Pathogens of first-episode pulmonary infection in 141 children with chronic granulomatous disease

To the Editor: Chronic granulomatous disease (CGD) is an inherited primary immunodeficiency disease characterized by recurrent life-threatening bacterial or fungal infection and tissue granuloma formation.[1] Pulmonary infection is the most frequent manifestation, affecting nearly 80% of patients,[2] and remains the major cause of morbidity and mortality in patients with CGD. Details of the pathogenic microorganisms responsible for first-episode pulmonary infection provide information on the spectrum of infection in patients with CGD, help guide the empirical choice of antibiotics or antifungal therapy, and contribute to the early identification and management of CGD. Aspergillus is the commonest causative agent of pulmonary infection, but the distribution of other pathogens varies among countries.[2–5] Few studies have investigated the etiology of pulmonary infection in CGD in a large cohort in China. In addition, isolation of causative pathogens from patients is often difficult,[4] and no studies have examined isolation methods for pulmonary pathogens in CGD patients. Here we described the causative microorganisms and corresponding detection methods for first-episode pulmonary infection in 141 children with CGD over 10 years in a single center in China. We also retrospectively reviewed the demographics, pulmonary symptoms, laboratory examinations, chest computed tomography (CT) images, and genotypes [Supplementary Material and Supplementary Tables 1–3, https://links.lww.com/CM9/B731]. The demographics and clinical characteristics are presented in Supplementary Table 1, https://links.lww.com/CM9/B731. There were 92 (65.2%) patients in the pathogen-positive group and 49 (34.8%) patients in the pathogen-negative group. The positive rate of pathogen identification was higher than that in previous reports.[5] The median hospitalization period was shorter in the pathogen-positive group compared with the pathogen-negative group (27.5 days vs. 30.0 days, P =0.024), but the incidence of severe pneumonia was higher in patients with isolated pathogens than in those with unknown pathogens (42.4% [39/92] vs. 22.4% [11/49], P = 0.018). Therefore, a microbiological diagnosis of the pulmonary infection is a key step in providing effective therapy, shortening hospitalization, and reducing medical expenses. Most CGD patients were symptomatic at the first-episode pulmonary infection. Fever was common (97.9%; 138/141), followed by cough (73.8%; 104/141), hypoxemia (44.0%; 62/141), and cyanosis (30.5%; 43/141). Among the 141 patients, extrapulmonary manifestations occurred in 87 patients (61.7%), and sepsis was the most common symptom [Supplementary Table 1, https://links.lww.com/CM9/B731]. Chest CT was abnormal in all patients. Nodules were the commonest findings (63.8%; 90/141), followed by consolidation (54.6%; 77/141), mass (51.8%; 73/141), and mediastinal or hilar lymphadenopathy (43.3%; 61/141). There was no statistically significant difference in clinical symptoms or chest CT findings between the pathogen-positive and pathogen-negative groups, except for hypoxemia, retraction, sepsis, consolidation, effusion, and mass [Figure 1A and Supplementary Table 1, https://links.lww.com/CM9/B731]. Pulmonary function was examined in 14 patients aged >5 years. The median forced vital capacity of predicted value was 71.4% and the forced expiratory volume in one second/forced vital capacity (FEV1/FVC) ratio was 80.2%, indicating restrictive hypoventilation as the main pulmonary dysfunction in CGD patients with pulmonary infection. Laboratory examinations revealed significant increases in peripheral leukocytes and neutrophils, C-reactive protein, and immunoglobulin E in both groups of CGD patients. There were no statistically significant differences in immunoglobulins or lymphocyte subsets between the two groups [Supplementary Table 2, https://links.lww.com/CM9/B731].Figure 1: Clinical, imaging and microbiologic features of 141 CGD patients with first-episode pulmonary infection. (A) Clinical features and chest CT findings in pathogen-positive (n = 92) and pathogen-negative (n = 49) CGD patients with first-episode pulmonary infection. * P <0.05. (B) Microorganisms and identification methods in 141 CGD patients with first-episode pulmonary infection. BALF: Bronchoalveolar lavage fluid; CGD: Chronic granulomatous disease; CT: Computed tomography; mNGS: Metagenomic next-generation sequencing.Among the 141 patients, 92 patients had positive etiologies. The distribution of pathogens and isolation methods are shown in Figure 1B. One pathogen was identified in 94.6% (87/92) of patients and two pathogens, suggesting mixed infection, were identified in 5.4% (5/92) of patients. The top three pathogens were Aspergillus (40.2%[37/92]), Burkholderia (13.0%[12/92]), and Mycobacterium (12.0%[11/92]). Aspergillus was the commonest causative agent in this cohort of CGD patients with first-episode pulmonary infection, in accord with reports from other countries.