Post-ESOphagectomy patients presenting for General Anesthesia INduction: A systematic review of the literature (PESO-GAIN-R)

Abstract

Purpose:

Immediately following esophagectomy, aspiration is responsible for most direct postoperative complications, largely attributed to anatomical alterations from the surgical procedure. The long-term risk of aspiration following successful esophagectomy has not been systematically investigated, and there are no guidelines for the anesthetic management of postesophagectomy patients who require elective surgeries involving general anesthesia (GA).

Methods:

PubMed/MEDLINE, Embase, and Web of Science were systematically searched from inception through January 1, 2024 to identify studies involving patients ≥90 days postesophagectomy and undergoing elective surgery unrelated to their esophagectomy status, where GA was required. Data on perioperative anesthetic management were extracted. The primary outcomes assessed were perioperative gastrointestinal and pulmonary complications. Secondary outcomes included the approach to induction of GA and author recommendations.

Results:

Of the 4097 studies, ten studies involving 131 patients met inclusion criteria. Intraoperatively, adverse events during induction occurred in 13.0% of 131 cases, including regurgitation of gastric contents in 5.3% and pulmonary aspiration in 9.9%. The airway was managed with an endotracheal tube in 95.6%. Induction was performed in the semi-Fowler position in 92.9% and performed using the rapid sequence and standard induction techniques in 58.7% and 41.3%, respectively.

Conclusions:

Patients who have undergone esophagectomy may remain at high risk for aspiration during GA even in the long term, necessitating enhanced vigilance in anesthetic management. We offer evidence-informed suggestions for elective induction of GA in patients with a history of successful esophagectomy to enhance safe practice for these patients.

Introduction

Esophagectomy is often indicated for esophageal cancer with approximately 8800 procedures performed annually in the United States (US).[1] This type of cancer and surgery carry significant morbidity and mortality. In the immediate postoperative period, more than 55% of complications can be attributed to aspiration, often as a result of sphincterotomy, vagotomy, gastric tube reconstruction, and/or recurrent laryngeal nerve damage.[2,3,4] In the late postoperative period, however, gastrointestinal (GI) symptoms are uncommon. Some literature studies suggest that these patients remain at an increased risk of regurgitation and aspiration during induction of general anesthesia (GA) for elective procedures unrelated to their prior esophagectomy.[3,4]

Advancements in chemoradiation and surgical techniques have substantially improved survival outcomes of patients with esophageal cancer undergoing esophagectomy, surpassing 50% and 20% at 3 and 5 years, respectively.[1,5] Because of the relatively small number of patients reaching this treatment phase, anesthesiologists may seldom encounter postesophagectomy patients presenting for GA induction (PESO-GAIN) in their daily practice, and there is currently no guidance for the approach to GA induction in these patients. We sought to perform a systematic review (PESO-GAIN-R) of aspiration risk during the induction of GA to complement a nationwide survey of anesthesiologists in the US (PESO-GAIN-S) and summarize existing evidence for safe practice.[6]

Methods

This study was considered exempt from review by the IRB. A systematic review of the literature was performed using the databases PubMed/MEDLINE, Embase, and Web of Science from the inception of each respective database inception until January 1, 2024. The Covidence Systematic Review Software was used for article screening and data extraction, and the search was performed according to the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines.[7] Search terms included (esophagectomy AND [intubation OR “rapid sequence induction” OR fiberoptic OR pneumonia OR “acute respiratory distress syndrome” OR ARDS OR “airway management” OR aspiration] AND [history OR after OR post OR following]). Article selection was performed by two reviewers (N.S and R.A.D) independently based on title and abstract screening. Disagreements were resolved by a team discussion (N.S, R.A.D, A.D.S, and R.S) and consensus. Following selection, a full-text review was performed for confirmation and subsequent data extraction.

