To the Editor:
Cough, one of the most common reasons patients seek medical attention [1], is categorised according to duration. Acute cough lasts <3 weeks, subacute cough lasts 3–8 weeks and chronic cough >8 weeks [2, 3]. Acute cough is usually due to viral infection; it is transient and self-limiting [4]. Chronic cough has a variety of causes that can be established in 88–100% of patients using a diagnostic algorithm [3]. In contrast, the aetiology of subacute cough remains poorly defined, although post-infectious cough, sinusitis, post-nasal drip and bronchial asthma are listed as possible causes [2, 3, 5, 6]. As subacute cough is often refractory to recommended therapies, a better understanding of its aetiology is needed. The present work examined a possible role for Mycoplasma pneumoniae infection in subacute cough.
Between July 2010 and June 2011, patients who visited the Dept of Respiration at the Affiliated Hospital of Academy of Military and Medical Sciences (AHAMMS, Beijing, China) on their own initiative were screened and enrolled if they met the following criteria: cough was the only major complaint; cough was not associated with haemoptysis; cough had lasted for 3–8 weeks (determined by patient recall); chest radiography was normal; and age ≥15 years. Exclusion criteria included inpatient status, a prior history of lung or other systemic disease that could account for cough, current or past smoking, taking angiotensin-converting enzyme inhibitors, known immunodeficiency or pregnancy. Control subjects were enrolled from healthy medical staff in the department, their family members and relatives of patients who accompanied the patients. Control subjects had no cough for ≥1 month and were subject to the same exclusion criteria as patients. Efforts were made to match controls with patients by age, sex and time of enrolment. The study was approved by the ethics committee of AHAMMS (protocol #2010-08-99-1). Written informed consent was obtained from all patients and control subjects. Demographic data and medical history were documented using a pre-designed questionnaire. All symptoms and signs detected in a physical examination were recorded.
Oropharyngeal specimens were obtained from the posterior oropharynx using a sterile cotton swab from which M. pneumoniae was cultured [7]. A real-time quantitative PCR method was used to measure bacterial load by identifying 16S rDNA of M. pneumoniae using the following primer and probe sequences: forward 5′-GCAAGGGTTCATTATTTG-3′, reverse 5′-CGCCTGCGCTTGCTTTAC-3′ and probe 5′-6-carboxyfluorescein-AGGTAATGGCTAGAGTTTGACTG-tetramethylrhodamine-3′. PCR was performed in 25-μL reactions containing 2× Premix Ex Taq (Takara, Dalian, China), 10 μM of both primers, 3 μM of the TaqMan probe and 3 μL DNA template (prepared from swab agitate samples using QIAmp DNA mini kit (Qiagen China, Shanghai, China)) with a CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). PCR conditions were 93°C for 2 min, 10 cycles at 93°C for 45 s and 55°C for 60 s, followed by another 30 cycles at 93°C for 30 s and 55°C for 45 s. Pure M. pneumoniae DNA (Da’an Gene Co Ltd, Guangzhou, China) was used to construct a standard curve for quantification.
To confirm the presence of M. pneumoniae, positive isolates, identified by both bacterial culture and real-time PCR, were subjected to nested PCR using two sets of primers directed against the organism-specific P1 gene sequence (P1 gene-based strain typing) [8] using M. pneumoniae strain FH (ATCC 15531) as a reference. PCR amplicons of the P1 gene were characterised by agarose gel electrophoresis and nucleotide sequence determination.
Isolates identified as positive by both bacterial culture and PCR methods were tested for susceptibility to erythromycin, azithromycin, midecamycin, ofloxacin, levofloxacin, moxifloxacin, tetracycline, minocycline, amikacin and etimicin. Minimum inhibitory concentrations were determined by broth microdilution [7].
Patient and control groups were compared for the presence of M. pneumoniae using Fisher’s exact test. Subgroup comparisons of clinical characteristics were performed using t-tests or the Chi-squared test (SPSS for Windows, version 17.0; IBM, Armonk, NY, USA). p<0.05 was considered significant.