[2,5]Aspergillus spores are ubiquitous in the air, especially during the wheat-harvesting period, wet weather, or plants-rotting autumn. The spores are easily inhaled, increasing the chance of pulmonary Aspergillus infection in CGD patients. Aspergillus fumigatus (A. fumigatus) was the commonest Aspergillus species (30/92), while Aspergillus nidulans (A. nidulans) infection was rare (4/92) but more likely to invade the adjacent bone. Aspergillus niger (A.niger) infection was found in 3 patients. Among 12 patients with Burkholderia infection, 10 patients were from cities with heavy rainfall or coastal areas, such as cities in Hubei and Guangdong provinces. Mycobacterium infection was identified in 11 patients, including seven cases of Bacillus Calmette–Guérin (BCG) pneumonia and four cases of pulmonary tuberculosis. BCG vaccination is mandatory for newborns in China, and tuberculosis is endemic. Other pathogens including Streptococcus pneumoniae (6.5%; 6/92), Haemophilus influenzae (6.5%; 6/92), and Candida albicans (5/92; 5.4%) were also found in our patients. Staphylococcus, Serratia marsescens (S. marsescens), and Nocardia were previously reported as prevalent pathogens in CGD patients[2,5]; however, we only found Staphylococcus in four patients and S. marsescens and Nocardia in three patients, respectively. Sputum, bronchoalveolar lavage fluid (BALF), and lung biopsy samples were used for bacterial and fungal cultures to isolate the pulmonary pathogens, and BALF was also used for metagenomic next-generation sequencing (mNGS). Among the 92 pathogen-positive patients, the pathogens were isolated from BALF (69.6%;64/92), sputum (7.6%;7/92), and lung biopsy cultures (6.5%;6/92), respectively, or were identified by mNGS using BALF (16.3%;15/92). We analyzed the sensitivity and specificity of the different pathogen-isolation methods, except lung biopsy due to the small sample size (six cases). BALF mNGS detected more pathogens than BALF or sputum culture, and is becoming an efficient approach for detecting lung pathogens, especially when microorganism cultures are negative and lung biopsy samples are unavailable. The combined use of various methods will identify more pathogens responsible for first-episode pulmonary infection in CGD. Aspergillus infection is often asymptomatic in CGD patients. Most patients in our cohort, however, were symptomatic with severe pneumonia at an early age. The clinical characteristics of pulmonary infection caused by Aspergillus, Burkholderia, and Mycobacterium are generally similar [Supplementary Table 3, https://links.lww.com/CM9/B731], making the early differential diagnosis of pulmonary infection difficult. Hypersensitive pneumonitis caused by Aspergillus has been reported in CGD patients,[6] with most patients previously misdiagnosed with miliary pulmonary tuberculosis. Among the three groups (Aspergillus, Burkholderia, and Mycobacterium infection), cough (P = 0.009) and pleural effusion (P =0.002) were more common, and C-reactive protein levels were higher (P = 0.001) in the Burkholderia group compared with the other two groups. Mediastinal or hilar lymphadenopathy was the commonest symptom of tuberculosis infection (P <0.001), and chest wall invasion was prominent in Aspergillus infection. These characteristics aided the differential diagnosis of infectious pneumonia before pathogen detection. All CGD patients were hospitalized. Most patients (96.5%; 136/141) recovered from pulmonary infection, but five patients (3.5%) died, including three of Aspergillus infection (two of A. fumigatus infection and one of A. nidulans infection), one of Burkholderia cepacia (B. cepacia) infection, and one of unknown pathogen infection. In conclusion, pathogens could be identified in about two-thirds of CGD patients with first-episode pulmonary infection, allowing the use of appropriate antibiotics targeting the responsible pathogens. BALF culture is more reliable compared with BALF mNGS, sputum culture, and lung biopsy culture, and mNGS of BALF increases the positive-detection rate. Aspergillus, Burkholderia, and Mycobacterium infections were the main causes of morbidity and mortality in CGD patients in our cohort. Declaration of patient consent The authors certify that they have obtained all appropriate patient consent forms. In the form, the patients' parents have given his/her/their consent for images and other clinical information to be reported in the journal. This study was approved by the Ethics Committee of Beijing Children's Hospital (No. 2020-k-70). Acknowledgments The authors thank the children and their families who participated in this study and all the physicians for their help in accomplishing this work. Funding The study was supported by a grant from the Respiratory Research Project of the National Clinical Research Center for Respiratory Diseases (No. HX2X-202103). Conflicts of interest None.