Epidemiological characterization of respiratory tract infections caused by Mycoplasma pneumoniae during epidemic and post-epidemic periods in North China, from 2011 to 2016

Mycoplasma pneumoniae (M. pneumoniae) is a common pathogen of respiratory tract infections (RTIs) in humans, especially in children and young adults [1, 2]. M. pneumoniae infections rangle clinically from mild, self-limiting upper respiratory symptoms to radiographically confirmed pneumonia requiring hospitalization and account for 10–30% of community-acquired pneumonia (CAP) cases [2–5]. Real-time polymerase chain reaction (PCR) has emerged as the primary techniques for detection of M. pneumoniae in surveillance programs and clinical practices in China [6, 7], due to its increased sensitivity and specificity compared with culture-based and serological methods.

M. pneumoniae is the smallest self-living and cell wall-less bacterium. Droplet infection during close contact mediates transfer of this pathogen from person to person and can lead to an epidemic. Epidemics of M. pneumoniae infections are documented worldwide with durations of 3–5 years [8–10]. The fluctuations in the incidence of M. pneumoniae infections are considered due to a decline in a population’s immunity and an increase of the immunologically naïve population [11], or changes in the proportion of individual strains with specific genotypes. Many investigations have focused on the epidemics, individual cases, and small clusters and outbreaks with various sizes [10, 12–14]; however, epidemiological features of M. pneumoniae infections during post-epidemic or non-epidemic periods, including age distribution, the seasonality and the hospitalization rate of the patients, have not been studied in detail.

In this study, we retrospectively analyzed patients of all age with M. pneumoniae RTIs from 2011 to 2016, covering an epidemic (2011–2013) and a post-epidemic period (2014–2016), and explored the epidemiological patterns of the infections during the two periods.

From 2011 to 2016, we observed that 14.4% (1127/7835) of patients with RTIs were positive for M. pneumoniae, which was comparable with the global incidence as 12% (range 11–15%) from the Atypical Pathogens Reference Laboratory Database [15]. Similar to previous reports [16, 17], the positive rates rose during the late summer and autumn and returned to low values during late winter and spring. The M. pneumoniae positive rates of children was significantly higher than those of adults (19.7% vs. 8.9%, P< 0.001) nearly in every month (Additional file 3: Table S1), and the highest rate was found in the school children group aged 7–17 years (32.9%, 443/1348). Similarly, Kogoj et al. reported the proportion of M. pneumoniae RTIs were 19 and 7% in children and adults, respectively, and the highest proportion was found in patients aged 6–16 years (26%; 565/2162) [18]. These observations were consistent with the fact that the majority of outbreaks had occurred within a community or in closed or semi-closed settings such as military bases or schools. Moreover, compared with other publications on hospitalized patients [17, 19] and pediatric patients [17, 20–22], these results were more comprehensive to the patients with different severity and age because we reported on patients with 43.7% (3427/7835) as outpatients and adults to children ratio as 1:1 (3852:3983).

We firstly characterized the epidemiological features of M. pneumoniae infections during the epidemic and post-epidemic periods. Slightly different seasonality was displayed between the two periods: the positive rates peaked from late summer to autumn in epidemic period, and from fall to winter in the post-epidemic period. A positive correlation between increases in temperature and the occurrence of M. pneumoniae infections was reported by Onozuka and colleagues [23, 24], which might help explain the higher numbers that occur during warmer months.

The positive rate trends among age groups were similar between the two periods, with higher rates in patients aged 4–44 years old. The distribution corresponds to the findings from England and Wales [4, 25]. However, M. pneumoniae positive rate was significantly higher in pre-school and school children during post-epidemic period (P< 0.001; Table 2), and no statistical difference was found during epidemic period (P=0.801; Table 2). Furthermore, most of older and elderly cases occurred during the epidemic period. A single genotype of M. pneumoniae is not the probable cause of the epidemic, since other studies have shown that both endemic and epidemic spreads of M. pneumoniae may be polyclonal [11, 26, 27]. Therefore, the potential reason for the observed pattern of age distribution during the two periods may be related to interactions between the pathogen and the immunological status of the human population [28]. A mathematical model of this process was recently reported [29]. During epidemics, due to the initial lack of protective immunity, M. pneumoniae infections can occur in humans of all age. Epidemics fade when the susceptible population gains protective immunity against the pathogen. However, younger individuals, who had not been exposed and thus did not gain protective immunity, were consequently infected during the post-epidemic time period. Further epidemiological investigations are warranted to yield more serological evidence for populations from different regions and periods and elucidate the biological and epidemiological reasons that contribute to this phenomenon.

Fulminant pneumonia accounted for 0.5–2% of all M. pneumoniae pneumonia cases and primarily affected young adults [30]. Two Chinese reports found that higher ICU admission rates (32.8%) and fatal M. pneumoniae pneumonia were be found in younger children [21, 31]. Moreover, in USA and Europe, the rates of ICU admission of hospitalized patients with M. pneumoniae pneumonia were reported as 10 and 16.3% [19, 32]. Here, during the whole period, only 14/1127 (1.2%) M. pneumoniae patients were treated in ICUs, with a median (IQR) age of 34 (37.75) years (range 15–83 years). Two factors may have contributed to the underreported number of ICU admissions: 1) nearly half of the detected patients were mildly symptomatic and treated as outpatients; 2) the severely ill children patients were immediately transferred to Children’s Hospital. A special phenomenon was observed that all the ICU patients positive with M. pneumoniae were admitted during the epidemic period (Table 3), along with the significantly higher incidence of M. pneumoniae RTIs in elderly and very elderly patients (Table 2). Although Khoury et al. retrospectively investigated M. pneumoniae patients from 2007 to 2012, covering the epidemic 2010–2012, the yearly distribution of the M. pneumoniae cases were not described [32]; thus these results do not enable a comparison of ICU admission rates during the epidemic. Nevertheless, both previous and our results indicated that M. pneumoniae should be considered as a possible pathogen in pneumonia patients admitted to the ICU in the setting of epidemic.

Limited by the retrospective design of the study, multivariable analysis, and time trend analysis could not be performed since more medical data (e.g., history of drug use, recent hospitalization, socio-economical factors and etc) could not be obtained for most of the patients. No information collected to analyze the effect of sources of the patients on the results is another limitation, since parts of the studied patients came from cities out of Beijing. Considering the most reliable diagnosis for M. pneumoniae RTIs would come from a combination of two or more separate laboratory methods, such as serology and PCR [33–36], another limitation of the study is the lack of serological results. Due to the much lower sensitivity, serological testing is actually not widely performed in our hospital, especially for adults [7]. Moreover, although all analyzed patients presented with RTIs, we still could not exclude the possible carriages in the upper respiratory tract [37]. However, the results of studies on carriages are inconsistent, from a low rate (< 3%) [38] to a relatively high proportion (13.5 and 4.6%) [39]. And the larger size and six-year consecutive detection strengthen our study.

Different patterns of age distribution and seasonality of M. pneumoniae RTIs between epidemic (2011–2013) and post-epidemic (2014–2016) periods were firstly reported. Most of older patients and all of ICU admissions infected with M. pneumoniae occurred in epidemic period.