Post-Bentall procedure ‘pseudo’ pulmonary embolism, cardiac tamponade in disguise: a case report

Abstract

Background

Post-operative cardiac complications require rapid evaluation, which may be hindered by various challenges. Sudden shortness of breath with persisting haemodynamic failure after cardiac procedure is often associated with cases of pulmonary embolism or cardiac tamponade that have contradicting therapies. Anticoagulant therapy is the treatment of choice for pulmonary embolism; however, it could worsen pericardial effusion where bleeding control and clot evacuation are the mainstays of treatment. In this study, we present a case of late cardiac complication due to cardiac tamponade mimicking signs of pulmonary embolism.

Case summary

A 45-year-old male with aortic dissection DeBakey type-II, 7 day post-Bentall procedure, presented with sudden shortness of breath and persistent shock despite therapy. Initial assessment directed towards pulmonary embolism was supported by hallmark imaging signs from X-ray and transthoracic echocardiography evaluation. However, computed tomography scan results were suggestive of cardiac tamponade mainly accumulating at the right side of the heart, compressing the pulmonary artery and vena cava confirmed by transoesophageal echocardiography, thus mimicking the findings of pulmonary embolism. After clot evacuation procedure, the patient improved clinically and was discharged the following week.

Discussion

In this study, we highlight a case of cardiac tamponade with classical findings of pulmonary embolism after an aortic replacement procedure. Physicians should carefully analyse a patient’s clinical history, physical, and supporting examinations to direct and change the course of therapy accordingly, as these two complications have opposite therapy principles and may exacerbate the patient’s condition.

Introduction

Post-operative cardiac complications require rapid assessment that may be hindered by various challenges. Sudden shortness of breath with persisting haemodynamic failure after cardiac procedure is often associated with cases of PE or cardiac tamponade.¹ Anticoagulant therapy is the treatment of choice for PE; however, it could worsen pericardial effusion where bleeding control and clot evacuation are the mainstays of treatment.² Early post-operative complication is more common due to cardiac tamponade; however, we found a rare case of late cardiac complication due to pericardial effusion mimicking signs of PE.³ With contradicting treatment principles, delayed or misdiagnosis may lead to fatality. We present a case of a 45-year-old male who developed cardiac tamponade post-Bentall procedure with some classical findings of PE.

Timeline

Case presentation

A 45-year-old obese male with Stanford A DeBakey type-II aortic dissection, atrioventricular re-entrant tachycardia, and atrial regurgitation was complaining of chest discomfort accompanied by palpitations, nausea, and cold sweats. Patient had a history of controlled hypertension with nebivolol 2 × 5 mg, as well as pericardial effusion and right bundle branch block 8–10 years prior. Previously, the patient was treated with anticoagulants and anti-arrhythmic medications. The patient was an ex-smoker for 16 years before quitting 6 years ago without any history of cardiac disease in the family. His vital signs on arrival were stable with normal jugular venous pressure, widened pulse pressure 99/59 mmHg similar in both arms, and soft diastolic aortic heart murmur. Laboratory findings were elevated for white blood cells, erythrocyte sedimentation rate, D-dimer, and cardiac marker. Plain chest X-ray showed aorta elongation with cardiomegaly and basal infiltrate on the right lung and middle segment of the left lung. Transthoracic echocardiography (TTE) evaluation (Figure 1) revealed aortic root and ascending aorta aneurysm with mild atrial regurgitation secondary to dilated aortic root. Based on the assessments, the patient gave consent for Bentall procedure to replace the dissected aorta.

Ascending and hemiarch replacement Bentall procedure was conducted successfully despite severe adhesion between epicardium and pericardium layer along with resolved complication of sinus bradycardia. The tear started at the level of aortic valve commissure extending to the ascending aorta and middle arch without sign of rupture. Circulatory arrest time was 30 min. Post-intervention, patient was prescribed antibiotics and antifibrinolytic drugs. Anticoagulant medication warfarin for prosthetic valve maintenance was postponed due to concerns of high bleeding risk from aortic dissection and Bentall procedure.

