see also:

• The febrile adult presenting to the ED
• wikipedia

• non-infective endocarditis or marantic endocarditis is due to thromboses on previously normal heart valves and occurs in hypercoagulable states including pregnancy and may cause emboli but not the septic or inflammatory characteristics of infective endocarditis which is the topic of this article.
• infective endocarditis is a potentially fatal condition but fortunately quite an uncommon one with an annual prevalence of only 10-40 per million population with more than 50% of cases occurring in those older than 50 years
• nevertheless, there should be a low index of suspicion in any iv drug user or patient with known heart valve disease or prosthetic valves who present with undifferentiated fever, a new heart murmur or high ESR.
• in patients with prosthetic heart valves, the main organism is usually Staph. aureus, but other less virulent organisms may be involved, and infection may occur “early” - within 60 days of valve implantation and accounts for more than 50% of endocarditis in these patients, or “late” - after 60 days.
• in patients with native valves, mortality rates are in the range of 16-27% and infection may result in either:
  • acute endocarditis generally due to virulent organisms such as Staph, aureus or group B Strept. and run an aggressive course
    • may occur on normal valves associated with Staph septicaemia and in iv drug users
    • in iv drug users with endocarditis, the commonest organism is Staph. aureus and its virulence means even normal heart valves may be infected and result in a rapid course of acute endocarditis which usually affects the tricuspid valve
      • it is said that endocarditis is the cause of fever in 10-15% of ivdu patients who present to ED with a fever!¹⁾
  • sub-acute endocarditis (SBE) generally due to less virulent organisms such as dental organisms infecting underlying valvular problems, and run a more indolent course
• endocarditis in patients with native valves not related to iv drug use usually occurs in those with underlying structural valve disease where eddies due to turbulent flow allow bacteraemic seeding to occur in the heart:
  • 30% cases occur in those with rheumatic heart disease
  • 15% cases occur in those with congenital heart disease
  • 20% of cases occur in those with mitral valve prolapse with murmur
  • other cases occur in those with Marfan's disease, syphilis or calcific aortic stenosis due to bicuspid valves
  • an increasing cause of endocarditis in that due to intravascular devices such as:
    • central venous catheters, PICC lines, implanted pacemakers or defibrillators, haemodialysis shunts, chemotherapy and hyperalimentation lines
  • 70% are Strept. sp - usually dental (viridans Strept.) or bowel or urinary organisms
  • 25% are Staph. sp.
  • fungal endocarditis is found primarily in iv drug users and ICU patients who receive broad-spectrum antibiotics.
  • other causes of culture negative endocarditis include Chlamydia, Q fever (Coxiella), Bartonella and HACEK group of oropharyngeal organisms

• historically, antibiotic prophylaxis during “at risk” procedures such as dental treatment has been advocated for high risk patients (those with valvular lesions)
• however, studies such as the 1998 Philadelphia study²⁾ have suggested antibiotic prophylaxis during dental procedures would not substantially reduce the incidence of endocarditis.
• although some 40% of cases of endocarditis are due to dental organisms, less than 25% of cases actually had dental treatment within 3 months prior to endocarditis developing, suggesting that daily brushing of teeth in patients with valvular lesions is just as likely to be the cause, and that the presence of valvular lesions is a far higher risk factor than dental treatment per se.
• this is supported by this 2008 study³⁾ which showed that brushing teeth results in positive blood cultures at 60 minutes in 9% of patients vs 2% of patients following dental extraction, implying that the daily brushing of teeth, whilst important in preventing gingivitis and load of oral organisms, is probably more likely to contribute to endocarditis risk in susceptible patients than visits to a dentist. Furthermore prophylactic antibiotics were not 100% effective in preventing bacteraemia.
• risk of endocarditis from a dental procedure ranges from 1 in 100,000 if prosthetic valve to 1 in a million if mitral valve prolapse
• it may be though that it is the extent and duration of bacteraemia from dental treatment that may be more important rather than just frequent lower “dose” bacteraemias of teeth brushing, and thus many dentists will still prefer to advise prophylactic antibiotics to at risk patients, and indeed, current recommendations are for prophylactic antibiotics for high risk patients only ⁴⁾
• see also this review article Infective Endocarditis Prophylaxis for Dental Procedures in 2009: What Has Changed? Hellenic J Cardiol 2009; 50: 493-497 (pdf)

• focus is on gaining a specific diagnosis, thus 3 sets of blood cultures prior to starting antibiotics
• manage complications such as congestive cardiac failure, etc
• commence empirical antibiotics according to whether native valve, intravenous drug users (IVDU) or injection drug use (IDU) or prosthetic valve
• admit for further Ix and Mx

• it would seem the far majority of patients in whom there was a defined precipitating event such as dental work, develop symptoms within 2-4 weeks (within hours in some cases) with a median of 10 days, but a median time to diagnosis of 6 weeks although some may not be diagnosed for 6 months or more.⁵⁾
• untreated, SBE may be fatal in 6 weeks to 12 months from onset.

