Clinical characteristics of severe influenza virus-associated pneumonia complicated with bacterial infection in children: a retrospective analysis

Pneumonia due to influenza virus infection is a common cause of hospitalisation among children. In severe cases, patients may experience respiratory failure, septic shock, or even life-threatening conditions [3, 4, 6]. One study reported that > 10,000 children around the world aged ≤ 5 years die from influenza virus-associated pneumonia annually, with most being paediatric patients with influenza virus-associated pneumonia complicated with bacterial infection [7]. Therefore, clinicians should pay greater attention to concomitant influenza virus-associated pneumonia and bacterial infection in children, particularly in children with severe illness. In our study, the average age of the observation group was significantly lower than that of the control group, indicating that younger age is associated with a higher risk of secondary bacterial infection in severe influenza virus-associated pneumonia. We also found that children aged ≤ 5 years comprised 95.5% (42/44) of the patients in the observation group, which was similar to the findings reported by Zhou et al. [1]. Besides, the observation group had a higher proportion of patients with underlying diseases than the control group. There was also one death in the observation group, which occurred in a patient with a history of epilepsy.

Given these findings, the possibility of concomitant bacterial infection in patients aged ≤ 5 years with severe influenza virus-associated pneumonia should be taken seriously. Relevant guidelines developed by experts in China and other countries have stated that children aged ≤ 5 years are not only susceptible to influenza virus infection but are also a high-risk population for secondary bacterial infection [8, 9]. This susceptibility may be explained partly by the fact that influenza virus replication in the respiratory tract reduces ciliary beat frequency in the airway, and the haemagglutinin and neuraminidase proteins of the virus alter the surface protein receptors of infected cells, thereby providing binding sites for bacterial adhesion [10]. Another key cause is immune cell dysfunction induced by the influenza virus or host factors. Previous research has indicated that immune cells such as alveolar macrophages, neutrophils, and natural killer cells form the second line of defence of the human body against bacterial invasion. However, influenza virus infection can destroy this line of defence, resulting in the weakening of phagocytosis, chemotaxis, and intracellular killing; such a mechanism is especially prominent in infected patients with reduced immune function [11]. In addition, these findings also suggest that children with underlying diseases may be more prone to developing concomitant bacterial infection when they have severe influenza, and the presence of underlying diseases is a risk factor for progression to severe disease or even death in paediatric patients with influenza [12]. Therefore, particular attention should be paid to paediatric patients with influenza and underlying diseases in clinical practice.

The reported incidence rates of influenza virus infection complicated with bacterial infection have varied over recent years. In a 2009 multicentre study conducted at 17 hospitals in China, 14.0% of children with influenza virus infection had a concomitant bacterial infection [13]. A 2018 study involving 838 paediatric patients with influenza in 35 PICUs in the United States reported the presence of concomitant bacterial infection in 274 (32.7%) patients, [14] a higher incidence rate than that reported in the 2009 study. Our results showed that the proportion of patients with influenza virus infection and with concomitant bacterial infection was 68.74% (44/64), which was higher than the aforementioned incidence rates. Certain biases may exist in our study due to its single-centre design and small sample size; however, the overall increasing trend in the proportion of paediatric patients with severe influenza virus-associated pneumonia complicated with bacterial infection cannot be ignored, especially in severe paediatric cases. We also observed clear changes in the bacterial spectra of patients with concomitant influenza and bacterial infection. The previously mentioned multicentre study concerning patients with influenza and bacterial coinfection in China found that gram-positive bacteria accounted for > 60% of all cases [13]. The results of a meta-analysis of > 3000 cases of influenza and bacterial coinfection in various countries in 2003–2014 also showed that gram-positive bacteria predominated, with Streptococcus pneumoniae and Staphylococcus aureus infections being the most common [15]. Our results differed from those of the aforementioned studies in that gram-negative bacterial infections accounted for 75% (33/44) of patients with severe influenza virus-associated pneumonia complicated with bacterial infection, with H. influenzae and M. catarrhalis accounting for 40.9% (18/44) and 27.3% (12/44) of patients, respectively. Possible reasons for the predominance of gram-negative bacteria in the present study are as follows. First, in recent years, changes have occurred in the bacterial spectra of lower respiratory tract infections in children, causing the incidence rate of gram-negative bacterial infections to exceed that of gram-positive bacterial infections. A 2015 Chinese study by Li et al. [16] reported that gram-negative bacteria accounted for approximately 74.1% of lower respiratory tract infections in children, which was significantly higher than the incidence rate for gram-positive bacterial infections. In 2019, Lim et al. [17] reported that, in Korea, H. influenzae accounted for 25.6% of cases of community-acquired pneumonia with viral and bacterial coinfection and that the rate of coinfection exhibited an increasing trend, which is consistent with the H. influenzae infection rate of 40.9% (18/44) observed in our study. Second, differences in bacterial strains causing secondary bacterial infections may exist between patients with severe influenza virus-associated pneumonia and those with common influenza virus infection. Kim et al. [18] reported that secondary bacterial infections in mechanically ventilated patients with influenza type A infection were most commonly due to Acinetobacter baumannii, Klebsiella pneumoniae, and Pseudomonas aeruginosa, which are all gram-negative bacteria. Third, factors such as differences in regional climate and sample size may also affect the conclusions obtained in different studies. Nonetheless, paying attention to changes in the bacterial spectra of concomitant bacterial infection in paediatric influenza virus-associated pneumonia is of practical significance for early diagnosis confirmation and timely administration of targeted antibacterial treatment.

