Tasty Morsels of Critical Care 096 | Inf...

Welcome back to the tasty morsels of critical care podcast.

For the next couple of posts we’re going to focus on infective endocarditis. We have had something of a run of these over the past few months with every other cardiac surgery seemingly being done for IE. Not all of this will be relevant to every intensive care unit but as I work in an ICU with lots of cardiac surgery we will be covering the relevant aspects. This episode we’ll look at the diagnostic process and next time we’ll look at management.

IE is a common consideration in patients with infection being admitted to the hospital. But we’re going to very much focus on what the ICU presentation might look like. Often they’ll be coming in with the diagnosis already made but a lot of the time you’ll be looking after a generically sick patient and only over time does the the diagnosis of IE raise its head.

3 main types of presentations of undiagnosed IE may present themselves to the critical care doctor (and there are of course others)

1. sepsis. this is of course bread and butter to us in ICU but IE needs to remain in the differential for septic patient in the ICU, rather than just pinning it on the first minor infiltrate we see on the CXR
2. the embolic presentation. The most common would be an intracranial event, either stroke or intracranial bleeding from a mycotic aneurysm or embolus. Certainly anything suggesting embolic stroke (like bilateral strokes for example) should you make you consider IE (along with of course, the much more common LAA thrombus in A fib). Outside the brain the main emboli that might show up for us will be obvious on CT scans with splenic and renal infarcts in left sided disease or cavitating pulmonary lesions in right sided valve disease. You will also get the occasional dramatic bleeder where some intrabdominal vessel has a mycotic lesion that bleeds dramatically needing an IR intervention
3. heart failure. this is usually because the vegetation has eroded through something important like a valve leaflet and now there is torrential insufficiency of this valve. these tend to be fairly dramatic presentations with wet lungs and cardiogenic shock in the left sided lesions.

How do we formally make the diagnosis of IE? Well, especially as this is exam focussed we need to be able to drag out Duke’s criteria. There have been some recent updates in 2023 and while it would seem unfair to expect a trainee to regurgitate all the details it would be helpful to have an overview.

You will probably have in your memory something about major and minor criteria. Major criteria can be split into

1. imaging major criteria – with a vegetation on echo still being the main sign and TOE preferred over TTE. But CT has some important findings that can fit in here too
2. micro major criteria – growing certain common IE bugs in the blood would be the commonest here though it does allow for some serologies of the weird culture -ve bugs in this category
3. surgical major criteria – if, at operation the surgeon sees vegetation or abscess or perforation consistent with IE then that counts as a major criteria

The minor criteria include many less specific features that I’m not going to go through here. A bit like failing your driving test you can get IE by having 2 major criteria , or 1 major and 3 minors or even 5 minors on their own can make the diagnosis.

There are pathological criteria which are considered the gold standard and that involves isolating the bug from tissue, typically a valve removed at surgery but typically that’s a diagnostic standard that most patients will not reach

I think it is helpful to have an idea of the common bugs involved. From European wide data it seems we could give the following prevalence of bugs

• Staph Aureus (30%)
• oral streps (17%)
• coag -ve Staph (11%)
• e faecalis ( remember 95% faecalis, 5% faecium)

Of note bacteraemia is constant (you do not need cultures just at time of fever) and most cultures should be +ve when IE is the cause.

Most culture -ve IE is due to the patient already being on antibiotics. You can expect blood culture -ve IE in ~10% but the vast majority of this are just standard bacteria in patients that someone had already started on antibiotics and now you can’t grow it. For the true “culture -ve” crowd  the following bugs should be on your list (and this is not a complete list

• coxiella burnetti [Q fever]
• bartonella,
• brucella

When it comes to imaging TOE remains the test of choice as it just gives you much better pictures. Even if you can see a vegetation on TTE then the guidelines recommend that most patients should still have TOE to properly assess valve complications and other valve involvement not obvious on TTE. The exception to this might be a TV endo with good images. Vegetation size has some prognostic implications and should be measured in its longest plane.

But we should also be thinking about other tests. CT in particular has a role in looking at aortic root abscesses, especially in prosthetics. PET scan is also becoming a more important test but i find getting this for an ICU patient somewhat impractical.

In terms of what bacteraemias should trigger an echo, the guidance would suggest staph aureus, e faecalis and classic streps as an indication. But not every bug in the blood needs an echo which I think is a message worth communicating.

Reading:

– Fowler, V. G. et al. The 2023 Duke-International Society for Cardiovascular Infectious Diseases Criteria for Infective Endocarditis: Updating the Modified Duke Criteria. Clin. Infect. Dis. 77, 518–526 (2023).

– Delgado, V. et al. 2023 ESC Guidelines for the management of endocarditis. Eur. Hear. J. 44, 3948–4042 (2023).