed and therefore underreported.

The recognition of gastric material in the pharynx does not alone support a diagnosis of aspiration. Despite the evident regurgitation, pulmonary aspiration may not have occurred. Even if foreign material is seen below the level of the true vocal cords, there may be a wide spectrum of clinical consequences. Silent aspiration may occur, wherein the patient exhibits no signs or symptoms of aspiration and there are no disruptions of physiological parameters. Indeed, it has been reported that asymptomatic aspiration may occur in up to 45% of normal subjects during sleep, and as many as 70% of people who have a blunted level of consciousness and responsiveness.⁵

Aspiration may become symptomatic, with cough and audible wheeze. Acute lung injury may be associated with tachypnea, increase in alveolar-arterial (A-a) gradient, hypoxemia, and radiological evidence of lung injury—with infiltrates and/or atelectasis. Frank respiratory failure may ensue, with the development of acute respiratory distress syndrome (ARDS), the need for ventilatory support, and (rarely) death.

Although many papers have reviewed the topic of aspiration, there has been a notable absence of consistent end points. In 1993, Warner⁶ published a retrospective review of 215,488 general anesthetics over a period of 6 years. Aspiration was defined as:

Of the anesthetics included, 202,061 were elective and the remaining 13,427 were emergency cases. There were 52 and 15 aspirations in these two groups, respectively. The overall incidence of aspiration was 1:3216. Aspiration occurred in 1:3886 of elective surgeries and 1:895 of emergency procedures. Sixty-seven cases of significant aspiration were recognized; one patient died from surgical causes intraoperatively. Of the remaining 66 patients, 42 (64%) experienced no obvious sequelae from the aspiration and 13 (20%) required mechanical ventilation with 6 (9%) needing prolonged (>24 hours) mechanical support. Of the latter group, three died of complications from their aspiration, giving a death rate of 1:71,829.

Mellin-Olsen et al.,⁷ in 1996, reported a prospective review of 85,594 cases over a 5-year period. They defined aspiration as “what the anesthetist has interpreted as such during, or immediately after the anesthetic procedures, based on clinical signs like gastric content, in the pharynx/larynx/trachea and a drop in O₂ saturation.” In their study, a total 25 cases of aspiration were recorded; 52,650 patients had received a general anesthetic, with the remainder undergoing either regional or IV sedation. All cases of aspiration occurred in the general anesthetic population, giving an incidence of 1:2106. The incidence of aspiration was 1:3303 in elective procedures under general anesthetic and 1:809 for emergency procedures, both rates similar in magnitude to those reported by Warner. In the patients who had aspirated, there were similar complications to those described by Warner.⁶ No deaths occurred and 22/25 patients had either no or minimal sequelae; three experienced more serious consequences. One patient required ventilation for 7 days, but made a complete recovery from his lung injury.

Olsson et al.⁸ reported a similar retrospective review, with an incidence of aspiration of 1:2131 and a mortality rate of 1:46,000. Finally, Sakai et al.⁹ conducted a 4-year retrospective review of 99,441 anesthetics at a single American university hospital. There were 14 cases of confirmed pulmonary aspiration for an incidence of 0.014% or 1:7103 overall. Seven cases of aspiration occurred in gastroesophageal procedures and patients in whom aspiration occurred had one or more identifiable risk factors. Six patients developed pulmonary complications related to the aspiration and one died. All these studies reaffirm that aspiration is a relatively uncommon occurrence and the mortality due to aspiration in the perioperative period is rare.

Kluger and Short¹⁰ reported, in 1999, the analysis of data from the Australian Anaesthetic Incident Monitoring Study (AIMS). AIMS is a voluntary, anonymous reporting system of anesthesia-related incidents, collected in a central database. Unfortunately, the nature of this type of study does not allow for a denominator (i.e., the total number of anesthetics performed by all reporting clinicians) and the incidence is unavailable. Of the 5000 reported events, 133 dealt with aspiration. Aspiration was deemed to have occurred if “any obvious nonrespiratory secretions were suctioned via a tracheal tube, there was chest X-ray evidence of new pathology after an incident, and/or there were signs of new wheeze or crackles after an episode of regurgitation or vomiting.” In this group of 133 aspirations, 5 deaths were recorded. Of interest, aspiration did occur in a number of patients undergoing regional anesthesia in the AIMS study (7 out of the 133), whereas none were reported to have occurred in almost 31,000 regional and sedation cases in the paper by Mellin-Olsen et al.⁷

The American Society of Anesthesiologists (ASA) Closed Claims Database reviewed the incidence of aspiration as a cause of liability to anesthesiologists and in 2000, Cheney¹¹ reported that aspiration represented 3.5% of all claims as a primary or secondary event; and in half of those it was the offending event leading to a claim. Seven percent of the aspiration claims were during regional anesthesia. In an updated report published in 2011, reviewing 8954 claims registered from 1970 to 2007, aspiration had become the third most common adverse respiratory event leading to anesthesia claims, now representing nearly one in five events leading to a claim.¹²

In conclusion, the incidence of aspiration in a surgical population is consistently estimated at between 1:2000 and 1:4000, depending on the population studied, and when the data had been reported. Emergency surgery increases the risk of aspiration fourfold, or more, compared to elective operation yet overall mortality remains low.

What Are the Risk Factors that Contribute to Aspiration?

In published studies measuring the incidence of aspiration, the most common association was with emergency surgery.^6–8,10,11 Olsson et al.⁸ reported that the timing of surgery also correlated with an increased incidence of aspiration, with a sixfold increase in the rate of aspiration between 18:00 and 06:00 hours. The causative factors that relate to the increased risk of aspiration with emergency surgery are not outlined in the studies, but numerous issues, such as lack of fasting, stress, depression of GI motility, less staff availability, higher dependence upon less-experienced anesthesia staff, and fatigue may all play contributing roles. A preponderance of the cases of aspiration reported by Olsson et al.⁸ occurred in patients who had abdominal surgery performed, with esophageal and upper abdominal procedures predominating. The incidence of aspiration in cesarean section was 1:661. Interestingly, Warner et al.,⁶ roughly 10 years later, reported no cases of aspiration during cesarean section. In another paper reviewing the incidence of aspiration in children, Warner et al.¹³ reviewed 63,180 anesthetics in children under the age of 18 and also found the incidence of aspiration in the emergency patient (1:373) to be significantly higher than that in the elective situation (1:4544).

