Due to the invasive nature of bronchoscopy, data on the presence of pathogens in the lower airways are very scarce. A unique set of 573 BAL fluid samples was collected in children between January and December in 2018, in Shenzhen, Southern China. This allowed us for the first time to study the presence of pathogens in the lower respiratory tract on a relatively large scale. Furthermore, this study has enabled us to compare the routine clinical diagnostics with a new PCR-based diagnostics approach.
PCR-based diagnostics resulted in detection of respiratory pathogens in 76.1% of all cases. A similar percentage (78.7%) as was previously found in a smaller study recently conducted in Zhengzhou (Northern China) [9], but higher than previously found in studies performed in Suzhou and Shanghai (Eastern China) [3, 16]. While the routine clinical diagnostics approach in this study resulted in the detection of respiratory pathogens in 36.1%. Which clearly indicates that PCR-based detection is far more sensitive than the routine clinical diagnostics as currently applied. PCR-based detection also revealed that co-infection was found in 16.4%. Most prevalent of all co-infections were viral-bacterial co-infections (11.2%). The co-infection rate was lower than that in Zhengzhou (30.6%) and similar to that in Shanghai (11.4%), also determined by PCR-based detection.
Interestingly, HADV and M. pneumoniae showed similar clinical characteristics, including the fact that 90% of patients infected with these pathogens developed fever. However, the median age of children infected with HADV, was significantly younger than children infected with M. pneumoniae. Strikingly, this study confirms that HADV is a common pathogen in Southern China (Guangzhou, Hunan) as high prevalence was earlier found in other studies conducted in the same region [17–18]. This was not the case for other areas in China, which is concluded based on studies conducted in Suzhou and Chongqi [19–20]. However, M. pneumoniae seems highly prevalent in different parts of China. This striking regional difference within China demands for further study.
HRV was detected in 13.96% of the patients and was the most common pathogen in patients with severe pneumonia. HRV detected in co-infection with other pathogens was higher than as single pathogen.
H. influenzae and S. pneumoniae were the most common bacterial causes of LRTIs. The qPCR method for detection of bacterial pathogens was more sensitive than bacterial culture.
Strikingly not RSV but HADV was the most important viral pathogen in this study as opposed to previous studies [11, 17]. This could be explained by, method of detection, study population and geographical location.
This study has some limitations. Firstly, none of the diagnostic methods were able to detect fungal pathogens, while fungi might be important pathogens of LRTIs in part of the patients. Secondly, we cannot exclude that we might have missed some of the RNA viruses because of RNA degradation during storage and sample processing. Thirdly, the amount of fluid retrieved and sampling conditions could have influenced the detection rate. Fourth, in 10.3% of the children the BAL sample was taken more than 1 week after admission because of different reasons. A combination of the above mention factors could possibly explain why in 23,9% of the cases no viral or bacterial pathogen was found. Nevertheless, the molecular methods used in this study significantly increased the diagnostic accuracy (from 36.1 to 76.1%).
In conclusion, lower respiratory tract infections are complex with a diverse clinical phenotype and caused by many different pathogens. We found that viruses were the major pathogens detected in the BAL fluids. Viruses and bacterial pathogens were detected in younger children. This study gave us insight into the etiology of LRTI in children from different ages from Southern China and will help to improve the diagnosis. GeXP-based multiplex PCR with a much broader range of pathogen detection which clearly showed to be superior showing promise for the near future of LRTI diagnostics.