Intern Report 6.5

Case Presentation by Dr. Laura Smylie

History of Present Illness:

50 year old male complaining of being “sick” for the past three to four days. He stated he began feeling generally poorly with an associated pain in his left shoulder. He states the pain is sharp in nature and he denies trauma to the area. The pain radiates to the axilla and is sharp and pulling in that area. He denies chest pain. He has been feeling hot and cold but has not taken his temperature at home. He has been diaphoretic. He has a mild nonproductive cough. He denies abdominal pain, nausea or vomiting. He denies shortness of breath.

Review of systems negative except as detailed in the HPI.

Past Medical History: No hypertension

Past Surgical History: none

Medications: None

Allergies: No known drug allergies

Family History: Negative for hypertension.

Primary care physician: None

Social History: Positive for tobacco use and IVDA. Denies alcohol use.

Physical Exam:

Vitals: T 39.5 rectal, HR 98 regular, BP 180/80, RR 24, Pulse ox 97% room air

Constitution: Thin male, alert, in no apparent distress.

Head: Normocephalic, atraumatic

Eyes: Pupils equal, round and reactive to light. Sclera nonicteric. Conjunctiva pictured below.

ENT: Mucous membranes slightly dry. No erythema in the posterior pharynx or tonsillar exudates.

Neck is supple. Trachea midline. No JVD.

Respiratory: Normal respiratory effort. Clear inspiration. Few scattered rales in bilateral bases, no wheezing. Good air exchange.

Cardiovascular: Regular rate and rhythm. 3/6 systolic murmur at the sternal edge. No gallop. Pulses were equal in the extremities.

Chest wall: Nontender.

Abdomen: Soft, nontender, nondistended. No rebounding or guarding.

Musculoskeletal: Right side extremities unremarkable. He has full range of motion at the right shoulder. No erythema or warmth over the left shoulder. No swelling, no tenderness of the clavicle. No tenderness over the AC joint. Some pain with passive motion but full range of motion present.

Skin: Warm and dry. No rashes or lesions.

Neurologic: Alert and orient to person, place and time. Cranial nerves II-VII intact. Gait is narrow and steady. Normal Speech, Strenght 5/5 upper and lower extremities both proximal and distal muscle groups

Picture 1:


 Medical Course:

IV access was established. Blood was sent for analysis. Blood cultures were sent and are pending.

CBC: 7.4\14.9/139

Electrolytes: 139 | 92 | 27

3.5 | 22 | 1.0


1)Which one test is most likely to solidify the diagnosis:

a) ECG

b) 2 view chest x-ray

c) Echo – usually needs to be TEE (where do blood culture fall in the diagnostic workup?)

d) This is a clinical diagnosis

2) Given this patients most likely diagnosis, what is the most commonly found physical finding :

a) Murmur

b) Conjunctival hemorrhage/petechiae

c) Splenomegaly

d) Fever

3) The most appropriate initial empiric therapy for the patient in this case is:

a) Penicillin G + nafcillin

b) Vancomycin alone

c) Vancomycin + gentamicin

d) Vancopime





  1. C. echo
  2. D. fever
  3. B vancomycin aloneNICE guidelines published in 2008 made an even more radical departure from the past.22 They do not recommend antibiotic prophylaxis for dental, or non-dental procedures (eg, respiratory, gastrointestinal, and genitourinary)




This patient’s history of IV drug abuse and fever should bring infective endocarditis near the top of the differential diagnosis. The physical exam finding of cardiac murmur should further heighten suspicion. The finding of conjunctival petechiae should reinforce this suspicion – although this vascular finding is not pathognomonic for endocarditis, it is one of the minor criteria that can help make the diagnosis. The most common physical finding is fever – up to 30% of those with endocarditis do not present with murmur, although the vast majority do have a murmur at some point in their disease course.






Conjunctival hemorrhage – septic emboli




Infective endocarditis is an infection of the endocardial surface of the heart, most commonly one or more of the cardiac valves, but can also include a septal defect or the mural endocardium itself. The pathophysiology of this disease is classically a valve with mechanical damage at which a sterile thrombus initially forms, which then becomes infected when a subclinical bacteremia is induced (ie, with IV drug injection, after dental procedures, endoscopy, cystoscopy, etc). IV drug users may have damaged valves secondary to contaminants in drugs such as talc. Valves also can become damaged with age secondary to degenerative changes.




Incidence and Risk:


The incidence of infective endocarditis varies widely depending on which part of the country in which you practice – it is very dependent on the prevalence of high risk indiviuals in the specific population. What is clear is that over 50% of those diagnosed with endocarditis are over the age of 60.


Presentation is most often a febrile patient who is high risk. Those at high risk include:


  • IV drug users (as in the case presented here),


  • Patients with prosthetic heart valves,


  • Unrepaired cyanotic congenital heart disease (which includes palliative shunts and conduits),


  • Completely repaired congenital heart defects with prosthesis during the first 6 months after the procedure,


  • Repaired congenital heart disease with residual defect at or adjacent to the site of the prosthetic device, and


  • Cardiac valvulopathy in a transplanted heart.




