Dysphagia aortica

Background Dysphagia aortica is an umbrella term to describe swallowing obstruction from external aortic compression secondary to a dilated, tortuous, or aneurysmal aorta. We performed a systematic literature review to clarify clinical features and outcomes of patients with dysphagia aortica. Materials and methods We searched PubMed, EMBASE, Web of Science, and the Cochrane Library. The terms “aortic dysphagia,” “dysphagia aortica,” “dysphagia AND aortic aneurysm” were matched. We also queried the prospectively updated database of our esophageal center to identify patients with aortic dysphagia referred for diagnosis and treatment over the past two decades. Results A total of 57 studies including 69 patients diagnosed with dysphagia aortica were identified, and one patient from our center was added to the database. The mean age was 72 years (range 22–98), and the male to female ratio 1.1:1. Of these 70 patients, the majority (n = 63, 90%) had an aortic aneurysm, pseudoaneurysm, or dissection. Overall, 37 (53%) patients received an operative treatment (81.1% a vascular procedure, 13.5% a digestive tract procedure, 5.4% both procedures). Thoracic endovascular aortic repair (TEVAR) accounted for 60% of all vascular procedures. The postoperative mortality rate was 21.2% (n = 7/33). The mortality rate among patients treated conservatively was 55% (n = 11/20). Twenty-six (45.6%) studies were deemed at a high risk of bias. Conclusion Dysphagia aortica is a rare clinical entity with high morbidity and mortality rates and no standardized management. Early recognition of dysphagia and a high suspicion of aortoesophageal fistula may be lifesaving in this patient population.


Introduction
Dysphagia is a common symptom reported by 10-33% of elderly individuals in the community and nursing home settings [1,2], although the true prevalence is likely underestimated because many patients adapt through behavioral changes [3]. The most frequent causes are neurogenic, mechanical obstruction, primary motility disorder, or external compression. The term dysphagia aortica was first introduced by Pape [4] in 1932 to describe dysphagia caused by external aortic compression from an aneurysmal, dilated, or tortuous aorta [5]. In 1997, Wilkinson wrote, "The condition of dysphagia aortica is reminiscent of the Churchillian paraphrase-a riddle wrapped in a mys-228 Dysphagia aortica K review tery inside an enigma" [6]. Dysphagia aortica is rarely mentioned in standard gastroenterological and surgical textbooks and has received little attention in the literature. Dysphagia arises when the aorta pushes the esophagus anterolaterally and against the crural diaphragm. Primary aortoesophageal fistula (AEF) is the most feared complication [7], typically in the setting of untreated thoracic aortic aneurysm (TAA) that occurs in 5-10 per 100,000 person years [8]. This may be asymptomatic and diagnosed incidentally, or it may present with symptoms due to mediastinal compression or with dissection or rupture in the worst-case scenario. Secondary AEF can occur after surgical or endovascular repair of thoracic aortic aneurysms. The typical presentation of AEF was first described by Chiari [9] as a triad of chest pain, sentinel hematemesis, and final massive hemorrhage with exsanguination after a symptom-free interval.
To date, several single case reports of aortic dysphagia have been reported, the majority in women over 70 years old with short stature, hypertension, and kyphoscoliosis [5], often in association with left ventricular enlargement and congestive heart failure [7]. The aim of the present study was to perform a literature review on dysphagia aortica, to add a case recently seen at our institution, and to highlight the diagnostic features and outcomes of this rare syndrome.

Materials and methods
A systematic literature review was conducted to identify patients with dysphagia aortica reported from 01 January 1997 to 31 December 2020 using the PubMed, EMBASE, Web of Science, and the Cochrane Library databases. The search was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement [10]. The following MeSH terms were used: "dysphagia AND aortic aneurysm," "dysphagia aortica," and "aortic dysphagia." Two independent investigators (SG and PM) performed the literature search to identify all English-written reports. The full text of the selected studies was assessed by one investigator (SG) and classified as relevant, not relevant, or unclear. The reference lists of eligible studies were manually searched to identify additional studies. The methodological quality of the studies was assessed according to Murad et al. [11], based on a global evaluation of the most critical factors that increase the risk of bias in the specific clinical context. Disagreements at either stage were solved by discussion and arbitrated by a senior author (LB).
Data extracted included first author name, country, year of publication, number of patients included in the report, age, sex, symptoms at presentation, diagnostic methods, imaging findings, characteristics of the aneurysm, type of treatment, and short-and longterm outcomes. The prospectively updated database of our tertiary care esophageal center was also queried to identify all patients with dysphagia as a predominant symptom referred for consultation between 2002 and 2021.

