| Home | E-Submission | Sitemap | Contact Us |  
Journal of Neurocritical Care > Volume 17(2); 2024 > Article
Mensah, Losso, Muram, Kumar, and Ogilvy: Discrepancies in computed tomography (CT) perfusion and CT angiography imaging following stroke in a patient post-heart transplant on extracorporeal membrane oxygenation: a case report

Abstract

Background

Extracorporeal membrane oxygenation (ECMO) is a critical life-support intervention for severe cardiopulmonary failure. Venoarterial (VA)-ECMO presents unique challenges in neuroradiological assessment due to hemodynamic alterations and the presence of imaging artifacts.

Case Report

This case report describes an adult post-heart transplant patient on peripheral VA-ECMO who developed left-sided hemiparesis, suggestive of cerebrovascular events. While our patient developed multifocal ischemic infarcts, the initial computed tomography perfusion (CTP) imaging indicated hypoperfusion of the entire right hemisphere. Given the discrepancy in the patient’s clinical assessment, which was suggestive of a more circumscribed right hemispheric involvement, and the CTP findings, subsequent digital subtraction angiography was performed, which demonstrated adequate vascular filling with no large vessel occlusion.

Conclusion

These findings highlight the diagnostic complexities of ECMO-related artifacts, even in the presence of neurological complications that can mimic true perfusion deficits and potentially lead to misdiagnosis.

INTRODUCTION

Extracorporeal membrane oxygenation (ECMO) is an advanced life-support technique for critically ill patients with severe cardiopulmonary failure [1]. Two primary types of ECMO have been described: venovenous (VV) ECMO, used primarily for severe respiratory failure, and venoarterial (VA)-ECMO, used for both cardiac and respiratory failures [2]. Neurological injury in patients on ECMO complicates detailed assessments because of intubation and sedation. Therefore, computed tomography angiography (CTA) and computed tomography perfusion (CTP) imaging are important diagnostic tools for evaluating these patients but can be affected by ECMO-related imaging artifacts due to altered hemodynamics [3]. These artifacts result from the mixing of the ECMO flow with native cardiac output, particularly when contrast agents are administered through conventional venous access. In VA-ECMO, peripheral cannulation, typically through the femoral artery, can cause retrograde blood flow to the aorta, resulting in asymmetrical contrast opacification and misleading results [3,4].
Here, we present a case report of artifactual hypoperfusion detected on CTA/CTP in a post-heart transplant patient on peripheral VA-ECMO. This case highlights the diagnostic challenges posed by ECMO-related artifacts and demonstrates the need for heightened awareness and expertise in interpreting neuroimaging studies in this unique patient population.

CASE REPORT

Presentation and initial management

A 54-year-old obese (body mass index 40.8 kg/m2) male with a history of non-ischemic cardiomyopathy complicated by heart failure with a reduced ejection fraction and left ventricular assist device (LVAD) placement was presented as a candidate for orthotopic heart transplant. The patient also had a history of pulmonary sarcoidosis, type 2 diabetes mellitus, and a recent admission for gastrointestinal bleeding due to an arteriovenous malformation that was treated with argon plasma coagulation.

Heart transplant and postoperative course

The patient underwent a repeat sternotomy for an orthotopic beating heart transplant. The procedure included removal of the LVAD and implantation of an implantable cardioverter-defibrillator. Due to primary graft dysfunction and an air embolism within the right coronary artery, the patient could not be weaned off the cardiopulmonary bypass and was placed on VA-ECMO via the left femoral vein and right axillary artery. Postoperatively, the patient required high doses of vasopressors (epinephrine, norepinephrine, and vasopressin) to maintain hemodynamic stability, and was managed with continuous renal replacement therapy for acute kidney injury and hyperkalemia. Despite these measures, the patient demonstrated poor cardiac contractility and was subsequently supported with an intra-aortic balloon pump for left ventricular unloading.

Neurological event

Two days after the transplant, while still on ECMO, routine neurological assessment revealed left-sided hemiparesis with brisk responses on the right side. These findings prompted us to suspect that a cerebrovascular event triggered the stroke alert. Further neurological examinations revealed an obtunded state characterized by minimal responsiveness and pinpointed pupils that were sluggishly reactive to light. There were no motor responses to verbal commands on the left side, but brisk responses in the right upper and lower extremities resulted in a National Institute of Health Stroke Scale score of 17. To further evaluate the patient’s neurological status and investigate the cause of hemiparesis, the patient underwent CTP and CTA of the brain.

Imaging findings

CTP studies were initially performed using contiguous non-contrast multidetector computed tomography (CT) axial imaging of the brain, followed by helically acquired rapid axial imaging from the aortic arch through the brain with contrast. Multifocal embolic-appearing infarcts were observed throughout the right middle cerebral artery, bilateral posterior cerebral artery territories, and cerebellar hemispheres. The entire brachiocephalic artery, along with the right subclavian and right common carotid arteries, was not visualized. CTA did not reveal any filling of the intracranial vasculature within the right hemisphere, while CTP demonstrated hypoperfusion of the entire right hemisphere (Fig. 1). The left common carotid, left subclavian, and left vertebral arteries were well visualized, with no evidence of occlusion within any of these vessels. Although the patient’s neurological examination was limited by sedation, the findings suggested subcortical dysfunction, which contradicted the CTP results. Four hours later, he was transferred to the angiography suite for digital subtraction angiography (DSA) to better assess the vasculature. There was no occlusion within the brachiocephalic, right subclavian, or right common carotid arteries as the entire right hemisphere was perfused (Fig. 2). The patient’s stroke was confirmed by CT 3 days later, which showed the expected evolution of multifocal subacute infarcts. While magnetic resonance imaging (MRI) would have helped differentiate between true hypoperfusion of the entire right hemisphere and widespread cerebral infarction, it was not feasible initially because of the patient’s condition. However, an MRI scan taken 1 month after this event demonstrated right high frontal and bilateral occipital lobe infarcts, which were now subacute to chronic. Additionally, there were punctate foci of increased fluid-attenuated inversion recovery and diffusion signals within the left frontal lobe without a corresponding apparent diffusion coefficient, suggesting subacute or chronic infarction with residual surrounding edema.

Patient management

The lack of large vessel occlusion suggested that the patient’s acute ischemic stroke was likely cardioembolic and secondary to his recent cardiac surgery. Given the patient's critical condition and the complexities of managing multiple organ support systems, the primary care team focused on optimizing ECMO flow, managing anticoagulation, and maintaining adequate perfusion pressures. The patient was successfully weaned off vasopressors and ECMO, with plans for rehabilitation to address left-sided hemiparesis.

DISCUSSION

ECMO is a life-saving intervention for patients with severe cardiopulmonary failure, but is associated with significant complications, including neurological issues. This case highlights the diagnostic challenges posed by ECMO-related artifacts in both CTA and CTP imaging, specifically in the context of VA-ECMO. Numerous case reports have reported neurological complications and artifactual CT findings in ECMO patients; however, we present the first case of concurrent ischemic infarction with exaggerated CTA/CTP findings in a patient with multiple organ support systems, highlighting the value of frequent neurological monitoring and careful interpretation of diagnostic imaging (Supplementary Table 1).
Neurological complications are common in patients on ECMO, with the incidence of ischemic stroke reaching 5% and 2% in VA-ECMO and VV-ECMO patients, respectively [5,6]. The risk factors for ECMO-related ischemic stroke include gastrointestinal hemorrhage, hemolysis, disseminated intravascular coagulation, and low perfusion brain states [7]. In VA-ECMO, hemodynamic alterations occur because of mismatched blood flow from the ECMO circuit and heart, potentially causing hypoxia in the regions supplied by the proximal aorta [8]. The cannulation method also influences the injury location. Central ECMO often results in reduced perfusion to the left hemisphere, whereas peripheral ECMO affects the right hemisphere, which is consistent with our case [9]. Impaired cerebral autoregulation, as indicated by elevated cerebral oximetry index values, disrupts the stable blood flow and leads to ischemia [10]. Additionally, the non-pulsatile nature of the VA-ECMO blood flow contributes to endothelial dysfunction, capillary leakage, and inflammation, further exacerbating brain injury [11]. While our case report demonstrated ischemic stroke in the right hemisphere, 70% of ischemic strokes in ECMO-treated patients occurred in the left hemisphere [12].
In VA-ECMO, the introduction of contrast agents can lead to significant imaging artifacts due to altered hemodynamics. As contrast-opacified blood is retrogradely reintroduced into the arterial system through the ECMO circuit, it can result in asymmetric opacification of the great vessels of the neck and head, potentially mimicking perfusion deficits [13]. This phenomenon was observed in the present case, in which the initial CTA/CTP study suggested hypoperfusion of the right hemisphere, which was later contradicted by the DSA findings.
The discrepancy between the CTA/CTP and DSA findings in patients undergoing ECMO presents a significant diagnostic challenge. In our case, the initial CTA/CTP study showed extensive hypoperfusion, raising concerns regarding severe ischemic events. However, DSA demonstrated adequate filling of the brachiocephalic artery and its branches, with no large-vessel occlusions, suggesting that the perfusion deficits were artifactual. Such discrepancies can lead to misdiagnoses and inappropriate interventions, highlighting the need for increased awareness and expertise in interpreting neuroimaging findings in patients on ECMO [4].
Several strategies have been proposed to optimize CTA imaging in patients receiving ECMO. These include temporary reduction of the ECMO flow rate, direct injection of contrast material into the ECMO outflow zone, and manual triggering of the CTA scans [13]. However, these techniques may not be feasible for all patients, particularly those with significant hemodynamic instability. In this case, manual triggering and adjustment of the ECMO flow rate were not viable options, necessitating the reliance on alternative imaging modalities and continuous monitoring.
Given the limitations of conventional imaging techniques in patients undergoing ECMO, alternative methods such as transcranial Doppler and near-infrared spectroscopy (NIRS) are valuable for neurological monitoring. For instance, transcranial Doppler can detect intracranial occlusions with high sensitivity and specificity and is portable and fast, making it suitable for use in critically ill patients [14]. NIRS allows continuous monitoring of cerebral oxygen saturation and offers real-time insights into cerebral perfusion [15].
This case highlights the complexity of diagnosing neurological events in patients due to imaging artifacts. Clinicians must be aware of these potential pitfalls and use a multimodal approach that incorporates alternative monitoring techniques to ensure accurate diagnosis and optimal patient care. Further research is required to develop standardized protocols and improve the diagnostic accuracy of neuroimaging in patients undergoing ECMO.

ARTICLE INFORMATION

Ethics statement
The requirement for informed consent was waived, and institutional approval was deemed unnecessary due to the retrospective nature of this single case report, in accordance with institutional policy. All patient-identifiable information was anonymized to ensure privacy.
Conflict of interest
No potential conflict of interest relevant to this article.
Author contributions
Conceptualization: CSO, SK. Methodology: all authors. Data curation: EOM, ML, SM. Visualization: ML, SM. Project administration: SM, SK, CSO. Writing - original draft: EOM. Writing - review & editing: all authors. All authors read and agreed to the published version of the manuscript.

Supplementary materials

Supplementary materials can be found via https://doi.org/10.18700/jnc.240028.
Supplementary Table 1.
Case reports with characteristic brain imaging findings in patients with ECMO
jnc-240028-Supplementary-Table-1.pdf

Fig. 1.
Computed tomography perfusion (CTP) imaging results. Initial CTP imaging displayed regions of hypoperfusion in the right hemisphere, indicated by the color-coded overlays. The pink areas indicate regions with severely reduced cerebral blood flow (CBF <30%), while the green areas represent regions with prolonged time to maximum (Tmax >6.0 seconds). The mismatch volume between these regions indicates potential penumbral tissue that could be salvageable. The imaging artifacts caused by the extracorporeal membrane oxygenation hemodynamics are evident, complicating the differentiation between true perfusion deficits and artifactual findings.
jnc-240028f1.jpg
Fig. 2.
Digital subtraction angiography (DSA) results. DSA results of the same patient demonstrated adequate vascular filling and perfusion in the right hemisphere, with no evidence of large vessel occlusion.
jnc-240028f2.jpg

