Mycoplasma pneumoniae infection in hospitalized adult patients with community-acquired pneumonia in China

Introduction: This study aimed to investigate the prevalence, clinical and radiographic features, and antibiotic responses of Mycoplasma pneumoniae (M. pneumoniae) infections in hospitalized adults with community-acquired pneumonia (CAP) in China. Methodology: Serum specimens collected from 189 CAP patients in both acute phase and convalescence were tested for IgG, IgA, and IgM mixed antibodies specific to M. pneumoniae. The clinical and radiographic characteristics and efficacy of three antibiotic regimens were compared between patients with M. pneumoniae infection and those without. Results: Among 189 CAP patients, 88 (46.6%) were positive for M. pneumoniae infection. Compared to the negative patients, patients with M. pneumoniae infection were significantly younger, had higher rates of dry cough, and had white blood cell counts of <10/L, but had less purulent sputum. Radiography further showed more centrilobular nodules, ground-glass opacities, tree-in-bud patterns and thickened bronchovascular bundles, but less pleural effusion and larger tracts of real opacities in patients with M. pneumoniae infections. Among the three regimens used, patients with moxifloxacin required significantly shorter fever abatement, treatment, and hospitalization times than those with azithromycin plus ceftriaxone and ceftriaxone only. Conclusions: M. pneumoniae infection was present in almost half of the CAP population in east China, with some distinct clinical and radiographic features. Moxifloxacin was an effective antibiotic for this infection.


Introduction
Mycoplasma pneumoniae (M.pneumoniae) is one of the main pathogens causing community-acquired respiratory tract infections, especially in children and young adults, although it also causes about 15% of pneumonias in patients older than 40 years of age [1].In recent years, community-acquired pneumonia (CAP) has become a common infectious disease, especially among the elderly and those with chronic diseases [2].Despite substantial progress in therapeutic options, CAP remains a significant cause of morbidity and death worldwide [3], including in China [4], where M. pneumoniae infection is associated with about 20% of CAP cases, and is most likely co-infected with other atypical pathogens.In fact, up to 30% of all pneumonia cases are reportedly caused by M. pneumoniae in the general population [1].However, diagnosis of the cause of CAP is difficult due to the co-infections with mixed pathogens.Thus, identification of unique clinical and radiographic features in CAP patients caused by different pathogens is critical to help improve the diagnosis of the pathogenic cause of CAP, and thus guide the treatment of the disease.Whereas clinical and radiographic features of CAP caused by more common pathogens, including Haemophilus influenzae (H.influenzae) and Staphylococcus aureus (S. aureus) [5], have been categorized, little is known about the clinical and radiographic features of CAP associated with M. pneumoniae infection.
In a community-based population, the most common clinical consequences associated with M. pneumoniae infection are acute bronchitis, pharyngitis, and otitis; a small percentage of patients may develop severe neurologic, hematologic, or dermatologic disease such as erythematous lesions [6].M. pneumoniae infection can result in a wide spectrum of symptoms ranging from a classic presentation with fever, cough, and sputum production to more subtle and nonspecific manifestations such as fatigue, malaise, and myalgia [1].Thus, generally, clinical manifestations of M. pneumoniae infection appear non-specific.However, whether CAP patients with M. pneumoniae infection also present with unique symptoms and signs remains unclear.More importantly, radiographic evaluation prior to laboratory investigation, such as sputum smear, culture, PCR, serology, etc., would prove to be a very valuable method in diagnosing M. pneumoniae infection in CAP patients if the unique features of M. pneumoniae-associated CAP are identified and characterized.Several studies [1,3,7] reported on the radiographic features, including computed tomography (CT); however, there is little information on the clinical and radiographic features in Chinese CAP patients with M. pneumoniae infection.Therefore, in the present study, we investigated the prevalence, clinical and radiological features, and antibiotic response of M. pneumoniae infection in hospitalized adult patients with community-acquired pneumonia in China.

Patients
Between January 2009 and December 2010, adult (over 18 years of age) patients who were hospitalized due to the diagnosis of pneumonia according to the international guideline for definition and criteria for enrollment of CAP [8] and the Chinese diagnosis and treatment manual for CAP [9] (based on the presence of selective clinical features and image of chest radiography) at the Third Affiliated Hospital of Wenzhou Medical University were enrolled.Patients who were confirmed with human immunodeficiency virus (HIV) infection, hospital-acquired pneumonia, viral or fungal infection, tuberculosis, lung tumors, chronic obstructive pulmonary disease, who had an organ transplant, serious complications, and/or were treated with immunosuppressive drugs, were excluded.
All eligible patients were invited to participate in the study, and gave written informed consent.This study was approved by the Ethics Committee of the Third Affiliated Hospital to Wenzhou Medical University (No. 2009001).

