M Tavakoli and E J C Kellar, TWI, Cambridge, UK
D Nassiri and A E Joseph, St George's Hospital, London, UK
Paper published in Medical Device Technology March 2006
*Professor Mehdi Tavakoli Consultant and Technology Manager, Advanced Material & Processes Group, TWI Ltd, and Programme Manager, Health Technologies Knowledge Transfer Network,Cambridge CB1 6AL, UK.
Dr Ewen J C Kellar Prinicipal Project Leader, TWI, Cambridge, UK.
Dariush Nassiri, Professor of Medical Physics, St George's Hospital, London, UK.
Anton E Joseph, retired Consultant, St George's Hospital, London, UK
*To whom all correspondence should be addressed.
A dynamic novel coating, consisting of a hydrophilic polymeric matrix and a bubbling agent, has been developed for precise entry and positioning of needles, and accurate collection of biopsy samples. These coated biopsy needles have been successfully used in vitro trials. The results are discussed here. Potential application areas include vascular, cardiovascular and orthopaedics.
The significant interest in minimally invasive surgery has increased the need for enhanced visibility and imaging of diagnostic tools as well as precision when following the passage, location and fate of short- and long-term implants and devices. A variety of techniques are available for detecting or tracing the presence of a device in a tissue or blood environment. These include ultrasound, X-ray and magnetic resonance imaging.
Most hospitals use ultrasound guidance to perform fine needle aspirates, biopsies and drainage procedures. Currently, the devices used for these procedures rely on scoring or roughening of the area around the needle tip to improve visualisation. However, the visibility of the tip alone does not allow the clinician to accurately gauge the entry angle of the needle. Often the process has to be repeated before a successful outcome is achieved, which can result in unnecessary discomfort for the patient and increased expense in terms of the number of needles used.
Bosley et al.  described an echogenic device that comprised a material with acoustic impedance that differs from the surrounding medium of biological tissue and fluids. The material applied to the device consisted of a polymer such as polyethylene or adhesive or a silver solder with sound-reflective particles such as glass microspheres of approximately 5µm in diameter. A radiopaque material such as barium or tungsten was also included to enable the device to be radiographically and ultrasonically imageable.
Development of a medical device such as a needle with an ultrasonically reflective coating was also described in a European patent.  The coating included a matrix of gas bubbles contained in a polymeric material. The polymeric material used in this case was described as a foam that contained a blowing agent, which is activated during melt processing such as moulding of the polymer or by introducing a gas into polymer melt creating a cellular core structure.
In a series of studies, [3-5] a novel polymeric coating has been developed to enhance ultrasound images of biopsy needles and other medical devices. The coating consists of a hydrophilic polymeric matrix and a bubbling agent. This bubbling agent reacts with tissue fluid as the needle penetrates and produces bubbles within and on the surface of the coating; this increases the back-scattering capacity of the coating and generates a brighter image of the device under ultrasound.
Prior to coating with a water-based primer, an 18-gauge pink needle (a) and a 21-gauge green needle (b) were wet blasted with 60-mesh alumina grit for 60-90 s at 4 bar pressure. A third needle, a 21-gauge blank cannula (c), was solvent cleaned and then dip-coated with solvent-based primer.
Layers of coating material were deposited onto the needle surface by dipping the needle into Tetrahydrofuran (Aldrich Chemical Co., www.sigma-aldrich.com) solutions containing Hydrothane, a hydrophilic polyurethane from Cardio Tech International Ltd, www.cardiotech-inc.com) and reagents. The thickness of each layer was estimated to be ~30µm with a total coating thickness of up to 120µm.
The coated plates and needles were supplied to St George's Hospital, London, UK, for analysis using an HDI 100 Apogee ultrasound system (Advanced Technology Laboratories) set at the lowest gain and fitted with a 3.5-MHz annular sector probe. The probe was fixed just under the surface of the water in a thermostatically controlled water bath maintained at 37 deg.C. The angle of inclination of the probe and the depth to which it was immersed were kept constant throughout the experiments. The test samples were placed in a fixed position relative to the probe and supported on a steel block immersed in the water bath. This ensured that scatter from the generated bubbles and the coated surface, rather than specular reflection, were responsible for the returning echoes.
The mean brightness of a region of interest of each of the digitised images (512 x 512 x 8 bit format) was calculated and expressed as
- mean brightness versus time
- percentage change in mean brightness versus time
- change in mean brightness in dB against background versus time.
Needles coated with the combination of materials that demonstrated the greatest change in brightness were evaluated under ultrasound (as described above) in a tissue-mimicking phantom and in an isolated ovine liver. Uncoated needles served as controls. The phantom consisted of tissue-mimicking material (TMM) with acoustic properties similar to those of soft tissue, for example, a liver.
The TMM was submerged in a water bath at 37°C and the selected biopsy needles inserted into it. The ultrasound probe was held in place above the phantom with the transducer elements submerged in the water. Images were acquired from the region of interest, digitised and stored, as described above, for further analysis. The data acquisition process was calibrated so that the relationship between the brightness of the digitised image and actual reflectivity/backscatter (dB) was expressed by the following formula:
Gain in dB = 0.4 x brightness in ROI + C 
Two sets of experiments were conducted using coated needles. The first set investigated the activity of the coating in relation to different gauges of needle versus control systems. It also enabled a quantitative analysis to be made of the activity ( Figure 1). The second set utilised the latest generation of coating, which had been improved in terms of responsiveness (rate of initial bubble generation) and brightness. In addition, the second set was compared with a commercially available coating that has been claimed to exhibit similar properties.
