Endgames Picture Quiz

An ECG that changed in a febrile patient

BMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f585 (Published 06 February 2013) Cite this as: BMJ 2013;346:f585

This article has a correction. Please see:

  1. Thomas Joseph Ford, cardiology registrar1,
  2. Greg Cranney, director of cardiology1,
  3. Annette Li, final year medical student2
  1. 1Department of Cardiology, Prince of Wales Hospital, Sydney, NSW 2031, Australia
  2. 2Kensington Campus, University of New South Wales, Kensington, NSW 2052, Australia
  1. Correspondence to: T J Ford thomasjford{at}gmail.com

A 50 year old man with a history of injecting drug misuse was admitted as an emergency with a reduced level of consciousness and left arm weakness after five days of swinging fever, rigors, and headache. His medical history included hepatitis C infection, paroxysmal atrial fibrillation, and a recent hospital admission with discitis and psoas abscess. His temperature was 38.4°C, but he was haemodynamically stable and alert. He had a soft early diastolic decrescendo murmur along the left sternal border. Splinter haemorrhages were present, along with a small digital infarct of the left third toe and mild upper motor neurone pattern weakness in the left arm. Laboratory investigations confirmed a normocytic anaemia and leucocytosis and blood cultures grew meticillin resistant Staphylococcus aureus. Computed tomography of the brain showed a hypodense area in the left frontoparietal area consistent with infarction.

Although the admission electrocardiogram (ECG) was normal, ECG changes developed early in the admission (fig 1).


Fig 1 Lead II rhythm strip on a 12 lead electrocardiogram


  • 1 What is the unifying diagnosis?

  • 2 What is the main abnormality on the electrocardiogram?

  • 3 What is the next definitive cardiac investigation?

  • 4 How should this condition be managed?


1 What is the unifying diagnosis?

Short answer

Infective endocarditis with an aortic root abscess.

Long answer

Infective endocarditis with an aortic root abscess. The two most important local complications of this condition are spread of infection, which may cause perivalvular abscess and cardiac failure through valvular insufficiency. Local spread of infection has a predilection for the weakest portion of the aortic valve annulus near the membranous septum and atrioventricular node (fig 2). This accounts for the propensity of an aortic root abscess to disturb conduction.1


Fig 2 Schematic representation of the atrioventricular bundle of His. Note the proximity of the aortic valve annulus to the membranous septum and conduction system running from the atrioventricular node (arrow)2

Periannular extension and abscess formation complicate 30-40% of cases of native valve endocarditis and are more common in prosthetic valve endocarditis. It is important to recognise these patients because they have a higher rate of systemic embolisation, heart failure, and mortality.3 4 The two most important independent risk factors for developing perivalvular extension are aortic valve involvement and injecting drug use. Infection with S aureus also makes abscess formation more likely.5

Clinicians should suspect perivalvular extension of infection when patients with infective endocarditis develop new conduction disturbance, have persistent fever despite appropriate antibiotics, have new onset valvular regurgitation, or develop pericarditis.6

2 What is the main abnormality on the electrocardiogram?

Short answer

First degree atrioventricular block, which suggests the development of an aortic root abscess.

Long answer

The rhythm strip is consistent with the changes on 12 lead ECG and demonstrates first degree atrioventricular block. This is defined as a delay or interruption in the transmission of an impulse from the atria to the ventricles owing to an anatomical or functional impairment in the conduction system that results in a PR interval longer than 200 ms or greater than 210 ms at slow heart rates (fig 3).


Fig 3 Lead II rhythm strip on a 12 lead electrocardiogram showing fixed prolongation of the PR interval to about 400 ms (as measured from the large to the small arrow)

3 What is the next definitive cardiac investigation?

Short answer

Transoesophageal echocardiography (TOE).

