A Retrospective Study of Coinfection of SARS-CoV-2 and in 11 Hospitalized Patients with Severe COVID-19 Pneumonia at a Single Center

Background

Since December 2019, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus had caused over 3 million cases of coronavirus disease 2019 (COVID-19) and more than 150 000 deaths in the U.S. [1]. As the pandemic evolves, the scientific and medical community accumulates more evidence of the disease’s clinical characteristics; however, information regarding bacterial coinfections is still minimal and sometimes overlooked [2].

Between 1918 and 1919, the “Spanish” influenza pandemic caused around 50 million deaths worldwide [3]. Preserved lung tissue sections and autopsy analyses have shown that most of these deaths likely resulted directly from secondary bacterial pneumonia by common upper-respiratory tract bacteria, with Streptococcus pneumoniae playing a significant role [4]. This lethal synergism between the influenza virus and S. pneumoniae is not limited to pandemic strains; during the seasonal influenza period, 41% of bacterial coinfections are attributed to S. pneumoniae [5].

In contrast, bacterial coinfection is considered relatively infrequent in hospitalized patients with COVID-19, and the co-prevalence of S. pneumoniae is even lower [6–10]. Lin et al. documented for the first time the coinfection of the SARS-CoV-2 with multiple common viral respiratory pathogens taking place in the community [6]. Subsequent cohort studies documented bacterial coinfections, with only a few studies reporting S. pneumoniae coinfection [7,10], generally with good outcomes. Therefore, this retrospective study aimed to describe 11 hospitalized patients with a coinfection of SARS-CoV-2 and S. pneumoniae diagnosed with severe COVID-19 pneumonia at a single center.

Material and Methods

We retrospectively analyzed clinical characteristics and outcomes of the patients subsequently admitted to AMITA Health Saint Francis Hospital between March 1 and June 30, 2020, with documented SARS-CoV-2 infection by RT-PCR or isothermal nucleic acid amplification performed on nasopharyngeal throat swab specimens (Abbott™ RealTime™ SARS-CoV-2 assay and Abbott™ ID NOW COVID-19™ assay) and hospitalized with symptomatic COVID-19 illness.

The Abbott™ RealTime™ SARS-CoV-2 assay is a dual-target realtime reverse transcription-polymerase chain reaction assay for the RdRp and N genes of the SARS-CoV-2 virus. In contrast, the Abbott™ ID NOW COVID-19™ assay is a rapid molecular in vitro diagnostic test that utilizes an isothermal nucleic acid amplification technology intended for the qualitative detection of nucleic acid from the SARS-CoV-2 viral RNA. Both tests were provided an Emergency Use Authorization (EUA) by the U.S. FDA for use by authorized laboratories. S. pneumoniae infection was diagnosed in patients with positive blood or sputum cultures or urinary antigen. For cultures, B.D.™ Taxo™ P Discs for differentiation of pneumococci were used, with bile solubility as a confirmatory test (B.D.™ Desoxycholate™ Reagent). The urine antigen was detected with a rapid immunochromatographic test (Abbott™ BinaxNOW™ Streptococcus pneumoniae Antigen Card Test Kit).

The study and creation of this de-identified dataset were approved by the Institutional Review Board of AMITA Health System (2020-0128-02). The Ethics Commission waived the requirement for informed consent given this research involves no more than minimal risk to subjects. Descriptive statistics were used to summarize results. The data were analyzed using SPSS Version 23.0 (IBM Corp, Armonk, NY).

Results

We identified 11 patients with S. pneumoniae coinfection: 9 patients with positive urine antigen, 1 with positive blood cultures, and 1 with positive sputum cultures. Detailed clinical characteristics and outcomes of the patients are shown in Table 1. The median age was 77 years (interquartile range [IQR], 74–82 years), 45.5% (5/11) were males, 54.5% (6/11) were white, and 90.9% (10/11) were long-term care facility (LTCF) residents. The most common comorbidity was hypertension (9/11, 81.8%), and the most common symptom on admission was shortness of breath (9/11, 81.8%). The median temperature at presentation was 38.8°C (IQR, 38.1–39.5°C), the median systolic blood pressure was 103 mmHg (IQR, 80–145 mmHg), the median respiratory rate was 28 rpm (IQR, 22–38 rpm), the median heart rate 106 bpm (IQR 98–119 bpm), and the median lowest documented oxygen saturation in the emergency department was 91% (IQR, 85–96%). Unilateral infiltrates were initially found in 45.5% (5/11) of the patients, whereas 54.5% (6/11) had bilateral infiltrates. All the patients received intravenous (IV) antibiotic therapy and only 3 patients were ultimately intubated. The median length of stay was 7 days (IQR, 6–8 days). Among 11 patients, 4 patients were discharged in stable condition, and 7 had died, resulting in an inpatient mortality rate of 64%. The Intensive Care Unit (ICU) admission rate was 36%, and the invasive mechanical ventilation (IMV) rate was 27%.