The Duke Criteria are widely accepted to stratify patients with suspected infective endocarditis into three categories: Definite, Possible, and Rejected. These criteria have been validated multiple times after the original study of 69 patients. Although there have been proposed updates to these criteria (including any s. aureus bacteremia in high risk patients as a major criteria), the original Duke Criteria are 95% sensitive and 99% specific.


Major Criteria include:


  • Positive blood cultures (of typical pathogens) from >=two sites
  • Definitive evidence of endocardial involvement by echo, including endocardial vegetation, paravalvular abscess, new partial dehiscence of prosthetic valve or new valvular regurgitation
  • *It has been proposed to add S.Aureus positive bacteremia


Minor Criteria include:


  • Predisposition defined as a predisposing heart condition (as described above) or IVDA
  • Vascular phenomena including arterial emboli, septic pulmonary infarcts, mycotic aneurysm, conjunctival hemorrhages, or Janeway lesions.
  • Immunologic phenomena including Osler’s nodes, Roth’s spots, and rheumatoid factor
  • Microbiologic evidence, ie single positive blood culture
  • Echocardiogram findings “consistent with endocarditis but do not meet major criteria” *(with the advancement of technology and widespread availability of TEE, it has been proposed to take this off)




 Roth spots – retinal hemorrhage with central pallor





Splinter hemorrhage – septic emboli seen under nails




Osler’s nodes – Painful lesions seen on finger/toe pads (caused by immune complexes)


Janeway Lesions – nontender lesions on the palms/soles (septic emboli)




Mycotic aneurysm of MCA seen on noncontrast head CT – (although any artery can be affected, intracranial arteries are most often involved followed by visceral arteries and then arteries supplying extremities.)



Definite Endocarditis – any one of the following combinations of clinical findings:


  1. Two major clinical criteria
  2. One major and any three minor criteria
  3. Five minor criteria


Possible Endocarditis – any one of the following combinations of clinical findings:


  1. One major and one or two minor criteria
  2. Three minor criteria


Rejected Endocarditis – any one of the following are present:


  1. A firm alternative diagnosis
  2. Resolution of clinical manifestations occurring after <= 4 days of antibiotic therapy
  3. Clinical criteria for possible or definite endocarditis is not met.




Antimicrobial therapy:


Most common pathogens: S. aureus (32%); s. viridans (18%); enterococci (11%); coagulase negative staph (11%), Streptococcus bovis (7%); other strep (5%); non-HACEK gram-negatives (2%); fungi (2%); HACEK – Haemophilus aphrophilus, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens, Kingella kingae (2%)




Choice of empiric antibiotics depends on the patient’s main risk factor for endocarditis. For patients with a native valve who are not IV drug users, the best choice is:


  • Penicillin G 5 million units IV q4h + nafcillin 2g IV q4h


  • Vancomycin 15mg/kg IV q 12h


  • Gentamicin 1mg/kg IV q8h


For patients with a native valve who are IVDA:


  • Vancomycin 15mg/kg IV q12h


For patients with prosthetic valve:


  • Vancomycin 15mg/kg IV q12 + Gentamicin 1mg/kg IV q8h




  •  Heart failure is linked to infective endocarditis and is a complication associated with poorer outcomes. Whether the patient’s CHF is a result of endocarditis or a separate preceding entity, patients tend to have increased mortality.
  • Extension of endocarditis beyond the valve’s annulus is also associated with a significantly higher mortality.
  • Mycotic aneurysms, if present intracranially, are associated with a 60% mortality.






Prophylaxis to prevent Endocarditis during procedures in the ED:


  • NICE (National Institute for Health and Clinical Excellence) guidelines published in 2008 do not recommend antibiotic prophylaxis for dental, or non-dental procedures (eg, respiratory, gastrointestinal, and genitourinary)
  • The American Heart Association also published guidelines against prophylaxis even in high risk populations in 2007.



Clinical Pearls:

  • In patients presenting with fever and IVDA or some cardiac history, keep infective endocarditis high on the differential and look for clues on physical exam to confirm diagnosis
  • Infective Endocarditis can be definitively diagnosed in the ED! An echo positive for valvular vegetation and significant physical findings with and IVDAer and fever.
  • However, ALWAYS SEND AT LEAST TWO BLOOD CULTURES (some sources say three from three different sites) – this will help our colleagues tailor the therapy appropriately once the pathogen has been identified and susceptibilities are available.


“Diagnosis, Antimicrobial Therapy, and Management of Complications: A Statement for Healthcare Professionals From the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease, Council on Cardiovascular Disease in the Young, and the Councils on Clinical Cardiology, Stroke, and Cardiovascular Surgery and Anesthesia, American Heart Association—Executive Summary: Endorsed by the Infectious Diseases Society of America.” Circulation. 2005; 111: 3167-3184

Garg SJ, Sivalingam A, Bolling J, Goldberg R, Sivalingam J, Magargal L. Ocular Abnormalities in Acquired Heart Disease. Duane’s Ophthalmology. 2006. Vol 5.

