Investigation on the ability of an ultrasound bubble detector to deliver size measurements of gaseous bubbles in fluid lines by using a glass bead model.

Detectors based on ultrasonic principles are today's state of the art devices to detect gaseous bubbles that may be present in extracorporeal circuits (ECC) for various reasons. Referring to theoretical considerations and other studies, it also seems possible to use this technology to measure the size of detected bubbles, thus offering the chance to evaluate their potential hazardous effect if introduced into a patient's circulation. Based on these considerations, a commercially available ultrasound bubble detector has been developed by Hatteland Instrumentering, Norway, to deliver bubble size measurements by means of supplementary software. This device consists of an ultrasound sensor that can be clamped onto the ECC tubing, and the necessary electronic equipment to amplify and rectify the received signals. It is supplemented by software that processes these signals and presents them as specific data. On the basis of our knowledge and experience with bubble detection by ultrasound technology, we believe it is particularly difficult to meet all the requirements for size measurements, especially if these are to be achieved by using a mathematical procedure rather than exact devices. Therefore, we tried to evaluate the quality of the offered bubble detector in measuring bubble sizes. After establishing a standardized test stand, including a roller pump and a temperature sensor, we performed several sets of experiments using the manufacturers software and a program specifically designed at our department for this purpose. The first set revealed that the manufacturer's recommended calibration material did not meet essential requirements as established by other authors. Having solved that problem, we could actually demonstrate that the ultrasonic field, as generated by the bubble detector, has been correctly calculated by the manufacturer. Simply, it is a field having the strongest reflecting region in the center, subsequently losing strength toward the ECC tubing's edge. The following set of experiments revealed that the supplementary software not only does not compensate for the ultrasonic field's inhomogeneity, but, furthermore, delivers results that are inappropriate to the applied calibration material. In the last set of experiments, we were able to demonstrate that the signals as recorded by the bubble detector heavily depend upon the circulating fluid's temperature, a fact that the manufacturer does not address. Therefore, it seems impossible to resolve all these sensor related problems by ever-increasing mathematical intervention. We believe it is more appropriate to develop a new kind of ultrasound device, free of these shortcomings. This seems to be particularly useful, because the problem of determining the size of gaseous bubbles in ECC is not yet solved.