7 days post-operation, the patient unintentionally strained himself while defecating causing sudden shortness of breath, right chest pain, Visual Analogue Scale (VAS) 5–7, and cold sweats with persistent hypotension (81/51 mmHg) and tachypnea (24 breaths/min), despite fluid support and vasoconstrictors. His electrocardiogram reading was normal. Laboratory tests showed insignificant erythrocyte reduction within predicted value (from 13.1 to 11.1 g/dL in 2 days), raised troponin-T marker (341 pg/mL), hypercoagulable state, and acute kidney injury (estimated glomerular filtration rate 65.9 mL/min/1.73 m²). Transthoracic echocardiography examination showed middle right ventricle free wall hypokinesis with apical sparing. Additionally, chest X-ray evaluation revealed right lung oligaemia and abrupt tapering of the pulmonary artery (Figure 2). All these findings were suggestive of PE for which the patient was prescribed anticoagulant heparin.

However, pulmonary computed tomography (CT) angiography (Figure 3) showed an enlarged heart with fluid collection in the right anterolateral-superior pericardial space of density 50–60 Hounsfield unit (HU) pushing the heart laterally while compressing posterior part of superior vena cava, as well as bilateral pulmonary vein and left atrium stenosis due to extraluminal cause indicating cardiac tamponade. Location of the compression at the level of vena cava and pulmonary vein supposedly gave rise to the findings that mimic that of PE. Anticoagulant therapy was immediately ceased, and the patient was scheduled for a surgical pericardiotomy.

Transoesophageal echocardiography (TEE) evaluations were taken for comparison (Figure 4). Right upper pulmonary vein (Figure 4A) and pulmonary artery (Figure 4C) were narrowed due to an external compression compared with after release (Figure 4E and G). The velocity–time integral (VTI) was also shown to be improved after the clot evacuation.

During the operation, active bleeding was found at the distal anastomosis of the ascending aorta with 600 mL of blood clots pressing against the right ventricle, right atrium, superior vena cava, and main pulmonary artery (Figure 5). After the procedure, his haemodynamics returned to normal and was eventually discharged the following week after close monitoring. At 2 month follow-up, the patient was asymptomatic with improving exercise tolerance. Echocardiography revealed no sign of periaortic leakage from the repaired aorta.

Discussion

Imaging modalities have their advantages and limitations when it comes to post-operative cardiac complications evaluation. The obstruction of pulmonary blood flow may be found in both PE and cardiac tamponade depending on the location of blood accumulation. In our case, the site of leakage was predominantly at the right side of the heart, which compressed the pulmonary artery and veins. Hence, classical findings belonging to PE such as Westermark sign, knuckle sign, and McConnel’s sign were also captured.⁴

Anticoagulant prescription post-operation for our patient is postponed considering the bleeding risk data from mechanical aortic valve studies. Study by Daniels et al.⁵ revealed that 3 month cumulative incidence of bleeding event (3.6%) is greater than thromboembolism (0.9%) among 556 patients with mechanical aortic valve undergoing elective procedures. Similarly, Schulman et al.⁶ also studied 117 patients who underwent bridging using low-molecular-weight heparin for invasive procedures and found higher bleeding incidence (11%) without occurrence of thromboembolism episode in either group.⁶ The 2020 ACC/AHA guidelines recommend weighing patient’s bleeding risk to determine bridging therapy for thromboembolism prevention among patients with mechanical heart valve.⁷ Given our patient’s high bleeding risk from aortic dissection with regard to the complexity of this procedure, we decided to postpone prescription of anticoagulants until after discharge, which also made clinical judgment towards bleeding less likely in our patient.

Serial erythrocyte count laboratory evaluation can be used to exclude occurrence of bleeding events. A rapid reduction of erythrocyte count in a short amount of time would support clinical suspicion towards bleeding. However, in our initial evaluation, the patient’s erythrocyte count reduction was insignificant and still ranges within predicted value. Afterward, follow-up erythrocyte count evaluation showed fast reduction (−1.8 g/dL) in just a few hours after the initial evaluation, which by then also supported the diagnosis of obstructive shock due to pericardial effusion compression.