There are reports that examine the incidence of aspiration in the pre-hospital and emergency department (ED) settings. In the pre-hospital setting, the rates of occurrence of aspiration vary from 6% to 90%, depending on the study populations and facilities, and whether the study considered survivors, non-survivors, or postmortem examinations.^14,15 Often, aspiration had already occurred prior to attempted intervention and provision of care. In some papers, the incidence of failure to intubate the trachea is as high as 47%.^16,17 In the study by Gausche et al.,¹⁶ patients were randomized into intubation group versus transport by bag-mask. There was an intubation success rate by the paramedics of only 57% and the only aspirations occurred in the intubation group. The outcomes between the two groups, in terms of survival, were similar. In the pre-hospital literature, the source of aspirate in trauma patients differs from that of the emergency surgical population. In the study by Lockey et al.,¹⁴ 34%, or a total of 18 of the trauma patients, had evidence suggestive of aspiration. Of these, 15 had blood contaminating their airway, while only 3 had evidence of gastric contents. All had significant head injuries, with a Glasgow Coma Scale of 8 or less.¹⁴

Taryle reviewed 43 consecutive intubations in the ED in a major teaching institution and reported a total of 38 complications in half of the patients (22/43). Aspiration occurring prior to airway manipulation was not included. There was a total of eight aspirations, the second most common complication after prolonged intubating time.¹⁸ Sakles et al.¹⁹ reported a 1-year review of all intubations in an ED that had a census of 60,000 patients per year. In 610 consecutive intubations, 49 patients had a total of 57 immediate complications. Although there were 10 cases of vomiting, they did not report any occurrences of aspiration. Mort,²⁰ reporting on airway complications during emergency intubation occurring outside of the operating room, noted that fewer aspirations occurred in the ED than on the wards or the medical intensive care unit (ICU). The obstetrical population will be discussed later in this chapter.

A review of aspiration in the pediatric population, published by the Pediatric Sedation Research Consortium, looked at the incidence of aspiration and major adverse events occurring during procedural sedation in patients known to have been nil per os (NPO) versus those not fulfilling NPO guidelines. This prospective review of over 139,000 encounters showed a very low incidence of aspiration in either group, with no difference between fasted and non-fasted patients (0.97 vs. 0.79 per 10,000). Major adverse events did not appear to correlate to the NPO status either, with an occurrence of 5.57 versus 5.91 per 10,000, respectively, echoing the findings of Brady et al.²¹ in their 2009 Cochrane review on the same topic. As the literature has indicated in the past, the incidence of aspiration and adverse events were associated with age (neonates and infants), ASA status, emergency versus nonemergency, and procedure type.²²

What Happens During an Aspiration That Determines Its Severity?

The consequences of aspiration can occur as a result of a chemical injury to the airway mucosa from either acid or bile. Injury may occur from particulate material in the aspirate causing either airway obstruction, or an inflammatory response. Finally, there may be pneumonia secondary to contamination from bacteria in the stomach or upper airway.

Injury resulting from aspiration is often that of an acute chemical burn and is a function of both volume and pH of the aspirate. Subsequent release of inflammatory substances, such as cytokines and interleukins from injured tissue, provokes neutrophil migration to the affected areas and further airway reaction. Airway edema, as well as capillary leak in the alveoli, can increase airway resistance and worsen lung compliance. The end result is ventilation–perfusion mismatching and hypoxemia, as well as inflammatory infiltration and/or atelectasis.

The initial chemical burn effect occurs within seconds, followed by neutralization of the acid within 15 seconds. The sudden onset of bronchospasm and laryngospasm may occur.

Histologically, rapid inflammation occurs, with alveolar wall thickening and cellular infiltration into the alveoli and interstitium. Hemorrhage can occur into the interstitium and alveoli. Alveolar wall disruption and rupture can follow as well as significant intra-alveolar edema and fluid.

Full evolution of injury can take several days. Repair of the injury is of the order of 3 to 7 days. Particulate materials can induce a local reaction themselves and lead to a pneumonia. Indeed, attempts to neutralize the gastric pH with particulate antacids may aggravate reaction in the lung due to the particles rather than from the acid itself. Particulate materials of sufficient size or number can produce substantive airway obstruction in their own right.

Pneumonia may follow, related to organisms from the upper airway, esophagus, and stomach. This is usually of a mixed flora, with anaerobes and aerobes present. Depending on the organism and the premorbid condition of the patient, this may progress to lung abscess. This is unlikely in the healthy patient. Differentiating between inflammatory pneumonitis and pneumonia may be difficult. The clinical and radiological picture may be similar. Pneumonitis is usually acute, resolving in hours to a day. If the presentation is one of progression without resolution, lasting days, with fever and purulent sputum, a diagnosis of pneumonia is more likely.^23–27

What Is the Relevance of the Issues of Gastric Volume, pH, and Constituency of the Gastric Contents?

Roberts and Shirley³ concluded that a pH of <2.5 and a gastric volume of 25 mL (or 0.4 mL·kg⁻¹) correlated with aspiration and resultant pneumonitis. In their study, an acid solution was injected directly into the bronchus of a monkey and extrapolations were made in regard to the volume and pH that would place humans at risk. These conclusions have been challenged by numerous investigators.^28,29 Schreiner²⁹ in 1998 pointed out that over 30% to 60% of patients have a gastric fluid volume of greater than 0.4 mL·kg⁻¹ (median 0.3, but as high as 4.5 mL·kg⁻¹), yet the incidence of aspiration is quite rare. Indeed, it has been demonstrated that the incidence of gastroesophageal reflux (GER) is not associated with residual gastric volume (RGV).³⁰ Rather, it has been shown that GER during anesthesia is related to episodes of straining on an endotracheal tube when inadequate anesthesia has been provided.³¹

Maltby et al.²⁸ argues that the risk of aspiration is due to loss of the barrier pressure at the gastroesophageal sphincter (GES), also referred to as the lower esophageal sphincter (LES). Normally, stomach contents are prevented from refluxing into the esophagus by the pressure exerted by the LES. The difference between LES pressure and intragastric pressure is the barrier pressure. The stomach is a very compliant structure and intragastric pressure can remain stable until volumes greater than 1000 mL are present.³² Indeed, as intragastric pressure rises, because of its anatomical design, so does LES pressure, maintaining the barrier pressure. In one study, measurements of the intragastric pressure and LES pressure during laparoscopy demonstrated that a rise in mean gastric pressure from 5.2 to 15.7 was matched by a rise in LES tone from 31.2 to 47.0 cm H₂O.³³

There is a clear association between aspiration and vomiting or gagging.^1,6–8,10 With active vomiting, or gagging, the sudden onset of high intragastric pressure is associated with relaxation of both the lower and upper esophageal sphincter mechanisms. This combination enhances the risk of pulmonary aspiration.

