Search for a command to run...
Acute chest syndrome (ACS) is an acute pulmonary complication of sickle cell disease (SCD) and is the leading cause of morbidity and mortality [1-3]. It can develop de novo or during hospitalization for other complications (e.g., vaso-occlusive pain episodes), necessitating a high index of suspicion for diagnosis. Current evaluation and management recommendations include imaging of the chest, maximizing pain control, initiation of broad-spectrum antibiotics, respiratory support as indicated, and consideration of red blood cell transfusion, but large scale epidemiologic studies regarding the implementation of these practices for ACS are limited [3-8]. This correspondence reports the current epidemiology, diagnostic evaluation, and management patterns of ACS in children with SCD when presenting to pediatric hospitals in the United States. Additionally, we explored patient and disease characteristics that may be associated with an increased risk for developing ACS. Briefly, we analyzed data from the Pediatric Health Information System (PHIS) database for SCD encounters from January 2018 through December 2023 for patients aged 0–25 years based on codes from the 10th revision of the International Classification of Diseases (ICD-10 codes) that pertained to SCD (Figure S1). Patient encounters were analyzed for demographic data, performance of chest imaging, type of chest imaging, and stratified by diagnosis of ACS and/or pneumonia. Encounters with an ACS diagnosis were further analyzed using age, gender, sickle cell genotype, existing asthma diagnosis, and length of hospital stay. In addition, data regarding diagnostic imaging modalities and clinical interventions such as antibiotic usage, level of care (general inpatient versus intensive care unit), use of mechanical ventilation, and utilization of packed red blood cell (pRBC) transfusions were collected. Analyses were performed using a multivariable mixed effects logistic regression model, with a compound symmetry covariance structure to best model the repeated measures aspect. A quadratic term for age was included in the model to account for the non-linear natural rise and fall of risk of ACS in young patients over time. The multivariable model included the independent variables admit age (years), admit age (years) squared, gender, asthma status, genotype, and ACS within 6 months. Patients with non-binary gender were excluded due to low sample size. Subjects with multiple genotypes recorded were excluded as this was either a record error or an undesirable comparison. Subjects with sickle cell trait were also excluded. Antibiotic and transfusion status were similarly estimated with univariable repeated measures models with compound symmetry covariance structure and a fixed effect of either age quadrant or geographic location. Data were analyzed using Statistical Analysis System (SAS) version 9.4 (SAS Institute, Cary, NC). Overall, encounters involving 41 456 unique patients (118 784 encounters) from 44 pediatric hospitals met the inclusion criteria. Encounters included children, adolescents, and young adults (median 10 years, range 0–25, IQR 4–16). Just over one half of encounters were male patients and half of encounters were for patients with a HbSS genotype. Most encounters (74.5%) included patients with public insurance. Only 10.7% of patients had an ICD10 code consistent with ACS or SCD with pneumonia. Of the 31.2% of encounters that included some form of chest imaging, the most common imaging modality was chest x-ray (CXR) (Table 1). When further stratified to include only patients with a diagnosis of ACS/pneumonia, CXR remained the most common imaging modality, but half of all encounters with a diagnosis of ACS (50.6%) did not have any imaging performed at the PHIS contributing pediatric hospital (Figure S2). The majority of encounters associated with an ACS code were inpatient (Figure S3). Over 1900 patients required management in an intensive care unit with 278 patients requiring mechanical ventilation and 16 patients requiring extracorporeal membrane oxygenation (ECMO). pRBC transfusions were utilized in only 53% of patients (Figure S3) with transfusions being more common in older patients (p < 0.0001) and varied by geographical location (Table S1). Transfusion usage also increased based on severity of clinical status, with 76.6% of ICU patients receiving a transfusion (Figure S4). Only 70% of encounters with a diagnosis of ACS included the combination of a cephalosporin and a macrolide, with almost 10% (n = 1251) not receiving either (Figure S5). Additionally, while only 10.7% of all encounters had a diagnosis of ACS, 12.8% of patient encounters were prescribed both ceftriaxone and azithromycin. Antibiotic administration also varied based on age (p < 0.0001), with younger patients more likely to receive antibiotics. Analysis of patient demographic and clinical data yielded a patient profile at highest risk for ACS. Figure 1 depicts the patient demographic profile at highest risk for development of ACS when considering sickle cell genotype, sex, age, diagnosis of asthma, and diagnosis of ACS within the prior 6 months. Despite improved survival rates for children with SCD, they continue to experience significant morbidity, as evidenced by the high rate of encounters noted in our study. Our data further reveal that the burden of morbidity varies according to SCD genotype, with a higher proportion of patients with Hb SS genotype presenting for care (39.5%) and a slight predominance of male patients as compared to female patients. While comparative data are limited, our data are similar to previously reported epidemiologic data [9]. CXR is the predominant imaging modality used for identification of ACS. It is a reliable and accessible tool in identification of lung consolidations, as is required for diagnosis of ACS. However, it