Despite being insensitive in differentiating PE and cardiac tamponade, chest X-ray may help rule out several differential diagnoses. Serial cardiothoracic ratio (CTR) evaluation prior and post-procedural might indicate periaortic leakage. Unfortunately, the patient’s CTR was already elevated >60% during initial admission; thus, changes in the CTR were not as distinct in the event of a pericardial effusion. Mediastinal widening can also be associated with pericardial effusion; however, it is quite common in patients undergoing open heart surgery. Moreover, findings of lung regional oligaemia are indicative of obstruction at the level pulmonary artery, although unable to discern whether the obstruction was of external or internal nature. Westermark sign has a high specificity (92%) but low sensitivity (14%) and positive predictive value (PPV) (38%) for PE.⁸ Given the low PPV for PE, multimodality imaging modalities were employed to increase diagnostic accuracy.

For PE, TTE is suitable as a rapid initial non-invasive imaging to look for signs of increased right ventricular afterload that was apparent in our case. In theory, middle right ventricle free wall akinesis with apical sparing is a distinct finding of acute PE termed as ‘McConnel’s sign’ with high specificity (96%) and PPV (77%), despite low sensitivity (16%).^9–11 However, in our case, localized compression on pulmonary vessels gave rise to similar findings. Besides, TTE is known to be operator-dependent with probe angle being highly influential. Ultrasound wave is incapable of penetrating the sternum bone where the blood accumulated.¹² Furthermore, despite xiphoid view commonly recommended to evaluate pericardial effusion, practically, the presence of drain placement may challenge the evaluation.

Although not routinely used, TEE may aid evaluating post-operative cardiac complications. In this study, TEE was focused to confirm nature of compression being the suspected clot. Reduced pulmonary blood flow was found to be due to an external compression instead of thrombus/emboli as shown from TEE images. Reduction of VTI was normalized after clot evacuation, thus indicating compression was of external nature.

Computed tomography scan has been regarded as the gold standard in imaging of PE and cardiac tamponade.¹³ In our case, despite the visualization of regional lung tissue infiltrates as indirect PE sign, the direct sign of pulmonary flow obstruction by embolus was not captured. Instead, fluid accumulation of 50–60 hounsfield unit (HU) surrounding the heart and ascending aorta were found, indicative of cardiac tamponade.¹³ In cardiac CT measurement, mean HU for cardiac muscle is +38; meanwhile, blood ranges from +50 to +70 which support the pericardial blood accumulation.¹⁴ Vena cava and cardiac chamber compression were also appreciated from CT images supporting cardiac tamponade diagnosis.

Most importantly, PE and cardiac tamponade complication management contradicts one another. For PE, the vast majority of patients recovered well after given anticoagulant therapy.^2,15 Meanwhile, blood clot formation is necessary in preventing excessive bleeding in cardiac tamponade cases. When clotted haemopericardium is expected, surgical drainage is more recommended as applied here.² Hence, extra precaution is needed as the treatments are of opposite principle and may exacerbate the patient’s condition.

Conclusion

This study highlights the importance of being critical towards utilization of various imaging modalities to confirm post-operative cardiac complications. Cardiac tamponade may mimic a case of PE, presenting with multiple imaging hallmark signs depending on the site of blood accumulation. Differentiating cardiac tamponade with PE complication is especially crucial as the nature of management is of contradicting principle and adjustment to therapy must be made swiftly when necessary.

Lead author biography

Ario Soeryo Kuncoro earned both his medical degree and cardiovascular specialization from Universitas Indonesia in the years 2000 and 2006, respectively. He has a keen interest on the utilization of echocardiography for cardiac evaluation and pursued subspecialization on that topic. He is currently a cardiologist consultant in Indonesian national cardiovascular centre Harapan Kita Hospital and Heartology Cardiovascular Center, Jakarta, Indonesia.

Supplementary material

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

Acknowledgements

We would like to acknowledge Semuel Manangka, Anas Alatas, and all clinical team in our institution for their contribution towards the management of our patient.

Slide sets: A fully edited slide set detailing this case and suitable for local presentation is available online as Supplementary data.

Consent: The patient described in this case signed a standard institutional consent form in accordance with COPE guidelines. The consent contains permission to publish their case for scientific purposes. Every effort has been made to anonymize the data being presented.

Funding: None declared.

Data availability

The data that support the findings of this study are available from the authors upon reasonable request.

References

Author notes