Esophageal Foreign Bodies

Anatomy

The esophagus is a muscular tube 20 to 25 cm in length. There are three anatomic areas of narrowing in which FBs are most commonly entrapped: in the upper esophageal sphincter, which consists of the cricopharyngeus muscle; in the midesophagus at the crossover of the aortic arch; and in the lower esophageal sphincter (LES) ( Fig. 39.1 ). The LES is the narrowest point of the esophagus and the entire GI tract.

Epidemiology

There are almost 100,000 cases of FB ingestions reported annually in the United States, and they are the eighth most common cause of calls to poison control centers. Patients with retained esophageal FBs generally fall into one of the following categories: pediatric patients, psychiatric patients, prisoners, and adults who either are edentulous or have underlying esophageal pathology.

Children account for 75% to 85% of esophageal FBs seen in the ED, with the peak incidence occurring at the age of 18 to 48 months. The incidence is equal in boys and girls. Inquisitive children frequently place objects in their mouth and unintentionally swallow them. As a result, children most commonly ingest coins, but they also swallow buttons, marbles, beads, screws, and pins. Unlike adults, children who have entrapped, accidentally swallowed FBs do not normally have underlying esophageal disorders. However, this is not the case in children with esophageal meat or food impaction, and these patients will need further evaluation for underlying esophageal disease, the most common of which is eosinophilic esophagitis (55%), or stricture (27%).

Patients with an anatomic abnormality of the esophagus or a motor disturbance are more prone to FB entrapment. Anatomic abnormalities include strictures, webs, rings, diverticula, and malignancies. Motor disturbances include achalasia, scleroderma, and esophageal spasm. Adults who have dentures or underlying esophageal anatomic or motor abnormalities may accidentally ingest food boluses, chicken bones, fish bones, glass, toothpicks, fruit pits, or pills while in the act of eating.

Prisoners and psychiatric patients ingest a wide variety of objects, some of which may be quite unusual: spoons, razor blades, pins, nails, or practically any other object. Patients who intentionally ingest foreign objects frequently ingest more than one, with an average of 4 to 5.

Complications

Impacted FBs of the esophagus must be removed or dislodged. The time frame under which this mandate must be carried out varies widely and depends on many circumstances. In general, however, the esophagus does not tolerate FBs well or for prolonged periods because it is prone to pressure, edema, necrosis, infection, and eventually perforation. FBs can transit the esophagus in a matter of seconds or minutes or may adhere to the mucosa. Retained objects may become less symptomatic after time, and the clinician must resist the urge to allow esophageal FBs to “pass by themselves” or “dissolve.” Once FBs become stuck in the mucosa, they may become less symptomatic, but they rarely pass on their own. The one exception may be children with coins or other smooth round FBs such as small marbles, especially those lodged at the LES. Approximately one third of these objects may pass spontaneously within 24 hours, and some authorities have advocated an observational approach, although this tends to be more poorly accepted by parents. Button batteries must be immediately removed because of the risk of rapidly occurring esophageal erosion and perforation.

The literature is replete every year with a wide array of case reports and small reports of series of complications that can arise from retained esophageal FBs ( Box 39.1 ), including benign mucosal abrasions, lacerations, esophageal stricture, and necrosis from corrosive agents such as button batteries. Esophageal perforation can lead to life-threatening conditions such as retropharyngeal abscess, mediastinitis, pericarditis, pericardial tamponade, pneumothorax, pneumomediastinum, tracheoesophageal fistula, and vascular injuries, including injuries to the subclavian vein and aorta. Complications are more common when FBs are entrapped for longer than 24 hours and when they are sharp.

Clinical Findings

Esophageal FB impaction is usually an acute condition, particularly in adults who have a clear history of ingestion. Children also commonly remember an ingestion, but some will have a vague history or symptoms. As many as one third of children with proven esophageal FBs are asymptomatic on initial evaluation ; therefore a high index of suspicion is indicated, especially in children who were seen with an object in their mouth that subsequently disappeared. This is particularly true if transient coughing or gagging occurred, even though the actual ingestion was not witnessed. Poor feeding, irritability, fever, stridor, cough, wheezing, and aspiration can all be caused by an underlying esophageal FB in a child, especially a young infant.