Literature review
The search strategy identified 1252 articles (918 from registers and 318 records from databases). After duplicates were removed, 725 records were screened. Two reviewers independently screened the titles and abstracts of all papers, leading to exclusion of 984 records. A total of 57 studies were eligible for analysis (Fig. 1). There was a total of 70 patients, 33 women and 37 men, with a median age of 72 years (range 22-98). Dysphagia was associated with aortic aneurysm (n = 53), aortic dissection (n = 7), tortuous aorta (n = 5), or aortic pseudoaneurysm (n = 3). The main patient characteristics are summarized in Table 1. All patients complained of intermittent or chronic dysphagia associated with weight loss in 32.9% of cases, chest pain in 18.6%, and dyspnea in 15.7%. About half of the patients (n = 33, 47.1%) were considered unfit for any endoscopic or surgical approach due to elderly age and multiple comorbidities, and were mainly treated conservatively with antihypertensive therapy and a modified oral diet or through a feeding tube.
The majority (53%) of patients underwent some form of vascular, digestive tract, or combined endoscopic or surgical procedure (Table 2)  Digestive tract procedures consisted of percutaneous endoscopic gastrostomy (PEG; n = 2), endoscopic esophageal stent (n = 1), Maloney bougie dilation (n = 1), and laparoscopic Heller myotomy and crural myotomy (n = 1). The procedure was successful in 3 patients, 1 patient died, and 1 was lost to follow-up. Combined vascular and digestive procedures consisted of TEVAR and esophageal stent (n = 1) and TEVAR and esophagectomy. The latter was complicated by AEF and sepsis.
A barium swallow study revealed a marked extrinsic compression at the level of the lower third of the esophagus, with a filiform contrast flow and dilatation above. Esophagogastroduodenoscopy confirmed a pulsatile extrinsic compression with luminal narrowing from 38 cm to 42 cm from the dental arch (Fig. 2). A computer tomography (CT) scan performed with oral contrast medium showed distal esophageal compression due to a giant thoracic aortic aneurysm (Fig. 3a). Magnetic resonance angiography (MRA) confirmed a giant aneurysm extending from the ascending aorta to the infrarenal region, with signs of intravascular thrombosis and perivascular reaction. The diameter of the aorta was 51 × 57 mm in the ascending thoracic portion, 48 × 46 mm at the aortic arch, 57 × 62 mm in the mid-third of the descending aorta, and 36 × 35 mm below the level of the renal arteries (Fig. 3b). signs of right lung aspiration pneumonia. Antibiotic therapy was started (piperacillin/tazobactam 4.5 g four times a day) and oxygen therapy was increased (Venturi mask 35%, 8 L/min). Due to the increasing need of oxygen therapy, the patient was switched to continuous positive airway pressure (CPAP) therapy with significant improvement of SpO2. Oxygen flow was then progressively reduced and the chest X-ray after 2 weeks of antibiotic therapy revealed complete resolution of the clinical and radiologic pattern (Fig. 4).
The conclusion of a multidisciplinary board meeting including general and vascular surgeons was that the patient was unsuitable for vascular repair given his age, the multiple comorbidities, the low ejection fraction, and the aortic anatomy. The patient refused review to undergo esophageal manometry. Therefore, enteral nutrition through a nasogastric tube or percutaneous endoscopic gastrostomy was recommended, but the patient declined any invasive procedure. He was then discharged on a semi-liquid diet. The patient died at home 4 weeks after hospital discharge due to probable aneurysm rupture.