The higher the baseline intragastric pressure, the greater the tendency for GER and pulmonary aspiration. With an intestinal obstruction, for example, the intragastric pressure is high in association with the large RGV. This accounts for the high incidence of aspiration in this patient population and the finding that it is one of the most common factors associated with aspiration in most publications. By the same reasoning, patients with a documented hiatal hernia, or a history of GER disease (GERD), are also exposed to a higher risk of regurgitation and aspiration.¹⁰

Active vomiting, in association with an unprotected airway, is most likely to occur during induction of anesthesia, with airway manipulation prior to placement of the endotracheal tube, and at the end of a procedure as the patient is awakening and the airway is no longer protected. Inadequate levels of anesthesia at these times, as well as difficulty securing an airway, are the essential elements favoring the occurrence of aspiration. Two-thirds of aspiration events are reported during induction and extubation, equally divided between the two periods.⁶

How Important Is a History of Heartburn? Acid Taste or Burping? A History of GERD? How Much Reflux Is Significant?

Reflux occurs when the barrier pressure fails to prevent gastric contents moving from the stomach into the esophagus. Intuitively, those with a clear history of reflux should be at greater risk of aspiration. Kluger and Short¹⁰ found that a history of reflux and hiatal hernia were the ninth and tenth most common predisposing factors for aspiration, representing 7 and 6 cases, respectively, in the database of 133 total cases of aspiration. The patient with a history of acid reflux, with complaints of acid taste or choking at night, is deemed by many clinicians to be at greater risk of aspiration than one with only complaints of heartburn. The latter may simply suggest gastric mucosal pathology. However, no specific data are available to indicate that one symptom is more helpful than another in identifying who is at greater risk. Again the larger the volume of reflux, the more significant is likely to be the risk of aspiration.

What Clinical Situations and Characteristics Predispose to Aspiration?

Emergency Surgery

Emergency surgery is the most significant risk factor associated with aspiration in the studies outlined earlier, increasing the incidence of aspiration by four- to sixfold.⁶

ASA Physical Status

When Warner et al.⁶ compared the ASA status to the risk of aspiration in elective situations, the risk or aspiration increased by almost sevenfold as ASA status rose from I to IV or V. In emergency situations, the occurrence of aspiration increased from 1:2949 for ASA I patients to 1:343 for ASA IV and V patients, or almost a ninefold increment (Table 5–1). Olsson et al.⁸ also reported an increased risk of aspiration and increased morbidity with increasing ASA status. Most, if not all, of the reported aspiration-associated deaths occur in ASA IV and V patients.

Airway and Intubation Difficulties

Difficult intubation is associated with an increased risk of aspiration. Vomiting during airway interventions is frequently associated with aspiration, far more so than passive regurgitation. In Olsson et al.’s⁸ paper, out of 15 cases of aspiration in elective patients with no risk factors predisposing to aspiration could be identified from the chart review, 10 (67%) had difficulty with intubation preceding the vomiting and aspiration. In total, 58 out of the 87 patients who aspirated did so due to difficulty with intubation or, with airway manipulation. Warner et al.⁶ described aspiration in 69% of his patients in whom active vomiting or gagging occurred during intubation or extubation. Mort demonstrated that when the number of intubation attempts went from ≤2 to >2, a significant increase in complications occurred. The incidence of regurgitation rose from 1.9% to 22% and aspiration from 0.8% to 13%, directly correlated with an increase in the number of intubation attempts.²⁰ Sakai et al.⁹ reported that 5 of 16 reported aspirations occurred during laryngoscopy or airway interventions including the exchange of airway devices in at-risk patients.

In summary, there are multiple studies describing morbidity and mortality associated with airway interventions and difficulties.^20,34–36

Obesity

There was no correlation between obesity and aspiration in the studies by Warner et al.⁶ or Mellin-Olson et al.⁷ Olsson et al.,⁸ however, did find obesity to be a contributing factor for aspiration risk and obesity is frequently listed as an aspiration-associated factor in many other references. The association of obesity with an elevated risk of aspiration may relate to a high incidence of pertinent comorbidities. For example, delayed gastric emptying is known to be associated with diabetes, which is a more frequent finding in the obese. Other factors related to the obese include GER, difficult intubation, and inadequate anesthesia at the time of induction. This may account for the larger number of obese patients reported in aspiration populations.¹⁰

Obese patients have the same gastric emptying rate for liquids as non-obese patients. Depending on the meal content, the gastric emptying in obese patients for solids may be faster, slower, or the same as in the non-obese.^28,37–40 Maltby et al.²⁸ reported that obesity did not slow gastric emptying in the absence of other predisposing comorbid conditions and suggested that fasting guidelines should be applied to obese patients using the same criterion as for the non-obese. In their paper, obese patients, with no comorbid conditions, were randomized into fasting and non-fasting groups. The latter received a 300 mL clear fluid challenge preoperatively, with no difference in RGV demonstrated post-intubation between the two groups. A study reviewing anesthesia for electroconvulsive therapy in 50 obese patients reported no cases of aspiration in 660 procedures.⁴¹

Pregnancy

It has been well accepted that the obstetrical population is at increased risk of aspiration. This has been felt to be secondary to a number of factors. Hormones particularly progesterone, cause relaxation of the LES and impair gastric emptying. Mechanical effects of the gravid uterus alter the position of the stomach and, as term approaches, create a gastric “pinchcock” partially obstructing the gastro–duodenal junction. The gravid uterus also increases intra-abdominal pressure, which then increases intra-gastric pressure. It has been demonstrated that the intra-gastric pressure in pregnancy is increased to 17.2 cm H₂O from the nonpregnant level of 7.3 cm H₂O. Women experiencing heartburn in pregnancy have a drop in the LES tone from the normal in pregnancy of 44 cm H₂O to 24 cm H₂O. Heartburn in pregnancy is reported in some series to be between 45% and 70%, with 27% of these patients having hiatal hernias. The onset of labor with pain and stress, coupled with the presence of opioid analgesics, are independent factors associated with a reduction in gastric emptying. Increased difficulty with intubation occurs in the parturient related to hormonally induced mucosal edema and increased breast mass. For many reasons, the parturient is at an increased risk for aspiration.^3,42–44

Interestingly, however, this risk has significantly decreased since Mendelson reported a maternal death rate from aspiration during C-section of 1:667.^2,5 This may well be due to the increased use of regional anesthesia, as well as the application of rapid sequence induction (RSI) techniques, with CP and cuffed endotracheal tubes.⁴⁵ The use of pharmacologic interventions, although not proven to alter the incidence, may also be a contributing factor. The adoption of difficult airway practices that discourage persistent failing attempts at intubation may be an important factor.

A number of recent publications support the contention that the risk of aspiration has become a much smaller contributor to maternal morbidity and mortality. Mhyre et al.⁴⁶ reviewed all reported maternal deaths in the state of Michigan, USA between 1985 and 2003. Of the recorded 855 deaths over that time, 15 were felt to be associated with anesthesia in either a related or contributing form. Of the 15 cases, only 1 was felt to be due to aspiration. This occurred in the recovery area following caesarian delivery of a stillbirth. The rest of the deaths had no association with aspiration. Interestingly, all anesthesia-related deaths from airway issues occurred during emergence from general anesthesia or in the recovery room. None occurred during induction.

