Day #99 - PCP Pneumonia

Today we talked about a classic case of PCP pneumonia as a presenting illness for HIV.

Classically PCP presents with progressive dyspnea, usually initially with exertion then at rest, with dry hacking cough and fever. Classically the respiratory exam is relatively normal (sometimes there is a pleural rub) despite often remarkably abnormal chest x-ray. The classic x-ray, shown here, is of bilateral perihilar infiltrates.

LDH is usually elevated but can be normal in up to 10%. LDH can also be elvated in a number of other infections and malignancies.

Exercise induced desaturation is classic and one can do walking oximetry on their patients to see this effect.

The diagnosis is confirmed by direct calciflor stain or silver stain for organisms in the sputum, induced sputum, or BAL fluid.

There is some evidence that high resolution CT can be helpful in excluding the diagnosis and perhaps saving a bronchoscopy.

Treatment is ideally TMP/SMX (dosed 15mg/kg TMP component divided TID/QID) either IV or PO depending on status. Adjunctive steroids should be considered in patients who are hypoxemic on presentation (PaO2 <70,>35), who are in severe respiratory distress, or who have poor cardiopulmonary reserve. Ironically, this Toronto study showed that steroids did not influence outcome but did reduce septra intolerance. However, a cochrane metanalysis showed a NNT of 9 to prevent one death.

The use of HAART in acute PCP remains contraversial. You shouldn't stop someone's HAART if they are on it already (unless it is clearly failing); however, initiation of new HAART is unclear. Some studies show survival benefit, others mortality.

There have been many studies looking at prognostication. Obviously more severe disease at presentation is related to more adverse outcomes. This study found that age, previous history of PCP, treatment other than TMP/SMX, CMV PCR+ in the BAL, and previous use of PCP prophylaxis when diagnosed were all highly associated with mortality.

This simple prognostic score seemed to identify survivors from those who died.

I have previously written about HIV and cryptococcal meningitis here and here. These blogs link to the HIV treatment guidelines as well as the guidelines for the management of opportunistic infections.