Clinical Presentation and Co-Detection of Respiratory Pathogens in Children Under 5 Years with Non-COVID-19 Bacterial and Viral Respiratory Tract Infections: A Prospective Study in Białystok, Poland (2021-2022)

Background

Respiratory tract infections (RTIs) cause a wide spectrum of illnesses, from the common cold to severe pneumonia [1]. Understanding the etiology of these infections is important not only in everyday clinical practice, but also in public health. It enables the introduction of prevention methods (including vaccinations and early isolation of specific individuals), and the application of appropriate treatment, reducing the overuse of antibiotics. Most RTIs are caused by viruses, but it is clinically impossible to distinguish them from bacterial infections [2]. The seasonality and etiology of RTI was profoundly altered during the coronavirus disease 2019 (COVID-19) pandemic. Global influenza virus and respiratory syncytial virus (RSV) activity decreased markedly with the introduction of non-pharmaceutical interventions (NPIs) in the COVID-19 pandemic [3]. However, several non-SARS-CoV-2 respiratory pathogens were still circulating in communities [4]. Despite the ongoing implementation of community NPIs, the prevalence of rhinoviruses and respiratory enteroviruses rebounded rapidly and persisted [4]. Evidence from the initial wave of the COVID-19 pandemic in early 2020 suggests that co-infections of SARS-CoV-2 with other respiratory viruses were also observed. The incidence of these co-infections varied significantly, and their impact on clinical severity and prognosis remains unclear [5]. These findings highlight the ongoing prevalence and impact of other respiratory viruses during the pandemic, emphasizing the need for comprehensive surveillance and understanding of RTIs beyond COVID-19.

Conventional methods of pathogen detection, such as cultures of nasopharyngeal aspirates or throat swabs, serological titters, and antigen detection, often do not have sufficient sensitivity, are time-consuming, and require extensive equipment and qualified staff, which often makes them inaccessible in everyday practice [2]. The introduction of molecular methods in recent years has significantly improved the detection of pathogens causing infections, especially viruses. Multiplex polymerase chain reaction (PCR) tests simultaneously detect several pathogenic viruses and bacteria in a short time and with high sensitivity, so they can improve patient management [6,7].

Therefore, this study aimed to evaluate outcomes in 100 children under age 5 years diagnosed with non-COVID-19 bacterial and viral lower respiratory tract co-infections between October 2021 and May 2022 in Białystok, Poland.

Material and Methods

ETHICS STATEMENT:

The protocol was approved by the Bioethics Committee of the Medical University of Białystok prior to patient recruitment, and the study was conducted in accordance with the Helsinki Accords (protocol code APK.002.420.2020 approved on 17 December 2020). Informed consent was collected from parents/guardians of all study participants.

STUDY DESIGN:

This was a prospectively recruited observational cohort study done in Białystok, Poland between October 2021 and May 2022. We included 100 children under 5 years of age with symptoms of upper and lower RTI. To address potential sources of bias, the samples were collected from 1 family doctor clinic and 2 pediatric wards in Białystok, Poland. The number of nasopharyngeal swabs (NPSs) taken per week depended on the number of patients with RTI symptoms. To cover a wider time frame, the maximum number of samples collected per week could not exceed 10 NPSs. Clinical and demographic data were collected using a questionnaire completed by the consulting doctor. The NPS were obtained from patients on the day they were included in the study. The criteria for including patients in the study were the presence of signs of acute RTI (2 of the following: fever, runny nose, cough, dyspnea) and a negative test for SARS-CoV-2 infection in the GeneXpert assays (Cepheid; Sunnyvale, CA, USA). Routine blood tests (complete blood count and C-reactive protein) collected from hospitalized patients using standard methods were also recorded. Patients pre-treated with antibiotics, or diagnosed with chronic respiratory diseases or non-infectious diseases were excluded from the analysis.

MOLECULAR DETECTION OF RESPIRATORY PATHOGENS:

NPSs were collected by a standard method according to the instrument capacity and immediately put into a tube containing transport medium. Following the sample collection, nucleic acids were extracted from nasopharyngeal swabs with the use of magnetic bead method (MagPurix EVO 24 CE-IVD Nucleic Acid Extraction System, Zinexts Life Sciences Corp., Taipei, Taiwan). Furthermore, molecular screening for the particular infectious agents was performed utilizing the nucleic acid amplification test (NAAT) with subsequent real-time signal detection performed using the CFX96TM Real-Time PCR Detection System (Bio-Rad, Hercules, California, USA).

