Infective endocarditis

Infective endocarditis featured in Question 21 from the first paper of 2008. The question, apart from interrogating the candidate's knowledge of the manifestations of endocarditis, also ventures into cardiothoracic territory by asking about the indications for urgent valve replacement. The college also mentioned IE briefly in Question 29.1 from the second paper of 2014 and the identical Question 30.1 from the second paper of 2018,  where they wanted to know which non-murmur clinical signs are normally associated with it.

Clinical manifestations of infective endocarditis

An excellent review article form 2009 lists the following clinical features:

• Osler's nodes
• Janeway lesions
• Splinter haemorrhages
• Roth spots
• Focal neurological signs suggestive of embolic phenomena
• A new murmur or a worsening of an old murmur
• Splenomegaly
• Glomerulonephritis
• Arthralgia and arthritis
• Elevated ESR, CRP or rheumatoid factor
• Haematuria

The abovementioned article has a table (Table 3) which lists these manifestations according to their prevalence among a large patient cohort. The emphasis of the article is on endocarditis "in the 21st century", implying that the endocarditis of the previous centuries was substantially different. This is certainly true. One need only refer to the 1885 Gulstonian Lectures by William Osler to see what infective endocarditis looked like in the pre-antibiotic era (in short, it was uniformly fatal). For a relatively recent 20th century perspective, one can turn to a 1983 review of the state-of-the art diagnosis and therapy for the previous 25 years.

Duke criteria

These were proposed in 1994 on the basis of an analysis of 405 consecutive cases of infective endocarditis. In order to qualify for IE, one must have either

• two major and one minor criteria
• one major and three minor criteria
• 5 out of the 6 minor criteria

Major Criteria:

• Blood culture evidence of persisting bacteraemia:
  • Typical microorganisms in 2 separate blood cultures, or
  • Typical microorganisms in 2 blood cultures taken 12 hours apart, or
  • Typical microorganisms in 3 out of 4 blood cultures all taken within 1 hour.
• Positive echo findings:
  • Cardiac abscess
  • Vegetations
  • Partial dehiscence of prosthetic valve

Minor Criteria:

• Predisposing condition (eg. mechanical valve, IV drug use)
• Fever > 38.0°
• Vascular manifestations eg. Janeway lesions
• Immunologic manifestations, eg. glomerulonephritis
• Blood cultures which do not meet the Major Criteria
• Echo findings which do not meet the Major Criteria

A major limitation of this set of criteria is the fact that up to 20% of patients have "culture-negative" IE and end up being misclassified. Particularly, patients with Q-fever endocarditis would grow nothing in their cultures, and their diagnosis would be delayed. Some have used this to call for an inclusion of Q-fever serology among the major criteria.

Typical organisms

Typical valve-eating organisms may include the following:

• S.epidermidis and other coagulase-negative staphylococci
• Streptococcus viridans
• S.aureus
• Enterococcus
• Coxiella burnetii (Q fever)

A large scale cohort review lists microbial aetiology of IE in their patient group (their Table 5).

HACEK organisms

HACEK organisms are also discussed in the section on acronymous organisms. They are responsible for only about 3% of native valve endocarditis.

• Haemophilus species: H.aphrophilus, H.parainfluenzae and H.paraphrophilus
• Actinobacillus and Aggregatobacter species
• Cardiobacterium hominis
• Eikenella corrodens
• Kingella kingae

Management

There's a few different recipes around for the empirical management of IE, which makes perfect sense as these recommendations would differ according to the prevailing microbiome:

• The Sanford guide (Hong Kong) recommends vancomycin and gentamicin or vancomycin and ceftriaxone as empiric therapy for native valve endocarditis, or vancomycin, gentamicin and rifampicin "triple therapy" for prosthetic valves. This is one of the few cases when one might give 1mg/kg of gentamicin every 8 hours.
• The eTG (Australia) recommend benzylpenicillin, flucloxacillin and gentamicin, substituting vancomycin for the beta-lactams wherever MRSA is suspected or where the patient has sepsis or septic shock.

Historically, the CICM examiners seem to use eTG for their antibiotic choices. In Question 21 from the first paper of 2008, they only said "AB guidlelines suggest Vanc + gent" without specifying which "AB guidelines" they meant. The patient in that specific question was recovering from AVR, and was "unwell over a few days with fever and rigors", which warrants vancomycin according to the eTG. Since 2008, eTG have also added flucloxacillin to vancomycin, as there is thought to be some sort of synergistic effect (Tong et al, 2016)

Generally speaking, after the blood cultures finally bloom into significance, one can de-escalate this empirical therapy. Frequently, the patient will only require something like ampicillin or benzylpenicillin. In the words of a recent (2012) Lancet review, "present recommendations for antimicrobial treatment are based on old but efficient antibiotic drugs because most pathogens that cause infective endocarditis are still sensitive to them, even if the emergence of resistant strains is growing".

Valve replacement

Immediate valve replacement for IE has been practiced for decades.

Even then, the following criteria for urgent surgery were followed:

• Haemodynamic instability
• Aortic root abscess
• Ongoing embolic phenomena

A 1994 article reports that " surgical replacement of the infected valve led to significantly lower mortality (23%) as compared with medical therapy alone (56%)". However, IE recurrence was observed in 30% of patients after 30 days, and 69% of patients after 60 days. More recent studies (2012, NEJM) confirm that surgical management of infected valves decreases mortality largely by decreasing the risk of death from embolic phenomena.