Discussion
In the present systematic review, dysphagia aortica was associated with thoracic aortic aneurysm in most patients. Interestingly, 21 of 63 (33.3%) patients underwent TEVAR as a single treatment modality or combined with other vascular or digestive tract procedures.
The prevalence of dysphagia aortica is neither well reported nor well studied [7]. It has been suggested that external compression of the esophagus may not represent the major pathophysiological mechanism, but rather an incidental finding. As in dysphagia lusoria, an underlying esophageal motility disorder may be present in some of these patients, particularly in those without evidence of aneurysm [28,29]. It has also been speculated that long-lasting esophageal compression may evolve into esophageal pseudoachalasia, a rare condition accounting for less than 5% of patients with achalasia-like syndrome [30,72,78]. Reported findings at esophageal manometry are low-amplitude propagated peristaltic waves in the proximal esophagus and a localized high-pressure zone at the site of vascular compression. Wilkinson [6] investigated 5 patients complaining of dysphagia to solids associated with a localized high-pressure zone on esophageal manometry. None of the patients had an aneurysm, and videoradiographic assessment with a solid bolus supported the diagnosis of dysphagia aortica.
Considering the rarity of dysphagia aortica, there is no gold standard for diagnosis and therapy. A history of aortic aneurysm or prior aortic graft or TEVAR is key for diagnosis. Radiological and endoscopic imaging provides a high index of suspicion [5]. The diagnostic work-up should include chest X-ray, upper gastrointestinal endoscopy, barium or videofluoroscopic swallowing study, chest CT scan with oral and intravenous contrast, and esophageal manometry. No single diagnostic tool can definitively prove the diagnosis of dysphagia aortica. Radiographic findings may be inconclusive because a dilated and tortuous aorta is frequently seen in elderly patients in the absence of a true aneurysm. Upper gastrointestinal endoscopy has the potential to exclude other possible causes of upper gastrointestinal bleeding, and to detect signs of AEF such as small mucosal erosions, oozing from a pin-hole erosion, ulcer with adherent clot over a pulsatile mass, or graft exposure [31].
The treatment of dysphagia aortica depends on the severity of symptoms and the patient's comorbidi-ties. Most patients with mild and intermittent symptoms may be treated conservatively by a modified diet. Treatment of associated cardiac failure or arterial hypertension may also significantly reduce the burden of symptoms, especially in the case of small aneurysms [6,32]. Percutaneous endoscopic gastrostomy, endoscopic esophageal dilation with bougie [6,33], or an esophageal self-expandable stent have been used sporadically [34]. In the past, surgical procedures proposed to reduce esophageal compression included anterolateral transposition of the esophagus with posterior cruroplasty [33], and mobilization of the distal esophagus from the aortoesophageal decussation area with creation of a posterior pleural sling [35]. More recently, Heller myotomy with division of the right crus of the diaphragm to relieve esophageal compression [33][34][35][36] has also been reported. A proposed management algorithm for dysphagia aortica is shown in Fig. 5.
Aortic aneurysm is a progressive disease, and current practice guidelines recommend treatment of asymptomatic thoracic aortic aneurysms with diameter ≥ 55 mm. Aortoesophageal fistula is a rare complication, representing less than 10% of all aortoenteric fistulas [37]. In 1991, Hollander and Quick [38] reviewed 500 patients with AEF and found that an untreated thoracic aortic aneurysm accounted for 54% of the cases. Since then, secondary AEF have become an increasingly recognized complication of surgical or endovascular repair of thoracic aortic aneurysms.
Direct open surgical aneurysm repair and TEVAR [7,39] are the most common therapeutic options in patients who are fit for these procedures. Over the past 20 years, TEVAR has evolved into an upfront treatment option, showing the potential for preventing further aortic enlargement and ultimate aortic rupture. However, the fact that periaortic hematoma remains untreated may potentially aggravate dysphagia. In addition, further impingement of the esophagus by the hematoma or by an oversized stent, endoleak, stent migration, or a penetrating aortic ulcer can accelerate aortic rupture, AEF, and sepsis [40,41].
In a large multicenter survey [42] including 1138 patients treated with TEVAR over a 10-year period, 2 of 25 (8%) patients with aortoesophageal or aortobronchial fistula presented with dysphagia. The interval between the first reported episode of sentinel hemorrhage and the final diagnosis ranged from 2 h to 6 months. Thirty-day mortality and actuarial 2-year survival were 28% and 54.7%, respectively. A combined endovascular and surgical approach reduced infectious complications and recurrent fatal bleeding. Although the evidence was not strong enough to justify changes in clinical practice, the authors felt that this complication was underestimated even in large trials and questioned the utility of the endovascular approach as the exclusive therapeutic modality.

review
Since the occurrence of secondary AEF complicating TEVAR is unpredictable, it would be paramount to establish the criteria for an early diagnosis. Unfortunately, the association of dysphagia with thoracic aortic aneurysm remains elusive in most reported series, often because the symptom is mild, intermittent, or neglected by both the patient and the physician. Further studies are needed to establish the prevalence of subclinical dysphagia aortica by using specific symptom questionnaires before and after aneurysm repair. Moreover, dysphagia should be rightfully included in the Chiari's triad that originally reported chest pain as the initial manifestation of AEF [9], as well as in clinical practice guidelines [40,79]. Interestingly, the interval between the onset of dysphagia and bleeding from AEF exceeded 1 month in the few reported patients [48]. It is possible that with increasing worldwide adoption of the endovascular procedures, the reported incidence of dysphagia and AEF may increase as well [80]. This may temper the enthusiasm for TEVAR, which should instead represent a bridge to definitive aortic and esophageal reconstruction in patients who are fit for a staged procedure.
This review has several limitations, including reporting bias and the fact that all studies were case reports including up to 5 patients. Therefore, a significant gap in clinical evidence for both diagnostic and therapeutic outcomes remains due to the heterogeneity and the average low methodological quality of the case reports.

Conclusion
Dysphagia aortica is a rare entity with a high mortality rate and no standardized management. Lack of awareness and symptom underestimation may contribute to diagnostic delay. A thorough investigation is recommended to exclude other causes of dysphagia. With modern diagnostic technologies, dysphagia aortica should no longer represent an clinical enigma. One-or two-stage aneurysm repair is feasible in selected patients and may prevent AEF. Surveillance of patients with thoracic aortic aneurysms, early recognition of dysphagia, and a high suspicion of AEF may be lifesaving. Funding Open access funding provided by Università degli Studi di Milano within the CRUI-CARE Agreement.

Declarations
Conflict of interest S. Grimaldi, P. Milito, A. Lovece, E. Asti, F. Secchi, and L. Bonavina declare that they have no competing interests.
Ethical standards All procedures performed in studies involving human participants or on human tissue were in accordance with the ethical standards of the institutional and/or national research committee and with the 1975 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study. Internal review board approval HSD 2021-087.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.