Molecular screening covered RNA viruses (HMPV – human metapneumovirus; EV – enterovirus; HPIVs – human parainfluenza viruses 1–4; HRV – human rhinovirus A/B/C; HCoV 229E – human coronavirus 229E; HCoV NL63 – human coronavirus NL63; HCoV OC43 – human coronavirus OC43; FluA – influenza A virus with H1, H3, and H1pdm09 subtyping; FluB – influenza B virus; hRSV-A – human respiratory syncytial virus subtype A; hRSV-B – human respiratory syncytial virus subtype B), DNA viruses (HAdV – human adenovirus; HBoV – human bocavirus 1/2/3/4), and bacteria (Chlamydophila pneumoniae, Mycoplasma pneumoniae, Legionella pneumophila, Bordetella pertussis, Bordetella parapertussis, Streptococcus pneumoniae, and Haemophilus influenzae). Amplification and detection were performed utilizing AllplexTM Respiratory Panels 1–4 (Seegene, Inc, Seoul, South Korea). The results of the PCR test were available after the study was completed (May 2022) and thus had no impact on the clinical care.

STATISTICAL ANALYSIS:

Statistical analysis was done using TIBCO Software Inc. (2017) Statistica, version 13 (Palo Alto, CA, USA) and GraphPad Prism version 9.4.0 for Windows, GraphPad Software. The quantitative data are expressed as median, maximum, and minimum values. The qualitative variables are expressed as absolute frequency and percentage. Comparative statistics included the chi-square test and Mann-Whitney U test. Multivariable logistic regression models were used to identify variables that were associated with the risk associated with the detection of at least 2 pathogens. In these models the types of pathogens detected (bacterial/viral/viral-viral/bacterial-viral) were modeled against the presence of symptoms and demographic data of patients. Statistical significance was defined as a p value of less than 0.05.

Results

CHARACTERISTICS OF THE STUDIED POPULATION:

The median age of patients was 24 months, and 54% of patients were male. Almost half of the patients (45%) were toddlers aged 13–26 months. Among the participants of the study, 56% were hospitalized in pediatric departments and 44% were consulted in the outpatient clinic. Most of the children (57%) were attending day-care units (a nursery or a kindergarten). The demographic features of the studied group are shown in the Table 1.

DETECTION OF RESPIRATORY PATHOGENS IN THE NASOPHARYNGEAL SWABS:

Among 100 respiratory samples tested, 90 were positive and 10 were negative for pathogens. A total of 15 different pathogens were detected, and co-detections were common (54% of all patients). The highest incidence of RTIs was observed from November to January. However, we did not observe any apparent seasonality of detected pathogens (Figure 1). Viral pathogens were detected in 66% of outpatients and 75% of hospitalized children. Bacteria were detected in 59% of hospitalized and 59% of patients consulted in the clinic. Haemophilus influenzae (48%) and Streptococcus pneumoniae (30%) were the only 2 bacterial pathogens detected in this study. Human rhinovirus (32%) was the leading viral pathogen, followed by HAdV (16%) (Table 2). A single detection was seen in 36%, dual in 37%, triple in 15%, and quadruple in 2% of patients. Bacterial pathogens were co-detected with viruses in 40 children. This constitutes 68% (40/59) of patients in whom a bacterial pathogen was detected (Table 3). Rhinoviruses were the most common viruses co-detected with bacterial pathogens (n=15; 37.5%). Two different viruses were found in 14 children and the most common co-detection was adenovirus with rhinovirus (n=5; 36%). The influenza A virus was detected in 2 children (both in March 2022 only) and RSV A/B were identified in 5 patients (between December 2021 and February 2022). The following pathogens were detected in none of the patients: HPIV1, HPIV4, Flu A H1, Flu A H1pmd09, Flu B, Chlamydophila pneumoniae, Mycoplasma pneumoniae, Legionella pneumophila, Bordetella pertussis, and Bordetella parapertussis.

