Monte Minute‎ > ‎Monte Minute‎ > ‎

Endocarditis - 9/18/2012

posted Sep 18, 2012, 12:28 PM by Rohit Das   [ updated Dec 27, 2012, 6:51 AM by Purnema Madahar ]

Infective endocarditis (IE), an ever formidable topic to tackle, was the topic of discussion for intern morning report today. I’m going to do my best to review some key aspects of this important infectious issue.

  • Regarding native valve endocarditis, the incidence is around 2-6 cases per 100,000 person-years, but this number is not very accurate due to varying case definitions, different institutional definitions, and alterations made to the clinical criteria, which we’ll discuss below.
  • Though still favoring the younger crowd in patients with an IVDU history, the epidemiology has changed over recent years, with increased incidence in older adults. More than one-half of cases are in patients over 60 years old, and this is probably due to the decline in importance of RHD, and the increasing longevity of the western population overall.
  • Major risk factors include IVDU (incidence of 150-2000 cases per 100,000 person-years), prosthetic heart valves and structural valve disease (MVP being the most common in the developed world, whereas RHD remains by far the most common reason in most of the world). Hemodialysis and DM are other important associated risk factors. HIV, as an independent risk factor, is controversial.
  • From an etiologic standpoint, Streptococcus (particularly viridians and Bovis) species and Staph. aureus remain by far the most common causes, representing 70-80% of all cases in most age groups. Other common bugs include Enterococcus, Gram negative bacilli, and the “HACEK” organisms. Table 1 of the attached review article provides a nice overview.

Clinically, endocarditis can present with a myriad of complications. The most common cardiac complications include heart failure and perivalvular abscesses (with or without conduction issues). There are a number of embolic manifestations of endocarditis, which can affect basically every organ system in the body – lungs, brain, kidneys, bone…you name it. The review article provides a nice pictorial overview of physical exam signs we typically associate with endocarditis – including Janeway lesions, Osler nodes, conjunctival petechiae, and splinter hemorrhages. Importantly, there are defined risk factors for the risk of embolization:

·         Certain Streptococcus species (Bovis in particular) and Staph are more commonly associated with embolic disease. The presence of a large vegetation (>10mm, and those seen on both on TTE/TEE as opposed to just TEE) is also a risk factor.

·         Embolic disease is much more common with left-sided lesions, particular lesions of the anterior leaflet of the mitral valve. In one case series analysis, 25% of patients with mitral disease, and 37% of patients with specifically anterior leaflet disease, had embolic issues.

We commonly use the Duke’s criteria as a tool to help us with the clinical diagnosis of IE. Table 3 of the attached review article provides the most recent, modified, version of the Duke’s criteria – which was initially developed 1994 by people at…Duke! Some words about the accuracy of this commonly used tool:

  • In one retrospective case series of 52 patients with “rejected” endocarditis by Duke’s criteria, 1 patient had evidence of vegetations on autopsy, and 3 patients were reclassified as having “possible” endocarditis – overall giving an NPV of 92%. Though this study does have several limitations, it is reflective of the Duke’s criteria to fairly accurately “rule-out” IE in those with a low pre-test probability.
  •  In another study, 93 patients with pathologically (at surgery) confirmed IE were used to assess the ability of Duke’s Criteria as a clinical tool. Of note, 24% of this patient pool was classified as having “possible” IE by Duke’s Criteria, leading to an overall sensitivity of 76% in those with “definite” or “possible” IE. This serves to reinforce that those patients within the grey, “possible,” IE zone, Duke’s Criteria is not as sensitive. Since many of the “possible” cases were due to Q fever, Duke’s criteria was appropriately modified in 2000.
  • Also, regarding TTE vs TEE, the general rule is that TTE is not a sensitive test (60-70%), but does maintain a very high specificity (98%). TEE increases the sensitivity (80-95%, NPV of around 92%), and is particularly useful for mitral lesions, prosthetic valves, and detecting signs of extravalvular extension.

Ok – now that we’ve touched on the epidemiology, clinical aspects and diagnosis of IE, let’s talk a bit about management and prognosis. The evidence for surgical management of endocarditis is limited for specific clinical circumstances, and has been assessed via observational and retrospective studies (no randomized trials).

·      Heart failure is the leading cause of mortality in patients with endocarditis, and that mortality seems to significantly improve with combined medical and surgical therapy. Valvular CHF from endocarditis, without surgery, carries a mortality of 50-80%, as compared to 10-30% with surgery. Even in the absence of overt signs of heart failure, AR or MR with signs of hemodynamic compromise (i.e., elevated LVEDP, pulmonary HTN, etc.) should also be considered for early surgery. Operative mortality is directly related to the degree of heart failure – so, the earlier surgery is done, the better the outcome.

·         Endocarditis with resistant organisms is another reason for surgery – positive blood culture 5-7 days into therapy is generally a good indication that medical therapy isn’t working. Pseudomonas, fungi, Enterococci, Coxiella are some of the typical organisms that are usually not responsive to just antibiotics.

·         Regarding surgical management in patients with embolization and large vegetations, the data is not as great. Surgery in patients with recurrent embolic episodes and/or vegetations that are >10mm in size are relative indications for surgical management.

·    Extravalvular extension, manifesting as conduction abnormalities, pericarditis, or perivalvular abscess, is another good indication for surgery. The mortality associated with these issues is around 40-50% with medical therapy alone, and drops to <25% when medical therapy is combined with surgery.

The bottom line though, from the data above, is even with surgery endocarditis has a VERY HIGH in-hospital mortality of around 20%. Of note, left-sided endocarditis has a much worse prognosis that right-sided endocarditis. In one retrospective analysis of 220 patients with tricuspid endocarditis, the overall mortality was <10%, with vegetation size >2cm and a fungal etiology as the only major risk factors associated with an increase mortality.

The attached review article gives a decent synopsis. However, this is a pretty hefty topic and worth going into the textbooks to get a really good handle on it…Keep on reading!!!

Infective Endocarditis In Adults
Mylonakis et. al., NEJM 2001, Volume 354 (18): 1318-30

Dodds III et. al., Am J Cardiology 1996, Volume 77: 403-07

Habib et. al., Am J Cardiology 1999, Volume 33 (7): 2023-29