Inclusion and exclusion criteria

Literature involving patients with a history of esophagectomy (≥90 days following their esophagectomy) who were receiving anesthesia for an elective surgery requiring GA were included. Exclusion criteria were: 1) wrong patient population (e.g., no prior history of esophagectomy, patient with tracheostomy or tracheotomy, lack of general anesthesia), 2) surgery for esophagectomy-related complications (e.g., fistula, stenosis, stent replacement, reconstruction), 3) surgery ≥90 days post esophagectomy but within the same admission, 4) active esophageal metastatic disease and/or undergoing chemotherapy, 5) anesthetic management not reported, and 6) review articles. All articles with at least one eligible case were included regardless of language, study type, or nature of the surgical intervention. The selected publications were assessed for additional references that were not initially detected. For foreign language articles, we used Google Translate to obtain an English version.

Data extraction and outcomes

Using a predesigned collecting table, the data were extracted by N.S and R.S. Any discrepancies in the retrieved data were resolved by team discussion (N.S, A.D.S, R.S). Collected data included first author, publication year, sample size, baseline patient characteristics (e.g., age, relevant comorbidities), type of esophagectomy, time since esophagectomy, relevant history/symptoms (e.g., radiation therapy, presence of metastasis, tolerating oral intake), precautions taken prior to surgery, anesthetic considerations (e.g, position, induction/intubation technique), intraoperative complications (e.g., regurgitation, aspiration), and suggested management recommendations.

Due to the limited number and type of articles meeting inclusion criteria, descriptive analyses were performed, with reported calculations derived from manuscripts that provided individualized data. The cumulative results of included articles are reported as noted by the original authors.

Results

The systematic search generated 4097 citations. We did not find any randomized controlled or prospective studies, larger retrospective analyses, or any review articles on this topic. Ultimately, ten manuscripts met the inclusion criteria and consisted of the following: two retrospective studies (n = 120), five case reports (n = 6), and three letters to the editor (n = 5).[3,4,8,9,10,11,12,13,14,15] Figure 1 displays the publication selection flow diagram. The manual full text review did not provide any additional articles.

Figure 1.

Flow diagram depicting the systematic review process

The case reports and letters describing a total of 11 patients are notable for nine regurgitation/aspiration events, one postintubation drainage of copious GI content, and one uneventful intubation.

Demographics

In total, the articles reported on 131 patients undergoing a total of 148 surgeries with an average age of 64.5 years (range 42–75 years). The time elapsed since esophagectomy was on average 37.4 months (range 2 days to 335 months). Due to the aggregate nature of data reporting in the two retrospective studies, some patients were included in this analysis who had an esophagectomy less than 90 days prior and some had surgery for GI pathology related to their original disease.[4,15] Symptoms concerning delayed gastric emptying were reported in only 4 (3.1%) cases and included food regurgitation or toleration of small feeds only (n = 3) and dysphagia (n = 1). Table 1 summarizes the characteristics of patients included in this review.

Table 1.

Characteristics of included studies

Study (year)	Age (years, absolute or mean±SD)	Type of esophagectomy	Time since surgery (months)	Relevant history/symptoms	Procedure type
Watanabe (2021)[15] n=79*	67.1±8.3	Retrosternal (72) and posterior mediastinal (6) routes Cervical (72) and intrathoracic (7) anastomoses 3-field (71) and 2-field (8) lymphadenectomy	32 (0.4 – 335)	Radiation therapy (3) Primary lung cancer (39) Esophageal cancer with lung metastasis (34) Recurrent laryngeal nerve paralysis (0)	Lobectomy (34) Sublobar resection (45)
Myoga (2020)[4] n=41†	65 (42 – 77)	Posterior mediastinal (36), anterosternal (3), and retrosternal (2) routes Cervical anastomosis (41)	15 (0.1 – 132)	Radiation therapy (9) Hiatal hernia (6)	Surgery for other malignancies (34) Lung cancer resection (9) Endoscopic submucosal resection for remnant esophageal/gastric cancer (8) Bladder cancer (7) Abdominal cancer (4) Pharyngeal/tongue cancer (3) Breast cancer (2)
Tang (2020)[14]	55	Posterior mediastinal route Partial gastric tube reconstruction Cervical (left) anastomosis	6	None	Lobectomy
Koh (2018)[13]	74	Intrathoracic anastomosis	180	COPD Suspected hypopharyngeal carcinoma	Video-assisted mediastinoscopic lymphadenectomy/neck dissection
Yamazaki (2017)[12]	66	Sub-total, otherwise unspecified	4	Radiation therapy (3 months ago) GERD (only in the supine position) Poor cervical mobility	Lobectomy
Raghunandan (2013)[11]	53	Intrathoracic anastomosis	36	Radiation therapy Food regurgitation (only in the supine position for up to 3 h following PO intake)	Spinal decompression/vertebroplasty
Kiyohara (2010)[9]	73	Anterosternal	60	Asthma Food regurgitation (only in the supine position for up to 30 min after PO intake)	Lumbar fenestration
Study (year)	Age (years, absolute or mean±SD)	Type of esophagectomy	Time since surgery (months)	Relevant history/symptoms	Procedure type
Borsari (2009)[10]	58	Gastric reconstruction Cervical anastomosis	NR	Scleroderma GERD Multiple bronchoscopies without complications ASA III/Mallampati II Edentulous	Surveillance bronchoscopy
Borsari (2009)[11]	53	Gastric reconstruction (following failed colonic transposition) Cervical anastomosis	168	Dysphagia due to pyloric diverticulum, requiring pharyngostomy tube ASA III	EGD/Pharyngostomy tube replacement
De Souza (2009)[3]	59	Intrathoracic anastomosis	1.5	Tolerating small feeds	Dental extraction
Black (1999)[8] n=3	NR	NR	NR	NR	NR