RESPIRATORY PATHOGENS ASSOCIATED WITH CLINICAL SYMPTOMS:

Wheezing (75% vs 22%; risk-ratio [RR] 10.8; 95% confidence interval [CI], 2.0–57.7; p=0.005) and dyspnea (63% vs 11% RR, 13.7; 95% CI, 2.8–66.0; p=0.001) were more likely to be present in infections caused by the HBoV, compared to other etiologies. Infections with HAdV were typically characterized by the presence of fever (88% vs 58%; RR, 5.0; 95% CI 1.1–23.4; p=0.04), but cough was slightly less common (56% vs 80%; RR, 0.3; 95% CI, 0.1–1.0; p=0.05), than in infections caused by other pathogens (Figure 2). The clinical picture of the disease was not associated with the presence of other pathogens, multiple detections, or with the detection of bacterial pathogens, as revealed by regression models. However, we found that children with a total number of symptoms equal to or higher than median (4 symptoms) were more likely to test positive for at least 1 virus (RR, 2.96; 95% CI 1.21–7.20; p=0.015). Detection of 2 different viruses or mixed viral-bacterial detection was not associated with the number of observed symptoms.

FACTORS INFLUENCING DETECTION OF BACTERIAL PATHOGENS AND MULTIPLE DETECTIONS:

Streptococcus pneumoniae and H. influenzae were detected in 35.1% and 57.9% of day-care attendees and in 23.3% and 34.9% of children staying home. In the univariate analyses, the risk of S. pneumoniae detection was not increased in day-care attendees, but the day-care-associated risk of H. influenzae detection was 2.57 (95% CI, 1.13–5.82; p=0.024). Boys attending day-care had an increased risk of testing positive for H. influenzae (RR, 4.25; 95% CI, 1.37–13.16; p=0.01) compared to girls. Among children staying home, girls were more likely to test positive for S. pneumoniae (RR, 5.37; 95% CI, 1.15–25.11; p=0.033) than boys. The presence of siblings who attend day-care was not associated with elevated risk of detecting bacterial pathogens for children staying home or for children attending day-care.

The risk of detecting bacterial pathogens was dependent on the time spent in a day-care unit. In the univariate analysis, the RR of detecting S. pneumoniae and H. influenzae in children spending over 40 h weekly at a day-care unit was 3.66 (95% Cl, 1.16–11.56; p=0.027) and 7.11 (95% CI, 2.00–25.24; p=0.002), respectively, compared to children who spent there less than 20 h weekly or did not attend day-care units at all. In the multivariable analysis, we found that longer hours at day-care units was the only significant factor associated with RR=11.02 of detecting any bacterial pathogens in NPSs (95% CI, 1.16–104.36; p=0.035). The risk of multiple detections was also higher in this group of children (Figure 3).

The risk of detecting none of the viruses was associated with attendance at day-care units. In 14 children, 2 different viruses were detected. This co-detection had no impact on the clinical symptoms. None of the analyzed demographic factors increased the risk of viral coinfection.

CLINICAL PRESENTATION IN HOSPITALIZED PATIENTS:

Compared to ambulatory care, children who were hospitalized more often had wheezing (42.9% vs 4.5%; p<0.001), malaise (28.6% vs 9.1%; p=0.016), fatigue (23.2% vs 6.8%; p=0.026), dyspnea (26.8% vs 0%; p=0.016), and poor appetite (42.9% vs 22.7%; p=0.035), but less often had runny nose (69.6% vs 88.6%; p=0.023). There were no differences in terms of age, sex, attendance at day-care, or exposure to tobacco smoke. Hospitalized children were less likely to test positive for HRV (RR, 0.27; 95% CI, 0.11–0.65; p=0.003). No other differences were found in the type and number of detected pathogens between children who were and were not hospitalized.

Concentrations of C-reactive protein (CRP) over 40 mg/L were seen in 13 (23%) of hospitalized patients. Children with elevated concentrations of CRP were more likely to test positive for HAdV (RR, 5.29; 95% CI, 1.32–21.23; p=0.019). The RR of elevated CRP in children with a mixed bacterial/viral detection was 3.86 (95% CI, 0.99–15.07; p=0.052).

USE OF ANTIBIOTICS:

Antibiotics were used in 45% of patients. Patients who tested positive for HAdV were more likely to receive antibiotics (RR, 4.17; 95% CI, 1.09–15.91; p=0.037), and those who tested positive for H. influenzae were less likely to receive antibiotics (RR, 0.23; 95% CI, 0.08–0.65; p=0.006). Demographic factors, such as age or time spent in a nursery/kindergarten, were not associated with antimicrobial treatment.

