We performed a randomised trial on 37 hips (33 patients) with early-stage osteonecrosis (ON). After the initial clinical evaluation, including plain radiography and MRI, 18 hips were randomly assigned to a core-decompression group and 19 to a conservatively-treated group. All the patients were regularly followed up by clinical evaluation, plain radiography and MRI at intervals of three months. Hip pain was relieved in nine out of ten initially symptomatic hips in the core-decompression group but persisted in three out of four initially painful hips in the conservatively-treated group at the second assessment (p < 0.05). At a minimum follow-up of 24 months, 14 of the 18 core-decompressed hips (78%) and 15 of the 19 non-operated hips (79%) developed collapse of the femoral head. By survival analysis, there was no significant difference in the time to collapse between the two groups (log-rank test p = 0.79). Core decompression may be effective tin symptomatic relief, but is of no greater value than conservative management in preventing collapse in early osteonecrosis of the femoral head.