The thermal effects of intramedullary reaming

Objective: To study the effect of tourniquet control on intramedullary reaming. Design: An experimental prospective nonsurvival animal study was performed using 5 mongrel dogs. A pneumatic tourniquet was randomized to either the right or left hind limb. Tibial intramedullary reaming was performed with progressively larger reamers. Cortical temperatures were measured using thermocouples inserted into the tibial diaphyseal cortex. Thermocouples were connected to an analog to digital converter that output continuous data that was collected on a computer. Upon completion of the procedure, the animals were killed. Results: The peak and low temperatures for each thermocouple with each reamer passage were recorded. Reamer sizes larger than the internal diameter of the intramedullary canal produced higher peak temperatures. The mean delta t (peak temperature minus low temperature) was calculated for each reamer passage. This measurement represents the overall amount of heat generated during each reamer passage. There were no significant differences between the 2 conditions (P = 0.8, paired t test). Temperatures decreased in between reamer exchange but did not return to baseline levels. Conclusions: Because similar temperatures were measured both with and without a tourniquet, the risk of thermal necrosis appears to be related more to the process of intramedullary reaming than to the tourniquet. Higher temperatures were measured with reamer sizes larger than the internal diameter of the intramedullary canal. Increasing the time interval between the passage of successive reamers may allow heat to dissipate and decrease the risk of thermal necrosis. The clinical practice of limited reaming ("ream-to-fit") should minimize the occurrence of this complication.