REFERENCES

1. Richardson AS, Tonna JE, Nanjayya V, Nixon P, Abrams DC, Raman L, et al. Extracorporeal cardiopulmonary resuscitation in adults: interim guideline consensus statement from the Extracorporeal Life Support Organization. ASAIO J 2021;67:221-8.
crossref pmid pmc
2. Hadaya J, Benharash P. Extracorporeal membrane oxygenation. JAMA 2020;323:2536.
crossref pmid
3. Viamonte MA, Yu J, Echevarria FD, Nagae L, Youn TS, Tahsili-Fahadan P, et al. Caveats to acute imaging for acute stroke in the setting of venoarterial extracorporeal membrane oxygenation. Neurol Clin Pract 2021;11:e959-61.
crossref pmid pmc
4. Acharya J, Rajamohan AG, Skalski MR, Law M, Kim P, Gibbs W. CT angiography of the head in extracorporeal membrane oxygenation. AJNR Am J Neuroradiol 2017;38:773-6.
crossref pmid pmc
5. Lorusso R, Barili F, Mauro MD, Gelsomino S, Parise O, Rycus PT, et al. In-hospital neurologic complications in adult patients undergoing venoarterial extracorporeal membrane oxygenation: results from the Extracorporeal Life Support Organization registry. Crit Care Med 2016;44:e964-72.
crossref pmid
6. Lorusso R, Gelsomino S, Parise O, Di Mauro M, Barili F, Geskes G, et al. Neurologic injury in adults supported with veno-venous extracorporeal membrane oxygenation for respiratory failure: findings from the Extracorporeal Life Support Organization database. Crit Care Med 2017;45:1389-97.
crossref pmid
7. Zhang H, Xu J, Yang X, Zou X, Shu H, Liu Z, et al. Narrative review of neurologic complications in adults on ECMO: prevalence, risks, outcomes, and prevention strategies. Front Med (Lausanne) 2021;8:713333.
crossref pmid pmc
8. Cove ME. Disrupting differential hypoxia in peripheral veno-arterial extracorporeal membrane oxygenation. Crit Care 2015;19:280.
crossref pmid pmc pdf
9. Sansevere AJ, DiBacco ML, Akhondi-Asl A, LaRovere K, Loddenkemper T, Rivkin MJ, et al. EEG features of brain injury during extracorporeal membrane oxygenation in children. Neurology 2020;95:e1372-80.
crossref pmid
10. Tian F, Farhat A, Morriss MC, Tweed J, Li X, Huet B, et al. Cerebral hemodynamic profile in ischemic and hemorrhagic brain injury acquired during pediatric extracorporeal membrane oxygenation. Pediatr Crit Care Med 2020;21:879-85.
crossref pmid
11. O’Neil MP, Fleming JC, Badhwar A, Guo LR. Pulsatile versus nonpulsatile flow during cardiopulmonary bypass: microcirculatory and systemic effects. Ann Thorac Surg 2012;94:2046-53.
crossref pmid
12. Cho SM, Farrokh S, Whitman G, Bleck TP, Geocadin RG. Neurocritical care for extracorporeal membrane oxygenation patients. Crit Care Med 2019;47:1773-81.
crossref pmid
13. Lee S, Chaturvedi A. Imaging adults on extracorporeal membrane oxygenation (ECMO). Insights Imaging 2014;5:731-42.
crossref pmid pmc pdf
14. Mattioni A, Cenciarelli S, Eusebi P, Brazzelli M, Mazzoli T, Del Sette M, et al. Transcranial Doppler sonography for detecting stenosis or occlusion of intracranial arteries in people with acute ischaemic stroke. Cochrane Database Syst Rev 2020;2:CD010722.
crossref pmid pmc
15. Hunt MF, Clark KT, Whitman G, Choi CW, Geocadin RG, Cho SM. The use of cerebral NIRS monitoring to identify acute brain injury in patients with VA-ECMO. J Intensive Care Med 2021;36:1403-9.
crossref pmid pdf
Editorial Office
Department of Neurology and Critical Care, Seoul National University Hospital,
101, Daehak-ro, Jongno-gu, Seoul, 03080 Korea
TEL: +82-2-2072-0743 (10:00 AM ~ 4:00 PM) / +82-2-2072-1810   FAX: +82-2-3672-7553    E-mail: office@e-jnc.org

Copyright© Korean Neurocritical Care Society.                Developed in M2PI