Secondary bacterial infections
Secondary bacterial infections are a known complication of RSV bronchiolitis, of which UTI is the most extensively studied. However, many of these studies have used various UTI definitions with differing levels of sensitivity and specificity. For instance, a recent meta-analysis reported that 3.1% of children with bronchiolitis were also diagnosed with a UTI. Many studies included only children under three months of age and not 24 months. Early detection of UTI can profoundly impact the clinical outcome of hospitalized infants.
In our cohort, 23 patients (1.2%) were diagnosed with UTI, with a relative risk of 11 for the high CRP group in a multivariable analysis. However, this association was highly sensitive but not specific. The highest AUC was in children 3-12 months old with a threshold of 3.42 mg/dL (AUC 0.79), although further investigation is needed since UTI incidence in our cohort was very low.
Lobar/alveolar pneumonia is another important complication of RSV bronchiolitis, but its definition varies in the literature and is based on clinical and radiographic features, often assisted by laboratory markers. A recent study from our center has demonstrated that RSV is associated with community-acquired pneumonia in a multivariable model in 50% of cases, most notably in children under 2 years of age (62%), indicating a significant role in its pathogenesis. This finding is supported by the decline in pneumonia and RSV prevalence during the COVID-19 pandemic despite persistent nasopharyngeal carriage of S. pneumoniae.
Several studies have assessed the association of elevated CRP with pneumonia in the context of RSV. One multinational study reported that increased CRP levels (>4mg/dL) were associated with bacterial rather than viral pneumonia in otherwise healthy children. However, another study showed only a slightly elevated CRP level in bacterial pneumonia (1.06 Vs. 0.62 mg/dL). In a prospective study of children with bronchiolitis in a PICU setting, elevated CRP and procalcitonin levels were significantly associated with pneumonia or sepsis. These results are similar to our study in the association of CRP with severe disease, but differ in the study population, which only isolated RSV in 65% of patients and only included severe cases, who often have other predisposing conditions that excluded them from our study. Our study found that elevated CRP was associated with secondary pneumonia (9% in the high-CRP group vs. 4.3% in the low group) with a relative risk of 1.98 in a multivariable analysis. However, we found that CRP had limited predictive ability, as the AUC for the suggested cutoff was low.
The main strength of our research is the large cohort size, which is larger than previous studies that have assessed CRP in acute bronchiolitis in children. Additionally, being a single tertiary medical center, SUMC was able to maintain a consistent clinical approach, with standardized protocols for hospitalization, discharge, antibiotic stewardship, and laboratory and radiologic evaluation, which further strengthens the reliability and validity of our findings.
Limitations of this study include its retrospective nature, as the inclusion and exclusion criteria may have resulted in the exclusion of mild cases who were likely discharged without CRP measurement, as well as critically ill patients with severe risk factors. Also, our definition of secondary infection was based on the electronic medical records, which were determined by the treating physicians and not based on interpretation of radiological findings, with fixed criteria. As such, there may be potential for under or overdiagnosis. Furthermore, an elevated CRP level may have influenced the given pneumonia diagnosis, which was also based on inflammatory markers and not merely based on radiologic findings. Lastly, our medical documentation system prevented us from further analyzing data on the use of high-flow nasal cannula in the pediatric ward setting, which was implemented in 2020.
This study was conducted with the aim of providing clinicians an easy and accessible tool to aid decision making and management of hospitalized children with RSV bronchiolitis. Specifically, we sought to determine the usefulness of CRP levels measured at ED admission as an early detection method for severe disease. Our findings demonstrate that CRP can be a valuable tool for informing decisions related to hospitalization, diagnostic evaluation, and communication with parents about the anticipated duration of hospitalization.