Discussion

In this retrospective study, we describe 11 hospitalized patients with a coinfection of SARS-CoV-2 and S. pneumoniae who were diagnosed with severe COVID-19 pneumonia in a community hospital located in the northern Chicago area, Illinois, U.S. The majority of patients in this series presented with acute distress, namely fever and shortness of breath, had elevated inflammatory markers, showed unilateral or bilateral infiltrates in chest X-rays, and required oxygen support. The mortality rate among this group was very high (64%). However, although all these patients carried an elevated disease burden at baseline, almost all of them were LTCF residents. All of them had 2 or more comorbidities, which has to be taken into consideration. Additionally, most patients presented with advance directives; therefore, aggressive therapy was not pursued, as demonstrated by the low critical care utilization rate.

S. pneumoniae is a gram-positive diplococcus that colonizes the upper respiratory tract of up to 30% of healthy adults. It is the most frequently isolated bacterial pathogen in community-acquired pneumonia and the predominant copathogen in influenza coinfection [4,5]. Experimental models have demonstrated the lethal synergism between the influenza virus and S. pneumoniae, with mechanisms that include pulmonary epithelial damage associated with increased bacterial adherence, sensitization of cells to secondary infection via expression of toll-like receptors, suppression of γδ T cell production of interleukin-17 by type I interferon favoring bacterial colonization in the lungs, functional impairment of neutrophils and macrophages (i.e., impaired phagocytosis), and worsened inflammatory disease through the release of neutrophil extracellular traps and a cytokine storm as a result of significant neutrophils influx triggered by viral and bacterial toxins [5].

Bacterial coinfection is considered uncommon in patients with COVID-19 who are newly admitted to the hospital [6–10]. Initial Chinese reports found a 5.1% to 38.9% incidence rate of secondary or hospital-acquired bacterial coinfections, and later reports in Western countries showed coinfection rates of 4.8% to 27.4% [2]. A retrospective cohort study that included 836 hospitalized patients with confirmed SARS-CoV-2 found that 27 (3.2%) patients had early confirmed bacterial isolates identified (0–5 days after admission), rising to 51 (6.1%) throughout the hospital stay. However, no patients with pneumococcal coinfection were identified [11]. Another large cohort study with 989 hospitalized patients identified a total of 88 non-COVID-19 infections in 72 patients (7.3%). Among those, 74 were bacterial, with S. pneumoniae being the most commonly isolated bacteria (12 cases). However, the outcomes of these patients were not reported [7]. In contrast, Adler et al. identified 5 patients with pneumococcal coinfection among 195 patients with COVID-19, all of which survived to hospital discharge [10]. Furthermore, a meta-analysis by Lansbury et al. of 30 studies that included 3834 patients found that 7% of hospitalized patients with COVID-19 had a bacterial coinfection, increasing to 14% in studies that only included ICU patients. The most common organisms were Mycoplasma pneumoniae (47% of 27 confirmed bacterial isolates), Pseudomonas aeruginosa (12%), and Haemophilus influenzae (12%), with little evidence of S. pneumoniae having a significant role [12].

Despite the low frequency of coinfections reported in patients with COVID-19, they are associated with poor outcomes. Up to 50% of patients who have died of COVID-19 had secondary bacterial infections [8]. Furthermore, in the review mentioned above by Lansbury et al., the pooled analysis of crude odds ratios for death indicated that COVID-19 patients with a coinfection were 5.82 times more likely to die than patients who did not have a coinfection [12].

The present study has several limitations. First, this was a single-center study, and these findings may be subject to selection bias, and generalizability to larger populations of patients is limited. Second, given the situation’s novelty, our institution did not have standard practices regarding testing for other respiratory pathogens. This was subject to each attending physician’s discretion; hence, there may have been high microbiological sampling variability with only sicker patients being tested for other pathogens. Third, 90.9% of the patients studied were long-term care facility residents. These patients present a higher burden of underlying conditions and low baseline functional status, which may account for the higher mortality observed compared to other cohorts. Fourth, given the study’s observational nature, causality cannot be inferred from an uncontrolled observation. Whether there is a lethal synergism between SARS-CoV-2 and S. pneumoniae similar to that of the influenza virus needs to be explored in further basic and clinical research.

Conclusions

At our center, 11 patients with COVID-19 pneumonia who had confirmed infection with SARS-CoV-2 were diagnosed with Streptococcus pneumoniae infection while in hospital. All patients had pneumonia confirmed on imaging and a nonspecific increase in markers of inflammation. The in-hospital mortality rate of 64% (7 patients) was higher in this group than in previous reports. This study highlights the importance of monitoring bacterial coinfection in patients with viral lung infection due to SARS-CoV-2.