In a recent prospective observational study, McDonnell et al.⁴⁷ reported on the incidence of problems associated with airway management in the parturient, during the period 2005 to 2006, in 13 hospitals with just under 50,000 deliveries per year. During that period, 1095 general anesthetics were performed. In that series, eight cases reported regurgitation (0.7%), with one case of aspiration confirmed (0.1%). Two of the regurgitation cases occurred in elective caesarian sections. Of the eight regurgitation cases, four occurred during induction, and five at emergence (one patient regurgitated at both times). Interestingly, the incidence of difficult intubation was 3.3%, with failed intubation of 0.36%. The latter were managed with a laryngeal mask airway (LMA).

The low incidence of aspiration is supported by the Closed Claims Analysis of the ASA.⁴⁸ From 1990 onward, only two cases of aspiration were implicated in a maternal death or permanent brain damage, with one of these being in association with general anesthesia, the other with regional anesthesia.

In conclusion, the incidence of aspiration in obstetrics continues to decline as a source of morbidity and mortality. Increased use of regional anesthesia, a larger number of skilled practitioners comfortable with airway management options, better monitoring, and the use of prophylaxis may all be contributing factors.

Age

Although Warner et al.⁶ did not find age to be an independent risk factor for aspiration, Olsson et al.⁸ reported that extremes of age increased the risk of aspiration. Warner et al.¹³ reviewed 63,180 anesthetics in children under the age of 18. The incidence of aspiration in that population was not dissimilar to adults, except that there was an increased incidence of aspiration in patients less than 3 years of age. Over 91% of the patients who experienced aspirations in this population had either a bowel obstruction or ileus perhaps skewing the incidence in young children, although there is also some uncertainty as to the effectiveness of the LES in this population. Distended stomachs from both fluids and air entrained during crying or using a pacifier predisposes to gastric reflux when these infants cry or gag. The efficacy and method of application of CP during RSI in small children has not been defined.

Borland et al.⁴⁹ found that the incidence of aspiration in the pediatric population was 10.2:10,000, higher than Warner et al.’s¹³ reported 3.8:10,000.

Decreased Levels of Consciousness and Neurological Disease

It is recognized that the incidence of aspiration of extraglottic (e.g., blood) and lower GI tract (e.g., stomach contents) contaminants is increased in patients with a reduced level of consciousness.^8,14 In these patient populations, loss of function of the LES and upper esophageal sphincter and delayed gastric emptying combine with a reduction of upper airway protective reflexes to promote both regurgitation and aspiration.^50,51 This has relevance for the post-anesthetic period as well, as it has been shown that patients left in the supine position with a reduced level of consciousness have an increased incidence of aspiration.

Patients with other underlying neurological diseases, such as Parkinson’s disease and multiple sclerosis, are also at increased risk of aspiration due to impairment of their protective airway reflexes.⁵² The diabetic with autonomic neuropathy has been demonstrated to have delayed gastric emptying, sometimes manifested by early postprandial satiety, but it is usually asymptomatic. Diabetics have a theoretically increased incidence of difficult laryngoscopy and intubation due to glycosylation of collagen in the joints of the cervical spine resulting in diabetic stiff joint syndrome.⁵³ In spite of speculation that diabetics ought to be at increased risk of regurgitation and aspiration,^54,55 no studies have found diabetes to be an independent risk factor for aspiration.^{1,6–8,10,11,13,49}

Bowel Obstruction or Other Gastrointestinal Pathology

As discussed in the section “What Is the Relevance of the Issues of Gastric Volume, pH, and Constituency of the Gastric Contents?” in this chapter, increased gastric volume predisposes patients to an increased risk of aspiration. Gastric obstruction, and/or ileus, is one of the commonest clinical findings associated with aspiration.^8,10,13 The incidence of aspiration in esophageal endoscopy is 1:188, and appendectomy 1:751.

Full Stomach

Even in the absence of bowel pathology, recent ingestion of a meal has been documented to be a risk factor for aspiration.^6,13 Guidelines for fasting have been developed for the elective population. However, fasting does not guarantee an empty stomach and RGVs can be quite variable.

Is There a Difference in Gastric Emptying in Emergency Patients or Those Who Have Received Opioids?

Trauma patients have been shown to have delayed gastric emptying up to a week after injury. Patients who are critically ill, in ICU, also have significantly delayed gastric emptying.⁵⁶ Neurological injury, either head or spinal cord, is associated with significant delays in gastric emptying related, in part, to catecholamine surge.⁵⁷

Opioids, irrespective of the manner of administration, have been shown to decrease gastric emptying significantly.^58–60

Finally, alcohol ingestion, which is not uncommon in the population presenting urgently after trauma, significantly delays gastric emptying; this is true even in when the ingested alcohol is of relatively small volume (300 mL) and of low alcohol content (4%–10%).⁶¹

What Are the Current Fasting Guidelines? What Evidence Supports Their Use?

The American Society of Anesthesiologists has published an updated set of fasting guidelines,⁶² generated from a review of the available literature and expert opinion these guidelines are consistent with those advanced in 1999.⁶³ These guidelines were developed with the healthy patient in mind, booked for elective surgery. They are not intended to be applied to patients with comorbidities that would increase the risk of aspiration. There is a striking paucity of evidence around the relationships between fasting times, gastric volume, gastric fluid pH, and the risk of pulmonary aspiration. The consensus guidelines recommended the following:

There is controversy as to the application of these guidelines in the parturient,⁶⁶ a complex group, as there is always a potential need for emergency surgery. It would seem reasonable to allow the moderate intake of clear fluids or ice chips for the low-risk parturient. However, the high-risk parturient, either due to comorbidities or at increased risk of requiring an operative delivery, should be fasted.⁶⁷ For elective caesarean sections, a fast of 8 hours following solids is recommended.

What Role Do Pharmacological Agents Play at Minimizing Aspiration Risk?

Although evidence exists to support the contention that pharmacological agents reduce gastric acid production, gastric volume, or both, no evidence exists to support their use in preventing or reducing the incidence of aspiration or improving outcome if aspiration was to occur. Furthermore, the administration of many of the pharmacological agents could not be justified on the basis of cost-benefit analysis. Subsequently, the ASA Task Force has not recommended the routine use of any pharmacological interventions for the prevention of aspiration in patients with no apparent increased risk for aspiration including gastrointestinal stimulants, histamine-2 receptor antagonists, or proton pump inhibitors.

Similarly, there are no recommendations regarding the “at-risk” patient, other than the use of non-particulate antacids in the obstetrical population. The majority of studies looking at the efficacy of these drugs have been carried out in the healthy, low-risk populations.

