Discussion
This study reported that children between 1-60 months with viral coinfection increased the risk of admission to the ICU for HFNCO/BiPAP/IMV assistance. While RSV and HRV (except for between 25-60 months) coinfections were associated with severe disease in all ages, PIV3 (7-24 months) and HBoV (7-12 months) coinfections led to severe LRTI in early childhood. Besides, our study showed that influenza A coinfection was independently a risk factor for severe LRTI in children between 7-12 months and 25-60 months.
In our study, children with viral coinfection constituted 28.3% of the study population, and HBoV was the most common virus after RSV and HRV. Previous studies conducted in Barcelona and Rome20,21, where the Mediterranean climate is dominant, like Izmir, had similar results to that presented here. On the contrary, other studies reported from Canada and Japan7,22, where the continental and subtropical climate is dominant, found that viral coinfection was 17.2% and 15.8% of the patients, with the third common virus as influenza A and HMPV, respectively. These results emphasize that the seasonal distribution and frequency of viruses may vary in different countries according to their geographical and meteorological characteristics. Furthermore, the annual circulation of infections and the composition of the study population and method used in virus detection may also affect the viral coinfection rate. For example, the studies conducted in infants or hospitalized patients may determine a higher viral coinfection rate than those held in adults or population-based studies because of the more frequent detection of multiple-viruses in these patients21. The authors should interpret the viral coinfection rate carefully among different studies. Children with viral coinfection were 3.4 times more at risk of admission to the ICU than those with a single virus infection, concordant with the previous studies6,23. Several possible explanations for these results are that the direct interactions among viral genes or gene products and host immune system, the indirect interactions of host environmental changes, and the immunological interactions may determine the course of LRTI in children with viral co-infection24,25. The first virus may boost viral superinfection by consuming host defense, similar to bacterial infection26. Moreover, the coinfection of two distinct virus types may have affected the natural course of LRTI in children.
Although HRV generally causes a common cold in children27, it reduces cell proliferation and bronchial epithelial cells’ self-repair capacity28. Thus, any secondary virus infection may increase the risk of severe illness. Besides, HRVs are a large group of genetically diverse RNA viruses, and HRV sub-types, particularly HRV-A and -C, have been shown to cause severe LRTI in children29. Unfortunately, we could not determine HRV sub-types in the respiratory specimens because of our virus detection method. These sub-types might have infected some severe cases. This issue stands as an interesting subject for further research. Richard et al.6 found that infants with RSV coinfection were at 2.7 times higher risk of admission to the ICU, while Semple et al.23 determined a 10-fold higher risk of IMV in children under two years of age who had RSV and HMPV coinfection. Additionally, our results suggested that RSV coinfection remains an essential agent for severe LRTI not only in children younger than 24 months but also in those beyond 24 months. Since there is no existing vaccine for RSV, the consideration of passive immunoprophylaxis during RSV season may protect children with an underlying medical condition from severe RSV co-infections17.
Coinfection with other viruses is very common during influenza infection30. In recent meta-analyses31,32, COVID-19 coinfection was identified more commonly with influenza A and RSV. Although it was suggested that children with COVID-19 had a better prognosis than adults, influenza A coinfection with COVID-19 may inhibit the host’s immune system, increase antibacterial therapy intolerance, and be harmful disease’s prognosis31. Besides, secondary bacterial infections may exist at a rate of 40% in influenza A infections33. Therefore, we note that any secondary bacterial infection, which was not detected, might increase influenza A’s severity. However, a comprehensive and extensive sample size study should be needed. Although there is no licensed influenza vaccine for children under six months, alternative strategies, including maternal vaccination during pregnancy and household vaccination, might reduce severe influenza infections. Increasing influenza and bacterial vaccination, with environmental precautions such as frequent hand washing, decontamination of hands, and cleaning concrete surfaces with water and disinfectants, are essential to prevent transmission of the respiratory viruses and reduce the severe disease burden. The efficacy of neuraminidase inhibitors in healthy children is limited and does not recommend general treatment34. Nevertheless, early initiation of neuraminidase inhibitors is associated with shorter symptoms, decreased complications, and hospitalization34.
