Management of Infective Endocarditis in the Current Era

Podcast Transcript

Announcer:

Welcome to Cleveland Clinic Cardiac Consult, brought to you by the Sydell and Arnold Miller Family Heart, Vascular and Thoracic Institute at Cleveland Clinic.

Haytham Elgharably, MD:

I'm going to talk today about some of the new trends, and updated data from multiple studies have been doing about the management of endocarditis in the current era. I'm going to go through some basic to translational science of the microbiology of endocarditis, some of the current trends we have learned from multiple studies we have been doing here at the clinic, and also what's coming up in the future or what areas of opportunities we need to work on for improvement.

For this slide, this is an overall outcome of infective endocarditis patients, surgical or nonsurgical, over the last maybe 25 years or so. Obviously, mortality is 100% if endocarditis was untreated. The in-hospital mortality in overall patients across the country is around 25%, and five-year mortality is still up to half of these patients will die within five years. Interestingly, almost 50% of these patients will still require intervention. You cannot control them with antibiotics alone. There really has been no significant improvement of the one-year mortality over two decades, despite all the advances of antimicrobial therapies.

To put things in context, I'll just go through a case scenario typical for prosthetic valve endocarditis that come here. A 60-year-old patient with history of bicuspid aortic valve stenosis. He had an aortic valve replacement. It was a prosthetic tissue valve, a couple years ago. He had a skin cut while he was working in the garden. He had an infected wound. Then he's admitted with fever and fatigue. Blood culture shows MSSA bacteremia at that time. He had a workup at an outside hospital. The valve had small vegetation, but the valve was working fine. They just started an antibiotic course, based on the blood culture sensitivity. He had improvement of symptoms, blood culture became negative. Then they sent him home on six weeks of antibiotics.

However, two weeks later, he came back with recurrent fever and short of breath. His blood culture grew MSSA again. Now the echo shows enlargement of the vegetation. Also the dehiscence of the prosthetic valve was peripheral leak. In echo, there is a posterior space behind the aortic root, suggestive for aortic root abscess. So, he came to our center. We did a redo surgery. We explanted the old prosthetic aortic valve. We debrided the abscess. We did an aortic root replacement with the homograft. When we send the valve for culture, it still grow MSSA.

So this case raised some questions. Why did the infection relapse after the patient completed the antibiotic course? And why can antibiotic clear the blood cultures, but they cannot sterilize the prosthetic valve vegetation? And why during that few weeks while the patient is still on antibiotic? The pathology still progressed within a few weeks and there was tissue invasion and destruction of the dehiscence of the valve.

Another important observation is that native valve endocarditis most commonly presents with localized infection. Less commonly, invasive. However, on the other hand, prosthetic valve endocarditis typically presents with aggressive invasion, tissue destruction around the valve, abscess formation and fistulas.

We don't really have a good explanation for that difference. So it seems like there's a gap in knowledge, that the outcomes would improve for almost two decades. Also, when it comes to the pathogenesis of endocarditis, we may not have a complete understanding of the microbial behavior. Why the antibiotics alone failed to control the infection and why there is more aggressive pathology and invasion with prosthetic valve endocarditis compared to native.

So going back to the basics. Endocarditis infection starts with access of bacteria to the bloodstream. Then, the bacteria can have protein lesions on their surface that can attach to either a native valve or a prosthetic valve, and then they can colonize the mesh of platelets and fibrin from the host on the surface of the valve and form the vegetation.

This is a cool electron microscope picture we took from one of the infected valves, yellow or green color. All these cocci, they clump together on the extracellular matrix on the surface of the valve. What's compromised these, or constitutes these vegetations? But why in these clusters, we cannot really kill with the antibiotics?

So going back to the microbiology, back in the 90s, there was a concept grown or identified it in the microbiology world called biofilm. Bacteria can exist in two forms. One is planktonic single cell bacteria. Or when they attach to a surface, they grow together as condensed micro-colonies or clusters with an extracellular matrix or a slime. And that's called biofilm.

While they're inside the biofilm, they still can seed out planktonic bacteria to cause infection around them. This is a hypothesis we proposed back in 2016, it was Dr. Petterson. We proposed that maybe when the bacteria the attach to the cardiac valves, they grow in these condensed microcolonies of biofilm. That's why we cannot kill them with antibiotics. We need surgical intervention in most of these cases.

So why are they resistant? Why is the biofilm resistant to both host immune response and antibiotics? There are multiple strategies that the bacteria will use to resist the antibiotics and evade the host immune response. So first of all, this biofilm matrix or slime will slow penetration of the antibiotics or prevent penetration by immune cells.

Also, the bacteria inside the biofilm changes the gene expression. They have more upregulated genes responsible for stress response. In antibiotic resistance, the environment inside that slime matrix is altered so the antibiotics won't work or, it's not feasible for the mechanism of action for most antibiotics. The most interesting part is at the in the core of the biofilm, the bacterial cells are in a dormant growth state. They’re not really replicating, so they can skip the antibiotic effect, which is relying on the rapid bacteria growth.

So the question becomes, is it really a biofilm infection or not? Is endocarditis a biofilm infection or not? 80% of our patients with endocarditis present with all gram-positive cocci that all can form biofilm in vitro work. Also the vegetation of the endocarditis fulfills the biofilm infection criteria that was published based on in vitro experiments. They are microcolonies, aggregate, that adhere into a surface. They are encased in a slime or extracellular matrix. They are resistant to both host immune response and antibiotics.

The current evidence supports that hypothesis, based on clinical observation. We learn from endocarditis patients, in vitro microbiology work or animal models of infective endocarditis, or histopathological imaging of explanted cardiac devices that show the aggregation of bacteria attached to the valve.

But we don't really have direct, in vivo evidence that when the bacteria attach to the valve, they will change their gene expression and switch to biofilm from planktonic form. So in a pilot study we tried to test the hypothesis. The planktonic cell attached to the valve, the change of the gene profile, they produce gene expression responsible for biofilm formation or biofilm mode of growth, and also production of other virulence factors that can cause tissue damage that we see in prosthetic valve endocarditis.

In this study, we collected samples from six patients with staph aureus endocarditis. Three were native and three were prosthetic. We collected preoperative blood cultures with bacterial isolates from these patients. During the surgery, we also collected vegetation samples. We ran an RNA sequencing, which is something has not been published before in the literature, that some groups took vegetation and looked at the gene profile or gene expression of the bacterium.

Just to clarify, there is a difference between RNA sequencing or the valve sequencing that we use. We send our sample outside to look at bacteria, what is called universal PCR. So universal PCR looks at one gene which is 16x ribosomal RNA, which is unique for prokaryotic or all microorganisms. That's what we use to differentiate different strains, different types of bacteria.

But it doesn't really tell us what other genes that the bacteria express. For that, we use RNA sequence. So it's a high-input sequencing tool that looks at all the transcription levels of the bacteria. Exactly how many genes they change when attached to the valve. So that's the approach we use in the study.

Announcer:

Thank you for listening. We hope you enjoyed the podcast. We welcome your comments and feedback. Please contact us at heart@ccf.org. Like what you heard? Subscribe wherever you get your podcasts or listen at clevelandclinic.org/cardiacconsultpodcast.