Long answer

TOE is the imaging modality of choice for diagnosing aortic valve endocarditis and specifically identifying the presence and extent of a valvular abscess (fig 4).7 8 Early detection of abscess formation can be challenging, but it is an important diagnosis to make before the onset of major tissue destruction.9


Fig 4 Images from the patient’s transoesophageal echocardiography (TOE). (A) Long axis view of aortic valve in cross section. (B) Short axis basal view through the trileaflet aortic valve. The blue arrows indicate a large vegetation. The red arrow in (B) highlights the location of the posterior aortic root abscess identified at surgery; importantly, this is not evident on the TOE study

TOE is more invasive than transthoracic echocardiography (TTE), but it is much more sensitive in detecting abscesses (76-100% v 18-63%).10 11 Importantly, a negative TOE result does not exclude abscess formation. In our patient, one TTE study and two subsequent TOE studies did not clearly demonstrate an abscess of the aortic root. Posterior root abscesses form adjacent to the left atrium and are usually obvious on TOE; however, our case shows that small abscesses (particularly those on the tricuspid side of the aortic valve) can be more difficult to visualise. In this case, the development of new atrioventricular block on ECG was a crucial indication of aortic root abscess. Missed abscesses are associated with delayed surgical intervention, and this may have important effects on clinical outcome.12 New atrioventricular block has an 88% positive predictive value for abscess formation but a low sensitivity (45%).13 There is an increasing role of axial imaging techniques, including cardiovascular magnetic resonance imaging and computed tomography, in providing complementary information on the site, structure, and extent of perivalvular involvement.14 15

4 How should this condition be managed?

Short answer

Intravenous antibiotics (based on BSAC guidelines) should be combined with aortic valve replacement and abscess debridement. Atrioventricular block in this setting requires urgent cardiac pacing.

Long answer

Early surgical debridement and valve replacement should be combined with intravenous antibiotics under the guidance of a microbiologist. Our patient was managed according to BSAC guidelines with early intravenous meropenem and vancomycin until his blood culture sensitivities were known.16 Compared with antibiotics alone, early surgery in patients with large vegetations reduces the risk of embolic events and death from any cause by decreasing the risk of systemic embolism.17 Perivalvular extension of infection manifesting as conduction disease or fistula formation is a strong indication for surgical intervention. Although few randomised clinical trials have evaluated management options for patients with complicated infective endocarditis, observational studies suggest a mortality of 40% in medically managed patients.18 Operative mortality increases with the patient’s age, staphylococcal infection, renal failure, and fistula formation.19

Patients require regular reassessment with ECG and TOE to monitor for progression. C reactive protein (CRP) is now included in the minor criteria of the modified Duke’s score for diagnosis of infective endocarditis, and it can also be monitored to assess treatment response.20 Evidence on the role of CRP monitoring in patients with perivalvular extension of infection is limited.21 Our patient had a CRP of 250 mg/dL (reference value <3 mg/dL; 1 mg/dL=9.52 nmol/L) at presentation, which dropped to less than 50 mg/dL one week after surgery.

A temporary right ventricular pacemaker must be inserted in the presence of new or worsening atrioventricular block because sudden asystole is a real possibility when the interventricular septum is affected. The pacemaker should be inserted when ongoing infection is excluded to minimise the risk of pulse generator or endocardial lead infection.22

Patient outcome

Our patient’s condition deteriorated, with numerous embolic phenomena including stroke, embolisation to the coronary circulation with resultant myocardial infarction, and cardiac arrest. He developed abdominal pain secondary to splenic and bilateral small renal infarcts. Despite the lack of a definite aortic root abscess on serial TOE studies, the development of new atrioventricular heart block raised clinical suspicions that were confirmed during open heart surgery, where emergency surgical valve replacement was combined with debridement of an aortic root abscess. Afterwards he needed prolonged intravenous vancomycin and insertion of a permanent pacemaker. Nevertheless, after six weeks as an inpatient, he was discharged into an outpatient drug rehabilitation programme with preserved cardiac function and no serious functional limitations.


Cite this as: BMJ 2013;346:f585


  • Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Not commissioned; externally peer reviewed.

  • Patient consent obtained.