To the Editor: Community-acquired pneumonia (CAP) in adults refers to inflammation of the lung parenchyma involving the alveolar wall. It occurs in the community environment (outside the hospital), including pneumonia caused by pathogens with definite latency after admission. CAP is still the main burden of medical and health resources. With the use of broad-spectrum antimicrobial agents, the aging population, and the emergence of new pathogens, the pathogen spectrum of CAP has changed. In addition to the traditional methods of microbiological cultivation and immunology detection, nucleic acid testing is playing more and more important roles in pathogen detection, which has greatly expanded the detection scope of pathogen. Presently, there is no epidemiological survey of the pathogenic changes in Fujian province. Thus, we carried out retrospective epidemiological studies on CAP pathogens in four centers in different regions of Fujian, and analyzed the effects of age and season on the distribution of pathogens, in order to provide a theoretical basis for regional CAP diagnosis and treatment. This study enrolled the data from 2012 to 2018 except those in 2019–2022 because coronavirus disease 2019 (COVID-19) epidemic may pose certain impact on the distribution of pathogens of CAP.

This study was approved by the Ethics Committee of Fujian Provincial Hospital (No. K2019-12-032). All patients signed informed consent form. From March 2012 to December 2018, patients with CAP in four hospitals (Fujian Provincial Hospital, Wuyishan Municipal Hospital, Putian University Affiliated Hospital, and Sanming First Hospital Affiliated to Fujian Medical University) in various regions of Fujian, a total of 791 people, were included in this retrospective study. The inclusion criteria were as follows: (1) ≥18-year-old adults requiring hospitalization with an onset within 7 days; (2) respiratory symptoms, such as fever, cough, or sore throat; (3) chest radiograph or chest computed tomography (CT) showing pneumonia; (4) signing the informed consent and passing the review of the medical ethics committee of hospitals; and (5) excluding non-infectious lung diseases. Continuous airway samples were collected from the enrolled patients. The first choice for respiratory samples was bronchoalveolar lavage fluid (BALF) or tracheal secretions, and the minimum requirement was deep induced sputum. The collection process should follow sterile principles to avoid contamination of the environment and human body symbiotic bacteria or miscellaneous bacteria. The first collection time was within 48 h of admission. The samples were collected every 2 days; at least two consecutive samples were required. Qualified sputum specimens were screened by microscopy (squamous epithelial cells <10 cells/low power field [LPF], polynuclear leukocytes >25/LPF, or the ratio of the two was <1:2.5). And the respiratory samples were inoculated on blood agar, chocolate, and Chinese blue plates for bacterial culture. NucliSENS® easyMAG™ (bioMérieux ref 280140) (bioMérieux SA, Firenze, Italy) automated nucleic acid extraction equipment was used to obtain nucleic acid from respiratory tract specimens and Bio-Rad fluorescent quantitative polymerase chain reaction (PCR) instrument (model CFX96) (Bio-Rad Laboratories, Inc., Singapore) and Fast-Track Diagnostics (FTD) (Premedical Laboratories Co., Ltd., Malta) kit were used to detect the nucleic acid of pathogens in the specimens. These mainly included influenza A virus (Flu-A), human metapneumovirus (hMPV), human parainfluenza virus (hPIV), respiratory syncytial virus (RSV), adenovirus (ADV), Streptococcus pneumoniae (S. pneumoniae), Klebsiella pneumoniae (K. pneumoniae), Staphylococcus aureus (S. aureus), Mycoplasma pneumoniae (M. pneumoniae), and Chlamydia pneumoniae (C. pneumoniae). Criteria for positive results: (1) The same pathogen was cultured twice in a row and with bacterial colony ≥ +++; (2) There were positive and negative quality controls in the nucleic acid test. When the negative control product had no amplification curve, the positive quality control product had an amplification curve and cycle threshold <35, the test sample had a typical "S" type amplification curve and cycle threshold <35, and it was judged as positive. When the pathogen culture is inconsistent with the nucleic acid detection, the clinical manifestations and treatment response should be considered. All analyses were carried out using SPSS version 23.0 (SPSS Inc., Chicago, IL, USA). Normally distributed quantitative variables were evaluated by Student's t-test, while categorical variables were assessed by Pearson's chi-squared test or Fisher's exact test. The bootstrap method was used to calculate the 95% confidence interval (CI). P-value <0.05 was considered statistically significant.

