An assay designed to distinguish between bacterial and viral infections of the lower respiratory tract appears effective and shows promise for helping hospital physicians reduce overprescribing of antibiotics to children, a study showed.
“It is often not possible to differentiate between bacterial and nonbacterial disease on the basis of clinical judgment alone, [so] antibiotics are prescribed almost twice as often as required in children with acute respiratory tract infections in the USA,” wrote Chantal B. van Houten of the University Medical Centre Utrecht (the Netherlands) and associates in a study published in the Lancet Infectious Diseases.
The assay in question is called ImmunoXpert, which uses three biomarkers – tumor necrosis factor–related apoptosis-inducing ligand (TRAIL), interferon-gamma–induced protein-10 (IP-10), and C-reactive protein (CRP) – to determine if a lower respiratory tract infection has a viral or bacterial origin. A total of 777 subjects, aged 2-60 months, were recruited from four hospitals in the Netherlands and two hospitals in Israel between October 16, 2013, and March 1, 2015 (Lancet Inf Dis. 2016 Dec. doi: 10.1016/S1473-3099(16)30519-9).
The patients all had fevers with unidentified sources when they presented, and had a follow-up assessment carried out 28 days after baseline. Blood samples and nasal swabs were collected within 24 hours of presentation for assay analysis. Additionally, every subject was diagnosed as “bacterial” or “viral” by a three-member panel of pediatricians, whose diagnoses were based on the data available from the follow-up assessment and from clinical and laboratory data. The panel diagnosis for each subject was used as the reference standard.
Of the 777 subjects initially recruited, 200 were excluded from the final analysis for various reasons. Of the 577 who remained, the panel diagnosed 435 as having a viral infection and 71 as having a bacterial infection; 71 were deemed “inconclusive.” The panel was unanimous in 354 of these cases, and a majority of the panel (two of the three experts) agreed in 443 of these cases. In unanimous cases, the sensitivity of distinguishing between viral and bacterial cases correctly was 87.8%, with a specificity of 93.0%. The panel’s positive and negative predictive value were 62.1% and 98.3%, respectively.
The assay’s sensitivity rate in distinguishing between viral and bacterial infections was very close: 86.7%, with a specificity of 91.1%, which the authors noted was “promising diagnostic accuracy.” The positive predictive value of the assay was 60.5%, while the negative predictive value was found to be 97.8%.
Regarding the 71 cases that were deemed “inconclusive,” Dr. van Houten and coauthors acknowledged that “such inconclusive cases are inherent to studies without a gold standard, and this was taken into account when calculating the sample size.” Additionally, they noted that follow-up studies should take into consideration the costs of utilizing assay testing like ImmunoXpert, in order to better assess the financial implications that adopting the technology would have on a health care facility.
Nevertheless, the investigators concluded, “our findings [support] the need for implementation research to examine the added clinical utility of ImmunoXpert to diagnose bacterial infection in clinical care for children with lower respiratory tract infection and fever without source presenting at the hospital.”
Funding for this study was provided by MeMed Diagnostics. Dr. van Houten and coauthors did not report any relevant financial disclosures.