Values are presented as median (range), mean±standard deviation, or n. *Includes n = 34 patients with metastatic esophageal cancer, for which the data is not separately reported. ^†Forty-one patients underwent 58 procedures. Includes 8 procedures for remnant esophageal/gastric cancer and 7 patients who underwent 8 procedures within 90-days, for which the data is not separately reported. ASA, American Society of Anesthesiologists, COPD, Chronic obstructive pulmonary disease; EGD, Esophagogastroduodenoscopy; GERD, Gastroesophageal reflux disease; NR, Not reported; PO, Per oral

Esophagectomy characteristics

The majority of esophagectomies were performed via the retrosternal approach (n = 74; 56.5%), followed by posterior mediastinal (n = 43; 32.8%) and anterosternal (n = 4; 3.1%). Gastric reconstruction was reported in three (2.3%) patients. Most patients had cervical anastomoses (n = 116; 88.5%), and a fewer had intrathoracic anastomoses (n = 10; 7.6%). Two-field lymphadenectomy was performed for eight patients.

Types of procedures performed

Where reported, the 148 surgeries consisted of lung resection (n = 89; 60.1%), surgery for other malignancies (n = 34; 23.0%), endoscopic submucosal resection (n = 8; 5.4%), bladder cancer (n = 7; 4.7%), abdominal cancer (n = 4; 2.7%), pharyngeal/tongue cancer (n = 3; 2.0%), and breast cancer (n = 2; 1.4%).

Anesthetic management

Prior to surgery, nil per os (NPO) data were reported in 92 (62.2%) of 148 cases, of which 54 (58.7%) fasted for an average of 22.6 hours (ranging from 9 to 48 hours). We assumed standard NPO times (≥8 hours) for 31 patients for whom no specific NPO times were reported prior to their surgery. Seven patients presented as an emergency without a prescribed fasting option. Nasogastric tube placement prior to induction was reported in five cases. In two cases, antacids were provided preoperatively, and another patient was compliant with their antacids for the treatment of gastroesophageal reflux at the time of surgery.[4,10,11] Otherwise, the use of preoperative GI chemoprophylaxis was not explicitly reported.

The type of airway device employed was reported for 68 (45.9%) of 148 procedures. These included endotracheal tubes (n = 65; 95.6%), nasotracheal tubes (n = 2; 2.9%), and a supraglottic airway in one case (1.5%). Three cases explicitly reported the use of video laryngoscopy during intubation.

An induction strategy was reported for 46 (31.1%) of 148 procedures, including rapid sequence induction (n = 27, 58.7%) and standard induction (n = 19, 41.3%). Where reported (n = 85; 57.4%), patient positioning during induction was most commonly semifowler (n = 79, 92.9%), followed by supine (n = 4; 4.7%), head-up tilt (n = 1; 1.2%), and semiseated (n = 1; 1.2%).

