Case collection and pathogen detection
Adult patients diagnosed to have C. psittaci pneumonia by mNGS at Anhui Provincial Hospital between January 1, 2018 and December 30, 2023 were retrospectively identified. All cases were confirmed by mNGS, and their clinical data, laboratory results, prognosis, and other information were recorded for statistical analysis (Fig. 1).
Flowchart showing the procedure used to select patients for inclusion in the study. CT, computed tomography; mNGS, metagenomic next-generation sequencing
Inclusion Criteria: (1) Age greater than or equal to 18 years old; (2) mNGS shows positive detection of C. psittaci; (3) Complies with the diagnosis of community-acquired pneumonia. Exclusion criteria:1.Incomplete case data; 2. loss to follow-up.
Diagnostic criteria and grouping
The diagnostic criteria for C. psittaci pneumonia were as follows: meets the diagnostic criteria for adult community-acquired pneumonia; C. psittaci gene fragments identified by mNGS of bronchoalveolar lavage fluid (BALF), blood, sputum, or lung tissue and criteria for a positive mNGS result met; and research results consistent with the clinical diagnosis and treatment efficacy by two senior physicians. Severe community-acquired pneumonia was diagnosed if the patient met either of the major criteria outlined by the American Thoracic and Infectious Diseases Society [5], namely, requiring intubation and mechanical ventilation or septic shock requiring vasoactive agents after aggressive fluid resuscitation, or if ≥ 3 of the following minor criteria were met: respiratory rate ≥ 30 bpm; oxygenation index ≤ 250 mmHg (1 mmHg = 0.133 kPa); infiltration of multiple lobes; impaired consciousness and/or disorientation; blood urea nitrogen ≥ 7.14 mmol/L; and systolic blood pressure < 90 mmHg requiring aggressive fluid resuscitation. There were 17 cases in the severe pneumonia group and 42 in the non-severe pneumonia group. All data represent the worst values recording within one week of admission. Collection and use of relevant information fully safeguarded patient privacy and complied with ethical norms.
Specimen collection and mNGS
mNGS was performed in patients with pneumonia who showed no significant improvement after treatment or developed severe pneumonia. Typically, 3–4 mL of BALF and/or sputum samples are collected for mNGS. When there is no sputum or BALF available, peripheral blood specimens are sent for testing. All samples are collected and preserved in strict accordance with aseptic principles. Peripheral blood specimens are stored in cell-free DNA storage tubes at room temperature, while other body fluid specimens are stored at 4 °C [6]. Samples are processed in accordance with the mNGS processing procedure [7, 8]. During the study period, specimens were delivered on dry ice to the MatriDx Biotechnology Company (Hangzhou, China) for mNGS detection. DNA sequencing was performed for each patient; RNA sequencing was conducted only for patients suspected of having an RNA viral infection. Upon receipt of the specimens, 1.2–1.5 mL of homogenized samples were centrifuged at 12,000 rpm for 3 min at 4 °C. Next, 400 µL of supernatant were added to the prepared matching cassette. The cassette was then placed into an NGSmaster™ automated workstation for processing, the details of which are as follows. The internal ambient temperature of the machine is maintained at 4 °C. DNA libraries were constructed by automated extraction of nucleic acids, reverse transcription (only for RNA), fragmentation of nucleic acids, terminal repair, A-tailing addition, primer ligation, and purification. The finished libraries were quantified by real-time PCR and sequenced by a shotgun approach using the Nextseq™ high-throughput sequencing platform (Illumina, San Diego, CA, USA).
Statistical analysis
Continuous data that were normally distributed are shown as the mean ± standard deviation and were compared between groups using independent samples t-tests. Continuous data with a skewed distribution are shown as the median (interquartile range) and were compared between groups using non-parametric tests. Categorical data are shown as the number (percentage) and were compared between groups using the chi-squared test or Fisher’s exact test. Multivariate unconditional logistic regression was used to identify risk factors. The statistical analysis was performed using SPSS version 26.0 statistical software (IBM Corp., Armonk, NY, USA). The significance level was set at α = 0.05.
