AUTHORS
Matthew D. Nguyen, MD, Ethan Pan, Dr.2, Sidharth Srinivasan3, Benjamin W. Friedman, MD4, Ammar Hamid5, Ellen W Friedman, MD6
AFFILIATIONS
1 Montefiore Einstein, Department of Internal Medicine, Bronx, United States
2 Montefiore Einstein, Department of Internal Medicine, Bronx, United States
3 Montefiore Einstein, Department of Internal Medicine, Bronx, United States
4 Montefiore Einstein, Albert Einstein College of Medicine, Emergency Medicine, Bronx, New York, United States
5 Montefiore Einstein, Department of Internal Medicine, Bronx, United States
6 Montefiore Einstein, Albert Einstein College of Medicine, Department of Oncology/Medicine, Bronx, United States
ABSTRACT
Introduction
Acute chest syndrome (ACS) is the leading cause of mortality in sickle cell disease (SCD). Diagnosis requires a pulmonary infiltrate plus compatible symptoms, but chest radiographs can be normal early in the disease course despite evolving ACS. Whether patients with this delayed radiographic pattern differ clinically or have worse outcomes than those with infiltrates present on admission remains poorly understood.
Methods
We performed a retrospective cohort study of adults ≥ 18 years with a diagnosis of ACS at three hospitals in the Bronx, NY, from January 2016 to January 2026. Patients were classified as having early ACS (positive chest radiograph ≤24 hours from ED presentation) or delayed ACS (initial negative radiograph with a new infiltrate appearing >24 hours after ED presentation). The primary outcome was progressive ACS, a composite of ICU transfer, non-invasive (BiPAP) or invasive mechanical ventilation, and in-hospital mortality. We constructed multivariable logistic regression models to evaluate the association between delayed ACS and our primary outcome, adjusting for age, hydroxyurea use, baseline lung disease, and prior ACS history.
Results
Of 104 patients, 42 (40%) had delayed ACS; in this group, the median time to radiographic confirmation was 78 hours. Despite initially negative imaging, 68% had ACS-compatible symptoms (chest pain, shortness of breath, cough) and 79% had abnormal vital signs (tachycardia, hypoxia, fever) within the first 24 hours of ED presentation. Similarly, 79% of patients in the early ACS group also had abnormal vital signs within the first 24 hours of ED presentation. Median time to transfusion, including either simple or exchange transfusion, was significantly longer in the delayed ACS group (46.0 hours [IQR 26.5–124.5] vs. 24.5 hours [IQR 13.2–39.8]; p=0.003), despite similar overall transfusion rates between groups (76% vs. 64%; p=0.30). Length of stay was significantly longer in the delayed ACS group (10.0 days [IQR 7.0-15.0] vs. 6.5 days [IQR 5.0-10.0]; p<0.001).
Delayed radiographic presentation was independently associated with the primary composite outcome of progressive ACS (aOR 4.06; 95% CI 1.66–10.50; p=0.003). In secondary outcome analysis, delayed ACS was independently associated with BiPAP or mechanical ventilation (aOR 3.41; 95% CI 1.24–10.00; p=0.020). No statistically significant differences were observed for in-hospital mortality (aOR 9.51; 95% CI 0.87–337; p=0.11) or 30-day readmission (aOR 2.63; 95% CI 0.73–10.40; p=0.15). Notably, all progressive ACS events coincided with ICU transfer (n=35), as every patient requiring ventilatory support or who died in-hospital was admitted to the ICU.
Conclusion
40% of adults with ACS presented with initially normal chest radiographs despite most having respiratory symptoms and/or abnormal vital signs within the first 24 hours. Delayed radiographic presentation of ACS was independently associated with over four-fold increased risk of respiratory deterioration and delayed transfusion. These findings suggest that an initial negative radiograph should not exclude ACS and support close monitoring and prompt repeat imaging when clinical suspicion remains high.
DISCLOSURES
Matthew D. Nguyen, MD: No relevant conflicts of interest to declare.