Abstract
Chronic infections of primary ciliary dyskinesia (PCD) airways are strongly associated with morbidity and mortality; however, relatively limited data exist about their microbiology. Therefore, this study was aimed to determine whether there was a correlation between sputum and bronchoalveolar lavage (BAL) microbiology in PCD cases. This was accomplished by comparing the lower airway bacteriology in PCD with the results of sputum samples. We retrospectively analyzed the microbiology of BAL and sputum samples of 114 patients who were diagnosed as PCD and followed up at Hacettepe University Hospital Department of Pediatric Pulmonology. At the time of sampling, patients had no pulmonary exacerbation, and sputum samples were obtained within 3 months of BAL. The mean age of the patients was 13.5 ± 4.3 (range 3–21 years) years. Microbiological analysis revealed microorganisms in 47.4% of sputum samples and in 63.2% of BAL samples. In both BAL and sputum samples, Haemophilus influenzae was the most frequent pathogen, followed by Streptococcus pneumoniae. Pseudomonas aeruginosa was detected more frequently in children at older ages. Aspergillus spp. was detected only in BAL samples. The microbiological yield of BAL samples was higher than the sputum samples. There was no significant correlation between the patient age and positive microbiology of BAL and sputum samples. In PCD patients, there was a higher microbiological yield in BAL samples than the sputum samples. Therefore, we conclude that negative sputum microbiology results should be evaluated cautiously, as the rate of bacterial and fungal yield was higher in BAL specimens. The results of this comparative study support the view that sputum cultures do not always reflect the lower airway microbiology in the management of pulmonary disease in PCD.
Introduction
P
While chronic infections of the PCD airways are strongly associated with morbidity and mortality, relatively little is known about their bacterial composition. Although there is a lack of evidence-based management guidelines for PCD, it is recommended that all PCD patients have surveillance cultures of expectorated sputum or oropharyngeal (OP) cough swabs performed 2 to 4 times annually.6,7 The most frequently isolated pathogens in PCD cases are Haemophilus influenzae, Staphylococcus aureus, and Streptococcus pneumoniae. Pseudomonas aeruginosa and nontuberculous mycobacteria (NTM) have also been reported, although predominantly in adults.5,8
Bronchoalveolar lavage (BAL) is the most reliable sampling method in infants and young children with PCD who are unable to expectorate. 9 Although sputum culture is a simple and noninvasive measure that can be used to evaluate airway microbiology in the routine clinical follow-up of children with PCD, BAL sampling is the gold standard for evaluating lower airway microorganisms. Because BAL is an invasive and costly procedure that requires the use of general anesthesia, the culture of sputum or OP swabs remains the standard practice in PCD patients.10–12
This study aimed to analyze the lower airway microbiology in PCD. Specifically, we aimed to compare the results of BAL and sputum samples to evaluate whether a correlation exists between sputum and BAL microbiology in PCD cases.
Materials and Methods
Subjects
Patients with PCD were retrospectively recruited from the Hacettepe University Hospitals Department of Pediatric Pulmonology between January 2013 and December 2015. During this period, 250 patients were followed up with the diagnosis of PCD. Among this population, 114 patients who had both BAL and sputum culture (within 3 months of each other) were enrolled in the study. This study was approved by the local institutional review board.
All of the 114 patients with PCD had a definitive diagnosis based on presentation of the characteristic clinical phenotype, the presence of ciliary ultrastructural defects (visualized by electron microscopy), and the presence of abnormal ciliary function (as determined by high-speed video microscopy), and/or a genetic mutation recognized to cause PCD. 13
All of the patients or their parents signed a written informed consent form before undergoing flexible bronchoscopy. BAL specimens were obtained from at least 2 different lobes of the lung parenchyma, while the patients were clinically stable. Sputum samples or deep throat swabs (from patients who could not expectorate sputum) were obtained within 3 months of the BAL procedure. Microbiological evaluation of sputum samples was performed at least every 3 months; this is the routine follow-up procedure for patients who regularly attend our outpatient clinic. Patients were treated with the antibiotics indicated according to their sputum culture in routine outpatient clinic.