Specimen collection
Blood samples were collected from the patients on the second day after hospitalization and then at a three-or four-week intervals.Serum samples were separated and kept at -80°C until use.The laboratory tests, including serological diagnosis and physical examinations including radiological diagnosis, were performed and the results were recorded at each time point.Pathogenic tests such as bacterial culture were performed on the sputum and bronchoalveolar lavage fluid of some patients.

Serological diagnosis
The serum specimens in both the acute phase and convalescence were used for detecting M. pneumoniae infection using a passive agglutination test (Serodia-Myco II, Fujirebrio Inc., Japan), which measures the mixed antibody titer of M. pneumoniae IgG, IgA, and IgM, as described previously [10,11].M. pneumoniae infection was defined as a serum sample that showed a fourfold change of antibody titer during the acute or convalescence period with a mixed antibody titer of ≥ 1:160 [9].

Radiological diagnosis
The CT or chest X-ray images were reviewed and the reports were verified by a radiological expert and a respiratory physician.Patients with the presence of a sheet or patchy shadow of invasion in the lung, or changes of interstitial lung diseases by chest X-ray or CT, were diagnosed with pneumonia.

Treatment of pneumonia with confirmed or suspected M. pneumoniae infection
The treatment of pneumonia is empirical based on the Chinese diagnosis and treatment manual for CAP [9].Patients with confirmed M. pneumoniae infection received one of the following empirical regimens: moxifloxacin (group A), azithromycin plus ceftriaxone (group B), or ceftriaxone (group C) for 48 to 72 hours at the treating physicians' discretion.The treatment was replaced with other antibiotics (piperacillin/tazobactam for group A, or moxifloxacin for groups B and C), if there was no improvement in body temperature and clinical symptoms.Patients who were finally confirmed as having pneumonia with M. pneumoniae infection were included in the analysis on the efficacy of treatment regimens.

Statistical analysis
Numerous data were expressed as mean ± standard deviation (SD), where appropriate, and categorical data were expressed as percentage.Statistical analysis was performed using SPSS 13.0 software (SPSS Inc., Chicago, USA).The incidence of underlying conditions, clinical findings, and radiographic findings were analyzed using the Chi-square test or Fisher's exact test.The mean age of patients and laboratory data were compared using the Student's t test.Statistical significance was defined as p < 0.05.

Prevalence of M. pneumoniae infection in patients with CAP
One hundred and eighty-nine CAP adult patients were hospitalized in the Third Affiliated Hospital of Wenzhou Medical University between January 2009 and December 2010.Patients between 18 and 86 years with a mean age of 40.6 ± 14.1 years were enrolled in the present study.Overall, 88 (46.6%) patients were serologically positive for M. pneumoniae infection and were thus diagnosed as having pneumonia with M. pneumoniae infection.Of the 101 (53.4%) patients negative for M. pneumoniae infection, 15 (7.9%) were found to be infected with Streptococcus pneumoniae, 10 (5.3%) with H. influenzae, 5 (2.6%) with Moraxella catarrhalis, 4 (2.1%) with Klebsiella pneumoniae, 3 (1.6%) with Escherichia coli, 2 (1.1%) with S. aureus, and 1 (0.5%) with Pseudomonas aeruginosa.No pathogens were found in the remaining 61 (32.3%) patients.

Clinical characteristics of patients with M. pneumoniae infection
The detailed comparison of patients with without M. pneumoniae infection are listed in Table 1

Radiographic characteristics of patients with M. pneumoniae infection
The radiographic characteristics of the two groups are shown in Table 2. Centrilobular nodules (Figure 1A), ground-glass opacities (Figure 1B), tree-in-bud patterns (Figure 1C), and thickened bronchovascular bundles (Figure 1D) were more commonly presented in patients with M. pneumoniae infection than in those without M. pneumoniae infection (all p < 0.001).However, pleural effusion and large opacity were significantly less in patients with M. pneumoniae infection than in those without M. pneumoniae infection ( 2 = 8.84, p = 0.003; and  2 = 5.74, p = 0.017, respectively).

