Philadelphia protocol

<p id="H4">Philadelphia protocol&#160;—&#160;Another group described an eight-year experience with 747 infants 29- to 60-days of age with a rectal temperature ≥38.2ºC (100.8ºF). Low-risk criteria included patients who were well-appearing with:

&#160;</p>

<ul >

<li>WBC &#60;15,000/microL</li>

<li>Band-neutrophil ratio &#60;0.2</li>

<li>UA &#60;10 WBC/hpf and a negative urine Gram stain</li>

<li>CSF &#60;8 WBC/microL and a negative CSF Gram stain</li>

<li>Chest radiograph lacking an infiltrate if one was obtained</li>

<li>Stool without blood and few or no WBCs on the smear in infants with diarrhea</li>

</ul>

<p>&#160;

All high-risk patients were hospitalized and treated with empiric antibiotics. Low-risk patients (19 percent of the study population) were discharged if follow-up was assured within 24 hours. In contrast to the Boston group, low-risk patients were not treated with empiric antibiotics.

The sensitivity for identifying patients with SBI was 98 percent (95% CI 92-100 percent); the specificity was 42 percent (95% CI 38-46 percent); the positive predictive value was 14 percent (95% CI 11-17 percent); and the negative predictive value was 99.7 percent (95% CI 98-100 percent), suggesting that these criteria did not miss many infants with SBI. However, many well infants underwent excessive laboratory testing, and many were managed as inpatients.</p>