of a portion of the bolus because of weakening of the amplitude of primary peristalsis at the transition zone between the striated and smooth muscle portions of the esophagus, a clinically trivial phenomenon known as proximal escape [136].
Varices
Esophageal varices may be classified as “uphill” or “downhill.” Uphill varices are caused by portal hypertension with increased pressure in the portal venous system transmitted upward via dilated esophageal collaterals to the superior vena cava. In contrast, downhill varices are caused by superior vena cava obstruction with downward flow via dilated esophageal collaterals to the portal venous system and inferior vena cava. Uphill varices are much more common than downhill varices. Whether uphill or downhill, varices are important because of the risk of upper gastrointestinal bleeding.
Figure 6.76 Diffuse esophageal spasm. (A) Prone single‐contrast view shows multiple lumen‐obliterating, nonperistaltic contractions that compartmentalize the esophagus, producing the classic corkscrew appearance associated with diffuse esophageal spasm. Note the presence of a small hiatal hernia (black arrow). (B) Prone single‐contrast view in another patient shows multiple nonperistaltic contractions of mild‐to‐moderate severity (white arrows) with tapered, beak‐like narrowing of the distal esophagus (black arrow) secondary to lower esophageal sphincter dysfunction. This patient has diffuse esophageal spasm with impaired opening of the lower esophageal sphincter. Again note the presence of a small hiatal hernia.
Uphill varices
Uphill esophageal varices usually develop as a result of portal hypertension or other causes of portal venous obstruction. Varices appear on barium studies as serpiginous or tortuous longitudinal filling defects in the distal half of the thoracic esophagus [139] (Figure 6.77). They are best seen on mucosal relief views of the collapsed or partially collapsed esophagus using a high‐density barium suspension to increase mucosal adherence [139]. The differential diagnosis for varices includes submucosally infiltrating esophageal carcinomas (so‐called varicoid carcinomas) and esophagitis with thickened folds caused by submucosal edema and inflammation.
Esophageal varices are characterized on CT by a thickened, lobulated esophageal wall containing tubular structures that enhance markedly after intravenous administration of contrast material [139]. Additional varices may be seen elsewhere in the abdomen at other sites of communication between the portal and systemic venous circulations. Angiography of the celiac or superior mesenteric arteries can be used to confirm the presence of varices in and around the distal esophagus. However, the need for portal venography for presurgical planning of portosystemic shunts has decreased with the widespread use of transjugular intrahepatic portosystemic shunting procedures.
Figure 6.77 Esophageal varices. Single‐contrast view shows multiple large serpiginous defects in the lower third of the esophagus in a patient with portal hypertension and uphill esophageal varices.
Downhill varices
One of the most common causes of downhill varices is bronchogenic carcinoma with mediastinal metastases and superior vena cava obstruction [139]. Additional causes include other primary or metastatic tumors involving the mediastinum, mediastinal irradiation, sclerosing mediastinitis, substernal goiter, and central catheter‐related thrombosis of the superior vena cava. Most patients with downhill varices present clinically with the superior vena cava syndrome.
Downhill varices typically appear as serpiginous longitudinal filling defects, which, unlike uphill esophageal varices, are confined to the upper or mid esophagus [139]. Venography may be performed to confirm the presence of superior vena cava obstruction, and chest radiographs or CT may be performed to determine the underlying cause.
Foreign body impactions
In adults, esophageal foreign body impactions most commonly are caused by inadequately chewed pieces of meat. Most of these foreign bodies pass spontaneously into the stomach, but 10–20% require some form of therapeutic intervention [140]. The risk of perforation is less than 1% during the first 24 h, but this risk increases substantially after 24 h because of ischemia and pressure necrosis at the site of impaction [140]. Affected individuals typically present with acute onset of dysphagia and substernal chest pain.
Figure 6.78 Esophageal food impaction. (A) On the initial barium study, an impacted bolus of meat in the distal esophagus appears as a polypoid defect (arrows) with complete obstruction at this level. (B) A repeat study 10 days after endoscopic removal of the bolus reveals a lower esophageal ring (arrow) as the cause of the impaction.
In the past, barium studies were often performed on patients with suspected esophageal food impaction, and if an impaction was present, the fluoroscopist sometimes attempted to relieve the impaction by administration of an oral effervescent agent, intravenous glucagon, or both. Because endoscopy is a more effective technique for relieving esophageal food impactions, and because residual barium above an impaction can impede endoscopic visualization or retrieval of the impacted food bolus, endoscopy has become the diagnostic and therapeutic test of choice for these patients [141].
Nevertheless, contrast studies are occasionally performed in patients with suspected food impaction to confirm the presence of obstruction, determine its level, and rule out esophageal perforation. An impacted food bolus typically appears as a polypoid defect with an irregular meniscus superiorly [140] (Figure 6.78A). Because of the degree of obstruction, it may be difficult to assess the underlying esophagus at the time of impaction. It is therefore prudent to perform a follow‐up barium study after the impaction has been relieved to determine whether the impaction was caused by a pathologic area of narrowing (Figure 6.78B). The most common causes are Schatzki rings and peptic strictures [140].
Fistulae
Esophageal–airway fistulae most commonly result from direct invasion of the tracheobronchial tree by advanced esophageal carcinoma (Figure 6.79). Such fistulae have been reported in 5–10% of all patients with esophageal cancer, often occurring after treatment with radiation therapy [140]. Other causes of esophageal‐airway fistulae include esophageal instrumentation, trauma, foreign bodies, and surgery. Affected individuals typically present with violent episodes