Out of 202 consecutive cough patients, 85 met the inclusion criteria. Two declined to participate. Excluded patients had the following illnesses: pneumonia (n=50), post-flu/cold cough for <2 weeks (n=31), interstitial lung disease (n=15), bronchiectasis (n=10), pulmonary hypersensitivity pneumonitis (n=6) and tuberculosis (n=5). 24 of the patients enrolled had prior consultations for cough; eight of them were prescribed a cephalosporin antibiotic for 3–5 days. 80 healthy controls were enrolled with a response rate of 100%. No significant difference existed between patient and control groups with respect to age, sex or seasonal variation (table 1).
Samples from 63 (76%) out of 83 patients were culture-positive for M. pneumoniae, and 61 (74%) out of 83 were found to be positive by real-time PCR (specificity ∼80%). 46 (55%) samples from patients were positive by both culture and PCR methods, and all were confirmed as M. pneumoniae by P1 gene-based strain typing. In contrast, samples from four (5%) out of 80 control subjects were culture-positive, and six (7.5%) were PCR-positive for M. pneumoniae. All four positive isolates identified by both culture and PCR were confirmed as M. pneumoniae by P1 gene-based strain typing. The M. pneumoniae detection frequency among subacute cough patients was significantly greater than among healthy controls (55% versus 5%, p<0.0001). The difference would have been even more significant if either culture- or PCR-positive samples were counted as M. pneumoniae-positive samples (94% versus 7.5%, p<0.0001), but such analysis would not exclude contribution from false-positives.
Quantitative real-time PCR revealed that 98% (45 out of 46) M. pneumoniae-positive samples from the patient group had a bacterial DNA load of 105–107 gene copies per mL of swab agitate suspension; 50% (23 out of 46) of these samples exhibited very high bacterial load (≥106 gene copies per mL). In contrast, the four M. pneumoniae-positive control samples gave only 103–105 gene copies per mL.
Of the 46 subacute cough patients identified as positive for M. pneumoniae by both culture and PCR methods, adolescents accounted for 33% (15 out of 46), middle-aged adults 52% (24 out of 46) and older adults 15% (seven out of 46). The 46 positive samples came from 15 males and 31 females. Seasonal effects were apparent: 21 (46%) positive cases in the patient group were detected during the winter season, 14 (30%) occurred in spring, 10 (22%) in autumn, and one (2%) in summer. Thus, infection with M. pneumoniae was significantly lower in summer (p=0.01).
Antimicrobial susceptibility data revealed that most isolates were resistant to macrolides and aminoglycosides. However, they were largely susceptible to tetracyclines and fluoroquinolones.
The present study showed, for the first time, that M. pneumoniae is highly prevalent in patients with subacute cough. Our conclusion is unlikely to be confounded by an M. pneumoniae outbreak, since an epidemiology study performed in the same area and overlapping most of our study duration revealed only 18% infection frequency among 500 community-acquired pneumonia patients [9]. Almost all samples from patients exhibited high M. pneumoniae DNA load (≥105 16S gene copies per mL). Among the factors analysed, season is the only risk factor that significantly affected M. pneumoniae detection rate; prevalence was significantly lower in summer (p=0.01). Antimicrobial susceptibility assays indicated that most M. pneumoniae isolates were nonsusceptible to both macrolides and aminoglycosides but highly susceptible to tetracyclines and fluoroquinolones. Thus, M. pneumoniae infection may play a significant role in subacute cough, at least among Chinese populations in the Beijing area. Although the American College of Chest Physicians evidence-based clinical practice guidelines assert no role for antibiotic therapy [10], because subacute cough is not thought to arise from bacterial infection, our work encourages the investigation of treatment benefit with tetracyclines or fluoroquinolones.
Acknowledgments
We would like to thank the enrolled subjects of the study and Y. Li, W-K. Niu, X.Y. Shang, Z.Z. Heng, Y.H. Liu, Y. Liang and Q. Liu (Dept of Respiration, Affiliated Hospital of Academy of Military Medical Sciences, Beijing, China) for their help with the collection of specimens and clinical data. We also thank K. Drlica (Public Health Research Institute, New Jersey Medical School, Rutgers Biomedical and Health Sciences, Newark, NJ, USA) for his critical comments on the manuscript.
Footnotes
Support statement: This study was supported by a grant from the National Natural Science Foundation of China (grant no. 30670927).
Conflict of interest: Disclosures can be found alongside the online version of this article at www.erj.ersjournals.com
- Received April 16, 2013.
- Accepted October 31, 2013.
- ©ERS 2014