Discussion

In this study we found that during the COVID-19 pandemic, the detection of non-SARS-CoV-2 respiratory pathogens in RTI children was high and reached 90%, which was similar to the pre-pandemic period [8,9]. In contrast to before the COVID-19 pandemic, when RSV and flu predominated, in this study rhinoviruses were the most commonly detected viruses. Co-detection of 2 different respiratory pathogens was common, but was not associated with clinical presentation of the infection.

RTIs are among the most common reasons for pediatric primary care consultations and hospital admissions, posing a significant health burden [10–12]. Despite greater knowledge about the etiology of infection, in practice it is difficult to unequivocally identify the pathogen responsible for RTI, and thus to apply appropriate antimicrobial treatment. The main pathogens detected in this study were H. influenzae, HRV, S. pneumoniae, and HAdV. Haemophilus influenzae and S. pneumoniae are commonly found in the upper respiratory tract of humans [13]. The prevalence of nasopharyngeal carriage of S. pneumoniae and H. influenzae varies among populations and is influenced by factors such as age, living conditions, and vaccination status [14–16]. Understanding the dynamics of nasopharyngeal carriage is essential for developing effective strategies to prevent and control invasive diseases in children [17]. Non-pharmaceutical interventions, like reduced person-to-person contact, masking, and washing hands, not only were helpful in containing the spread of COVID-19, but also have had a huge impact on transmission of other infectious diseases [18–23]. In a Polish study done before the pandemic between 2016 and 2020, S. pneumoniae was detected with PCR in 39.1% of children between 1–5 years old [24]. In this study we observed a similar detection rate of S. pneumoniae. Although national-level behavioral restrictions remained high throughout 2021, from April 2021 onward day-care units were fully opened in Poland, what might have affected our results. In line with previous studies [13,16], in this study we found that horizontal transmission of bacterial pathogens among children staying in day-care may be an important factor for obtaining carrier status.

The most frequent viruses reported in our analysis were HRV and HAdV, which were commonly detected in patients with RTIs before [25,26] and during the COVID-19 pandemic [27,28]. On the other hand, the RSV and influenza viruses were rarely detected in this study compared to the pre-pandemic period [7,29]. We found that RSV and influenza virus were actively circulating in early 2022 in Poland, possibly leading the influenza and RSV outbreaks in late 2022. These outbreaks of non-COVID-19 respiratory infections were observed across the world [30–32]. A decrease in the number of COVID-19 cases was followed by an increase in prevalence of non-COVID-19 pathogens, indicating interactions between respiratory pathogens [34]. A recent study from Austria showed that the surge of RSV infections observed in 2022 was caused by lineages already present before the pandemic [33].

Bacterial-bacterial, bacterial-viral, and viral-viral co-detections were common, but did not add to the severity of disease in this study. Previous studies have suggested that polymicrobial infections enhance the RTI severity in children [35,36]. In contrast to our observations, in these studies the prevalence of RSV and influenza infections was high. Possibly, interactions between RSV and influenza viruses with bacteria in the pediatric respiratory tract contribute to greater disease severity than when other viruses are present.

Our study has several important limitations. First, the sample size was small. A larger group could have allowed for a better understanding of factors affecting detection of respiratory pathogens in RTI children. Second, we did not recruit healthy controls for comparison, so we could not determine whether the detected organisms were the causes of RTI, since those organisms are commonly found in healthy children. Third, this study was done in a single hospital and a single outpatient clinic, so the detected pathogens might not be representative of the true epidemiology. Moreover, although sensitivity of molecular methods used in this study is considered to be high, we cannot rule out the possibility that some of the pathogens were missed because of the limitations of this approach, including inhibition of the amplification reaction or the lack of a specific target [37].

Conclusions

The findings of this study show that rhinoviruses and other non-SARS-CoV-2 viral and bacterial pathogens were commonly detected in children under age 5, despite ongoing implementation of non-pharmaceutical interventions to contain the COVID-19 pandemic. Co-detection of 2 different bacterial and viral respiratory pathogens was common, but did not add to the severity of acute respiratory infections. Possible mechanisms that drive these bacterial-viral interactions require further investigation.