P012 - EARLY SURGERY IN ENDOCARDITIS WITH ISCHEMIC AND HEMORRHAGIC STROKES

Case background: Infective endocarditis (IE) has high morbidity, mortality, and clinical decision-making is complex. Embolic complications are common and include ischemic stroke, mycotic aneurysm (with or without rupture), and intracerebral hemorrhage. The timing of cardiovascular surgery in patients with indications is nuanced—weighing the risk of brain hemorrhage against delaying cardiac intervention.
We present a patient with infective endocarditis complicated by both ischemic stroke and subarachnoid hemorrhage (SAH) who subsequently underwent early surgical intervention with a successful outcome.

Case:

A 65-year-old patient presented to the emergency department with a 4-day history of flu-like symptoms, generalised weakness, and abnormal language.
His examination was significant for tachycardia at 113 beats per minute, tachypnea at 22 breaths per minute, requiring 3 litres of oxygen via nasal cannula, and presence of Janeway lesions on his palms and feet. Neurological exam showed a National Institutes of Health Stroke Scale (NIHSS) score of 2.
A computed tomography (CT) of his brain with and without contrast showed a wedgelike area of hypodensity and loss of gray-white differentiation in the left anterior parietal lobe, consistent with subacute MCA infarct. There was a SAH in the left frontal distribution not initially reported; size, although not reported, was estimated at less than 30mls. A CT Angiogram of the carotid arteries and circle of Willis did not demonstrate a mycotic aneurysm.
An hour later, he developed new neurological deficits, now scoring 5 on the NIHSS. Repeat CT imaging reported worsening of the prior SAH in the left frontal region with new SAH in the right frontal and bilateral parietal lobes. Magnetic Resonance Imaging (MRI) of the brain correlated with the CT findings along with acute embolic appearing infarct in the right occipital, left parietal, left frontal, and left occipital lobes, and right cerebellum. Additionally, there was presence of blood product in multiple locations including the locations of the known SAH, and some of the larger infarcts, as well as some small punctate lesions. Although reported as hemorrhagic transformation, the pattern appeared to be more in keeping with laminar necrosis in the larger areas of infarct.
His blood cultures were positive for methicillin-sensitive Staphylococcus aureus.
Transthoracic echocardiogram showed thickened/myxomatous mitral leaflets with mild prolapse of mitral leaflets with at least moderate eccentric mitral regurgitation (MR) and suspected vegetations. It also showed a mildly dilated, globally hypokinetic left ventricle with an ejection fraction of 30-35%. Transesophageal echocardiogram confirmed small mobile lesions involving the tip of posterior mitral valve leaflet at 0.6cm, as well as a large coaptation gap with severe MR.
The patient’s neurological symptoms improved until day 5, when he unfortunately developed new deficits, elevating his NIHSS score to 9. He subsequently underwent an emergent CT of his head which showed a new left M1 occlusion.

Management Challenges: Management:
Given the above clinical features, a diagnosis of IE was established and he was initiated on appropriate antibiotics.
After the new infarct on day 5 of his admission, he underwent emergent endovascular thrombectomy. It was felt by our local multi-disciplinary team (MDT) including Stroke Neurology, Cardiology, and Cardiovascular surgery, that recurrent cerebral embolism should prompt us to address the mitral valve as quickly as possible. Thus, despite expanding SAH, albeit small, and reports of hemorrhagic transformation, the decision was made to pursue bioprosthetic mitral valve replacement and reconstruction of the mitral annulus, 11 days after our patient’s initial hospital presentation. Blood cultures cleared on Day 11, and he was discharged home 17 days later to complete a 6-week course of antibiotics.

Discussion:
Our case report highlights the success of early surgical intervention in patients with infective endocarditis complicated by ischemic and hemorrhagic strokes.
The pathophysiology of intracranial hemorrhage (ICH) in patients with endocarditis and ischemic strokes can vary from ruptured mycotic aneurysms, ruptured septic emboli, pyogenic arteritis, hemorrhagic transformation of ischemic infarcts, or inflammatory small vessel disease [1]. The favoured mechanism in our case was inflammation and leakage of mycotic thrombus into the vessel, without development of aneurysm (as none were seen on CTA or angiogram). Although hemorrhagic transformation was described in our patient’s MRI report, the small volume of bleed and absence of mass effect was actually more consistent with scattered petechiae (Heidelberg Classification HI1a), as well as laminar necrosis. Thus, the significance of ICH related to the prior stroke was likely overstated. This underscores the importance of MDT, especially with Stroke Neurology, to avoid overestimation of risk of hemorrhagic transformation during cardiac surgery [1].
Current evidence would support early surgical intervention in our patient with IE given persistent bacteremia > 7 days despite appropriate anti-microbial therapy [2,3]. Moreover, after multi-disciplinary discussion, it was noted that our patient had a persistently high subsequent embolic risk given type of organism (MSSA), involvement of anterior mitral leaflet, and recurrent embolism.
In patients with transient ischemic attacks and ischemic stroke, observational data support surgical intervention without delay if patients are not comatose. In patients complicated by hemorrhagic stroke, there is mixed evidence surrounding surgical timing, guided mostly by etiology, size and appearance on serial imaging [2,4,5].
The 2020 American College of Cardiology/American Heart Association guidelines [3] suggest delaying surgery >4 weeks in patients with ICH as long as they remain hemodynamically stable. The 2023 European Society of Cardiology guidelines [2] list urgent surgery following hemorrhagic stroke with volume < 30 mL or NIHSS < 12 as a class I indication, but recommend delaying cardiac surgery >1 month following larger hemorrhagic strokes as a class IIa recommendation. Urgent or emergent surgery is suggested in patients with high risk of embolic strokes as a class IIa indication. Overall, they recommend case-by-case review based on the mechanism of intracranial hemorrhage, volume of bleed, and NIHSS score.
Timing of surgery in the setting of repeated cerebral insults is not well addressed in the guidelines. Our patient was selected for urgent surgery after MDT discussion, guided by existing published recommendations around size of bleed and perceived risk of subsequent cerebral embolism. We believe MDT discussion significantly enhanced decision-making for our patient and highly recommend this approach when surgical decision-making is nuanced by recent stroke and hemorrhage.