Article Text

Download PDFPDF

Aorto caval fistula—the “bursting heart syndrome”
  1. S Leigh-Smith,
  2. R C Smith
  1. Surgical Department, Falkirk and District Royal Infirmary NHS Trust, Falkirk
  1. Correspondence to: Mr Leigh-Smith, Emergency Department, Royal Infirmary of Edinburgh, Edinburgh EH3 9YW (e-mail: simon{at}


Aorto caval fistula is one of the less well recognised complications of abdominal aortic aneurysm seen in accident and emergency departments. It presents in a number of different ways the commonest of which is high output congestive cardiac failure with warm peripheries. Initial diagnosis is based on the index of suspicion of the clinician. However, early diagnosis by the emergency physician and early surgery can markedly improve the patients prognosis.

  • aorta caval fistula

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Case report

A 79 year old man was seen in the accident and emergency (A&E) department five hours after a brief syncopal episode. He described a sudden onset of fast palpitations, dizziness, nausea, one episode of vomiting, sweating and a “feeling as though his heart was going to burst.” These all subsided spontaneously within five minutes. He was a smoker and was soon due to undergo an elective graft replacement for a 9 cm abdominal aortic aneurysm (AAA). His main complaint on presentation was lethargy since his earlier “funny turn.”

On examination he was pale but warm and well perfused. His jugular venous pressure (JVP) was increased, he had a pansystolic flow murmur, a tachycardia of 105 and BP 105/55. Abdominal examination revealed a large but non-tender pulsatile mass and a fullness in his right loin. Abnormal investigations included a mild hypoxia on air, an ischaemic ECG with left axis deviation and a mild neutrophilia.

An initial differential diagnosis of (a) arrythmia, (b) myocardial infarction, (c) leaking abdominal aneurysm was made. Blood was cross matched, a myocardial infarction screen was started, he was put on telemetry and a fluid challenge was performed.

Over the next few hours he became oliguric and shocked with no further evidence of myocardial infarction or arrythmia. He was therefore taken to theatre for repair of a suspected leaking aneurysm. An aorto caval fistula was surprisingly discovered and successfully repaired along with insertion of an aorto bi-femoral graft. He did well postoperation after transient worsening of renal function and bilateral leg oedema that spontaneously improved.


Spontaneous aorto caval fistula is one of the less well recognised complications of an atherosclerotic AAA and yet is more common (10% of ruptures) than aorto duodenal fistula (2% of ruptures), which may be an easier diagnostic challenge.1 Although described as rare in most references, its quoted incidence is very variable from as low as 0.22%2 to as much as 10%1, 3 of all AAAs. Spontaneous rupture of an atherosclerotic plaque in an existing AAA is the commonest cause (80%) with trauma (15%) and iatrogenic after lumbar disc surgery (5%) less common causes.4The incidence of all AAAs is increasing and therefore so will the incidence of its complications.5

The prognosis of this condition is very dependent on how early it is diagnosed and particularly if this is done before operation. Although survival up to two months without surgery has been reported7 it is generally accepted that prompt surgery improves survival.6 Diagnosis and surgery before development of shock can double the chances of survival from 25% to 50%.8 Diagnosis before surgery is desirable as it allows preparation by the surgeon for appropriate operative techniques,6 care by the surgeon not to dislodge debris into the inferior vena cava causing a pulmonary embolism,5 insertion of a pulmonary artery catheter for the difficult haemodynamic control intraoperatively,6, 9 and avoidance of early fluid overload worsening the cardiac failure.1 In one series mortality was 15% if diagnosis was made before surgery in contrast with 100% mortality if it was not.10

Early diagnosis is hence the key to improving patient outcome in this condition and that is dependent on the physicians awareness of it.11 The problem is the different ways in which it can present. In fact three authors describe “classical” presentations all of which vary slightly.1, 4, 5 Pain is even described as being absent,12 or always present.8

Symptoms and signs may be attributable to the high venous return and arterial insufficiency to other structures caused by the fistula itself or attributable to associated intraperitoneal or retroperitoneal rupture. This sudden increase in venous return to the heart along with decreased peripheral vascular resistance can lead to cardiac arrest, but more commonly leads to an acute compensatory phase.1

Review of the medical literature shows the commonest symptoms and signs to be1, 4, 5, 8, 1215:

  1. High output cardiac failure (dyspnoea, increased JVP, pulmonary oedema and widened pulse pressure)

  2. Abdominal bruit and thrill

  3. Palpable abdominal aneurysm

  4. Oliguria

  5. Consequences of regional venous hypertension (leg oedema with/without cyanosis, haematuria and rectal bleeding)

  6. Variable symptoms and signs (shock, abdominal pain, chest pain, low back pain, scrotal oedema, tenesmus, priapism, and poor peripheral pulses)

Once the diagnosis is suspected there are various options open to confirm it providing the patient is stable. Central venous blood may have high oxygen saturations.6 Doppler ultrasound in A&E will show the AAA and may even demonstrate the fistula.1, 16 Angiography is considered the gold standard but only if there is no renal impairment or shock.4 Computed tomography, magnetic resonance imaging and radioisotope studies have all been used to make the diagnosis.4, 13, 1619 Local resources and expertise are probably the most important factors in choice of diagnostic modality.12


The condition may not be as rare as expected. Considering it as a diagnosis in “arteriopaths” with acute onset of cardiac failure and listening for a bruit in all patients with a ruptured AAA20 may increase the diagnosis rate in the A&E department.

Our case with his description of a sensation of his “heart bursting” also shows that the patient can be describing exactly what is happening to him and we should bear this in mind when using structured closed questions in our history taking!



Dr S Leigh-Smith was responsible for the initial assessment, investigation and resuscitation of the patient and conducted the background literature review. Mr R C Smith was responsible for the subsequent assessment, operative procedure and subsequent management of the patient. Both authors contributed to the text of the article. Mr R C Smith is the guarantor.



  • Funding: none.

  • Conflicts of interest: none.