The aim of the present study was to establish a threshold for the initial displacement of a spiral tibial shaft fracture beyond which its retention in an acceptable position cannot be guaranteed by plaster immobilization. We reviewed the records and radiographs of 131 plaster cast-treated patients with spiral tibial shaft fracture, initially displaced 50% or less, for patients whose fracture had either lost its acceptable retention or consolidated in an unacceptable position. The fractures were classified, according to the true initial displacement as measured on the first radiographs, into four pairs of categories using cut-off points of 10, 20, 30 and 40% of the diameter of the tibial diaphysis. Comparison was then made of the proportions of failed treatments between each of these pairs. Plaster cast treatments failed in 28% when the true initial displacement was 30% or less, and in 46% when the true initial displacement was more than 30%. The risk of failed plaster cast treatment increased when true initial displacement exceeded 30%. In all patients whose plaster cast treatment was interrupted the true initial displacement was more than 30%. We therefore conclude that diaphyseal fractures of the tibia for which the initial displacement exceeds 30% are not suitable for plaster cast treatment.