Case report: role of multimodality imaging in diagnostics and follow-up of a giant intramyocardial dissecting haematoma

Abstract

Background

Intramyocardial dissecting haematoma (IDH) is a rare life-threatening event usually complicating an acute myocardial infarction. Poor data exist about diagnosis, management, and outcome.

Case summary

We reported a case of giant IDH managed conservatively, thanks to stable clinical status and haemodynamics, which evolved towards resorption. Echocardiography and second-level imaging tools, like computed tomography scan and cardiac magnetic resonance, helped in differential diagnosis and studying the haematoma evolution over time, especially providing data about dimension, connection with the left ventricular cavity, consolidation, and resorption. The course is influenced by many factors including localization, edge integrity, and antithrombotic therapy on board. In this case, IDH resorption was observed despite the huge size and anticoagulant therapy on board, used for secondary cardioembolic protection, under close imaging follow-up.

Discussion

Intramyocardial dissecting haematoma management depends on clinical stability, and imaging provides key data about diagnosis and evolution.

Introduction

Intramyocardial dissecting haematoma (IDH) consists in the penetration of blood within the myocardium creating a dissecting space, which follows the muscle planes tangential to the cardiac cavity, limited by epicardium on the outer side and endocardium on the inner.¹ It differs from pseudoaneurysm, which has a transmural path barely contained by the pericardium. Intramyocardial dissecting haematoma is more likely found as complication of myocardial infarction (MI), often in patient with single-vessel disease and absent preconditioning or collateral circulation, where the necrotic tissue is prone to haemorrhagic damage even after reperfusion.² Less frequently, it may complicate chest trauma or percutaneous procedures, e.g. coronary angioplasty. Interstitial tissue frailty can be the substrate, and microvessel rupture, the triggering event, after which intracavitary pressure forces blood propagation and widening of the dissection plane.¹

Summary figure

Case presentation

Hospital admission and urgent imaging

A 74-year-old woman with a family history of ischaemic heart disease arrived at the emergency department complaining of fatigue and dry cough in the last 10 days. Clinical evaluation was unremarkable except for basal pulmonary crackles. The haemodynamic status fell within the definition of Stage B (beginning) of the shock Society for Cardiovascular Angiography and Interventions classification.³ Interstitial lung congestion was evident at chest X-ray. The 12-lead electrocardiogram demonstrated sinus rhythm with Q waves and persistent ST elevation in leads V1–4. She was admitted to the cardiology unit with diagnosis of subacute anterior MI complicated by heart failure. Fast transthoracic echocardiography (TTE) showed akinesia of the whole left ventricular (LV) apex, a big mass occupying two-thirds of its volume, severe functional mitral regurgitation (MR), and mild pericardial effusion. Differential diagnosis concerned organized intraventricular thrombus vs. IDH (Table 1). The urgent thoracic computed tomography (CT) angiography demonstrated a voluminous IDH extended from mid-portion to LV apex, with hyperdensity compatible with unclotted blood [30 Hounsfield units (HU) in unenhanced acquisition], thinned myocardial walls without rupture, diffuse mild pericardial effusion, and no signs of aortic dissection (Figure 1). Blood tests confirmed the subacute setting: white blood cell count was 11.46 10⁹/L (URL 10.00), haemoglobin 12.4 g/dL (range 12.0–15.5), high-sensitivity troponin-T 45–73 ng/L (female URL 10), creatinine 1.33 mg/dL (URL 1.0), spontaneous international normalized ratio 1.65 (normal value 1.0), bilirubin 2.55 mg/dL (URL 1.2), C-reactive protein 153.7 mg/L (URL 5.0), and N-terminal pro–B-type natriuretic peptide (NT-proBNP) 14 583 ng/L. Coronary angiography showed proximal left anterior descending artery occlusion, with faint collateral circulation. Given the risk of IDH expansions or free-wall rupture, no antithrombotic therapy was administered, but diuretics, angiotensin-converting enzyme inhibitors, statin, and mineral corticoid antagonists were started. Considering a EuroSCORE II of 20% in the case of combined LV repair and coronary bypass grafting procedure or 13% for isolated repair, the heart team deemed the patient at very high operative risk and excluded an indication for urgent surgery. Two further elements supported this decision: the relative haemodynamic stability and the huge extent of the IDH, which would have required resection and repair of a very large area of ​​the LV.

Early in-depth investigations

On the second day, contrast TTE confirmed akinesia of LV apex and mid-segments, LV ejection fraction (LVEF) 20%, and increased end-diastolic diameter. The IDH, largely obliterating the LV, had heterogeneous structure and no connection with the remaining cavity and was delimited by a hyperechogenic continuous fluctuating endocardial layer (Figure 2A–C). Thanks to clinical stability, cardiac magnetic resonance imaging (CMRI) was performed. Cine steady-state free precession sequences showed severe LV dilation, LVEF 20%, apical and mid-segments akinesia, a giant low-signal mass surrounded by a continuous endocardial layer, and thinned epicardial infarcted area (Figure 3A and B). At T1-weighted fat sat images, the mass was hyperintense compared with blood and heterogeneous (Figure 3C). Short-time inversion recovery T2-weighted images did not show a high wall signal, making the presence of myocardial oedema unlikely, whereas the mass appeared hyperintense and heterogeneous (Figure 3D–F). At early gadolinium enhancement and at late gadolinium enhancement (LGE) sequences, the mass appeared diffusely hypointense. Moreover, late acquisitions showed transmural LGE in all mid-apical segments, namely Numbers 7–17 of the standard American Heart Association guidelines bull’s-eye plot used for LV assessment (Figure 4A–E). The clinical course was complicated by new-onset aphasia and right hemiplegia. After neurological consultation, thrombolysis was excluded due to the high haemorrhagic risk. Baseline, 24-h brain CT scans, and angio-CT of the extra- and intracranial vessels were negative. Given the quick and complete recovery, the diagnosis was transient ischaemic attack with possible cardioembolic origin, likely favoured by slow flow inside the small remaining LV cavity. Therefore, a reduced dose of subcutaneous low-molecular-weight heparin (LMWH) was started, with daily echocardiographic monitoring. Gradually, LMWH dose was increased to full anticoagulation and, before discharge, imbricated with vitamin K antagonist (VKA).

