University of Michigan
Department of Pediatrics
Evidence-Based Pediatrics Web Site

Question

• A 2-month-old boy is seen for a well-child exam. A new systolic murmur is noted. The baby is asymptomatic and growing well. In patients like this, is chest x-ray helpful in determining whether or not there is structural heart disease?

Clinical Bottom Lines

1. One hundred children, ages 1 month to 15 years, presenting with a new heart murmur were evaluated with chest x-ray (CXR), EKG, and cardiac ECHO.¹ The ECHO was taken as the gold standard.
2. The sensitivity of CXR compared to ECHO was 43% and the specificity was 82%. Therefore, the positive likelihood ratio (LR (+)) is 2.5 and the LR (-) is 0.75.
3. For patients with an abnormal x-ray, the test is minimally useful, but for the result we really care about (a negative CXR), the test is basically useless.
4. Thus, for patients with an asymptomatic heart murmur, referral to a cardiologist for consideration of a cardiac ECHO is the best option.

Summary of Key Evidence

1. One hundred patients had initial history and physical, followed by CXR, EKG, and cardiac ECHO. Children with symptoms of heart disease, anemia, or hospitalized for other reasons were excluded.¹
2. The gold standard (ECHO) was applied uniformly to this group of patients and results of the other evaluations were blinded from those reviewing the ECHOs.
3. The children were classified after the H&P into three groups: no heart disease, possible heart disease, and definite heart disease. Adding the CXR and EKG did not improve the evaluation of these children.
4. Three children had abnormal EKGs and all were classified correctly as definite heart disease after the initial evaluation.
5. The prevalence of heart disease in this population was 23%. With the LR (+) above, the post-test probability would be 42%. With the LR(-) above, the post-test probability of underlying heart disease would still be 18%.

Additional Comments

• Swenson, et al. claim that the addition of EKG and CXR to the initial evaluation can still help the evaluation of murmurs.² However, this study is limited significantly in two ways. First, they included patients who were being evaluated because of chest pain. Second, they did not perform ECHOs on all patients; this limits the validity of their results.
• Using the same data collected for the initial study, Birkbaek and colleagues investigated the reproducibility and accuracy of the chest x-rays.³ They found that the reproducibility and accuracy were quite low, limiting the usefulness of these tests even further.

Citation

1. Birkbaek NH, Hansen LK, Oxhoj H. Diagnostic value of chest radiography and electrocardiography in the evaluation of asymptomatic children with a cardiac murmur. Acta Paediatr 1995; 84:1379-81.
2. Swenson JM, et al. Are chest radiographs and electrocardiograms still valuable in evaluating new pediatric patients with heart murmurs or chest pain? Pediatrics 1997; 99: 1-3.
3. Birkbaek NH, et al. Chest roentgenograms in the evaluation of heart defects in asymptomatic infants and children with a cardiac murmur: reproducibility and accuracy. Pediatrics 1999; 103(2): electronic pages.

CAT Author: John G. Frohna, MD

CAT Appraisers: <Reviewers>, MD

Date appraised: March 20, 1999