Patients’ background characteristics
Sixty-two patients had a confirmed diagnosis of C. psittaci pneumonia during the study period. After three exclusions for incomplete data or loss to follow-up, the remaining 59 patients were included in the study. The mean age was 60.78 ± 12.68 years, 32 patients (54.24%) were male, and only 23 (39.0%) had a clear history of contact with birds or parrots. Two of these 23 patients were a father and son, with the son becoming ill while caring for his father. Many patients had underlying diseases, including cardiovascular disease (30.5%), diabetes (18.6%), and cerebrovascular disease (15.3%). There was no significant difference in sex, age, or underlying diseases between the severe and non-severe pneumonia groups. However, complications were common, with the most frequent being abnormal liver function (43 cases, 72.9%), followed by hypoproteinemia (36 cases, 61.0%), coagulation disorders (20 cases, 40.7%), myocardial injury (15 cases, 25.4%), and acute kidney injury (10 cases, 16.9%). Some conditions, such as venous thrombosis (4 cases, 6.8%) had a lower incidence but still required attention from clinicians. Patients in the severe pneumonia group were more prone to myocardial injury, kidney injury, hypoproteinemia, and coagulation dysfunction (P < 0.001), as well as venous thrombosis (P = 0.007)(Table 1).
Clinical symptoms
ical symptom of C. psittaci pneumonia was fever, which was found in 93.2% of patients, with over 80% of patients having a maximum temperature above 38.5 °C. The next most common symptoms were cough (67.8%) and expectoration (54.2%). The cough was dry early in the course of the disease. Fatigue was reported by 40.7% of patients, chest tightness by 30.5%, and dyspnea and chills by 28.8%. Six patients (10.2%) first visited the neurology department complaining of dizziness and headache. One patient visited the orthopedic department with a complaint of arthritis. Five patients received venovenous (VV)-extracorporeal membrane oxygenation (ECMO), which was successfully withdrawn and followed by uneventful discharge in all cases. Clinically, all patients in the severe pneumonia group had a lower oxygenation index and required oxygen therapy. Patients with severe pneumonia were significantly more likely to have impaired consciousness (P = 0.020) while those with non-severe pneumonia were more likely to complain of fatigue (P = 0.022). All four deaths occurred in the severe pneumonia group. The interval between symptom onset and diagnosis was shorter in the severe pneumonia group (8 days vs. 14 days, P = 0.002) (Table 2).
Laboratory results
Sixty-four specimens from 59 patients were sent for mNGS, including 46 BALF samples, 8 whole blood samples, 6 deep sputum samples, and 2 lung tissue biopsy samples. Forty-two patients (71.2%) had a white blood cell (WBC) count within the normal range, 41 (69.5%) had an elevated neutrophil percentage, 55 (93.2%) had a lymphocyte count below the normal range, and 44 (74.6%) had a lymphocyte percentage below the normal range. Twenty-three (39%) of the 51 patients who underwent procalcitonin (PCT) testing had a PCT level above the normal range, but only 7 had a PCT level > 10 pg/mL. Twenty (95.2%) of the 21 patients in whom interleukin-6 (IL-6) was tested had a significantly elevated level. Fourteen (51.9%) the 27 patients who underwent myocardial enzyme testing had a creatine kinase level exceeding the upper limit of normal, and 23 (85.2%) had a lactate dehydrogenase (LDH) level above the upper limit of normal. The CD4/CD8 ratio in the 20 patients who underwent T-cell subset testing ranged from 1.02 to 8.23. Routine blood investigations showed significant increases in the WBC count and neutrophil percentage and decreases in the number and percentage of lymphocytes (all P < 0.001) in the severe pneumonia group. Although the platelet count was within the normal range overall, it was lower in the severe pneumonia group than in the non-severe pneumonia group (P = 0.024). PCT and IL-6 levels were higher in the severe pneumonia group; however, the increase was only statistically significant for PCT (P < 0.001). LDH, D-dimer, and prothrombin time (PT) were significantly higher in the severe pneumonia group (P < 0.05). There was no statistically significant between-group difference in activated partial thromboplastin time, FIB, CD3, CD3 + CD8, CD3 + CD4, CD3 + CD4CD3 + CD8, sodium, aspartate transaminase, alanine transaminase, or creatine kinase (Table 3).
Findings on CT imaging
CT scans were performed in all 59 patients and revealed bilateral lung involvement in 30 (50.8%) and lower lung involvement in 21 (35.6%). The most common radiographic finding was patchy shadowing (n = 49, 83.1%), followed by consolidation (n = 45, 76.3%), pleural effusion (n = 32, 54.2%), reticular shadowing (n = 5, 8.5%), and ground-glass opacities (n = 5, 6.8%). Bilateral lung involvement and pleural effusion were more common in the severe pneumonia group; although the between-group difference was not statistically significant, clinicians should be aware that bilateral lung involvement and pleural effusion may indicate a trend towards more severe disease. All findings, including uneven thickening of the interlobular septa, ground-glass opacities, streaky shadows, and honeycombing changes, are reported in detail in Table 4 and shown in Fig. 2.
