OBJECTIVE This study was undertaken to examine the accuracy of two hand-held devices commonly used to measure the pressures produced by extensible bandages. METHODS The performance of the pneumatic sensors of two devices, the Kikuhime and Picopress instruments was first examined in air in a compression chamber, then subsequently beneath multiple layers of bandages applied to standard cylinders with predetermined levels of tension. RESULTS In the compression chamber, both instruments provided readings that were typically within 1 mmHg of the reference value, but on curved formers in free air or beneath bandages the accuracy of both sensors was greatly reduced, influenced both by the curvature of the cylinders and the volume of air contained in the sensor capsule. The Picopress instrument recorded pressures up to 70% higher than predicted, particularly on the smaller cylinders and at the lower end of the pressure range (circa 20mmHg). At 40mmHg measured pressure were around 40% higher than predicted values in some instances. The accuracy of the Kikuhime was greatly influenced by the calibration technique but percentage deviations as high as 150% were recorded in some tests. CONCLUSIONS Pneumatic pressure sensors used to record the pressures developed beneath compression bandages are much less accurate than is commonly believed. Calibration studies using air chambers or water tanks have no relevance to the normal clinical use of these devices. These limitations should be considered by clinicians when making judgments about the performance of other medical devices such as bandages or stockings. They also call into question the validity of many published studies which rely upon such measurements. BACKGROUND This project was generously supported by 3M Healthcare who provided a small research grant to assist with the costs associated with the construction of the test equipment.