• most of the early features are quite non-specific and include:
  • fever - often low grade and intermittent in SBE
  • anorexia, weight loss, malaise, headache, myalgias, night sweats, low back pain
  • iv drug users often develop SOB, cough and chest pain presumably due to seeding into pulmonary vasculature from the tricuspid valve
• additional presenting features to the ED may include:
  • 90% will have a PUO
  • 85% will have a cardiac murmur and 10% will have a change in quality of a murmur
  • 50% will have one or more of the “classical” features:
    • petechiae - 20-40% of cases
    • splinter haemorrhages in nail beds
    • Osler nodes - tender s/c nodules on the volar pads of fingertips
    • Janeway lesions - non-tender macular lesions on palms and soles
    • Roth spots - retinal haemorrhages (only 5% of patients)
    • splenomegaly in 25-50%
  • 40% will have neurologic features due to septic emboli to the brain causing embolic stroke (CVA) with possible microabscesses, meningism, and delirium
  • acute left sided valvular insufficiency may result in congestive cardiac failure
  • mycotic aneurysms may occur - eg abdominal aorta, coeliac axis, sup.mesenteric artery
  • fungal endocarditis with its larger, more friable vegetations may cause limb ischaemia from septic emboli to the major limb arteries
  • disseminated intravascular coagulation (DIC) may occur
  • renal infarcts or abscesses may occur
  • arthralgias - particularly shoulder, knee and hip occur in under 20% of cases

• see calculator
• clinical criteria for infective endocarditis requires:
  • Two major criteria, or
  • One major and three minor criteria, or
  • Five minor criteria

• multiple positive blood cultures of the infecting organisms
• echocardiographic evidence of endocardial involvement or a new regurgitant murmur on physical examination

• predisposition - pre-existing valve disease or intravenous drug users (IVDU) or injection drug use (IDU)
• fever > 38.0 deg C
• vascular phenomena - either major arterial emboli, septic pulmonary infarcts, mycotic aneurysm, intracranial hemorrhage, conjunctival hemorrhages, or Janeway lesions
• immunologic phenomena - either glomerulonephritis, Osler's nodes, Roth spots, or rheumatoid factor
• microbiologic evidence - positive blood culture but does not meet a major criterion as noted above¹ or serological evidence of active infection with organism consistent with IE
• echocardiographic findings consistent with IE but not fulfilling a major criteria

• urinalysis
  • proteinuria, leuks and micro-haematuria occur in 50-60% of cases
• FBE, U&E, ESR, CRP
  • anaemia of chronic disease is present in 50-80% of patients with SBE⁶⁾
  • leukocytosis is common in acute endocarditis but not in SBE
  • leukopenia is present in 5-15% of SBE
  • ESR:
    • is said to be raised in more than 90% of cases although in this Swedish study⁷⁾ it was raised in only 72%
    • mean ESR in those who survive is ~78mm/hr, while those with aggressive disease due to Staph. aureus tend to have a paradoxically lower mean ESR of 40mm/hr
  • CRP:
    • was raised in 96% in this Swedish study⁸⁾ with a median level of 90mg/L
    • falls more quickly than ESR in response to Rx and may be a more useful indicator of infective complications following Rx
• blood cultures:
  • take 3 sets of blood cultures over several hours, preferably from different sites and prior to starting empirical antibiotics
  • a febrile state is not required as continual intravascular shedding of bacteria occurs
  • if no prior antibiotics given, one of the first two blood cultures will yield a positive result 90% of the time
• ECG
  • usually non-specific changes if any, although new 1st degree heart block or interventricular block suggests aortic valve disease and is a poor prognostic sign
• CXR
  • may show evidence of cardiac failure if left valve disease or pulmonary abscesses or emboli if tricuspid disease
• trans-thoracic echocardiography
  • 60-80% sensitivity for valvular lesions
  • visible vegetations suggests worse prognosis
• trans-oesophageal echocardiography:
  • 90% sensitive for valvular lesions
• adjunctive investigations not routinely performed in ED:
  • rheumatoid factor:
    • may be elevated in SBE - 50% of cases with duration > 6 weeks compared to 6% of cases with duration < 6 weeks
    • titer usually falls with Rx
  • serology for atypical organisms if culture negative

• a UK study of 220 patients published in 2002⁹⁾ showed:
  • duration of illness before admission, age, sex, valve infected, infecting organism, and left ventricular function were not predictors of adverse mortality
  • abnormal white cell count, serum albumin concentration, serum creatinine concentration, or cardiac rhythm, the presence of two major Duke criteria, or visible vegetation conferred a poor prognosis