Dysphagia is a common initial complaint with esophageal FBs. Drooling is suggestive of high-grade obstruction, and complete inability to handle oral secretions is a sign of total obstruction. Infants with a clandestine esophageal FB can exhibit wheezing or a chronic cough. They may appear to have bronchospasm and may be treated for asthma by a number of clinicians before an FB is suspected. Stridor from an FB can mimic epiglottitis.

The esophagus is well innervated proximally, and patients can typically accurately localize FBs in the oropharynx or upper third of the esophagus. However, scratches or abrasions of the esophagus can create a persistent FB sensation. Upper esophageal FBs often cause gagging or vomiting. In rare cases, an upper esophageal FB can impinge on the trachea, especially in children, and mimic infection by inducing wheezing, stridor, or frank respiratory distress. The lower two thirds of the esophagus is not as well innervated, and FBs in this location typically cause vague symptoms of discomfort, fullness, or nonlocalizing pain. Swallowed coins that lodge in the lower part of the esophagus in children may cause no overt symptoms until feeding is attempted.

The location of retained esophageal FBs is related to age and areas of physiologic narrowing of the esophagus ( Table 39.1 ). Children more typically have objects entrapped in the upper part of the esophagus at the level of the cricopharyngeus muscle, whereas adults more commonly have entrapment at the LES.

Evaluation

The most useful aspect of the evaluation is the history. The time of the ingestion, size and shape of the ingested object, and any current symptoms should be ascertained. Findings on physical examination are frequently normal in patients with esophageal FBs, unless complete obstruction is present. In this case they will be drooling, spitting, and unable to handle oral secretions. Even though a patient may be asymptomatic at initial encounter, transient coughing or gagging should raise the index of suspicion for an esophageal FB. Examination of the oropharynx, neck, respiratory system, cardiac system, and abdomen is essential in the evaluation of potential complications.

After attending to life-threatening conditions such as airway compromise, the goal of ED evaluation is to localize the FB to determine what, if any, interventions need to be undertaken to remove it or assist its transit into the stomach. Once an FB passes into the stomach, it has a greater than 90% likelihood of passing through the entire GI tract without any further problems. Even irregular, and seemingly dangerous FBs will often transit the entire GI tract with relative ease.

Background

Radiographic imaging of a patient with a suspected esophageal FB is a common practice and is particularly useful for detecting radiopaque FBs. Traditionally, an inability to quickly identify the object by physical examination encouraged the use of plain radiography in an attempt to verify and localize the retained FB. However, the limitations of plain radiography require that other diagnostic approaches also be considered.

Indications

Essentially every patient with a suspected esophageal FB warrants radiographic evaluation with some exceptions, including those patients who have an obvious complete obstruction with an inability to manage their secretions. These patients need endoscopy to relieve the obstruction. If there is any question of aspiration, a chest x-ray may be indicated. Patients with a history of sharp, nonradiopaque FBs such as toothpicks will also need urgent endoscopy. If there has been a delay in presentation in the case of a sharp FB and there are any signs of infection such as fever or chest pain, computed tomography (CT) may be indicated to evaluate for perforation or mediastinitis. Finally, those patients with an FB sensation that is visualized in the upper pharynx on exam and removed may forego radiographs. In nonverbal patients, including preschool children and those who are demented or debilitated, maintain a low threshold for screening radiography in cases with a suspicious history. Examples include a child seen with an object in the mouth that “disappeared” or a patient with symptomatology suggestive of an esophageal FB, such as drooling, gagging, or unexplained respiratory symptoms.

Plain Radiographs

Plain radiographs reliably verify and localize radiopaque FBs such as glass and metal of sufficient size, and are indicated as the main method of radiologic evaluation for these objects.

