Introduction

The novel coronavirus (now called SARS-CoV-2) initially discovered in Wuhan, China, has spread all over the world causing a global pandemic and representing a great medical challenge in terms of treatment, prevention and, not less important, diagnosis [1].

So far, there is limited knowledge on the rate of co-infections with other respiratory pathogens [2]. Likewise, data regarding bloodstream and respiratory bacterial and fungal infections among patients with COVID-19 are very scarce and generally overlooked [3].

While early reports from China and Spain described co-infection as a rare event (6/104, 5.7% and 3/103, 2.9%, respectively) [4, 5], Kim et al. reported the presence of one or more additional pathogen in 24 of 116 patients (20.6%) diagnosed with SARS-CoV-2 infection [6] whereas another study from China showed that up to 80% of SARS-CoV-2 infected subjects had IgM positivity against at least one respiratory agent, therefore highlighting how the detection of other respiratory pathogens cannot be used to rule out COVID-19 diagnosis [7]. Furthermore, it is still unknown whether co-infection with other pathogens, and in particular with intracellular atypical microorganisms, might play a role in determining the prognosis of SARS-CoV-2 infection.

In all the cases, viruses were the most representative agents [2,3,4,5,6,7,8,9]; on the other hand, scarce was the co-infection rate with Mycoplasma pneumoniae [10,11,12,13,14,15,16] and, interestingly, only two cases out of 1996 with Chlamydia pneumoniae has been described so far [17].

Herein, we describe patients where co-infection with Chlamydia pneumoniae or Mycoplasma pneumoniae and SARS-CoV-2 was detected in our teaching hospital in Rome, Italy. Furthermore, the updated literature regarding the co-infection between SARS-CoV-2 and these atypical pathogens is reviewed.

Cases description

We retrospectively analyzed data from clinical reports of all the patients admitted to Azienda Ospedaliero-Universitaria Policlinico Umberto I (Sapienza University) of Rome between 1 March and 30 April 2020 with documented SARS-CoV-2 infection. The study was approved by the local Ethics Committee (ID Prot. 109/2020). A total of 182 subjects were tested also for C. pneumoniae and M. pneumoniae. We found that seven patients (3.8%) were co-infected with SARS-CoV-2 and atypical microorganisms (five C. pneumoniae, 2.7%, two M. pneumoniae, 1.1%). Diagnosis of C. pneumoniae and M. pneumoniae infection was made based on the serologies (DIESSE Diagnostica Senese S.p.A., sensitivity 97.4% and 94.7%, specificity 94.1% and 92.6% for C. pneumoniae and M. pneumoniae, respectively) [18] whereas SARS-CoV-2 diagnosis was based on nasopharyngeal swab positivity by using polymerase chain reaction (PCR) [19]. Definition of pneumonia or severe pneumonia was based on the WHO interim guidance and included clinical signs of pneumonia (fever, cough, dyspnoea, fast breathing) with or without signs of severe pneumonia such as respiratory rate > 30 breaths/min, severe respiratory distress, or SpO2 < 90% on room air [19, 20].

Clinical and laboratory characteristics of patients are listed in Table 1, radiological findings are shown in Supplementary Figure 1. Among the patients, four were male and three female, the median age was 73 years (IQR 45–79). All but one patient underwent CT-scan of the lungs, one patient underwent only chest X-ray, which showed bilateral interstitial involvement. Lung CT-scan showed multifocal, bilateral and prevalent peripheral infiltrates in six patients (85.7%), ground glass in five patients (71.4%), subpleural consolidation in four patients (57.1%). No patient had pleural effusion. According to guidelines [19], severe pneumonia was observed in 2/7 (28.5%) cases. All patients underwent therapy with hydroxychloroquine and azithromycin, 5/7 with heparin (57.1%), 3/7 with corticosteroids (42.8%), 2/7 with lopinavir/ritonavir (28.5%), 2/7 with tocilizumab (28.5%). One patient received also teicoplanin that has been described as potentially active against coronaviruses [21, 22]. As for oxygen delivery, two patients (28.5%) received high-oxygen non-invasive support (one high-flow nasal cannula, one C-PAP), three (43.0%) were on Venturi masks and the remaining two (28.5%) were on room air. All patients were discharged after a median length of hospitalization of 28 days (IQR 13–34).

Table 1 Characteristics of patients with SARS-CoV-2 and Chlamydia pneumoniae (n = 5) or Mycoplasma pneumoniae (n = 2) co-infection
Full size table

Finally, when clinical outcomes (ICU admission and intra-hospital mortality) of 175 patients without M. pneumoniae or C. pneumoniae co-infection [median age 63 years (IQR 52–76), 71 (40.5%) females] were compared to those with co-infection, no differences were observed [1/7 (14.2%) vs. 24/175 (13.7%) and 0/7 (0%) vs. 25/175 (14.2%), respectively].