Influenza and bacterial coinfection lack specific manifestations in the early stage and can only be confirmed through a combined analysis of clinical symptoms, aetiology, and other laboratory examination results. Our findings showed that patients in the observation group were more prone to longer pyrexia duration and clinical symptoms such as gasping, seizures, and consciousness disturbance than patients in the control group. This suggests that the development of the aforementioned symptoms in children with severe influenza virus-associated pneumonia may serve as a warning sign of the possibility of concomitant bacterial infection. The observation group also exhibited significant increases in inflammatory indicators such as C-reactive protein and procalcitonin levels, indicating that these elevated levels in paediatric patients with severe influenza virus-associated pneumonia may assist in the early identification of bacterial infection. In addition, lactate dehydrogenase (LDH) levels were also significantly elevated in the observation group compared with the control group. A retrospective study of Japanese patients with pneumonia showed that the LDH level can serve as an indicator of lung tissue damage in children [19]. Our study results also suggest that LDH may also be used as a predictive indicator of the presence or absence of concomitant bacterial infection in children with severe influenza virus-associated pneumonia. However, further deliberation is required for the establishment of specific quantitative criteria.

In this study, the bacterial species most commonly detected in the patients were H. influenzae and M. catarrhalis. We found that H. influenzae was susceptible to cefotaxime, ceftriaxone, and meropenem but exhibited a resistance rate of 61.7% to ampicillin, which is close to the resistance rate of 69.19% reported by Mai et al. [20]. This result suggests that ampicillin may no longer be considered the drug of choice for the treatment of H. influenzae infection. Third-generation cephalosporins may instead be selected for initial empiric treatment. M. catarrhalis was susceptible to ampicillin/sulbactam, amoxicillin/clavulanate potassium, and ciprofloxacin, and exhibited resistance rates of 59.4% and 13.3% to ampicillin and co-trimoxazole, respectively. Therefore, we recommend the use of penicillin-based compound formulations as first-choice drugs for initial empiric treatment of M. catarrhalis infection in children. Ricketson et al. [21] reported that patients who did not receive standardised antibiotic treatment had a 5.6-fold increase in case fatality rates compared with a standardised treatment group. Similarly, we found that satisfactory therapeutic effects could be achieved with early administration of antibiotics against susceptible bacteria in the observation group.

This study had some limitations. First, this was a single-centre retrospective study that included patients from the past four years. Therefore, the data used for analysis were limited and may not fully reflect the characteristics of severe influenza virus-associated pneumonia complicated with bacterial infection in children. Second, all patients included in this study originated from Xiamen, Fujian Province, in Southern China, which may have created certain biases given the existence of genetic variation and differences in prevalent viral and bacterial strains among different geographical regions. In future research, we aim to address these shortcomings and increase the sample size to enhance the accuracy of our results.

In conclusion, severe influenza virus-associated pneumonia complicated with bacterial infection was found to be more common in paediatric patients aged ≤ 5 years. Younger patients with underlying diseases were more susceptible to bacterial infection. Secondary bacterial infections were mainly due to gram-negative bacteria, with H. influenzae and M. catarrhalis being the most commonly detected pathogens in the patients. The clinical manifestations of severe influenza virus-associated pneumonia complicated with bacterial infection lacked specificity. However, the following may serve as warning signs of the possibility of concomitant bacterial infection: severe clinical manifestations such as persistent high-grade pyrexia, consciousness disturbance, and gasping during the early stage; elevated C-reactive protein, procalcitonin and LDH levels; and non-improvement of disease condition or relapse following an initial response when antiviral treatment was administered. Once the presence of bacterial infection has been confirmed, empiric antibiotic therapy should be initiated as soon as possible.