When used, a complete oropharyngeal radiographic series includes the nasopharynx to the lower cervical vertebra in both lateral and anteroposterior views. Optimum-quality radiographs are mandatory. Patients should be positioned upright with the neck extended and the shoulders held low. Use of a soft tissue technique enhances the discrimination of weak radiopaque FBs. Phonation of “eeeee” during radiography prevents motion artifact from swallowing, distends the hypopharynx, and enhances soft tissue landmarks. As previously mentioned, FBs are most frequently entrapped at one of three locations of physiologic narrowing in the esophagus: the cricopharyngeus muscle ( Fig. 39.2 ), the aortic crossover ( Fig. 39.3 ), and the LES ( Fig. 39.4 ). Objects that become lodged in the middle portion of the esophagus most likely represent esophageal pathology, such as stricture related to tumor or eosinophilic esophagitis.

Posteroanterior (PA) and lateral views of the chest are used to evaluate the remainder of the esophagus. Both projections are indicated to identify multiple objects and FBs visible in only one plane. Esophageal FBs typically lie in the vertical plane and are differentiated from airway bodies or calcifications by their location posterior to the tracheal air column on lateral radiographs. As a rule, flat objects such as coins perch in the coronal plane in the esophagus and in the sagittal orientation in the trachea. Intraesophageal air and air-fluid levels represent indirect evidence of esophageal obstruction, and may aid in the verification of radiopaque FBs. Soft tissue swelling, extraluminal air, and aspiration pneumonitis can occasionally help identify complicated impactions radiographically.

In children, a film from the nasopharynx to the anus is frequently obtained to allow visualization of the entire nasopharynx, throat, and esophagus, in addition to the abdomen in case the FB has passed into the stomach or beyond. Radiation exposure can be minimized if adult-sized radiograph cassettes are used. Swallowed coins or other FBs may become lodged in the nasopharynx, usually after gagging or vomiting, and could be missed if this area is not included on the radiograph. In adults, if neck or chest films are negative, abdominal films are sometimes obtained for reassurance of the presence of the FB in the stomach.

Unfortunately, many ingested FBs are nonopaque, including nonbony food, plastic, wood, and aluminum. The visibility of low-opacity FBs can be improved using low-dose radiography, especially in neck films. Some pull tabs from beer cans may be seen if oriented in the coronal plane. A metal detector has been reported to help localize radiolucent aluminum pull tabs. Calcification of fish and chicken bones is often incomplete, but cooking alters the structure of bones and makes them radiolucent on plain films. The degree of bony calcification varies with the fish species and between different samples of the same species, thus preventing useful guidelines. For these reasons, plain films provide little substantive evidence in the majority of cases of fish or chicken bone dysphagia. Plain radiographs detect only 25% to 55% of endoscopically proven bones, and carry a high rate of false-negative and false-positive interpretations. Because of the lack of diagnostic value for detecting bones, many clinicians do not routinely order plain radiographs and instead initially opt for CT in cases in which radiographic evaluation is required.

Plain radiography of the neck is limited by the radiographic properties of ingested materials and the complicated anatomy of the upper aerodigestive tract. The base of the tongue, palatine and lingual tonsils, vallecula, and piriform recesses are common regions for entrapment of small, sharp objects, and deserve careful interpretive attention ( Fig. 39.5 ). Superimposition of the mandible contributes to suboptimal resolution of this region on lateral neck films. Calcified airway cartilage often masquerades as FBs and contributes to false-positive rates as high as 25%. Normal ossification of airway cartilage begins in the third decade and progresses with age. The typical curvilinear contour and well-defined margins of bony FB fragments may help distinguish them from normal laryngeal calcifications. The orientation of bony FBs is variable. The C6 vertebra approximates the level of the cricopharyngeus, a common site of FB impaction. Increased prevertebral soft tissue width, air within the cervical esophagus, and soft tissue emphysema are rare indirect findings that may help identify radiolucent objects.

Computed Tomography

CT of the neck and mediastinum is an easy, rapid, cost-effective, and accurate noninvasive means of detecting or excluding esophageal FBs ( Fig. 39.6 ) and has garnered support in the setting of suspected FB entrapment. CT further excels at localization and characterization of the impacted FB and identification of associated complications, such as perforation. CT has a sensitivity and accuracy of 97% and 98%, respectively, for the diagnosis of esophageal FBs compared to a sensitivity and accuracy of 47% and 52%, respectively, for plain films. CT provides improved diagnostic utility for fish bone FBs over plain radiography when obtained either with or without barium enhancement. The sensitivity of CT can be improved with 3D reformations. These allow better visualization of the FB and any esophageal injury. Use of CT in patients in whom clinical suspicion for a retained FB is high has the potential to reduce the number of unnecessary endoscopies.

