Urinary antigen testing for Streptococcus pneumoniae (PAgT) is simple, rapid and can still be used days after initiation of antibiotic therapy or when conventional methods are failing. PAgT is recommended by international guidelines in severe community acquired pneumonia (CAP). The test attains an excellent specificity (>90%) in adults but shows a varying sensitivity (60-85%). We aimed to analyze the PAgT sensitivity in a population with blood culture proven invasive pneumococcal disease(IPD) and to study its performance for the different pneumococcal serotypes.
PAgT (BinaxNOW®, Alere®) was introduced in 2009 in a large secondary care hospital in Ghent, Belgium. PAgT is requested by the attending physician or the clinical microbiologist in case of IPD suspicion. Pneumococci from blood are identified by standard methods (optochin susceptibility and bile solubility) and serotyped by the national reference center. Overall PAgT performance and test sensitivity for different serotypes were calculated.
Over a 9-year period, (2009-17), 235 bacteremia episodes in 234 patients were observed with an average of 26 episodes/year (range 12-36). 31/235 (13%) episodes occurred in pediatric patients. Most prevalent serotypes were 1, 12, 8, 3, 7, 9, 5 and 6 for the whole time period. PAgT was performed in 161/235 (69%), test execution for the individual most prevalent serotypes ranged from 55 to 86%. 99/161(61%) PAgT were positive. PAgT positive results varied according to the most prevalent serotypes: >70% for types 1, 3, 7and 5, 50% for type 9 and <50% for types 12, 8 and 6. From 2014 on, disappearance of serotype 1 and a significant decrease in serotype 7 were observed.
A 70% compliance to the diagnostic algorithm for IPD was observed. PAgT detects C-polysaccharide (teichoic acid) on the pneumococcal cell wall. Differences in concentration for the individual serotypes have been described and may account for the varying sensitivity in our dataset. Introduction of 10/13-valent childhood pneumococcal vaccines(2014) in Belgium has changed the overall serotype distribution, also possibly leading to a shift in PAgT performance. A dynamic validation of PAgT accuracy remains warranted.
A. M. Van Den Abeele,
C. Verfaillie, None
L. Cattoir, None