Cricoid pressure or like maneuvers were reported in 9 cases (17.3%) in articles representing 52 (39.7%) patients and included cricoid pressure (n = 6; 66.7%), sternal compression (n = 2; 22.2%), and compression at the cervical surgical scar site (n = 1; 11.1%). One study (n = 79) did not mention any such maneuvers.[15]

Perioperative complications

Intraoperatively, adverse events during induction occurred in 17 (13.0%) of 131 patients, including 7 (5.3%) patients who regurgitated gastric contents and 13 (9.9%) who experienced pulmonary aspiration. Two cases experienced both complications, mounting to an overall 15.2% induction event rate. There was one death of a patient who experienced pulmonary aspiration on elective induction and subsequently developed aspiration pneumonia, and an additional episode of pulmonary aspiration following emergency intubation during the same admission after a complicated hospital course.[10] No other perioperative mortality was reported.

Recommendations

Table 2 summarizes any recommendations from the extracted literature. Watanabe et al.[15] introduced a preoperative NPO algorithm for their patients and did not observe any aspirations in their cohort thereafter. It consisted of admission and a 48-hour NPO period in symptomatic patients and a 24-hour NPO period for asymptomatic patients. Preoperative review of clinical and especially imaging information pertaining to the prior esophagectomy was recommended by three articles, and another suggested the preoperative use of ultrasound to guide GI content assessment and guidance for the utility of specific additional maneuvers.[4,10,11,14] Further recommendations across the extracted literature included the use of preoperative oro/nasogastric decompression, administration of GI motility and antacid medications, routine rapid sequence induction technique, selective (depending on the surgical approach) use of cricoid pressure, head-up positioning, and awake fiberoptic intubation.

Table 2.

