A cooling effect for antibiotic fever

Testing for concentrations of serum procalcitonin could lead to more accurate infection diagnoses. Reducing antibiotic over-prescribing -- and resistance

By Owen Dyer

Nothing seems to stop doctors who should know better from prescribing antibiotics when they have no proof of bacterial infection. With powerful antibiotics being handed out like candy, it's hardly surprising that methicillin resistant Staphylococcus aureus (MRSA) and other superbugs are doing the hospital rounds with increasing frequency.

The classic scenario is the lower respiratory tract infection. Sputum can be a notoriously treacherous guide in identifying the culprit, because any sputum collected from the lower tract is liable to be corrupted by organisms in the higher tract on the throat swab. The easiest solution is to assume the worst and throw broad-spectrum antibiotics at the problem.

But respiratory bacterial infections do have a common marker. Procalcitonin, is known to be elevated in the serum of people suffering respiratory bacterial infections. A new sensitive assay to detect concentrations of serum procalcitonin has just been introduced. A Swiss team of researchers decided to put it to the test. Their research is published in the February 21 issue of The Lancet.

Two hundred and forty-three patients with suspected lower respiratory tract infections were randomized into a standard treatment group, and a procalcitonin-tested treatment group, at the University Hospital of Basel. The standard treatment group was given antibiotics only when warranted. The 124-strong procalcitonin group, by contrast, was generally only given antibiotics if procalcitonin concentrations exceeded 0.5 µg/L.

Final diagnoses were pneumonia in 87 patients (36%), acute exacerbation of chronic obstructive pulmonary disease(COPD) in 60 patients (25%), acute bronchitis in 59 patients (24%), asthma in 13 patients (5%), and other respiratory conditions in 24 patients (10%). Of 175 patients tested, 141 (81%) had serologic evidence of viral infection. Bacterial cultures were positive from sputum in 51 patients (21%) and from blood in 16 patients (7%).

Overall, 235 out of 243 patients survived their infection. Positive outcomes were equally common in both groups. But patients in the procalcitonin group were only half as likely to have taken antibiotics while recovering.

The 60 patients with COPD were the only group in which there appeared to be no close correlation between high procalcitonin levels and positive bacterial culture. Sixty percent of this group tested positive for bacterial infection on cultural assay. But among those in the procalcitonin group, the researchers took a chance on ignoring these findings and continued to base their prescribing on procalcitonin levels. The gamble paid off, because these patients had outcomes every bit as good as those in the standard treatment group, despite receiving fewer antibiotics.

It appears that COPD patients are always likely to test positive for bacterial infection, but their symptoms were in most cases actually due to viral infection. At the end of treatment, the bacteria might still be there, but their symptoms were gone.

"Procalcitonin guidance substantially reduced antibiotic use in lower respiratory tract infections," commented study author Dr Beat Mller. "Withholding antimicrobial treatment did not compromise outcome. In view of the current overuse of antimicrobial therapy in often self-limiting acute respiratory tract infections, treatment based on procalcitonin measurement could have important implications."