Inclusion criteria were as follows: (1) clinically stable PCD patients (no change in cough or sputum, no fever, no change in therapy for a period of at least 2 weeks, and change in FEV1 < 10% since the last measurement), and (2) patients who were not under antibiotic treatment at the time of sampling.
Exclusion criteria were as follows: (1) patients who had pulmonary exacerbation at the time of sampling and (2) interval between sputum culture and BAL sampling was more than 3 months.
Microbiological investigations
Semiquantitative cultures were performed on sputum and BAL samples, while deep throat swabs were analyzed by direct cultivation methods. BAL samples were also evaluated for the presence of NTM. All samples were inoculated onto 5% sheep blood agar, MacConkey agar, bacitracin containing 10% chocolate agar, and Sabouraud dextrose agar.
14
Organisms were identified by standard microbiological methods and by Matrix-Assisted Laser Desorption–Ionization-Time of Flight Mass Spectrometry (VITEK MS; bioMerieux, France).
15
For the microbiological evaluation of Mycobacteria spp., the samples were decontaminated and homogenizated with NaOH and N-acetyl
Statistical analysis
Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) version 19.0 (SPSS, Inc., Chicago, IL). Continuous variables were presented as mean ± standard deviation for normally distributed variables and as medians (min–max) for nonnormally distributed variables. The Mann–Whitney U test was used to compare nonnormally distributed measures and the t-test was used for normally distributed measures. The proportional frequency of positive isolates in paired samples was compared using the 2-sample t-test. The numbers of patients with positive and negative microbiology were compared using a chi-square test. Values of P ≤ 0.05 were considered significant.
Results
One hundred fourteen patients with a mean age of 13.5 ± 4.3 years (range 3–21 years) were enrolled in this study. Fifty-nine (51.8%) of the patients were male and 55 (48.2%) were female.
Microorganism growth was detected in 47.4% of sputum samples and in 63.2% of BAL samples (P = 0.005). Sputum microbiology revealed 35.1% H. influenzae, 10.5% S. pneumoniae, 1.8% P. aeruginosa, 1.8% S. aureus, 1% Moraxella catarrhalis, and 1% Candida spp. The distribution of the microorganisms in the BAL was 46.5% H. influenzae, 18.4% S. pneumoniae, 7.9% M. catarrhalis, 2.6% Candida spp., 2.6% P. aeruginosa, and 1.8% Aspergillus spp. Aspergillus spp. was detected only in the BAL samples. None of the BAL samples (n = 92) contained NTM. Results of the sputum culture and BAL culture were identical in 27 patients (23.6%).
The microbiological yield of BAL samples was higher compared with the sputum samples. Bacterial culture results of BAL and sputum samples obtained from the same patient are presented in Table 1. BAL samples revealed simultaneous growth of 2 microorganisms in 16 patients and 3 microorganisms in 2 patients. There was a significantly higher yield of all bacteria in the BAL compared to the sputum (P = 0.005, phi = 0.36). The distribution of the microorganisms in the BAL and sputum samples is shown in Table 2. P. aeruginosa was detected more frequently in older patients. There was no significant correlation between patient age and the positive microbiology of BAL and sputum samples (P = 0.61).
Discussion
Chronic airway infection begins in early childhood and is considered the leading cause of morbidity and mortality in PCD. 1 Typically, the most common bacteria isolated in respiratory cultures from children and adolescents with PCD are members of the OP microbiota, namely nontypeable H. influenzae, S. aureus, and S. pneumoniae.9,17,18 Older patients with more advanced lung disease are more commonly infected with P. aeruginosa.1,9,14 In this study, the most frequently isolated pathogens in PCD patients were H. influenzae and S. pneumoniae. Similar to previous studies, some of the patients in this study had more than one type of bacteria in their respiratory samples. 1 BAL samples revealed simultaneous growth of 2 microorganisms in 16 patients and 3 microorganisms in 2 patients.