Case and induction details, author recommendations

Study (year)	NPO information	Induction position	Induction technique	ETT type and choice of Laryngoscope (ease of intubation)	Regurgitation and aspiration events	Author recommendations for induction
Watanabe (2021)[15] n=79	NPO 12-24 h (38); NPO 48 h (10); Non-fasting (31)	Semi-fowler (79)	NR	NR	Aspiration pneumonia (5): 2 non-fasting cases required mechanical ventilation; 2 had an NGT placed intraoperatively; 2 were attributed to contralateral thoracotomy during esophagectomy and resultant connection between the pleural spaces.	NPO for 48 h (versus 12-24 h) for patients with postprandial reflux, recurrent laryngeal nerve palsy/hoarseness, sitting up for >2 h postoperatively after PO intake for all patients, and vigilance in cases requiring ipsilateral thoracotomy to the previous esophagectomy due to a higher risk of postoperative complications.
Myoga (2020)[4] n=41*	Antacid prophylaxis (1) NGT (5) Non-fasting (7) due to emergency	Head-up tilt (0), otherwise NR	RSI (22) Standard (17) No data (8) Manual compression of the sternum (1)	ETT: (single-lumen 44; double-lumen 11); nasotracheal tube (2); SGA (1)	Standard induction in patients (NPO 9-14 h) with retrosternal reconstruction (conduit lateral to the trachea) → regurgitation noted on video laryngoscopy (1)/direct laryngoscopy (1) → intubation → suction catheter/bronchoscope negative for aspiration → extubated on POD #0 → no respiratory complications	Preoperative CT scan review, RSI with head-up position, and no manual ventilation (to lower the risk of reflux from lung expansion/gastric compression) and no routine cricoid pressure.
Tang (2020)[14]	NPO (>17 h) hours	NR	Rocuronium (50 mg) Standard	Video laryngoscope Double-lumen ETT (unspecified)	Regurgitation noted on laryngoscopy → intubated → NGT placed, draining 100 mL → aspiration pneumonitis → prophylactic antibiotics → extubated on POD #2 → no respiratory complications	Pre-induction ultrasound to assess gastric content and guide cricoid pressure, preoperative NGT, and prophylactic antacid/stomach stimulants.
Koh (2018)[13]	NPO (>13 h) NGT	NR	Succinylcholine (1.5 mg/kg) Standard	Video laryngoscope Mallinckrodt laser-ETT intubation (complicated)	Regurgitation noted on laryngoscopy → intubation difficult at the vocal cords with aspiration noted upon manipulation → NGT drained 50-100 mL → extubated on POD #0 → no respiratory complications	Aspiration precautions in patients with prior esophagectomy, even in those without risk factors or uncomplicated intubations in the past.
Yamazaki (2017)[12]	NR	Semi-seated	Rocuronium (80 mg) RSI Cricoid pressure	Video laryngoscope Double-lumen ETT (complicated)	Intubation difficult passed the glottis, attributed to tracheal deviation by the cricoid pressure → neck flexion pressure facilitated passage into the trachea → no aspiration or respiratory complications	Description of video laryngoscope use in quasi sitting position, and forward head flexion during application of cricoid pressure in setting of limited neck extension
Raghunandan (2013)[11]	Antacid prophylaxis NPO (12 h) NGT (failed attempt due to uncooperative nature)	Head-up (30°) tilt	Rocuronium (40 mg) RSI Manual compression of the surgical scar (left neck)	ETT (easy)	Following intubation, NGT placed, draining 200 mL → regurgitation (200 mL) noted on prone positioning → extubated in a head-up tilt (45°) position on POD #0 → no respiratory complications	Review of preoperative CT scan to localize anastomotic site/deviation, antacid prophylaxis, pre-induction NGT, manual compression above the neck scar, and head-up laryngoscope position.
Study (year)	NPO information	Induction position	Induction technique	ETT type and choice of Laryngoscope (ease of intubation)	Regurgitation and aspiration events	Author recommendations for induction
Kiyohara (2010)[9]	NPO (12 h)	Supine	Rocuronium (25 mg) RSI Sternal/gastric compression	ETT	Regurgitation noted on induction → intubation → OGT placed, draining 300 mL→ metoclopramide → extubated POD #0 → no respiratory complications	Antacid prophylaxis, strict NPO, NGT, manual compression of the neo-esophagus in patients with anterosternal reconstruction, and awake intubation.
Borsari (2009)[10]	Antacid for GERD NPO (12 h)	NR	Succinylcholine (100 mg) RSI Cricoid pressure	ETT (easy)	Regurgitation noted on laryngoscopy → intubation→ prophylactic antibiotics → extubated POD #1 → no respiratory complications	Review of preoperative CT scan demonstrated deviation of the EGJ at the level of the cricoid to the left (patient #1) and right (patient #2) of the trachea. Preoperative CT scan/operative note/thoracic consult review to identify anatomy and determine the utility of cricoid pressure and/or manual application of pressure at the EGJ.
Borsari (2009)[10]	NR	NR	Succinylcholine (200 mg) RSI Cricoid pressure	ETT (complicated)	Regurgitation noted on induction → intubation → bronchoscopy revealed aspirated bile on the right bronchi→ respiratory failure → extubated on POD #11 → respiratory failure following embolization for retroperitoneal hematoma → emergency intubation POD #24 → aspiration → died POD #25
De Souza (2009)[3]	NPO (duration unspecified)	NR	NR	NR	Aspiration noted on induction → intubation → ARDS → resolved	Aspiration precautions (NGT, RSI, awake intubation in the head-up position, etc) and selective cricoid pressure/utility due to altered anatomy.
Black (1999)[8] n=3	NR	Supine	Cricoid pressure	ETT (easy)	Aspiration noted on induction/intubation (3)	Meticulous preoperative evaluation, upright position during intubation, low threshold for fiberoptic intubation in the seated position.

