Selective criteria may increase lumbosacral spine roentgenogram use in acute low-back pain

Abstract

Eleven clinical criteria have been proposed to limit use of lumbosacral spine roentgenograms in patients with acute low-back pain who are at risk for vertebral cancer, osteomyelitis, acute fracture, or herniated disk. We retrospectively applied the criteria to 471 patients with acute low-back pain in three teaching hospital walk-in clinics. Roentgenograms were obtained at the initial visit in 99 patients (21.1%); the number would have increased to 217 (46.1%) if the criteria had been used. The following four patient characteristics were associated with actual roentgenogram use: older age, longer duration of symptoms, reflex asymmetry, and point vertebral tenderness. Adoption of the 11 criteria studied herein may inadvertently increase roentgenogram use, thereby raising health care costs and exposing more patients to gonadal irradiation. The standard of practice in these three clinics seemed to entail use of less broad roentgenogram selection criteria. Other published guidelines for roentgenograms emphasize clinical follow-up, reserving further evaluation for patients who fail to improve after a trial of bed rest and analgesics.