H-2 antagonists, such as ranitidine, famotidine, and nizatidine, act by binding competitively to the histamine receptors on the gastric parietal cell. They are effective in increasing gastric pH and reducing gastric volume within 2 to 3 hours of administration. Unfortunately, the effect of the drug diminishes after a few days as tolerance develops.⁶⁸

Proton pump inhibitors interfere with the H+/K+ ATPase pump on the parietal cells. They are less effective than H-2 antagonists if the intent is to use them for a single dose as they require at least two doses to be effective, both the night before and the morning of surgery. Tachyphylaxis does not appear to develop with these agents.

Although reductions in gastric acid and volume have been shown with these agents, they do not reduce the harm from biliary fluid or particulate matter aspiration. Evidence from the animal literature suggests that pulmonary injury from bile is as significant, if not more so, than acid alone.⁶⁹ Bile, with a pH of 7.19, caused severe chemical pneumonitis and edema in one animal study.⁶⁹

Prokinetic agents are used to accelerate emptying of the stomach, thereby reducing RGV. The most commonly used of these is metoclopramide. It has multiple effects, including prokinetic properties, antiemetic properties, and finally, an effect on increasing the tone of the LES. It acts by antagonizing dopamine and serotonin receptors. Depending on the receptor subtype, it may also act as an agonist. Its antiemetic effects are largely due to the 5HT3 antagonism and the prokinetic effects from the 5HT4 agonism. Its prokinetic properties, however, are quickly inhibited by the presence of opioids and anticholinergic agents.

Of interest, the antibiotic erythromycin has been shown to be an effective agent in stimulating gastric motility, thereby increasing the rate of gastric emptying. It exerts its effect via the motilin receptor. Unlike metoclopramide, it has no extrapyramidal side effects and its prokinetic properties are not inhibited by opioids or anticholinergics. Doses of 1 to 2 mg·kg⁻¹ have been reported to be effective in reducing gastric fluid volume.^70,71

The use of antacids has been demonstrated to reduce the pH of gastric contents for variable lengths of time. As discussed earlier in this chapter, the administration of particulate antacids can be problematic. The use of clear non-particulate antacids has been in wide use for more than two decades, primarily in the obstetrical population. Sodium citrate is in routine use, prior to elective or emergency obstetrical procedures. Bicitra is a commercially available form of sodium citrate. The mechanism of action is by conversion to sodium bicarbonate. Rebound acidity will occur with prolonged use, increasing the volume of acid production.⁷²

What Is the Role of Ultrasound in Assessing Gastric Volume and Aspiration Risk?

Carp et al.⁷³ tested the ability of high-resolution ultrasound imaging to identify the stomach contents of nonpregnant volunteers as well as parturients and demonstrated that it was capable of identifying both the gastric volumes and content. Perlas et al.⁷⁴ evaluated the feasibility of using bedside ultrasonography for assessing gastric content and volume. They reported that the gastric antrum could be consistently identified and assessment of the antrum provided the most reliable quantitative information for gastric volume; the antral cross-sectional area (CSA) correlated with volume in a close-to-linear fashion, particularly when subjects were in the right lateral decubitus position. They concluded that that bedside two-dimensional ultrasonography could be a useful noninvasive tool to determine gastric content and volume. In a review of a number of studies on the topic, Van de Putte et al.⁷⁵ also reported that the antrum is the gastric region that is most amenable to sonographic examination and its evaluation accurately reflects the content of the entire organ.

Ultrasonographic measurement of antral CSA is feasible and reliable in the majority of critically ill patients. Hamada et al.⁷⁶ assessed the feasibility and validity of ultrasonographic measurement of antral CSA (us-CSA) in 55 critically ill patients who had an abdominal CT scan within an hour of the ultrasound assessment. Antral us-CSA measurements were feasible in 95% of cases and were positively correlated with gastric volume measured by the CT scan when performed in “good” conditions (65%) (r=0.43). There was good reproducibility of measurements and there was clinically acceptable agreement between measurements performed by radiologists and intensivists.

Mathematical models are available to predict gastric volume based on antral CSA in adults and they are thought to be both accurate and clinically applicable. A semi-quantitative three-point (grade 0 to 2) grading system has also been reported as a simple screening tool to differentiate low- from high-volume states.⁷⁷ This three-point grading system is based solely on qualitative evaluation of the gastric antrum that is scanned in both the supine and right lateral decubitus positions. A grade 0 antrum appears empty in both positions, and suggests no gastric content is present. A grade 1 antrum appears empty in the supine position, but fluid is visible in the RLD, consistent with a small volume (<100 mL) of gastric fluid. A grade 2 antrum is that clear fluid is evident in both patient positions consistent with a higher volume (>100 mL). Ultrasound assessment of gastric volume by clinical anesthesiologists is highly reproducible with high intrarater and interrater reliability. Kruisselbrink et al.⁷⁸ assessed the intrarater and interrater reliability of a method of gastric volume assessment based on gastric antral area and determined that reliability was nearly perfect and the two methods were essentially equivalent.

As a new diagnostic tool, gastric sonography needs to be characterized in terms of its validity (does it assess what it intends to assess, and how accurately), reliability (how reproducible are the results), and interpretability (i.e., what are the clinical implications of specific findings). Most studies to date deal with validity considerations and suggest that bedside ultrasound accurately determines RGV. Even though several descriptions of the type of content have been published, the sensitivity and specificity of a qualitative exam remain to be studied in a systematic manner. As data on the validity and reliability of gastric sonography become increasingly available, the next important question is correlate ultrasound assessment with aspiration risk so as to tailor anesthetic management in appropriate cases. Gastric ultrasound may ultimately prove to be very useful in aiding anesthesia practitioners in the determination of aspiration risk at the bedside and more appropriately guide anesthetic management.

What Is the Role of RSI in Airway Management?

RSI with CP has been deployed widely in anesthesia and emergency care for close to five decades and has been described as the standard of care in anesthesia for patients at risk for gastric regurgitation.⁷⁹ As well, it has been cited to be the most common method of airway management by emergency physicians for critically ill and injured emergency patients and is also the principal salvage technique when other oral or nasal intubation methods fail in the ED.⁸ Although it is characterized by a high success rate and a low rate of serious complications, its use has been challenged recently by authors in both anesthesia and emergency medicine who point out that the evidence base supporting its use is largely comprised of nonrandomized historical controls, case series, uncontrolled studies and expert opinion, and that the use of CP may complicate airway management.^19,80–83

The use of a rapid sequence technique results in fewer attempts, more rapid intubations, and higher success rates when compared to intubation with no sedation or sedation alone, both in hospital and in the pre-hospital setting.^84–86 Concerns about the role of rapid sequence techniques in the pre-hospital setting for the care of severely head-injured patients relate to the potential for hypoxemia under some circumstances. The major issues seem to be the occurrence of severe hypoxemia during induction in patients who were not hypoxemic before induction and the excess morbidity and mortality in those patients. An effective strategy for maintenance of oxygenation is needed before it can be concluded that rapid sequence intubation is of value in the out-of-hospital care of patients with serious closed head injury.^87,88

Describe the Technique of RSI

The patient should be placed supine at a height that is most convenient for the airway practitioner performing laryngoscopy. The head should be placed in the “sniffing” position with a firm pillow under the occiput. Although the benefit of the “sniffing” position compared to simple extension was recently challenged by Adnet et al.,⁸⁹ it did provide an advantage in patients who were obese or in whom there was at least one factor predictive of difficult intubation.