Human bocavirus was first identified in children’s respiratory tract in 2005; 75% of the HBoV infections are associated with multiple viral infections16. HBoV infection usually results in a mild, self-limiting respiratory tract infection and might even be asymptomatic4. However, several case reports reported that HBoV coinfection with other viruses could cause complications such as pneumothorax, pneumomediastinum, and severe respiratory failure requiring ICU/IMV35. Slow elimination of the viral antigens by the immature immune system might explain the coinfection with HBoV and the severity of the illness in infants. Considering the National Respiratory and Enteric Viruses Surveillance System study conducted from 1990 to 200436, PIV3 (52%) was the most frequently determined serotype. In the US, the estimated LRTI and hospitalization related to PIV3 were reported at 1.1million and 29.000, respectively37. Parainfluenza virus 3 leads to LRTI more common than other serotypes in neonates and infants and is clinically indistinguishable from RSV infection. Additionally, PIV and RSV belong to the Paramyxoviridae family, enveloped RNA, similar epidemiologic, and clinical outcomes22. Thus, children aged younger than 24 months with PIV3 coinfection may need equal medical attention to those with RSV coinfection regarding disease severity.
Our regression analysis determined that young age, prematurity, malnutrition, exposure to tobacco smoke, and atopy history were independent risk factors for severe LRTI, aligned with earlier studies6,15. Young age and prematurity cause more severe respiratory stress because of the limited aerobic respiratory capacity associated with the relatively smaller airway size, the immune system’s naivety, and less strength and respiratory muscle endurance24. Malnutrition may result in secondary immunosuppression, atrophy of the respiratory muscles, and inadequate muscle contraction because of the electrolyte disturbance15. Taking measures to prevent malnutrition, such as providing information and supporting the mother about breastfeeding, using prophylactic vitamin D and iron supplements, and a regular follow-up of the anthropometric measurements, may protect children from severe disease.
An unanswered question has been whether the presence of a pre-existing abnormality of the immune response in some infants leads to severe illness. A recently published study has demonstrated that both viral infection and allergic sensitization are strongly correlated with asthma development after six years old in children14. Furthermore, severe disease is found to be related to allergic sensitization before symptomatic HRV infection. Our results showing that allergic sensitivity increases the disease’s severity is one of our relevant findings, which are rarely supported by the literature but can shed light on studies investigating this relationship. Tobacco smoking continues to be a fundamental health problem worldwide, and a recently published meta-analysis has confirmed the effect of exposure to tobacco smoke on developing LRTI in children38. Increasing comprehensive bans on tobacco advertising, promotion, sponsorship, and tobacco taxes might reduce tobacco consumption. Additionally, ample pictorial or graphic health warnings, with hard-hitting messages, might convince smokers to protect the health of non-smokers by not smoking inside the home38.
In children with CLD, the increased inflammatory markers, impaired airway anatomy, and mucociliary clearance could support the progression to severe LRTI, whereas altered pulmonary mechanics, cyanosis, pulmonary hypertension, and ventilation-perfusion mismatch in children with CHD could increase the severity of the disease. Children with NMD expose severe illness because of impaired or inadequate mucociliary activity and immobility. These findings are essential when considering the increased survival of children with a chronic underlying condition. The presence of neutrophilia, lymphopenia, and high CRP values were independent risks for severe disease. Previous studies have shown that many neutrophils detected during influenza A and RSV infections are associated with a more severe condition, and lymphopenia increases virus replication39,40. The prognosis can be predicted to be more severe in children with neutrophilia, lymphopenia, and high CRP values. However, larger-population studies are required to monitor children with LRTI according to hematologic values.
Our study’s limitations include those related to retrospective studies, including bias regarding patient selection and accuracy related to the medical record. To minimize bias, we developed a standardized data form to guide data collection, only included patients with the medical records’ information, and the same experienced clinician performed data collection. Second, patients who visit the hospital later may present with more severe signs and symptoms than those who visit the hospital immediately after illness onset. However, we could not compare this potential effect on disease severity because of the study’s retrospective design. Third, including only children presented to the tertiary hospital might have resulted in the study population’s heterogeneity. Finally, a reliable test for bacterial co-detection was not available at the time of the study, which could cause that we might have underestimated the impact of bacteria on severe LRTI.
In conclusion, children between 1-60 months hospitalized with LRTI and detected viral coinfection were at about 3.5 times higher risk for HFNCO/BiPAP/IMV assistance. Respiratory syncytial virus and HRV (except for between 25-60 months) coinfections caused severe LRTI in all age groups, whereas PIV3 (4-24 months) and HBoV (7-12 months) coinfections were associated with severe LRTI in early childhood. Additionally, influenza A coinfection led to severe LRTI in children between 7-12 months and 25-60 months.