TGH "Case of the Week" July 14-18, 2008

30F with known HIV (CD4 unknown, not on treatment) presents with acute onset fever, chills, rigors, and cough with sputum. On exam she is febrile, tachycardic, hypotensive, hypoxemic on room air requiring hi-flow oxygen to maintain saturations >90% and in moderate respiratory distress.

The chest x-ray is taken and appears below:



Questions:

1) What is the most likely microbiological diagnosis?
2) What is the differential diagnosis?
3) What tests would you order to make your diagnosis?
4) What empiric treatment would you initiate in the ED?

BONUS: What "management strategy" would you employ in the ED and what journal was the study published in?

Thank you to the people who submitted answers. The "contest" is closed for this week. The winner will be notified in person.

"Answers"

1) The most likely etiologic agent to cause a lobar pneumonia in a patient with HIV is still streptococcus pneumoniae. In fact, patients with HIV are at an increased risk of getting pneumococcal infections and should all be vaccinated with the polysaccharide vaccine.

2) The differential diagnosis includes lobar pneumonia with the other organisms of community acquired pneumonia including haemophilus influenzae, moraxella cattharalis, and Staphylococcus aureus. Of particular concern in a patient who is rapidly deteriorating is community acquired MRSA necrotizing pneumonia.

Other causes would include Legionella pneumophillia and if risk factors such as underlying structural lung disease enteric gram negative organisms.

Upper lobe pneumonia should prompt concern for tuberculosis, although in this case the acquity and sepsis-syndrome argue more strongly in favor of bacterial pneumonia.

PCP tends not to be lobar and consolidative and the classic x-ray appearance looks more like bilateral peri-hilar interstitial infiltrates.

Fungal pneumonias like cryptococcal pneumonia, blastomycosis, invasive aspergillosis are unlikely and would not generally have this radiographic appearance and acute presentation.

3) Blood cultures should be sent ASAP (ideally before antibiotics) as they may be positive in up to 25% of cases of pneumococcal pneumonia. Sputum cultures should also be sent for conventional culture, legionella culture and TB culture. I would not recommend sputum for PCP in this case.

Urinary antigen detection for pneumococcal antigen is used in some centres (not here). Urinary antigen testing for legionella can help make this diagnosis.

CD4 testing in acute illness may not be helpful as the acute illness could cause a decrease in the counts; however, it would not be unreasonable as if the CD4 count was >200 rare causes are much more unlikely.

4) In a septic patient with community acquired pneumonia you need to cover broadly for the most likely pathogens. There are many ways to do this and in some cases what you choose will depend on recent antibiotic exposure.

In this case a combination of VANCOMYCIN 1gIV q12h and LEVOFLOXACIN 750mg IV/PO q24h would be appropriate. This would cover MRSA and quinolone resistant pneumococcus, the usual pathogens including the majority of pneumococcus, unusual pathogens like legionella, and many enteric gram negatives.

BONUS: The management strategy to be employed is "Early Goal Directed Therapy in Sepsis"