Imaging Practices of Children With Complicated Pneumonia Requiring Chest Tube

Patients with complicated pneumonia often require chest tube (CT) placement, and chest radiographs (CXRs) are commonly used for monitoring.1 Studies in patients in the pediatric intensive care unit have demonstrated that routine daily CXRs offer no significant differences in clinical outcomes when compared with clinically driven imaging.1 Multiple studies also demonstrate minimal benefit of CXRs following CT removal in asymptomatic patients.2,3 However, there are no guidelines for imaging pediatric patients with pneumonia who have CTs in place. In other conditions such as asthma, croup, and bronchiolitis, the Choosing Wisely campaign highlights the importance of imaging driven by clinical findings rather than routine practice.4 Overuse of CXRs could result in unnecessary radiation exposure, particularly in pediatric patients who are more sensitive to ionizing radiation than adults.5 Understanding current practice is critical to developing evidence-based strategies that optimize patient care while minimizing unnecessary interventions. The goal of this study was to assess institutional policies and practices for imaging pediatric patients with complicated pneumonia requiring a CT.A 29-item multiple choice survey created in REDCap was emailed to pediatric hospital medicine division directors of the 49 Pediatric Health Information System (PHIS) hospitals between September and November 2024. Additional details of survey development are described in the Supplement 1. Surveys were directed to the physician who could best describe typical institutional practices related to imaging hospitalized patients with a CT placed for complicated pneumonia. Surveys included hospital identifiers to ensure that only one was completed per institution. Respondents were offered a $10 gift card for participating. Descriptive analysis was conducted using Microsoft Excel. This study was approved by the institutional review board.Surveys were completed by 34 (69%) sites, most by hospitalists (n = 32, 97%) from free-standing children’s hospitals (n = 27, 79%) in urban areas (n = 29, 85%) (Table 1). More than half (n = 19, 56%) reported daily CXRs were typically obtained. Almost a third (30%) reported that they only obtained CXRs when clinically indicated. Half (n = 17) had a clinical pathway or order set for complicated pneumonia with more than a third of these (n = 6, 35%) recommending daily CXR. Of the 20 sites where routine CXRs were typically performed, half (n = 10, 50%) reported that the practice was not useful. To explore other factors that may influence imaging acquisition, our study asked about tissue plasminogen activator (TPA) use via CT. All hospitals reported using TPA, with most (n = 27, 82%) reporting TPA use did not affect the decision to obtain CXR. Although ultrasonography (US) use was common before CT placement (n = 27, 82%), only 2 sites reported using US to monitor the patient with a CT. Twenty-two (65%) reported that they typically obtained routine CXR after CT removal, and 10 (29%) reported obtaining CXR only for clinical indication. Table 2 presents additional data on CT policy and procedures.This study highlights that obtaining routine CXRs is a common practice for pediatric patients with complicated pneumonia following CT placement. Among the institutions reporting routine CXRs, 50% of respondents believed these imaging studies were not helpful. Additionally, more than half of the surveyed sites routinely obtained CXRs after CT removal, despite evidence demonstrating the limited value of this practice.2,3Although studies show US is superior to CXR in detecting pleural effusions, very few institutions reported using US for monitoring after CT placement.6 This limited use may stem from a lack of training or resources because most US studies were performed by the radiology department rather than at the bedside. Expanding the use of bedside US, particularly by hospitalists, could improve patient monitoring while reducing radiation exposure.This study has several limitations. Self-report may not precisely reflect actual practices of the institution. Response bias is possible because physicians with a particular interest in imaging practices might have been more inclined to participate. In addition, a high percentage of respondents were division or section chiefs who may not be as knowledgeable of current clinical practices. Distribution of the survey within the PHIS network may restrict the generalizability. Finally, the study did not capture the perspectives of surgeons who are often involved in CT management decisions.Daily CXRs are commonly obtained in children hospitalized with complicated pneumonia with CT. Future studies should focus on the impact of this practice on clinical management and outcomes and identify criteria in which imaging would optimize outcomes.