Values are presented as n (as reported by authors), unless otherwise indicated. *Forty-one patients underwent 58 procedures (elective: 51 (88%); emergency: 7 (12%)), of which 47 required general anesthesia. Both patients with intraoperative regurgitation were undergoing elective surgery. ARDS, Acute respiratory distress syndrome; CT, Computed tomography, ETT, Endotracheal tube; EGJ, Esophagogastric junction; GERD, Gastroesophageal reflux disease; NGT, Nasogastric tube; NR, Not reported; NPO, Nil per os; OGT, Orogastric tube; POD, Postoperative day; RSI, Rapid sequence induction

Discussion

Our systematic search of the literature with our keyword strategy produced 4097 citations. Upon detailed screening, only ten reports were relevant to our specific topic, of which eight consisted of individual case presentations. This finding indicates a research and knowledge gap in the population of interest. A total of 131 patients were represented in these manuscripts and in combination with the categories of articles, suggesting a very low level of scientific evidence on the topic of interest.[3,4,8,9,10,11,12,13,14,15] A consistent finding among the articles describing 11 individual cases was the regurgitation and/or aspiration of GI content on induction of GA in nine instances. Similar to our recent survey of US anesthesiologists, a range of approaches to induction of GA in patients with a history of successful esophagectomy were reported. To date, only one retrospective study assesses the risk of aspiration in individuals with previous esophagectomy. In this retrospective propensity-matched cohort study, Watanabe et al. (2021)[15] reported a twofold increase in the incidence of aspiration among patients with a prior esophagectomy undergoing lung cancer resection compared to those without.

In the general population, aspiration is a relatively uncommon occurrence during induction of GA, affecting 2–7 patients per 20,000 cases.[16] In the study cohort of this review, the combined regurgitation and aspiration event rate during induction of GA reached 15.2%, suggesting a significant risk far beyond that for the general population. Given the reports and authors’ experience, aspiration risk in this patient population is unpredictable and not reliably detected by GI symptoms. The true rate of aspiration events is unknown but likely higher than our data suggest due to underreporting of adverse events. Nevertheless, this risk is emphasized by the specific case reports we identified.

Esophagectomy itself is a highly morbid procedure with early postoperative aspiration accounting for 55% of postoperative mortalities, attributed to surgical alterations of the upper GI tract and laryngeal anatomy leading to delayed gastric emptying and dysphagia (e.g., vagotomy, gastric tube reconstruction, and relocation of the gastric cavity into the thorax).[2,15,17,18]

Depicted in Table 3, different surgical approaches to esophagectomy may result in different postoperative GI symptomatology and aspiration risk.[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38] For instance, postoperative dysphagia may occur more frequently with the retrosternal approach (versus posterior mediastinal), whereas horseness/vocal cord paralysis and reflux disproportionately impact those undergoing extensive three-field (versus two-field) lymphadenectomy and whole stomach (versus partial stomach) reconstruction, respectively [Table 3].[19,20,21] However, the current literature does not consistently favor any technique over another based on these postoperative outcomes as freedom from disease and recurrence is the main objective of the intervention, and long-term follow-up is sparse.

Table 3.

Clinical and surgical characteristics indicating aspiration risk in patients with prior esophagectomy

I. Signs and symptoms	II. Incidence	III. Mechanism	IV. Route	V. Anastamotic site	VI. Gastric size	VII. Lymphadenectomy
Dysphagia and/or impaired swallowing	3-years: 11.9 – 16.7%[22]	Damaged hyoid muscles, hyoid displacement, recurrent laryngeal nerve damage, impaired laryngeal elevation, and vocal cord immobility[23]	Retrosternal > Posterior mediastinal[19]	Cervical = Intrathoracic[22]	Whole stomach = gastric tube[20]	NA
Horseness and/or vocal cord paralysis	1-year: 22%[24]	Recurrent laryngeal nerve damage, delayed swallowing reflex, and vocal cord immobility[25]	Retrosternal = Posterior mediastinal[26]	Cervical =[19] or >[27] intrathoracic	NA	3-Field >2- Field[21]
						Right > Left[21]
Pharyngolaryngeal reflux and/or regurgitation	≥90 days: 68 – 70%[28]	Upper esophageal sphincterectomy/impairement[29]	Posterior mediastinal > Retrosternal[19,36]	Cervical = Intrathoracic[27,28]	Whole stomach > gastric tube[19,20]	NA
Gastroesophageal reflux and/or epigastric pain	3-years: 25 – 39%[22,30] 5-years: 24%[31]	Lower esophageal sphincterectomy with reintervention and complete recovery of acid output[32]	Retrosternal = posterior mediastinal[31,33]	Cervical =[22,27] or <[31,34] intrathoracic	Whole stomach >[19] or <[31] gastric tube	NA
Abdominal fullness and/or delayed gastric emptying	≥6 months: 22.8%[35]	Vagotomy, sphincterotomy, and elevation of the stomach into the negatively pressured thorax[35]	Retrosternal = posterior mediastinal[30,33,36]	Cervical = Intrathoracic[22,26]	Whole stomach >[36,37] =[20,27,38] or <[38] gastric tube	NA
			Conduit located partially in the right hemithorax > completely located in the mediastinum/right hemithorax[37]

