In some ways, our job consists of making the colourful spectrum of real life medicine fit the black and white confines of the hospital protocol or the textbook definitions. Though the presentation of this next case was as black and white as it gets, the way it was dressed up to the AMU could lead you up the wrong path. As you go down the case you will get more clues; see how early you can figure it out.

A 59 year old man originally from Eastern Europe had felt tired for the past 3 months. He had seen his GP, who discovered a normocytic anaemia (Hb 7.4) with a normal WCC and platelets. (The patient’s haematinics were normal, making a mixed iron/folate/B12 deficiency unlikely. His reticulocyte count and kidney function were also normal.)

The patient reported no source of bleeding, including normal bowel habits. He had no skin or joint symptoms. He felt there was no change to his appetite, but did notice about 5 kg of weight loss over this time. There was no fever, but yes, there were occasional drenching night sweats.

PMH revealed rheumatic fever as a child, which left him with a systolic murmur of some sort that the GP was unable to classify further.

He took no regular medications.

Apart from this he was normally a fit and well builder. He had been unable to go to work in the past month because of the tiredness. He lives with his wife. Social history was otherwise unremarkable.

On examination, his obs were Pulse 105, BP 134/90, RR 16, Sats 99% on air and T 37.8. The clerking FY1 noted this:

On palpation there was splenomegaly.

On auscultation there was a loud pansystolic murmur radiating to the axilla.

Bloods confirmed a normocytic anaemia with a WCC of 10.1 and a CRP of 53. His U&Es, LFTs and Coagulation were all normal.

A urine dipstick showed haematuria (+++).

A transthoracic echo showed a mitral valve oscillating vegetation, with severe mitral regurgitation and normal LVEF.

Three sets of blood cultures returned two days later which all grew strep viridans sensitive to benzylpenicillin.

What is actually going on in IE?

As the under review HPA guidelines on blood cultures states:

“IE is infection of the heart valves and/or other areas of the endocardium. It usually occurs at the site of a predisposing cardiac lesion or congenital defect where there is turbulent blood flow, encouraging endocardial damage and adhesion of platelets. A fibrin clot is deposited on the damaged endocardial surface and becomes colonised with organisms which have entered the bloodstream, so forming infected vegetation. Viable bacteria may be present deep within the vegetation as well as on the surface making antimicrobial treatment difficult.”

Why did this man get IE?

Intact valvular endothelium is normally quite resistant to colonisation. The rheumatic fever caused some degree of valve damage, which makes it easier for any bacteria that happens to be in the blood to make a camp there.

What bugs are most likely in this case?

According to the UK guidelines (2012), the most common causes of NVE in non-intravenous drug users are currently S. aureus (28%), coagulase-negative staphylococci (CoNS; 9%), streptococci (35%) and enterococci (11%); 9% are culture-negative.

The main times staph aureus dominates over streptococci is healthcare associated IE, IVDU associated IE and in the first year following a prosthetic valve placement. Staph aureus is the second most common cause of IE in prosthetic valves in the first year after coagulase negative staphylococci. In fact, the closer it is now to the time of the prosthetic valve placement, the more likely that any IE the patient is currently having on that valve is caused by coagulase negative staphloccoci.

How should cultures be taken?

According to the UK guidelines:

“In patients with a chronic or subacute presentation, three sets of optimally filled blood cultures should be taken from peripheral sites with ≥6 h between them prior to commencing antimicrobial therapy.”

It adds:

“There is no evidence to support the commonly perpetuated view that blood cultures should be taken from different sites. All skin surfaces are colonized by bacteria and adequate skin disinfection is key to reducing contamination. Taking blood cultures at different times is critical to identifying a constant bacteraemia, a hallmark of endocarditis.”

In practice, this means it would be 12 hours from your first suspicion of IE to giving your first set of IV antibiotics in a patient with a chronic or subacute presentation.

But the sepsis six! Each hour’s delay results in a 7.8% increase in mortality! What should I do if the patient is septic with IE?

According to the same guidelines:

“In patients with suspected IE and severe sepsis or septic shock at the time of presentation, two sets of optimally filled blood cultures should be taken at different times within 1 h prior to commencement of empirical therapy, to avoid undue delay in commencing empirical antimicrobial therapy.”

If you spoke to a sepsis champion, they would feel very uncomfortable about knowingly delaying giving antibiotics in any patient with severe sepsis or septic shock. In fact, the guidance from the sepsis world is to give fluids, oxygen and antibiotics ASAP and only take cultures before giving antibiotics if this would not delay giving antibiotics.

I’ll have to cross check what to do in this situation with a sepsis don.

What is the most common extracardiac manifestation of IE? 

Though this man had a splinter haemorrhage, this could be related to his occupation as a builder. The patient was unsure if this line in his nail was new or old. Splinter haemorrhages are a pretty rare feature of IE in the developed world, where the presentation is usually fairly early. Janeway lesions, Osler nodes and clubbing are all subacute signs, and are also quite rare in the developed world.

Cerebral complications are the most severe extracardiac complications of infective endocarditis, as well as the most frequent (occurring in 15 to 20% of patients). A haemorrhagic or ischaemic stroke may result. In other words, if a patient has an unexplained stroke, we may question whether or not there is underlying IE (after dealing with the stroke).

To TOE or not to TOE?

Same guidelines:

What should the first line antibiotics be whilst awaiting culture results in this case?

According to the same guidelines:

Note amoxicillin rather than the benzylpenicillin. Although you should follow your hospital policy, it seems that amoxicillin has better activity against enterococci, which cause 11% of native value IE. The newest versions of the BNF support this.  Fully sensitive streptococci might still be managed with benzylpenicillin.