does come with limitations including radiation exposure and delay in radiographic appearance of lung consolidation [3-5]. Importantly, patients with SCD have an increased radiation burden as compared to the general population, with studies indicating children with SCD may undergo greater than one hundred CXRs prior to the age of 18 years [10]. A study published in 2018 in general pediatric patients showed a decrease in CXR usage in all children from 30.4% to 18.6% of encounters for ED visits for fever or respiratory illness since 2010 [11]. This trend does not seem to extend to SCD, as our data show children with SCD who present to care in the ED with concern for ACS/pneumonia undergo chest imaging at a much higher rate, with CXR being the most common modality. This radiation exposure is substantial, especially in a patient population that already has an increased lifetime risk of malignancy [10]. Alternative lung imaging modalities, like lung ultrasound, may be one way to mitigate these risks. Use of lung ultrasound in the emergency room setting for SCD has shown promise for diagnosis of ACS, with good sensitivity and specificity for diagnosis and excellent negative predictive value [12, 13]. With additional benefits of bedside availability and its adaptability to patient clinical status and body habitus, utilization of lung ultrasound as a primary imaging modality for ACS, both in ED and inpatient settings, may be appropriate. Though NHLBI guidelines provide recommendations for ACS management, the implementation of these is inconsistent. In our study, there was notable practice variability in antibiotic usage, with patients being undertreated (up to 30%) or overtreated (up to 15%), and pRBC transfusions (approximately 50% of patients with ACS received pRBCs). When comparing these two interventions across age groups, there was also variability, with younger patients being more likely to receive antibiotics than older patients and older patients being more likely to receive transfusions than younger patients. Additionally, transfusions varied by geographical region, reflecting the lack of standardization of transfusion practices in ACS nationally. Risk stratification is one way to empower providers when caring for patients with a high-risk disease. Using this retrospective dataset, we were able to identify patient and disease characteristics associated with a diagnosis of ACS, including younger age, male sex, Hb SS genotype, a pre-existing diagnosis of asthma, and a diagnosis of ACS within the last 6 months. This is in line with previous international studies that have reported higher incidence of ACS in patients with HbSS and asthma [14, 15]. Clinical risk profiles like this can be validated prospectively and used to determine need for diagnostic imaging, treatment setting, and interventions for high-risk clinical diagnoses such as ACS. They may also be used in conjunction with biomarker profiles currently being studied to help in earlier identification of ACS and potentially prevent severe sequelae associated with ACS [16, 17]. There were limitations to our study due to its retrospective nature and the use of ICD-10 codes. For example, approximately 50% of encounters with a diagnosis of ACS did not have a record of chest imaging. PHIS does not include outside hospital clinical information or evidence of previous hospital transfer, so interventions performed at outside institutions were not captured. This could contribute to the variation in imaging and antibiotic administration seen. Additionally, ICD-10 codes are based on physician billing and are subject to human error, as evidenced by several patients having multiple coded genotypes. However, the large number of encounters minimizes the effect inaccurate billing codes may have on overall results [18]. This study highlights the persistent morbidity of ACS in children with SCD and demonstrates the significant variability in diagnostic and management practices across U.S. pediatric hospitals. Identifying patient populations at greatest risk for ACS and recognizing practice variability provides an opportunity to refine care delivery. Future work should focus on evaluating alternative diagnostic modalities, promoting evidence-based standardization of transfusion and antibiotic practices in ACS, and developing consensus-driven ACS guidelines to improve outcomes for children and adolescents with SCD. Concept and design: Seethal A. Jacob, Thomas C. Fisher-Heath, Benjamin Nti, Colin Rogerson, Matthew Hays. Acquisition, analysis, or interpretation of data: All authors. Drafting of the manuscript: Thomas C. Fisher-Heath, Seethal A. Jacob, Matthew Hays. Critical review of the manuscript for important intellectual content: All authors. Statistical analysis: Matthew Hays, Thomas C. Fisher-Heath. Obtained funding: Thomas C. Fisher-Heath, Seethal A. Jacob. Supervision: Seethal A. Jacob. This project was supported, in part, with support from the Indiana Clinical and Translational Sciences Institute funded, in part by Grant Number UM1TR004402 from the National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Sciences Award. Additionally, this project was supported, in part, by Grant Number K23HL143162 from the National Heart, Lung, and Blood Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors have nothing to report. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. Figure S1: Consort diagram of inclusion and exclusion criteria. Figure S2: Imaging modalities used for patients with ACS. Figure S3: Supportive care interventions utilized for patient encounters with a diagnosis of ACS. Figure S4: Transfusion rates based on medical intervention. Figure S5: Antibiotic interventions utilized for ACS patients. Table S1: Probability of transfusion in the setting of ACS diagnosis based on geographic region. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.