NA, Not available. <, = or > in columns IV-VII indicates a comparatively lesser, equal or higher incidence of the signs and symptoms described in column I

The common features of esophagectomy most often include replacing the esophagus with a gastric, colonic, or jejunal conduit, thereby disrupting normal peristaltic movement. Gastric reconstruction and pyloromyotomy may be additional surgical choices. The passage of oral intake becomes dependent on gravity, resulting in food stasis and symptoms of dysphagia and regurgitation. In addition, removal of the lower esophageal sphincter predisposes patients to reflux, exacerbating symptoms. Esophagectomy surgery may damage or intentionally interrupt the vagus nerve, particularly in oncologic resections, leading to delayed gastric emptying, pyloric stenosis, or rapid transit. Symptoms can range from early satiety and bloating to dumping syndrome.[23] Post esophagectomy, many GI symptoms improve over time. Dysphagia and reflux often recede within 6 months.[39] Beyond 5 years, only 20% of patients still experience surgery-related GI symptoms, the absence of which may not reliably predict safe intubation conditions.[40]

It is conceivable that anesthesiologists who do not routinely encounter (asymptomatic) patients with a history of successful esophagectomy may be unaware of the elevated aspiration risk in this population. The greatest risk factor for aspiration is the presence of their esophagectomy and associated GI changes. Additionally, a standard NPO regimen prior to elective surgery may be insufficient in these patients to assure a content-free upper GI tract. Just one study used a specific NPO strategy that included a prolonged fasting period in their patients and reduced aspiration events to zero.[15] The type of preoperative diet was not addressed in any study. Extrapolating from the nonesophagectomy population, it is well established that liquids transition faster in the GI tract compared to solid food.[41] Interestingly, the open comment section in PESO-GAIN-S provided information from a participant in a high-volume center, suggesting a 2-day liquid diet in these patients prior to elective surgery. These findings may be considered for our patient population to improve perioperative safety. None of the reviewed literature provided information on patients who may have a percutaneous GI feeding tube. It may be reasonable to perform preoperative drainage of such tubes when present.

The majority of procedures in the present systematic review were performed using an endotracheal tube (95.6%) and RSI technique (58.7%). A debated component of RSI is the application of cricoid pressure (Sellick maneuver), intended to compress the esophagus against the cervical vertebrae.[42] However, cricoid pressure does not appear to be consistently efficacious and may be detrimental for postesophagectomy patients, partly due to the posterolateral location of the esophagus, which may be further displaced in patients with cervical esophageal anastomoses.[10,43,44] In two articles of this systematic review, cricoid pressure complicated tracheal intubation.[10,12] A selected approach based on preoperative information seems prudent. The question of using a direct versus an indirect laryngoscopy for intubation was not sufficiently addressed in this review to draw definitive conclusions. However, recent studies repeatedly report superior intubation conditions and success rates with videolaryngoscopy compared to standard direct laryngoscopy. When available, this should be the default approach during induction of general anesthesia in these patients.[44]

Positioning during induction mostly included head elevation (e.g., semi-Fowler, head-up tilt, semiseated) in 95.3% of cases where this induction component was recorded. Proponents argue that these positions leverage gravitational effects to alleviate compression of the intestinal conduit, especially if located intrathoracic. However, while head elevation can lower the risk of regurgitation in these patients, it may carry a risk of catastrophic aspiration should regurgitation occur, underscoring the importance of sustained vigilance.[3,4,8]

Preoperative oro-/nasogastric tube placement was infrequent in our reported cases. Almost all patients who experienced adverse events during induction of GA subsequently received an oro-/nasogastric tube, resulting in notable drainage of intestinal content. There were no insertion complications reported postoperatively. The safety of oro-/nasogastric tube placement for the neoesophagus in esophagectomy patients remains an issue of concern for many practitioners, but recent surgical literature suggests less risk than previously assumed.[45,46,47] Only one letter to the editor from a high-volume esophagectomy center specifically describes the utility of preoperative oro-/nasogastric tube placement with the exception for cases of colonic interposition where the authors report placement as impossible.[48] The same authors suggest selective and targeted use of cricoid pressure and an experienced anesthesiologist to perform the intubation.