Treatment of patients with M. pneumoniae infection
Initially, of the 189 patients, 66, 62 and 61 patients were assigned into groups A, B and C, respectively, based on the differences in clinical treatment.Among the 88 patients with confirmed M. pneumoniae infection, 32 patients received moxifloxacin (group A), 30 received azithromycin plus ceftriaxone (group B), and 26 received ceftriaxone only (group C) (Table 3).There was no difference among the three groups in age, sex, and major clinical and laboratory parameters.When compared with groups B and C, group A had a significantly different fever abatement time, treatment progress, number of cases with antibiotic drugs switching, and average hospitalization time (all p < 0.001).Moreover, the proportion of cases with an antibiotic drug switch was significantly lower and the average hospitalization duration was significantly shorter in group B than in group C (both p < 0.001).

Discussion
Despite rapid economic development during the last decade in China, M. pneumoniae infection remains a great public health problem in the country.In the present study, the infection rate of M. pneumoniae detected by serology was 46.6% among hospitalized CAP patients.This rate was significantly higher than those (6.8%-29.6%)previously detected by various methods in most studies [12][13][14][15][16].However, this rate is consistent with the one recently reported by Tao et al. in Chinese patients with CAP, in whom M. pneumoniae infection was the most frequently identified pathogen, detected by serological tests in 38.9% of CAP patients and in 45.7% and 24.4% of young (< 65 years of age) and old patients, respectively [10].The high detection rates of M. pneumoniae infection in the present study and the study by Tao et al. might be caused by regional sporadic infection, because the incidence of M. pneumoniae infection has occurred approximately every three to five years, mostly during summer, autumn, and winter [17,18].The unrecognized community outbreak of M. pneumoniae infection is consistent with the observation in Rhode Island, USA [6].
Because serological diagnosis of M. pneumoniae infection requires specimens collected in both acute phase and convalescence, the diagnosis based on clinical symptoms and chest CT were more sensitive and specific than rapid serologic detection of IgM antibody of M. pneumoniae at the initial stage of CAP [19].In this study, patients with M. pneumoniae infection were younger than those without M. pneumoniae infection, and most of them (88.6%) were younger than 50 years of age (Table 1), suggesting that CAP patients < 50 years are most frequently associated with M. pneumoniae infection, which is in agreement with the findings of a previous study [17].Dry cough occurred more frequently, while purulent sputum occurred less frequently in patients with M. pneumoniae infection, compared with those without M. pneumoniae infection, suggesting that M. pneumoniae infection can cause dry cough with little phlegm.Baseline laboratory examinations showed that most of the patients (94.3%) with M. pneumoniae infection had normal WBC counts, while only 20.8% of patients without M. pneumoniae infection had normal WBC counts.Because there was no significant difference in increased levels of C-reactive protein (CRP) (p = 0.260) between the two groups (68.3% and 77.4%, respectively), this data indicated the effect of M. pneumoniae infection on CRP; it does, however, require further investigation.
In order to understand the radiological characteristics and improve the diagnosis of M. pneumoniae infection in patients with pneumonia, we compared the CT results of CAP patients with and without M. pneumoniae infection.We found that centrilobular nodules, ground-glass opacities, tree-inbud patterns, and thickened bronchovascular bundles were more common in patients with M. pneumoniae infection than in those without M. pneumoniae infection (Table 2), which likely resulted from M. pneumoniae-induced bronchiolitis [20].High rates (88%-93%) of centrilobular nodules, ground-glass opacities, and thickened bronchovascular bundles in patients with Chlamydia pneumoniae (C.pneumoniae) pneumonia were also previously reported [21].In addition, we found that pleural effusion was less common in patients with M. pneumoniae infection than in those without M. pneumoniae infection.In the present study, there was no specific difference in radiographic findings between M. pneumoniae and other atypical pneumonia; however, chest CT data are helpful for rapid diagnosis of M. pneumoniae infection because of the reported sensitivity of 73% and specificity of 85% for the diagnosis of M. pneumoniae infection based on the CT characteristics of chest highresolution computed tomography [19].
We also compared the prognosis of different treatment of M. pneumoniae infection.Table 3 shows that patients treated with moxifloxacin had significantly shorter fever abatement time, treatment progression, and hospitalization duration than did those treated with azithromycin plus ceftriaxone or ceftriaxone only.In addition, patients treated with azithromycin plus ceftriaxone showed a shorter fever abatement time and better treatment progress than those treated with ceftriaxone only (Table 3).Interestingly, no patients treated with moxifloxacin needed to switch to another antibiotic program due to any clinic symptoms, while 73.1% of patients treated with ceftriaxone needed to switch to another antibiotic regimen, which may be due to the fact that M. pneumoniae has no cell wall and is resistant to drugs such as beta-lactams, including ceftriaxone, that target the cell wall [22].Generally, M. pneumoniae is susceptible to macrolides and related antibiotics, tetracyclines, and fluoroquinolones [19].However, high resistant rates to macrolides including azithromycin (69%-92%) in M. pneumoniae isolates have been observed in pediatric and adult patients with respiratory tract infections in China [23][24][25], which may further explain our clinical findings, though susceptibility testing was not performed in the present study.No CAP patients with M. pneumoniae infection died in the present study, which supports the observation of a previous study in China that infection with M. pneumoniae as well as C. pneumoniae was associated with a low pneumonia severity index, indicating low CAP severity and low risk for death [26].
There are some limitations in the present study.First, we used a passive agglutination test measuring the mixed antibody titer of M. pneumoniae IgG, IgA and IgM.Ideally, serology, culture and fluorescence quantitative polymerase chain reaction (FQ-PCR) should be incorporated in the establishment of M. pneumonia infection in clinical practice [27].However, culture is slow and FQ-PCR is expensive, and both have low sensitivity [27].Therefore, we applied the serological method, which is rapid and practical, with considerable sensitivity, specificity and accuracy, and has thus been used in many previous studies [10,27,28].Second, although we further examined bacterial pathogens in samples of 101 patients without M. pneumonia infection, most of them (60.4%, 61/101) remained unclear.Thus, there may be heterogeneity in the group without M. pneumonia infection.However, the main focus of the present study was to determine the prevalence, clinical and radiographic features, and antibiotic response of M. pneumonia infection in hospitalized adults with CAP, and current analysis already produced some significant results; however, further sub-analysis may not be more meaningful, considering the relatively small sample size.Finally, we did not carry out susceptibility testing in vitro, due to technical issues, and thus it is unknown whether the responses to the antibiotic treatment were associated with the antibiotic susceptibility in the bacteria.