Fifteen-day transition

At 15 days, the IDH ultrasound study showed diffuse hyperechoic appearance dotted with anechoic lagoons, an evolutive pattern described in previous reports (Figure 2D–F).³ The patient status progressively improved. A second heart team evaluation excluded surgery.

Forty-day transition

Discharge TTE, 40 days after admission, confirmed a dilated ischaemic cardiomyopathy, LVEF 25%, and akinesia of LV apical and mid-segments. The IDH was largely resorbed, limited to the apex, without vacuolizations (Figure 2G–I). Therapy included oral anticoagulation, furosemide, beta-blocker, ACE inhibitor, mineral corticoid antagonist, and dapagliflozin.

Sixty-five-day transition

At 1-month follow-up, 65 days from diagnosis, she reported slight clinical improvement. Transthoracic echocardiography demonstrated no evidence of IDH but mid-to-apical LV aneurysm, LVEF 30%, and moderate MR (Figure 2J–L). Despite reduced blood pressure, a very low dose of sacubitril/valsartan was started. Re-evaluation was planned after 1 month, to consider primary arrhythmic prophylaxis.

Discussion

We reported the short-term favourable outcome of a giant IDH complicating a silent anterior MI. The certain diagnosis was done by colour Doppler TTE that remains the first diagnostic tool thanks to easy availability and, by angio-CT scan, feasible in the urgent setting, ruling out intraventricular thrombosis and LV free-wall rupture.⁴ Finally, CMRI confirmed the heterogeneous and haemorrhagic nature of the mass, intact borders, and absent viability of surrounding infarcted tissues.⁴ Serial multiple TTE studies showed the evolutive ultrasound pattern, somewhat described in other reports.⁵ Indeed, haematoma appearance may vary over time, since blood extravasation, thrombotic organization, expansion, or resorption may follow one another depending on tissue laxity, transmural wall tension, and antithrombotic therapy (Supplementary material online).⁶ During the long hospitalization, we could appreciate a change in IDH echo-structure at different time points (Figure 2). In the early phase, the IDH had a mixed echogenic appearance, compatible with recent bleeding and initial organization. Heterogeneity was the main characteristic, possibly related to the different state of its content, being organized thrombus hyperechoic and more recent bleed less echogenic (Figure 2A–C).⁵ Haematoma evolution can be further influenced by the reparative process of the infarcted area, characterized by interstitial oedema and leucocyte infiltration.⁷ Two weeks later, the structure appeared predominantly echo-dense, scattered with anechoic lagoons. Likely, the lesion had reduced oedema, and organized thrombus largely occupied the area. Anechoic lagoons might be due to limited and more recent bleeds, possibly favoured by ongoing anticoagulant therapy (Figure 2D–F).^5,8 At 40 days, the haematoma was still echo-dense but significantly reduced in size, and vacuolizations were no longer evident. This appearance may represent an advanced phase of resorption of extravasated material (Figure 2G–I). Finally, during outpatient follow-up, no evidence of IDH was found (Figure 2J–L). In our case, anticoagulant therapy was administered after a suspected cerebral embolism, and it may have affected the clinical and echographic course of the haematoma. On one side, anticoagulation may have favoured extravasation of blood within the infarcted area in the presence of microvascular injury. On the other, anticoagulant therapy might have contributed to haematoma resorption, along with preventing additional embolic events. Although published data have shown that a direct oral anticoagulant may be equally effective and safe compared with a VKA in the setting of LV thrombosis, its use is not approved for this indication.⁹

Conclusions

The favourable outcome of our patient, treated conservatively, was due to the lack of progression towards free-wall rupture, as demonstrated by pericardial effusion disappearance, and towards an overt low-output state, thanks to the progressive IDH resorption. This was possible despite ongoing anticoagulation and the initial giant size of the haematoma.^4,7

Lead author biography

Dr Gianni Dall’Ara graduated in Medicine and Surgery in 2007 and achieved his PhD degree in 2018 at the University of Bologna, Italy, where he currently has a position of assistant professor. His main areas of interest are interventional cardiology and acute cardiovascular care.

Supplementary material

Supplementary material is available at European Heart Journal – Case Reports online.

Consent: The patient provided written informed consent for data collection and publication. No personal identifying information is provided in this manuscript.

Funding: This work was supported by Fondazione Cardiologica M. Zito Sacco.

Data availability

Data underlying this article will be shared on reasonable request to the corresponding author.

References

Author notes