The use of intravenous (IV) contrast with CT is not indicated, unless there is concern for an inflammatory process (abscess, peritonitis, fistula formation), or vascular injury (aortic or caval perforation). Use of oral contrast with CT in the setting of suspected FB remains controversial. If used, it can impair visualization on endoscopy. Use of oral contrast is generally reserved for those rare cases where either esophageal perforation is suspected (water-soluble media should be used), or a nonradiopaque FB is suspected and the radiologist wants a small amount of contrast to “outline” the potential FB.

Contrast-Enhanced Esophagograms

Procedure

Esophagograms couple voluntary ingestion of an enteric contrast agent (Gastrografin [Bracco Diagnostics Inc., Monroe Township, NJ] or barium) and plain radiography. Immediately after ingestion, erect and horizontal radiographs are performed at right-angle projections (PA and lateral, or right and left anterior oblique). In addition to anatomic abnormalities, radiolucent FBs may be identified by contrast delineation or filling defects within the contrast column ( Fig. 39.7 ).

The initial choice of contrast agent is debated and should be individualized according to the threat of aspiration and perforation. Water-soluble Gastrografin is indicated first in most cases of suspected perforation because it causes less mediastinal inflammation when extravasated; however, it can give rise to severe chemical pneumonitis if aspirated, and is relatively contraindicated in patients with complete esophageal obstruction. Patients without evidence of complete esophageal obstruction are instructed to swallow progressively larger aliquots of contrast agent up to approximately 50 mL. If these films are normal, the procedure is repeated with half-strength and then full-strength barium to delineate small esophageal injuries. Note that water-soluble contrast material (Gastrografin) causes more pulmonary reaction than barium does when inadvertently aspirated, and should be used in small aliquots if aspiration or complete esophageal obstruction is a concern. Barium interferes with endoscopy and should not be used when endoscopy is anticipated.

Follow-up Radiographs

Often FBs are visualized in the stomach, or procedures performed in the ED (described later) result in the FB moving into the stomach. A variety of organizations have proposed recommendations for how to manage FBs in the stomach and intestines ( Table 39.2 ).

Conclusions

Diagnostic radiography for esophageal FBs is indicated in almost all cases with the exception of those with a complete obstruction where there is no concern for aspiration. Plain radiographs clearly assist the clinician in several situations: (1) screening of children, adults with dementia, and nonverbal patients with a history or symptoms suspicious for purposeful or inadvertent FB ingestion that can be assumed to be radiopaque; and (2) localization of known radiopaque FBs to clarify the necessity for and means of FB extraction. Conversely, attempts to verify radiolucent FBs, including bones, by plain radiography are often misleading. CT is more accurate and sensitive than plain films in diagnosing FBs, and in most cases, contrast is not needed.

Background

As the LES is the narrowest portion of the entire GI tract, most FBs that reach the stomach eventually move through the GI tract without further problems. Because a large number of entrapped esophageal FBs are lodged at the LES, especially in adults, several therapeutic maneuvers have been proposed to assist transit into the stomach, including pharmacologic relaxation of the LES. In theory, agents that promote smooth muscle relaxation should improve mobility through the LES. Although many clinicians use pharmacologic adjuncts for all esophageal FBs, objects lodged at the LES will probably benefit most from such interventions. Nonspecific pain relief, anxiolysis, vomiting, and spontaneous passage over time may account for the success attributed to many pharmacologic manipulations of esophageal FBs as there is little scientific evidence that any of these agents offer any benefit over placebo.

Indications and Contraindications

The indication for pharmacologic relaxation of the LES is the presence of a smooth or blunt FB such as a coin or food bolus. Angulated, abrasive, or sharp FBs should not be treated with pharmacologic modalities but instead should be removed by esophagoscopy. Analgesics and sedatives are routinely indicated if pain is present or the patient is excessively anxious.