Denitrogenation

The usual method for denitrogenating (often referred to as pre-oxygenation) patients involves having the patient breathe 100% oxygen at tidal volumes through a tight-fitting facemask for 3 to 5 minutes. An alternative strategy is to have patients take four vital capacity breaths, but there is evidence that the former methodology is preferable.⁹⁰ Having said that, two recent reviews have concluded that having the patient take eight deep breaths (8DB) in 60 seconds provides a similar duration of safe apnea as does 3 to 5 minutes of tidal volume ventilation.^91,92 For this reason, whenever possible, it is recommended that either the tidal-volume breathing or the 8DB technique be employed. The same reviews concluded that denitrogenation of obese patients in the head-up compared to supine position also provided a longer period of safe apnea after the induction of anesthesia.

Cricoid Pressure

The cricoid cartilage should be identified by the assistant during denitrogenation, before induction of anesthesia, and the accuracy of the landmark should be confirmed by the airway practitioner. Cricoid pressure (CP) may be gently applied at the start of the induction sequence and the pressure increased to the amount recommended concurrent with the induction of anesthesia. The pressure should not be released until the cuff is inflated and the intratracheal position of the tube has been confirmed.

Nasogastric Tubes and Gastric Evacuation

Sellick⁴ recommended evacuating the stomach content with a gastric tube and then removing the tube before induction of anesthesia. Stept and Safar⁹³ argued that there was little evidence to support an effect of the tube on esophageal sphincter competence and suggested that the risk would be outweighed by the advantage of continuous gastric decompression when the tube was left in situ. Satiani et al.⁹⁴ demonstrated no difference in the incidence of regurgitation with or without a nasogastric tube. Salem and colleagues⁹⁵ demonstrated the effectiveness of CP in preventing reflux in both infant and adult cadavers with nasogastric tubes in place. It is recommended that if a nasogastric tube has been placed to empty the stomach, it should be left in place and open to atmosphere to limit increases in intra-gastric pressure during induction of anesthesia.

The Choice of Sedatives/Hypnosis During RSI

Despite wide acceptance and use of RSI, no single agent has emerged as the drug of choice for sedation and hypnosis during RSI. A deeper plane of anesthesia may improve intubating conditions in emergency patients undergoing RSI by complementing incomplete muscle paralysis.⁹⁶

The use of etomidate, ketamine, a benzodiazepine, or no sedative agent prior to neuromuscular blockade is associated with a lower likelihood of successful intubation on the first attempt, as compared with thiopental, methohexital, or propofol.⁹⁶ The use of the benzodiazepine midazolam alone is associated with a prolonged delay to time of laryngoscopy and doses >0.1 mg·kg⁻¹ are associated with a dose-related incidence of hypotension.^96–99

Etomidate, thiopental, and propofol have a favorable effect on intraocular pressure (IOP) and intracranial pressure (ICP).^100–102 Barbiturates provide cerebral protective qualities against ischemia caused by elevated ICP. However, barbiturates may significantly lower mean arterial blood pressure and thereby lower cerebral perfusion pressure, potentially compromising collateral blood flow to ischemic regions of the brain.¹⁰⁰ Although propofol also reduces ICP, it reduces mean arterial pressure (MAP) more than barbiturates and can thus cause a significant reduction of cerebral perfusion pressure.¹⁰³ Etomidate, in contrast to barbiturates and propofol, reduces ICP to a similar degree while maintaining or increasing MAP and cerebral perfusion pressure, but there is evidence of neurotoxicity in experimental models.^104,105

Ketamine, when used in the presence of hypovolemic shock, can be unpredictable in its effect on the hemodynamic profile. It possesses both indirect sympathomimetic stimulation and direct myocardial depressant properties, which support or raise systemic blood pressure in the acutely injured and hypovolemic patient. However, a patient who has been physiologically stressed for an extended period may be depleted of endogenous catecholamines, thereby rendering indirect autonomic stimulation ineffective and allowing the direct myocardial depressant effects to dominate.

Although it clearly has some advantages in a compromised patient, a significant disadvantage of etomidate is that it does not blunt the sympathetic response to endotracheal intubation.¹⁰⁶ This may result in hypertension and tachycardia during endotracheal intubation secondary to sympathetic stimulation. This response may raise ICP and increase myocardial work. Mitigation of this effect is achieved with the use of 1.5 to 5 μg·kg⁻¹ of fentanyl in conjunction with etomidate.¹⁰⁷

Lidocaine is widely used as a pretreatment agent to decrease the magnitude of increase of ICP in patients with closed head injury (with or without increased ICP). In fact, the evidence that IV lidocaine reduces the magnitude of the increase in ICP with elective tracheal intubation or suctioning in patients with increased ICP is indirect, and there is no evidence that it renders this effect in head-injured patients undergoing RSI.¹⁰⁸ As well, when it was administered intravenously in a dose of 1 mg·kg⁻¹ in healthy patients receiving thiopental and succinylcholine for induction of anesthesia, lidocaine use was associated with a decrease in MAP of 30 mm Hg.¹⁰⁹

The Choice of Muscle Relaxant During Rapid Sequence Techniques

Succinylcholine is widely used in anesthesia and emergency medicine during rapid sequence techniques. Doses approximating 1 mg·kg⁻¹ have been conventionally used for intubation; the average time to return to 50% of twitch height following this dose is 8 to 9 minutes, and 10 to 11 minutes for a return to 90% twitch height. A number of authors have explored the use of smaller doses to decrease the time to recovery and to limit the dose-dependent sequelae. El-Orbany et al.¹¹⁰ assessed onset times and time to twitch recovery for succinylcholine in doses of 0.3, 0.4, 0.5, 0.6, and 1 mg·kg⁻¹ after anesthesia was induced with fentanyl and propofol. Onset times ranged between 82 and 52 seconds, decreasing with increasing doses of succinylcholine but not differing between 0.6 and 1 mg·kg⁻¹. Intubation conditions were often unacceptable after 0.3 and 0.4 mg·kg⁻¹ doses, but acceptable conditions were achieved in all patients receiving more than 0.5 mg·kg⁻¹; intubation conditions in patients receiving 0.6 and 1.0 mg·kg⁻¹ were identical. The times to twitch recovery and to regular spontaneous reservoir bag movements were significantly shorter in the 0.6 mg·kg⁻¹ dose group compared with patients receiving 1 mg·kg⁻¹. Naguib et al.¹¹¹ carried out a similar study administering succinylcholine 0.3 to 1.0 mg·kg⁻¹ after anesthesia was induced with fentanyl and propofol. Intubating conditions were acceptable (excellent plus good grade combined) in 30%, 92%, 94%, and 98% of patients after 0.0, 0.3, 0.5, and 1.0 mg·kg⁻¹ succinylcholine, respectively. The calculated doses of succinylcholine that were required to achieve acceptable intubating conditions in 90% and 95% of patients at 60 seconds were 0.24 mg·kg⁻¹ and 0.56 mg·kg⁻¹, respectively.

