Correlation of clinical and chest radiograph findings in pediatric submersion cases.

Abstract

Submersion injuries are a leading cause of injury death in children in the United States. The clinical course of a submersion patient varies depending on the presence of anoxic brain injury and acute respiratory failure. We studied changes in clinical findings and chest radiograph findings and determined the sensitivity/specificity of the presenting chest radiograph in predicting clinical improvement within the first 24h in pediatric submersion cases. We conducted a cross-sectional study of pediatric submersion patients through age 18years treated at a children's hospital from 2010 to 2013. We reviewed demographics, comorbidities, prehospital/hospital course and chest radiographic findings. Clinical improvement occurred when a child demonstrated normal vital signs and mentation. We compared radiographic findings among children based on clinical improvement up to 24h post submersion. Using odds ratios, we calculated associations between radiographic findings and clinical improvement. We studied the sensitivity/specificity of the presenting chest radiograph in predicting clinical improvement within 24h. One hundred forty-two of 262 (54%) patients had initial chest radiographs; 41% had follow-up radiographs. The odds of an abnormal initial chest radiograph were 4 times higher in children with respiratory distress or abnormal mentation at emergency department (ED) presentation compared to children without these findings (odds ratio [OR]=4.83; 95% confidence interval [CI]=2.1-10.85; P<0.001). Improvement in radiographic findings occurred in 85% of children within 24h. Children with an abnormal initial chest radiograph were 87% less likely to improve clinically by 24h (P<0.001). A presenting chest radiograph that was normal or with mild pulmonary edema/atelectasis predicted clinical improvement within 24h (sensitivity 95%, specificity 57%). Most chest radiographic findings improve in pediatric submersion patients who recover within the first 24h. An initial chest radiograph that is normal or with mild pulmonary edema/atelectasis satisfactorily predicts clinical improvement by 24h post submersion.