Thrombolysis for acute myocardial infarction always has a risk of causing haemorrhage.¹ Both gastro-oesophageal reflux disease and ischaemic heart disease are common,² and can have some overlap in their symptomatology. We present a patient who suffered an unusual complication of thrombolytic therapy.

CASE REPORT

An 80 year old man presented to his local emergency department with an acute inferior myocardial infarction. His only history of note was quadruple coronary artery bypass grafting at the age of 70, from which he had made a good recovery. He was assessed as appropriate for thrombolysis; tenectplase 37.5 mg, and aspirin 300 mg were given, and systemic heparinisation started. His electrocardiographic changes of myocardial infarction improved and reperfusion was thought to have occurred. An echocardiogram at 24 hours after admission reported moderate systolic dysfunction with infero-posterior akinesis, consistent with recent infarction. His troponin I concentrations were 0.64 μg/l, 14.1 μg/l, and 21.6 μg/l (normal range 0–0.1 μg/l) at 6, 12, and 24 hours after admission respectively. He was given regular β blockers, aspirin, and statin therapy, and remained well for the next 36 hours. He then complained of severe, sudden onset chest pain, which was associated with a small amount of haemoptysis and dysphagia. The chest pain was unrelieved by nitrates, and he complained of increasing difficulty with swallowing. This symptom was initially dismissed and thought to be a manifestation of the distress caused by his severe chest pain. His electrocardiogram at this time showed non-specific ST segment changes, which were considered suggestive of myocardial ischaemia. He was considered again for thrombolysis as it was felt that the most probable diagnosis was re-infarction. However, because of the non-specific ST segment changes he fortuitously underwent another trans-thoracic echocardiogram. This showed external compression of his left atrium, but no new wall motion abnormalities. A contrast computed tomogram of his thorax showed a large submucosal oesophageal haematoma (fig 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

 Contrast computed tomogram of the thorax showing a large submucosal oesophageal haematoma.

Thrombolysis was deferred, heparinisation stopped, and coagulation normalised with protamine sulphate and fresh frozen plasma.

He became increasingly agitated and confused and required intubation and ventilation. He was transferred to our hospital for further treatment.

We elected to keep the patient sedated, intubated, and ventilated and β block was continued intravenously. An upper gastrointestinal surgical opinion was obtained and it was decided to treat him expectantly.

He was kept sedated and ventilated for a further 48 hours, to allow good conditions for myocardial oxygenation and to reduce the likelihood of further oesophageal bleeding from retching or vomiting. He was extubated without incident at 72 hours after admission. He was initially parenterally fed and then permitted oral feed on day six of his admission to our hospital, when he was able to swallow his own saliva. Serial CT scans of the submucosal haemorrhage showed gradual resolution over the course of two weeks. He unfortunately suffered a cardiac arrest secondary to ventricular fibrillation on day 17 of his admission while awaiting coronary angiography. Cardiopulmonary resuscitation was unsuccessful.

DISCUSSION

Thrombolytic therapy has become an important advance in the treatment of acute myocardial infarction.³ There is evidence that the sooner that reperfusion is achieved the better the outcome is for the patient, and so it has become standard practice to try to administer thrombolysis in eligible patients as soon as possible.⁴

The clinical questions that are asked before thrombolysis are chosen to ascertain the risk of catastrophic haemorrhage. Major bleeding from thrombolytic treatment may occur in up to 4% of patients undergoing thrombolysis for acute myocardial infarction.⁵ The incidence of bleeding in the 96 hour period after thrombolysis for acute myocardial infarction with either streptokinase or recombinant tissue plasminogen activator has been studied. Bleeding occurred in 3.8% of thrombolysed patients overall. The leading types of bleeding of all patients affected were: perivascular access site (18.4%), gastrointestinal (6.4%t), skin/soft tissue/muscle (5.0%), urinary (3.4%), pulmonary (2.2%), systemic (1.9%), and oral (1.4%).⁶

The recombinant tissue factor plasminogen activator, tenecteplase was given to our patient. It is reported to have fewer major bleeding complications when compared with alteplase, and is simpler to administer as it is given as a once only bolus dose.⁷ For these reasons it has been promoted as the ideal agent for prehospital thrombolysis.⁸ The half life of tenecteplase is 90 to 120 minutes, so it is not clear as to whether this was the cause of his oesophageal haematoma, or whether it was attributable to the heparinisation and aspirin therapy.

It is also well described that the symptoms of myocardial infarction can be mimicked by gastro-oesophageal reflux disease,⁹ and indeed the two conditions can often coexist within the same patient, in part because smoking is strongly associated with both conditions.^10,11 What is perhaps not as well recognised is that other oesophageal abnormalities, in particular oesophageal rupture, can also present in a similar way to myocardial infarction.¹² The unusual development of a submucosal haematoma in our patient did lead to similar diagnostic confusion, particularly in the context of a recent acute myocardial infarction. It was fortunate that our patient had an echocardiogram, which gave an important clue to the cause of his dysphagia, for had further thrombolysis been given at this time, the immediate outcome would have been disastrous. It pays to listen to the patient, especially if they are complaining of unusual symptoms in the face of recent thrombolytic therapy.