Among 791 patients, 66.2% (524 cases) were males, and 33.8% (267 cases) were females, with an average age of 64 ± 16 years. The characteristics of patients are shown in Supplementary Table 1, https://links.lww.com/CM9/B479. In respiratory samples, BALF specimens accounted for 13.5% (107 cases), and sputum specimens accounted for 86.5% (684 cases). All respiratory samples were tested by PCR and pathogens were identified in 493/791 (62.3%) patients. Forty-seven cases were culture-positive at the same time, among which six cases were detected with pathogens inconsistent with PCR results. Combined with the clinical manifestations and treatment response of these patients, the cultured strains were considered as colonizing bacteria, and the PCR test results were in line with the clinical manifestations. The main pathogen composition is shown in Supplementary Figure 1A, https://links.lww.com/CM9/B479. And viral infection, bacterial infection, and atypical pathogen infection accounted for 32.2% (255/791), 39.2% (310/791) and 8.7% (59/791), respectively. Flu-A was the most common pathogen with a positive rate of 20.5% (162 cases), followed by S. pneumoniae 12.8% (101 cases), K. pneumoniae 8.6% (68 cases), S. aureus 6.4% (51 cases), and C. pneumoniae 5.2% (41 cases). 186 of 791 patients had mixed infections, accounting for 23.5% of cases [Supplementary Figure 1B, https://links.lww.com/CM9/B479]. The infection of S. aureus, C. pneumoniae, K. pneumoniae, S. pneumoniae, and RSV was mostly co-infected with other pathogens. Among patients infected with S. aureus (51 cases), 41 cases had mixed infections, accounting for 80.4%, and 73.2% (30/41) of patients infected with C. pneumoniae had mixed infections. M. pneumoniae and ADV were rarely co-infected with other pathogens [Supplementary Figure 1C, https://links.lww.com/CM9/B479]. Patients with severe community-acquired pneumonia (SCAP) accounted for 7.7% (61 cases), with a male-to-female ratio of 4.5:1. The 65- to 79-year-old group constituted the majority of the cohort, accounting for (39.3%, 24 cases). The top four pathogens detected in SCAP were K. pneumoniae (21.3%, 13 cases), Flu-A (19.7%, 12 cases), S. pneumoniae (16.4%, 10 cases), and S. aureus (9.8%, 6 cases) [Supplementary Figure 1D, https://links.lww.com/CM9/B479]. The epidede-mic peak of Flu-A was mostly in winter and spring. Despite a high detection rate of S. pneumoniae every year, it continued to decline from 2013 to 2017, while the detection rate of K. pneumoniae has increased year by year. The detection rate of atypical pathogens fluctuated between 7.2% and 14.3% from 2013 to 2017 [Supplementary Figure 1E, https://links.lww.com/CM9/B479]. Compared to the Fuzhou area, the detection rates of Flu-A, K. pneumoniae, S. aureus, and M. pneumoniae were higher in the non-Fuzhou area (P <0.001, = 0.001, 0.021, 0.035, respectively) [Supplementary Figure 1F, https://links.lww.com/CM9/B479].In the four age groups, Flu-A was the most common pathogen. M. pneumoniae were detected more frequently in the 18–49-year-old patients than the other three groups (7.9% [11/139] vs. 2.6% [17/652], χ2 = 9.448, P = 0.002). In addition, among patients <65 years old, the detection rate of ADV was sigificantly higher than that in patients aged ≥65 years old (3.1% [11/356] vs. 0.9% [4/435], χ2 = 4.957, P = 0.026) [Supplementary Figure 1G and Supplementary Table 2, https://links.lww.com/CM9/B479].