Table 4 summarizes evidence informed best practices for the elective induction of general anesthesia in patients with a history of successful esophagectomy. We formulated these suggestions by synthesizing the results of the evidence in the literature of this systematic review (PESO-GAIN-R), the characteristics of GI changes induced by esophagectomy, author consensus, and our recently published survey (PESO-GAIN-S).[6]

Table 4.

Evidence-informed best practices for post-ESOphagectomy general anesthesia induction

Based on a systematic literature review, structural changes to the GI tract and upper airway from the esophagectomy, and mindful interpretation of our findings, the following recommendations are proposed. Due to limited available data, esophagectomy technique-specific recommendations remain elusive. All patients with a history of successful esophagectomy, even in the absence of GI symptoms, should be considered at high risk for intestinal regurgitation and pulmonary aspiration during the induction of general anesthesia. →Aspiration precautions should be universally applied (see intraoperative management below). →Supraglottic airways should be avoided with the exception of use for airway rescue.
Pre-operative assessment and management
Medical history	GI symptoms
	Esophagectomy operative reports
Review Workup (as available)	Computed tomography (evaluate anatomy, residual GI content, etc)
	Upper GI studies
	Speech pathology consult
Fasting precautions	Consider liquid diet - 48 hours Consider extended Nil per os - 12 to 24 hours[15]
Intra-operative management
Pre-induction	Depending on preoperative assessment consider evacuation of any GI content via percutaneous GI tube if present±GI decompression with the naso/orogastric tube±Awake fiberoptic if otherwise indicated, or if the patient is unlikely to tolerate brief periods of hypoxemia/has limited lung reserve that would make aspiration especially detrimental (e.g., prior pneumonectomy) ± GI chemoprophylaxis
Induction	Reverse Trendelenburg and head-up position
	Consider rapid sequence intubation with or without use of targeted cricoid pressure
	Consider intubation by the most experienced anesthesiology practitioner
	Primary use of indirect laryngoscope[49]
Emergence	Reverse Trendelenburg and head-up position as tolerated
	Extubation when awake
	Consider pre-extubation GI decompression via gastric tube and naso/orogastric tube as present in select patients
	Head-up position for recovery

GI, Gastrointestinal

The overall low level of evidence found in this review suggests that prospective studies on this topic are needed. An alternative to ethically prohibitive randomized controlled trials for these patients may be the creation of a prospective and ideally international anesthesia registry and database with specific reporting elements for patients with a history of esophagectomy undergoing anesthesia.

Limitations

The nature and scarcity of publications on this topic resulted in a very low level of guiding evidence. However, it is also a call to action to improve anesthetic safety for patients with a history of esophagectomy. We were not able to completely ascertain a ≥90 day postesophagectomy status or absence of a GI-related indication for surgery for all patients of the two retrospective studies included because of the aggregate reporting in these studies.[4,15]

Conclusion

This review confirms that a range of GA induction techniques are used by anesthesiologists who may be unaware of the elevated regurgitation and aspiration risk in patients with a history of successful esophagectomy. This is especially relevant in the absence of patient-reported GI symptoms. Based on the available information, patients with a history of esophagectomy should be considered at high risk for pulmonary aspiration of GI contents and their airway should be rapidly secured with an endotracheal tube when presenting for general anesthesia. Specific recommendations for preoperative NPO management, patient assessment, and the induction process are included in this work and provide guidance for anesthesiologists encountering these patients. The low grade of scientific evidence found in this review is not surprising given its narrow topic and the relatively infrequent presentation of such patients in routine anesthesia practice.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.