Discussion and review of the literature

In the present report we described for the first time in Europe [2, 17] that patients with SARS-CoV-2 infection might be co-infected, among agents of atypical pneumonia, not only with M. pneumoniae but also with C. pneumoniae. These microorganisms can affect adults and children, are usually mild and only occasionally could represent life-threatening conditions. In particular, M. pneumoniae may cause epidemics and spread in close clusters. As the majority of symptomatic patients with SARS-CoV-2 infection develop an atypical pneumonia syndrome with fever, cough, and shortness of breath, co-infections with C. pneumoniae or M. pneumoniae are likely obscured, making therefore difficult the differential diagnosis only based on clinical presentation [19, 20]. The rate of co-infection with M. pneumoniae in SARS-CoV-2 pneumonia patients has been reported in the literature [10,11,12,13,14,15,16] whereas co-infection with C. pneumoniae has been reported only in two cases in a large US study involving 5700 patients with COVID-19 [17] (Table 2). In detail, Fan et al. described a case of a 36-year old male requiring Intensive Care Unit (ICU) admission and presenting with severe lymphopenia, low platelet count and cold agglutinin titer of 1:8 with M. pneumoniae antibody titer of 1:160 [14] whereas Ziang Gao et al. described a case of 49-year old female presenting with cough, expectoration and lung CT scan showing multiple ground-glass opacities in bilateral lower lobes [16]. Gayam et al. reported that six out of 350 patients (1.71%) with SARS-CoV-2 infection were also diagnosed with M. pneumoniae detected by serology [12] and, in a recent double-center Chinese study conducted at Qingdao and Wuhan regions and involving 68 patients with SARS-CoV-2 infection, the authors found a not-negligible rate of co-infection with common respiratory pathogens, with 8/68 (11.7%) of subjects showing also M. pneumoniae positive serology [7]. In the same study, a highly different distribution between the two regions (7/30, 23.3%, in Qingdao and 1/38, 2.63%, in Wuhan) was observed [7]. Although the whole rate of co-infection was far different, retrospective studies conducted in Spain and in the UK showed a similar number of SARS-CoV-2-M. pneumoniae co-infection (0.97% and 1.49%, respectively), the latter detected with multiplex PCR assays [5, 11]. In pediatric patients, co-infection with M. pneumoniae was surprisingly high, accounting for 16/34 (47.0%) of the total and a case report described the presence of COVID-19 infection with pleural effusion complicated by secondary M. pneumoniae infection in a 12-year old boy [13, 15]. As for C. pneumoniae, only one large US study which had the aim to describe the clinical characteristics and outcomes of 5700 hospitalized patients with COVID-19 found two C. pneumoniae cases out of 42/1996 positive samples tested also for respiratory pathogens panel [17]. Of note, and unlike our report, no clinical information of these two cases of C. pneumoniae and SARS-CoV-2 co-infection were available [17].

Table 2 Literature data on SARS-CoV-2 and Mycoplasma pneumoniae/Chlamydia pneumoniae co-infection
Full size table

Similarly to what has been reported in the literature, the majority of our patients presented with fever, cough and/or shortness of breath, showed bilateral infiltrates at the lung CT, received oxygen support and were treated with hydroxychloroquine and azithromycin.

The possible co-existence of pathogens other than SARS-CoV-2 in patients with COVID-19 infection focuses the attention on the real incidence of SARS-CoV-2 and other bacterial/viral or even fungal co-infections, which should be investigated to find whether co-infections might play a role in disease severity and/or mortality [2]. In our case series, only one patient needed ICU admission, no patients died and the median duration of hospitalization was 28 days.

The present report has several limitations. First, not all the hospitalized patients with SARS-CoV-2 infection were tested also for C. pneumoniae and M. pneumoniae; therefore, we could present only a part of patients with serological detection of atypical pathogens and infection with SARS-CoV-2 and the real incidence of co-infection cannot be truly established, requiring the need of testing always for pathogens other than SARS-CoV-2. Then, for the diagnosis of co-infections we could rely only on serology, since molecular analyses of respiratory samples specifically detecting M. pneumoniae or C. pneumoniae were lacking. In fact, although rarely, serology might be limited by possible false positive results, which should always be taken into account when deciding to exclude SARS-CoV-2 infection. One additional limitation is represented by the lack of paired samples to confirm prior serological results for the diagnosis of atypical pathogens.

However, with these limitations in mind, we reported for the first time the clinical characteristics of patients with C. pneumoniae, and not only M. pneumoniae, as a co-existing pathogen during SARS-CoV-2 infection. Therefore, the present report opens the path to additional studies investigating the real incidence of co-infections during SARS-CoV-2 epidemic and their possible impact on infection severity and mortality. Not less important, keeping in mind that in the future SARS-CoV-2 might be sporadic and not the cause of a pandemic infection anymore, we could infer that the serological detection of these atypical pulmonary pathogens in subjects presenting with respiratory symptoms cannot be used to rule out a diagnosis of COVID-19 [2, 4, 7, 23]. On the other hand, the reliability of serology for atypical bacteria should be considered when excluding the diagnosis of COVID-19 in patients with nasopharyngeal negative swabs (which has been demonstrated to occur in a not-negligible percentage of cases) [24], symptoms highly suggestive of SARS-CoV-2 infection and positive serology for other pathogens. Based on these considerations, physicians should assume that the presence of a pathogen other than SARS-CoV-2 does not ensure that a subject does not have also COVID-19.

In conclusion, SARS-CoV-2 infection might be associated with other common respiratory pathogens, including those causing atypical pneumonia. This finding should be considered in the near future, especially when ruling out the diagnosis of COVID-19. Therefore, the search for SARS-CoV-2 infection should be added to routine diagnostic testing even though other common respiratory pathogens are detected. Further studies are needed to evaluate the possible influence of co-infections on the severity of SARS-CoV-2 infection.

Data availability statement

The data used to support the findings of this study are available from the corresponding author upon request.