Conclusions
In conclusion, M. pneumoniae infection is present in almost half of the CAP population and has distinct clinical and radiographic features, which indicates that it is possible to screen M. pneumoniae infection in CAP patients for early effective antibiotic intervention.Moxifloxacin is an effective antibiotic in the treatment of M. pneumoniae infection.treatment, and manuscript preparation.Qing Feng Sun initiated the study design, participated in literature search, performed the statistical analysis and revised and finalized the manuscript.Ming Hua Zhang participated in the study design and patient treatment.Ji Guang Ding participated in the study design and patient treatment.Lin Gao Yi participated in the study design and management of antibacterial drugs.Yuan Tong Gao participated in chest CT readings and analyzed the imaging diagnosis.Ai Xia Zhan participated in antibody detection of M. pneumoniae and analyzed results of immunological diagnosis.Ren Guo Zhao participated in the study design and patient treatment.Xiao Ci participated in the study design and patient treatment.All authors read and approved the final manuscript.

Figure 1 .
Figure 1.Representatives of major radiological characteristics of pneumonia with M. pneumoniae as shown in computed tomography.A: Centrilobular nodules (arrows) in a 29-yearold woman with M. pneumoniae pneumonia; B: Ground-glass opacities (arrows) in a 35-year-old woman with M. pneumoniae pneumonia; C: Tree-in-bud pattern (arrows) and bronchial wall thickening (arrowheads) in a 25-year-old man with M. pneumoniae pneumonia; and D: Wall thickening (arrowheads) and ground-glass attenuation (arrows) and in a 42-year-old man with M. pneumoniae pneumonia.

Table 1 .
. Patients Comparison of demographic, clinical, and laboratory characteristics of patients with and without M. pneumoniae infections WBC: white blood cell count; CRP: C-reaction protein

Table 3 .
Efficacy of three antibiotic regimens in patients with M. pneumoniae pneumonia

symptoms and laboratory parameters at baseline (%)
Data are expressed as mean ± standard deviation or number (%), where appropriate.Group A, treatment with moxifloxacin; group B, treatment with azithromycin plus ceftriaxone; group C, treatment with ceftriaxone only.* P < 0.001, when compared with Groups B and C; ** P < 0.01, when compared with Group C.