Representative images of mild and severe pneumonia caused by Chlamydia psittaci. (A) Non-severe C. psittaci pneumonia showing patchy increased density in the left lower lung. (B) Non-severe C. psittaci pneumonia showing patchy increased density in the left lower lung and multiple ground-glass opacities and pleural effusion in both lungs. (C) Severe C. psittaci pneumonia showing extensive consolidation and ground-glass opacities in both lungs, predominantly in the right lung, with a small amount of pleural effusion. (D) Severe C. psittaci pneumonia showing an extensive increase in density in the right lung, predominantly consolidation.
Etiologies and distribution
mNGS combined with traditional microbiology and clinical manifestations revealed a mixed infection rate of 39.0% for C. psittaci pneumonia, which was significantly higher in the severe pneumonia group (P = 0.0002). Bacteria, particularly non-fermenting bacteria, were a common cause of infection in the severe pneumonia group, as were Aspergillus spp. (P = 0.034). No significant difference in viral infections was observed between the two groups (P = 0.444)(Table 5). The bacteria identified included Acinetobacter baumannii, Klebsiella pneumoniae, Pseudomonas aeruginosa, and Burkholderia cepacia, all fungi were Aspergillus fumigatus or A. flavus, and the viruses were either novel coronavirus 2019 (COVID-19) or influenza A virus. Candida, Epstein–Barr virus, cytomegalovirus, and other viruses were excluded because of low read counts and inconsistent clinical symptoms despite positive mNGS results. Specific pathogens included (A) baumannii (n = 10), (B) cepacia (n = 5), K. pneumoniae (n = 5), Haemophilus influenzae (n = 2), Streptococcus pneumoniae (n = 1), Mycobacterium tuberculosis (n = 1), Aspergillus (n = 5), COVID-19 (n = 4), and influenza A (n = 1). The COVID-19 and influenza A-positive cases were in the non-severe pneumonia group. One patient was a poultry farmer, and both blood and bone marrow cultures on admission yielded B. cepacia. Poultry farmers infected with (C) psittaci should also be tested for this pathogen. The distribution of pathogens is shown in Fig. 3.
Distribution of pathogens in C. psittaci pneumonia with co-infection.
Treatment plan and outcomes
Forty-eight patients (83.1%) had taken at least one dose of a quinolone (moxifloxacin or levofloxacin), or azithromycin before diagnosis. The proportion of patients who received an effective antimicrobial agent before confirmation of the diagnosis was higher in the non-severe pneumonia group (88.1% vs. 70.6%), but the between-group difference was not statistically significant (P = 0.215). The quinolones were the effective antimicrobial agents used most frequently (n = 43, 87.8%). Noteworthy is that the rate of use of tetracyclines before diagnosis was 0%. Furthermore, there were statistically significant between-group differences in the use of effective antibiotics (quinolones and azithromycin alone or in combination)(P = 0.0358). After diagnosis of C. psittaci, 39 patients (66.1%) received combination treatment with two of the three types of antimicrobial agents. There was no significant between-group difference in use of single or combination antimicrobial medication after diagnosis or in use of two-drug or three-drug combinations (p = 0.542)(Table 6).
During 6 months of follow-up, the pulmonary CT lesions were absorbed in 55 patients, with a small amount of striped shadow remaining in some patients. Eight patients who had been admitted to the intensive care unit reported sleep disorders during follow-up.
Analysis of factors related to severe pneumonia
Two sets of statistically significant indicators were selected for multivariable logistic regression analysis. The dependent variable was set as 1 for severe pneumonia and 0 for non-severe pneumonia. The independent variables included myocardial injury and lymphocyte percentage with adjustment for age, sex (1 male, 2 female), abnormal coagulation, neutrophil percentage, platelet count, interval between symptom onset and diagnosis, PT, and the D-dimer value. Myocardial injury and lymphocyte percentage were risk factors for severe pneumonia before adjustment for age and sex. However, after adjustment, only myocardial injury was identified as an independent risk factor for severe pneumonia caused by C. psittaci (Table 7).