While these results may have significant clinical implications, more studies are needed to examine the effectiveness of these smaller doses of succinylcholine in different patient populations, including obese, pregnant, pediatric, trauma, and critically ill patients. It must also be emphasized that the goal is “100% acceptable intubating conditions,” particularly in an emergency.

Since there are significant side effects associated with the use of succinylcholine, there is considerable enthusiasm in anesthesia practice to replace succinylcholine with a non-depolarizing muscle relaxant. At this time, rocuronium has emerged as the most likely non-depolarizer to fill this role. Rocuronium 1 mg·kg⁻¹ given after induction in a rapid sequence technique is clinically equivalent to succinylcholine 1 mg·kg⁻¹.¹¹² The incidences of clinically acceptable intubating conditions with rocuronium and succinylcholine were 93.2% and 97.1%, respectively. Clinically acceptable conditions occurred less frequently when rocuronium 0.6 mg·kg⁻¹ was used. The use of propofol 2.5 mg·kg⁻¹ combined with rocuronium 0.6 mg·kg⁻¹ results in satisfactory intubating conditions in 90% of patients within 61 seconds (range 50–81 seconds).¹¹³ The use of either thiopental (5.0 mg·kg⁻¹) or etomidate (0.3 mg·kg⁻¹) combined with rocuronium 0.6 mg·kg⁻¹ results in a longer time to achieve, as well as a lower incidence of satisfactory intubating conditions than that achieved with propofol/rocuronium combinations.¹¹⁴ The addition of alfentanil 10 μg·kg⁻¹ to these doses of etomidate and thiopental results in an increased likelihood of acceptable intubation conditions at 60 seconds following rocuronium administration.¹¹⁵

A recent review has concluded that a dose of succinylcholine of ≥1 mg·kg⁻¹ is required to ensure excellent intubating conditions and that smaller doses may not consistently provide such conditions.⁹¹ Rocuronium 1 mg·kg⁻¹ is a suitable alternative to succinylcholine during RSI, and although it will provide acceptable intubation conditions as often as equivalent doses of succinylcholine, it will provide excellent conditions less often. The time to full twitch recovery following paralyzing doses of rocuronium may be in excess of 1 hour, a factor that may exceed the comfort level of some practitioners. The availability and use of sugammadex (see section “Sugammadex” in Chapter 4) would be a potential solution in the situation where rocuronium is required for RSI, rather than succinylcholine but concern exists due to the prolonged neuromuscular block from the dose of rocuronium required.¹¹⁶

Positive Pressure Ventilation During RSI

In the eighteenth century, application of pressure in the cricoid area was advocated to allow for ventilation of the lungs without causing gastric distention.¹¹⁷ Sellick,⁴ in his description of CP, also recommended ventilating the lungs while awaiting onset of muscle paralysis. On the other hand, Stept and Safar⁹³ recommended against the use of ventilation after application of CP and conventional practice had until recently favored this recommendation. However, there is now evidence that not only is there a benefit to ventilating the patient’s lungs during the period of apnea but also that it can be done safely.

The average time to return to 90% of twitch height following an intubating dose of succinylcholine is considerably longer than it will take most patients to desaturate, even under ideal circumstances.¹¹⁸ Using a simulator model, Hardman et al.¹¹⁹ have identified the factors that shorten the time to desaturate with apnea. The factors that have a moderate effect are a reduced ventilatory minute volume preceding apnea and a reduced duration of denitrogenation. Those that have a large effect are increased oxygen consumption and reduced functional residual capacity. All of those factors are likely to be relevant in many instances of RSI.

There is a relationship between airway pressure and gastric inflation.^120,121 In subjects ventilated by bag-mask without CP, airway pressures below 15 cm H₂O rarely cause stomach inflation. Similarly, pressures ≤10 cm H₂O will not cause gastric distention but may provide insufficient ventilation; pressures ≥15 cm H₂O are generally adequate.¹²² Pressures between 15 and 20 cm H₂O will result in gastric insufflation in some patients, and pressures ≥20 cm H₂O do so in most patients.^120,122 Application of CP during BMV increases the maximum pressure that may be generated during mask-ventilation, without air entering the stomach, to about 45 cm H₂O.¹²¹

Petito and Russell¹²³ measured the ability of CP to prevent gastric inflation during BMV of the lungs. Fifty patients were randomized to either have or not have CP applied during a 3-minute period of standardized mask-ventilation. Patients who had CP applied had less gas in the stomach after mask-ventilation. However, more patients who had CP applied (36% vs. 12%) were considered more difficult to ventilate and these patients tended to have more air in the stomach when compared to those considered easy to ventilate with applied CP.

In summary, adequate ventilation of the lungs may be achieved with ventilation pressures of about 15 cm H₂O. As applied pressures increase to ≥20 cm H₂O, gastric insufflation will occur. The application of CP significantly reduces the volume of air entering the stomach at low to moderate ventilation pressures. It allows for continued ventilation of the lungs even in situations where past convention would have discouraged it, such as in RSI. Ventilating the lungs while awaiting the onset of muscle block is a prudent and useful maneuver to prevent oxygen desaturation, and there is an evidence base that supports this intervention. In order to prevent gastric insufflation, every effort should be made to ventilate the lungs at the lowest pressure possible.