The isolation rates of specific microorganisms usually differ in BAL and sputum samples, even in the same patient. 8 In this study, the mean isolation rates in BAL were 46.5% for H. influenzae, 18.4% for S. pneumoniae, 7.9% for M. catarrhalis, and 2.6% for P. aeruginosa. Pizzutto et al. reported 28.6% H. influenzae, 17.9% S. pneumoniae, 16.1% M. catarrhalis, and 7.1% P. aeruginosa in a group of patients with noncystic fibrosis (CF) bronchiectasis. 19 Li et al. also reported different isolation rates of microorganisms in the sputum and BAL of patients with non-CF bronchiectasis; 39% versus 51% for H. influenzae, 17% versus 22.1% for S. pneumoniae, 2% versus 3% for M. catarrhalis, 11% versus 14.4% for P. aeruginosa, and 4% versus 4.8% for S. aureus in the sputum and BAL, respectively. 20 In this study, the mean isolation rates in the sputum were 35.1% for H. influenzae, 10.5% for S. pneumoniae, 1.8% for P. aeruginosa, 1.8% for S. aureus, and 1% for M. catarrhalis. P. aeruginosa was isolated with a similar frequency in the sputum and BAL samples. However, the overall P. aeruginosa detection rate was very low in our study population, which may be attributed to the relatively younger age of the patients. Similar to this study, the study by Edwards et al. also reported a low (2%) overall P. aeruginosa detection rate in patients with non-CF bronchiectasis. 21 Davis et al. reported that the predominant pathogens were S. aureus (19%), H. influenzae (22%), S. pneumoniae (14%), and M. catarrhalis (8%) in the sputum samples of their PCD cohort. 7 Nonmucoid and mucoid P. aeruginosa strains were recovered in 6% and 3% of the subjects in that study, respectively. 7 Non-mucoid colony morphotype P. aeruginosa strain was isolated in this patient population.
We found higher microbiological yields in the BAL samples compared to the sputum samples. We conclude that sampling of BAL may be a better diagnostic approach in patients who cannot expectorate sputum. Likewise, Chang et al. reported higher microbiological yields, including fungi, in BAL of patients with non-CF bronchiectasis. 22 In contrast to these results, Kapur et al. did not detect fungi in BAL samples of patients with non-CF bronchiectasis. 23
The prevalence of NTM in PCD is reported to be similar to that seen in CF, with ∼15% of adult PCD patients having positive NTM cultures. However, a lower prevalence of NTM was reported in children with PCD.7,9 In this study, no NTM were isolated in BAL samples of PCD patients, which again may be related to their young age.
There was no significant correlation between age and positive microbiology of BAL and sputum samples in this study. In a study by Zampoli et al., culture-positive yield was significantly higher in sputum compared to OP swab in patients with CF. 24 Mussaffi et al. compared 29 simultaneously collected OP swabs and sputum in 20 children with CF and reported an increased, but nonsignificant, detection of microorganisms from sputum compared to OP swabs. 25 It should be noted that although BAL is an invasive procedure, it provides valuable information, especially in patients who are unable to appropriately expectorate. In addition, BAL samples are also useful in distinguishing lower airway infection and upper airway colonization.
In this study, the low rate of microorganism detection in sputum could be attributed to the antibiotic treatment of some patients before the sputum analysis. Since this was a retrospective study, the most important limitation was that the BAL samples and sputum samples were not simultaneously obtained from the same patient. The variability of airway microbiome both within and between PCD patients is an important consideration, and this points to the need for an extension of the analysis presented in this study to a larger patient population. Despite bacterial loads being lower in samples from PCD patients, they were still substantially above levels that might be expected in lower airway samples from healthy individuals, suggesting strongly that they represent chronic infection. 26
In conclusion, negative sputum microbiology results should be evaluated cautiously in patients with PCD, and the BAL procedure should be encouraged since the rate of identification of potential pathogens in the lower respiratory tract was higher in BAL specimens than sputum specimens. Appropriate specimen collection and proper identification of the potential pathogens are the mainstays of therapeutic interventions in all lower airway diseases, including PCD. The results of this study support the view that sputum culture does not always reflect the lower airway microbiology in the management of pulmonary diseases in PCD. Further prospective studies are needed for the comparison of sputum and BAL samples in patients with PCD, and this will help to reach a consensus about appropriate diagnostic sampling and also therapeutic approach for these patients.
Footnotes
Author Disclosure Statement
No competing financial interests exist.