The above experiments with the first set of needles were repeated using an isolated ovine liver to evaluate the performance of the coated needles and commercial needles with echogenic tips in solid tissue. The fresh liver was obtained from an abattoir from a recently slaughtered sheep, placed in a solution of heparinised normal saline and transported to St George's Hospital. The liver was placed in fresh heparinised saline under a gentle vacuum to reduce the number of gaseous inclusions trapped inside the liver. Coated needles were inserted into the isolated liver as described above for the TMM. As before, the commercial uncoated needles served as controls. For a further comparison, one of the coated needles was stripped of its coating at the end of the experiment to see if the roughening process used to promote adhesion of the coating contributed to the brightness of the needle being imaged.
Both the green- and pink-coated needles were readily visible in the resulting ultrasound image when tested in the tissue-mimicking phantom. Enhancement of the needles persisted throughout the imaging experiment (>100 s). The larger bore needle (pink) was more readily seen than the smaller bore needle (green). The increased visibility is probably a result of the increased area available for bubble generation compared with the thinner green needle. Figure 1quantifies the increased enhancement of the needles, compared with the uncoated controls. In particular, Figure 1 shows a peak mean backscatter of 25-30 dB, relative to the controls. The coated needles were visible within a few seconds of insertion and the rate of bubble generation, hence reflectivity, continued to increase, reaching a plateau after approximately 60 s.
Figure 2 shows the results of the second set of experiments where the novel coating and the commercial coating are clearly visible compared with the uncoated reference. However, the novel coating is more than twice as bright as the alternative system, which demonstrates the superior features offered by a coating that generates bubbles dynamically. These active bubbles are most graphically illustrated in Figure 3, which shows a coated needle that has been immersed in water for several seconds. The quantitative effect of this bubble evolution for the latest generation of bubble coating is demonstrated in Figure 4.
The other benefit of dynamic bubble generation is that the bubbles are deposited along the track of the needle leaving a memory of placement ( Figure 5). This enables subsequent placement of the needle in position different to that made in the first procedure.
The ovine liver
The coated needles were clearly seen in the images obtained in the isolated ovine liver ( Figure 6). A track of bubbles was still visible in the liver once the needles had been removed. Consequently, the clinician is able to track a bright line during needle placement. This provides information about the angle and position of the needle relative to the target, rather than the single bright spot afforded by the tip of the echogenic needle. Furthermore, when the coating was completely stripped from one of the needles (green), no background reflectivity was seen, which results from roughening the needle surface to promote coating attachment.
The main conclusions that can be drawn from this work are as follows:
- Incorporation of a gas-producing reagent mixture within a hydrophilic polyurethane matrix and employing it as a coating to medical devices such as biopsy needles can significantly enhance the ultrasound visibility of those devices.
- The in vitro experiment using a TMM showed that the coated needles were readily visible in the resulting ultrasound image, with a peak mean backscatter of 25-30 dB relative to uncoated control needles.
- Improvement in coated needle visibility was confirmed in an isolated ovine liver experiment.
- A new coating formulation with novel deposition techniques was developed that enable a thin, smooth, bubble-free coating to be applied to a medical device without undesirable premature reaction between gas-producing reagents.
- This coating has been shown to be significantly brighter than a commercially available coating and offers the additional benefit of producing a bubble-track 'memory', which enables operators to avoid sampling the same area of tissue.
The coating formulation and techniques developed as a result of this work will allow manipulation of the start of bubble formation from a few seconds to a few minutes. It is also possible to design a coating that can increase ultrasound visibility of a medical device to up to many minutes. The possibility of using this novel coating in a number of areas, including cardiovascular, vascular and orthopaedic applications is currently being explored. The research team welcomes discussion on the use of this novel coating technology for any specific application in collaboration with a device manufacturer.
- R W Jr Bosley, P G Thomson and L F Thomas, 'Echogenic Devices Material and Method' US Patent 51081, 997, (21 January 1992).
- P Lawrence, 'Instrument Having Enhanced Ultrasound Visibility', European patent application, 0624342 A1, 4/5/1994.
- E A Joseph, 'Instrument Having Enhanced Ultrasound Visibility', International Patent No. WO 98/18387.
- E J C Kellar and S M Tavakoli, 'Ultrasound Detectable Instrument', International Patent No. WO 00/66004, 9 November 2000.
- A E Joseph, E J C Kellar, S M Tavakoli and D Nassiri, 'The Enhanced Visualisation of Devices for Ultrasound Guided Interventional Procedures', Annual meeting of the British Ultrasound Society, Glasgow, UK, 8-10 December 1999.
The authors wish to thank BTG and Medilink for funding this project, Ruth McD Sutherland and Ian M Bingham of BTG and colleagues for their continuous support, and colleagues at St George's Hospital and TWI for their collaboration.