The Hand in Pathology. Stanford School of Medicine.

Lee W, Mossop P, Little A, Fitt G, Vrazas J, Hoang J, Hennessy O. Infected (Mycotic) Aneurysms: Spectrum of Imaging Appearances and Management. RadioGraphics, 28, 1853-1868. November 2008.

Li JS, Sexton DJ, Mick N, Nettles R, Fowler VG Jr, Ryan T, Bashore T, Corey GR. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clinical Dis. 2000 Apr; 30(4):633-8.

Infective Endocarditis.  Rosen’s Emergency Medicine.  7th edition.  2010.  Print.


9 Responses

  1. 1) my answer is _ However I want to argure the wording of this question.- Endocarditis is a clinic diagnosis (D), we start treatment for this condition on clinical exam. TEE and blood cultures are to identify what type of endocarditis it is. When you put “solidify the diagnosis” I took it as identifying the organism or type pf endocarditis but the diagnosis was already made on clinical exam.


  2. 1. blood cultures should be obtained in ED prior to administration of ABx. 3 sets from 3 diff sites. 2. 3.

  3. 1)
    3) **

    ** Most people will say that vancomycin alone is sufficient coverage for native valve endocarditis. BUT in IVDA, specifically in Detroit, up to 10% will be infected with pseudomonal endocarditisand so IVDA should be started on Vanco and cefepime until cultures grow out and the source is confirmed.

  4. 1) ; two to three sets of blood cultures (from separate venipuncture sites) should be obtained prior to administration of antibiotics.


    3) ; until an organism is identified, vancomycin alone is the recommended therapy

  5. I’m a little confused about the answer to question 3. It seems that this patient needs more coverage than vanc alone.

    In your excellent explanation, you present the Duke Criteria for infective endocarditis. The patient in the case has only two minor criteria (vascular phenomena and IVDA), which means that he does not yet meet criteria even for possible endocarditis.

    I agree that the clinical situation is very suspicious for endocarditis. However, at the point the case presentation ends (cultures sent, echo not yet done), all you really have is an IV drug user with 3/4 SIRS and unknown source. In that clinical context, I would argue that, of the choices given, the best answer to the question of “the most appropriate initial empiric therapy” would be vanc + cefepime. This provides the most broad coverage and is consistent with our institutional guidelines.

    Could you help clarify why the best approach is vanc alone?

    Thanks for a good case!

    • Sean I think this is a good example of a couple things to remember.

      First, when taking a test, do not base your answers on what we do here, rather what has been proven to be accepted on the national scale. Laura hit the current literature right on the head and how many people are now practicing when she talks about Vancomycin alone in the IVDA population. Would this be fully acceptable here? No. Would this be the correct answer on your board exam? Yes.

      The second thing alludes to Desean’s response. This is NOT simply IV drug user with 3/4 SIRS and unknown source. This IS endocarditis until proven otherwise. You can speculate on the source until the Lions win a playoff game or you can admit this pt to the ICU, floor, wherever they go and medicine physicians from all over the world can argue major and minor criteria based on whatever literature they want or know. However, you have a given presentation that is consistent with a clinical scenario and you really have no choice but to diagnose it and treat for it unless you find something else during your limited work-up time to explain this presentation.

      • I agree with everybody that this is probably endocarditis. For the sake of argument, however, I also want to point out our cognitive dissonance in making the findings support our hypothesis.

        We can’t really say that this IS endocarditis until proven otherwise and in the same breath also cite the Duke Criteria. The corollary to Deshon’s comment that endocarditis is a clinical diagnosis is that the clinical diagnosis should be based on the agreed-upon clinical definition. Based on that definition (ie. using the most liberal set of criteria–modified Duke), this patient only has possible endocarditis. That means it’s possibly something else, not “endocarditis until proven otherwise.” If it’s possibly something else and we start with coverage that is too narrow, there is no doubt that we increase morbidity and mortality. In that case, the national guidelines AND the local practice would demand we start broadly.

        Will we eventually get an echo or blood cultures that push us into the “definitive endocarditis” range? Almost certainly.

        Will the vanc+cefepime that I started be narrowed in the ICU, floor, or wherever they go? Probably.

        What is the downside to starting broad coverage until that happens, in accordance with the national (and local) guidelines? Very minimal.

        What is your acceptable miss rate for everything else? If the emergency physicians have premature closure on gram-positive endocarditis, and it ends up being almost anything else, we did the patient a tremendous disservice AND we fell victim to the doublethink of “classic textbook presentation” that doesn’t actually meet our agreed-upon definition.

  6. And just to show that I’m not just a proponent of throwing around the big guns all the time:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: