Evaluation of real-time PCR assay for Haemophilus influenzae detection in nasopharyngeal samples from healthy children

Abstract

 Haemophilus influenzae (Hi) colonizes the human nasopharynx and eventually can cause diseases in  susceptible individuals. The Hib conjugate vaccine provides direct protection against the diseases and  also reduces nasopharyngeal colonization by vaccine strain. Hi carriage studies allow the knowledge  of circulating strains and the accurate identification of the microorganism is important to estimate the  effect of the vaccine. Culture is the gold standard method to detect Hi in clinical samples, this is a  specific method, with variable sensitivity and demand more time for the complete identification of the  microorganism. The use of molecular techniques has helped in differentiating Hi from other species of  the Haemophilus spp. due to the high sensitivity and specificity to detect the etiologic agent. Different RTPCR  and PCR assays were developed for the diagnosis of Hi using several biomarker genes such as hpd  and fucK gene. These genes are highly conserved and are present in encapsulated and non-encapsulated  Hi strains, allowing the detection of both variants. The aim of this study was to evaluate the accuracy of  RT-PCR to detect Hi in nasopharyngeal samples of healthy children vaccinated against Hib comparing it  with the culture. A total of 410 nasopharyngeal swabs stored at -70°C in STGG medium were randomly  selected to evaluate RT-PCR assay targeting protein D (hpd#3). Considering the 410 nasopharyngeal  swabs, 161 (39.2%) samples were positive for Hi and 249 (60.8%) had negative RT-PCR. By culture,  166 (40.5%) samples were positive for Hi and 244 (59.5%) were negative. Thus, the hpd#3 RT-PCR  showed a sensitivity of 90.9% (95% CI: 85.3 - 94.7) and a specificity of 95.9% (95% CI: 92.4 - 97.9)  when compared to culture to detect Hi in nasopharyngeal swabs. The results showed no significant  difference (McNemar’s chi-square = 0.64 p>0.5) between RT-PCR and culture and the Kappa coefficient  showed an excellent agreement (0.873) between the two techniques. As complementary results, the  same samples was evaluated for the presence of the fucK gene by conventional PCR and only 99 (24%)  samples were positive and 311 (76%) were negative for the presence of this gene. Thus, the fucK PCR  showed a sensitivity of 57.8% (95% CI: 49.9 - 65.4) and a specificity of 98.8% (95% CI: 96.2 - 99.7)  when compared to culture. A lower positivity was obtained by PCR compared to culture (McNemar’s  chi-square = 0.64 p<0.05) and the Kappa coefficient showed a moderate correlation (0.605) between the  two techniques. According to the excellent agreement between the results of hpd#3 RT-PCR and and  culture, we have concluded that this technique is an important tool for detecting Hi in nasopharyngeal  samples and can be used in studies of population-based carriage to assess the effect of the Hi vaccine.