In our study, the results showed that the positive rate of pathogen detection was 62.8%. The viral detection rate was 32.2%, among which the Flu-A virus was the most common one. The bacterial detection rate was 39.2%, and the common bacterial pathogens were S. pneumoniae, K. pneumoniae, and S. aureus, but the detection rate of atypical pathogens was less. These findings were similar to the previous results of our study.[1] However, it was not the same as the CAP etiology survey >10 years ago in China, in which the detection rate of atypical pathogens reached nearly 1/3.[2–4] Further research found that K. pneumoniae was the most common pathogen in patients with SCAP, followed by Flu-A, S. pneumoniae, and S. aureus. K. pneumoniae ranked the first in the pathogen detection rate of SCAP in our cohort, which has not been found in previous studies, and may be due to more elderly patients than before. The status of K. pneumoniae in domestic SCAP surpassed that of S. pneumoniae, which requires sufficient clinical attention in a nowadays aging society. These results suggested the changes in the CAP pathogen spectrum over the past 10 years and could be attributed to the following reasons. Firstly, the increase in detection methods in recent years, especially nucleic acid testing assay, results in higher detection sensitivity than previous pathogen cultivation or serological detection methods. Secondly, the increase in underlying diseases due to the advent of an aging society and the widespread use of antibacterial drugs may cause changes in pathogen detection. A European study of 1166 SCAP patients admitted to the intensive care unit (ICU) showed that the common organisms were S. pneumoniae (28.6%), S. aureus (5.9%), Legionella pneumophila (L. pneumophila) (5.5%), and Haemophilus influenzae (4.8%), while the virus detection was less, accounting for only 1.4%, which was obviously different from our study.[5] Another foreign retrospective study of patients with SCAP from 1999 to 2013 found that S. pneumoniae (41.7%), L. pneumophila (6.3%), and viruses (5.4%) were the main pathogens.[6] It was also not similar to our results, suggesting that there were obvious differences in the pathogen spectrum of SCAP in China and abroad, which may be related to the different regions and the antibiotic use policy of different countries.

Nevertheless, the present study had some limitations: (1) The PCR assay mainly used in this study had more potential to be false positive although it had higher detection sensitivity than the other methods. Moreover, we detected only 10 pathogens, with some common CAP pathogens not included. (2) This study lacked data on antibiotic resistance and the application of antibiotics before admission, which might cause a certain deviation in the results. (3) Since this study was only conducted in four central hospitals in Fujian Province, which might have the risk of bias. Therefore, more centers should be covered in the future to ensure the validity of the study.

Funding

This study was supported by grants from the National Science and Technology Major Special Project (No. 2017ZX10103004), Natural Science Foundation of Fujian Province (No. 2019J01178), and Fujian Provincial Hospital "Creating Double High" Flint Fund project (No. 2019HSJJ11).

Conflicts of interest

None.

References 1. Wang DX, Chen YS, Li HR, Zhang W, Huang WS, Lin XH, et al. Epidemiological study on the respiratory pathogens in hospitalized patients with lower respiratory tract infection in Fujian Province. Int J Clin Exp Med 2017. Available from: https://e-century.us/files/ijcem/10/12/ijcem0056977.pdf. [Last accessed on April 25, 2022]. 2. Liu YN, Chen MJ, Zhao TM, Wang H, Wang R, Liu QF, et al. A multicenter investigation of the etiology of 665 cases of community-acquired pneumonia in Chinese urban adults (in Chinese). Chin J Tuberc Respir 2006;29: 3–8. doi: 10.3760/j:issn:1001-0939.2006.01.003. 3. Huang H, Zhang Y, Huang SG, Lu Q, Zhou X, Xiu QY, et al. Etiological investigation of community-acquired pneumonia in Shanghai area (in Chinese). Chin J Anti-infective Chemother 2003;3: 321–324. doi: 10.16718/j.1009-7708.2003.06.001. 4. Li Q, Li WM, Gao S, Tang FM, Yan H, Ma B, et al. Etiological investigation of community-acquired pneumonia in Sichuan area (in Chinese). West China Med 2008;23: 275–277. doi: 10.3969/j.issn.1002-0179.2008.02.041. 5. Walden AP, Clarke GM, McKechnie S, Hutton P, Gordon AC, Rello J, et al. Patients with community acquired pneumonia admitted to European intensive care units: An epidemiological survey of the GenOSept cohort. Crit Care 2014;18: R58. doi: 10.1186/cc13812. 6. Vallés J, Diaz E, Martín-Loeches I, Bacelar N, Saludes P, Lema J, et al. Evolution over a 15-year period of the clinical characteristics and outcomes of critically ill patients with severe community-acquired pneumonia. Med Intens 2016;40: 238–245. doi: 10.1016/j.medin.2015.07.005.