BACKGROUND Previous studies have demonstrated high rates of intraspinal anomalies in patients with congenital scoliosis; however, various authors have not considered the presence of an isolated hemivertebra to be sufficient reason for further evaluation with magnetic resonance imaging. Consequently, the rate of magnetic resonance imaging-detected intraspinal anomalies and subsequent neurosurgical intervention in patients with a single hemivertebra is unknown. Therefore, we studied all patients with a hemivertebra, after eliminating patients with a myelomeningocele, to compare those who had a single hemivertebra with those who had a complex hemivertebral pattern. METHODS A retrospective review of the history, physical examination findings, and magnetic resonance imaging findings for patients who had presented with at least one hemivertebra, excluding those who had a myelomeningocele, was conducted to identify the prevalence of intraspinal anomalies as seen on magnetic resonance imaging and the rate of subsequent neurosurgical intervention. Additionally, the diagnostic value of the history and the physical examination in predicting the presence of intraspinal anomalies was determined. RESULTS One hundred and sixteen patients with congenital scoliosis and a curve that included at least one hemivertebra were identified. Seventy-six of these patients had had magnetic resonance imaging and were included in the present study. The mean age of these patients at the time of presentation was 4.9 years, and the mean duration of follow-up was 7.7 years. Twenty-nine patients had an isolated hemivertebra, and forty-seven patients had a complex hemivertebral pattern. Eight (28%) of the twenty-nine patients with an isolated hemivertebra and ten (21%) of the forty-seven patients with a complex hemivertebral pattern had an intraspinal anomaly that was detected with magnetic resonance imaging. Overall, an abnormal finding on the history or physical examination demonstrated an accuracy of 71%, a sensitivity of 56%, a specificity of 76%, a positive predictive value of 42%, and a negative predictive value of 85% for the diagnosis of an intraspinal anomaly. Three patients with an isolated hemivertebra and five patients with a complex hemivertebral pattern underwent neurosurgical intervention. All eight patients who underwent neurosurgical intervention had had detection of an intraspinal anomaly with magnetic resonance imaging, whereas only four of these patients (two of whom had an isolated hemivertebra and two of whom had a complex hemivertebral pattern) had had an abnormal finding on either the history or the physical examination. CONCLUSIONS Patients who have an isolated hemivertebra and those who have a complex hemivertebral pattern have similar rates of intraspinal anomalies that are detected with magnetic resonance imaging and similar rates of subsequent neurosurgical intervention. The history and physical examination findings are not predictive of intraspinal anomalies. Therefore, a magnetic resonance imaging evaluation of the entire spine should be considered for all patients with congenital scoliosis, including those with an isolated hemivertebra.