Another method to reduce the onset of oxygen desaturation, or increasing the safe apneic time, is the use of apneic oxygenation. This occurs when a patent airway exists, between the lung and the upper airway. With a nasopharyngeal catheter insufflating O₂ at 5 L·min⁻¹ after onset of apnea following denitrogenation, Taha et al.¹²⁴ demonstrated that SpO₂ would remain at 100% for 6 minutes. In the control, with no O₂ flow via cannula, the mean time to SpO₂ to drop to 95% was 3.6 minutes. A similar result occurred with Ramachandran when he randomized obese males into two groups (n=15/group). The study group had nasal cannula with free flow of O₂ @ 5 L·min⁻¹ and the second was a control group. Both were denitrogenated in a standard fashion, induced with propofol/remifentanil/succinylcholine then underwent a simulated difficult laryngoscopy. Oxygen saturations were measured against time. The mean time to desaturation for the control group (SpO₂ <95%) was 3.49 minutes, whereas the study group had a mean time to 95% saturation of 5.29 minutes. In the study group, eight still had an SpO₂ >95% @ 6 minutes versus only one in the control population. Finally, the lowest saturation in the study oxygenated group was 94.3% versus 87.7% in the control population.¹²⁵

Discuss the Applied Anatomy of Cricoid Pressure

The cricoid cartilage is shaped like a signet ring with the narrow part of the ring being oriented anteriorly. The anterior arch of the cricoid cartilage is attached to the thyroid cartilage by the cricothyroid membrane. Laterally the cricothyroid muscles are situated in the cricothyroid gap (see Figure 3–23). The inferior horns of the thyroid cartilage articulate with the lateral surfaces of the cricoid cartilage. The cricoid cartilage is attached to the first tracheal ring by the cricotracheal ligament. The esophagus begins at the lower border of the posterior aspect of the cricoid cartilage. Sellick⁴ proposed the application of CP during induction of anesthesia, to prevent regurgitation of gastric or esophageal contents by compressing the esophagus between the cricoid and the cervical spine, obliterating the esophageal lumen. To perform the maneuver, the neck was extended, increasing the anterior convexity of the cervical spine and stretching the esophagus. Sellick hypothesized that this prevented lateral displacement of the esophagus when CP was applied. However, Vanner and Pryle¹²⁶ reported that contrast CT scanning in one patient revealed that when CP was applied, although the cricoid cartilage and cervical vertebrae were approximated, only part of the esophageal lumen was obliterated. Smith et al.¹²⁷ reviewed 51 cervical CT scans of normal patients to assess the anatomic relationships between the cricoid cartilage and the esophagus. Lateral esophageal displacement relative to the cricoid cartilage was evident in half (25 of 51) of the patients; 64% of those with lateral displacement had esophageal displacement beyond the lateral border of the cricoid cartilage. Smith subsequently reported on MRI taken of 22 volunteers with and without CP applied. The esophagus was again seen to be displaced laterally relative to the cricoid cartilage in 52.6% of the subjects; this increased to 90.5% with the application of CP. Lateral laryngeal displacement and airway compression were observed in 66.7% and 81% of the necks, respectively, with the application of CP.¹²⁸ Boet et al.¹²⁹ studied esophageal patency with and without CP in 20 conscious volunteers using MRI. Target CP was achieved in 16 of 20 individuals, corresponding to a mean percentage reduction in crico-vertebral distance of 43% (range 25%–80%). Incomplete esophageal occlusion was seen in 10 of 16, or 62.5% of individuals when what was deemed to be appropriate CP was applied. Incomplete esophageal occlusion was always associated with a lateral deviation of the esophagus. Rice et al.¹³⁰ investigated the anatomic impact of CP in 24 awake adult volunteers using MRI with and without applied pressure. With CP applied, the mean anterioposterior diameter of the hypopharynx was reduced by 35% and the lumen likely obliterated, and this compression was maintained even when the cricoid ring was lateral to the vertebral body. The location of the esophagus was irrelevant to the efficiency of the CP with regard to the prevention of gastric regurgitation into the pharynx. The magnetic resonance images showed that compression of the alimentary tract occurs with midline and lateral displacement of the cricoid cartilage relative to the underlying vertebral body. Finally, Zeidan et al.¹³¹ used real-time visual and mechanical means to assess the patency of the esophageal entrance with and without CP in 107 patients who were anesthetized and paralyzed. Attempts to insert two gastric tubes (GT) into the esophagus were made by a “blinded” operator without and with CP, the timing of which was randomized while images were recorded with a GlideScope. A successful insertion of a GT in the presence of CP was considered evidence of a patent esophageal entrance and ineffective CP, whereas an unsuccessful insertion of a GT was considered evidence of an occluded esophageal entrance and effective CP. Advancement of either size GT into the esophagus could not be accomplished during CP in any patient but was easily done in all subjects when CP was not applied. This occurred whether the esophageal entrance was in a midline position or in a left or right lateral position relative to the glottis. Esophageal patency was visually observed in the absence of CP, whereas occlusion of the esophageal entrance was observed during CP in all patients. The efficacy of the maneuver was independent of the position of the esophageal entrance relative to the glottis, whether midline or lateral.

There is a potential for lateral positioning and displacement of the esophagus relative to the cricoid cartilage and this may possibly explain case reports where, despite the application of CP during RSI, regurgitation and aspiration occurred.⁸¹ It is also possible that the failure to prevent aspiration was attributable in some of these instances to the improper application of the technique rather than the failure of the technique. The esophageal lumen may be occluded with the application of CP even when the esophagus rests partially lateral to the midline or is displaced laterally with the pressure application and some of the effect of the applied pressure in occluding the lumen of the upper gastrointestinal tract may occur at the hypopharynx.

What Is the Sellick Technique?

Sellick¹³² outlined a number of steps in his original description of CP applied concurrently with anesthetic induction. The patient was placed in the tonsillectomy position with the cervical spine in extension. Before induction of anesthesia, the cricoid was palpated and lightly held between the thumb and index finger; as induction commenced, pressure was exerted on the cricoid cartilage mainly by the index finger. As the patient lost consciousness, Sellick recommended firm pressure sufficient to seal the esophagus. CP was initially felt to be contraindicated by Sellick in the setting of active vomiting, in the belief that the esophagus may be damaged by vomit under high pressure. He later modified this stand, stating that he felt the risk of rupture to be almost nonexistent.¹³² Since his original description, the technique has been exposed to much study and critique. Data have now accumulated to provide evidence to support many of Sellick’s recommendations.

Does Cricoid Pressure Reliably Protect Against Regurgitation and Aspiration?

There are no outcome studies confirming the clinical benefit of CP when used either in anesthesia or resuscitation. Brimacombe and Berry⁷⁹ cite numerous case reports documenting the occurrence of aspiration despite the application of CP. There are also multiple studies documenting a negative impact of CP on patient interventions, usually relating to airway management. There is also a single case report in the literature attributing rupture of the esophagus to CP.¹³³ It involved an elderly female subjected to laparotomy after repeated episodes of hematemesis. The patient, who vomited on induction, was positioned laterally, CP was released, and the trachea was intubated after pharyngeal suctioning. At surgery, a longitudinal split was found in the lower esophagus. It was concluded, by the reporting authors, that the esophageal rupture represented an esophageal injury attributable to the CP. However, the diagnosis of rupture of the esophagus as a result of the repeated episodes of hematemesis represents as likely a cause, as the stomach adjacent to the area of esophageal injury was noted to be bruised and swollen during the surgery, suggesting a temporally more remote injury.