The ideal for intravenous drug delivery in the home is a small portable device that can be hidden discretely on the patient. The development and manufacture of a micro drug pump of this type are described.
By: R. Hoyle
MicroBridge Services Ltd, Cardiff, UK
Achieving the optimum effect
Many different types of drug delivery system have been developed aimed at meeting different requirements in terms of the actual drug to be administered, the form of the drug (powder, liquid or gas), the rate of delivery, the patient’s state of health and physical and mental capabilities, reliability and repeatability, product shelf life, cost of manufacture and many other considerations. Even for similar types of ailment, there are several different types of drug delivery systems; for example, asthma sufferers can choose from several different types of drug, drug form, delivery mechanisms and product delivery systems depending on their personal circumstances. Pressurised metered dose inhalers, breath activated inhalers, dry powder devices and nebulisers are examples of the available devices that have been developed to cater for everyone from children to frail and elderly patients. However, for all the different delivery systems there are the same basic requirements to achieve the optimum effect: drug delivery systems need to be able to deliver the correct amount of drug, to the required place and at the appropriate time.
Numerous diseases and associated treatments require drug delivery into the body by means other than inhalation. The most obvious method is orally in the form of liquids or solid pills and capsules and many medicines and treatments are designed to be administered in this way. A less frequently used drug delivery method is absorption through the skin, but this has the disadvantage that only relatively small amounts of drug can be administered over a longer period of time. However, through-the-skin methods are ideal for certain applications, including immunisation and allergy testing in which extremely small amounts of drug or substance needs to be administered to determine the immuno-responses of the patient.
Another common method of administering a drug is intravenous delivery in which a drug is introduced directly into the blood stream by a hollow needle or thin pipe. A different approach is to diffuse the drug into the vascular system via the skin so that the fine network of blood vessels under the surface can absorb the drug and deliver it to the blood stream. Intravenous drug delivery devices have been developed to a great degree of sophistication and a large range of product types. However, they all have a common theme in that they deliver drugs in liquid form into the body and this in itself presents material biocompatibility issues that need to be addressed. Drug delivery by this method has a fast response time and both large and small quantities can be delivered relatively efficiently. However, one of the main challenges in this method of delivery is dose control and errors in administered drug volumes account for several deaths each year. Controlling the dosing rate is something that hospitals have made great efforts to get right and this is possible in the controlled and regulated environment that is typical in healthcare services. This control system, however, is expensive and often requires extended patient hospitalisation, something that is not attractive to the patient or the healthcare provider. Systems have been developed for intravenous drug delivery in the home and these offer patients significant independence, but the ideal goal is to deliver drugs via a small portable device that can be hidden discretely about the patient. Devices of this type would be ideal for extremely small, almost continuous dosing of drugs so that unnatural or undesirable peaks and troughs in concentration can be avoided. Obvious applications of this type of drug delivery device would be certain chemotherapy treatments, insulin treatment for diabetics and immunisation treatments. The big advantage of this type of system to the patient would be the portable nature of the device, but this leads to several challenges in the design and manufacture of the device.
Development of a micro drug pump
| Figure 1. The micro drug delivery pump showing main body, the disposable drug reservoir and pump and the access-to-body patch. |
 |
Exploiting the potential of micro dosing of drugs was the aim behind the establishment of Starbridge Systems Ltd (Swansea, UK). It has developed a pump that incorporates microfluidic structures and wells suitable for precise delivery of small quantities of liquid drugs. The size of this pump (approximately 42 × 35 × 14 mm) means that it can be easily attached to discrete parts of the body or underclothing in such a way that it does not show as an unsightly bulge (Figure 1). It is light in weight and can administer amounts of drug for several days without need for replacement. The weight of the pump varies with drug volume, but is in the range 15 to 20 g and contains up to 3 cm3 of liquid drug. The pump outlet is attached to a plastic pipe, which is inserted into a blood vessel or into the subcutaneous layer, through which the drug is pumped. The main fluid handling part of the pump is disposable and is intended to be replaced every two or three days. For this part to be disposable, it must be cheap to manufacture, particularly in view of the large projected market for the product (millions per year in the medium term), thus the manufacturing process demands injection moulding. This, in turn, requires specific medical grade polymers, precise mould tool manufacture and moulding conditions. To meet these requirements, specialist micro manufacturing processes and techniques were needed.
| Figure 2. Part of thin wafer component in polycarbonate. Thickness of part is 250 µm except for the 3-lobed recess around the large hole, which is 125-µm thick. Small hole diameters are 300 µm. |
 |
The main disposable fluid handling part of the pump consists of a series of plastic components of different sizes and shapes, but characterised as being small and micro structured. The functional internal part of the disposable fluid hand-ling device is assembled from a layered assembly of flat components. Each component measures 15 × 7 mm and each layer plays a vital individual role: some form the pump mechanism and other layers handle and control the volume of liquid being pumped. One of the more challenging components from a mould tool manufacturing and moulding perspective is a thin polymer wafer with lots of small holes, each 300 µm in diameter (Figure 2). The thickness of this component is a maximum of 250 µm and it is recessed in parts and those recesses are 125-µm thick, which presents a challenge for moulding. Some of the other mouldings with micro channels present micro feature machining issues such as machining micro, high aspect ratio walls and bosses with low surface roughness. The initial stages of the product development project were satisfied by the manufacture of brass prototype mould tools and these were successful in producing 2000 prototype components for trial purposes (Figure 3). However, the projected expansion in numbers of more than one million components requires mould tools in a much harder material than brass, and tool steel is an obvious solution.
| Figure 3. Brass prototype mould tool for part in Figure 2 showing hole bosses. This cavity image is prior to the runner and injection gate being added. |
 |
The main difficulty with the thin wafer part was getting the mould to fill without short shots or flashing. A Mikrosystem 50 microinjection-moulding machine was used (Wittmann Battenfeld, www.battenfeld-imt.com). This proved to have an extremely narrow process window, but the brass prototype mould tool managed to survive more than 2000 shots before tool wear became unacceptable. The material used was a medically compatible polycarbonate (PC) and this proved to be more difficult to mould than other nonmedical grades of PC, probably because of the different proportions of plasticisers in the material.
| Figure 4. Four-impression steel mould tool. The total shot volume to fill the mould tool is less than one cubic centimetre. The wall structures that form the microfluidic channel in the polymer components can be seen on each component part on the right hand side half of the mould tool. |
 |
Other components were also moulded in PC and the design of these allowed the manufacture of a steel rather than brass mould tool (Figure 4). This proved more robust than the brass tool and several thousand shots were obtained without any problems. The moulding includes a small fluid channel on one side, which is in the order of 500-µm wide and 250-µm deep, that was formed by a wall feature on the right hand cavity plate.
Machining capability
The machining of the mould tools by MicroBridge Services Ltd exploited the micromachining capabilities developed at the Manufacturing Engineering Centre (MEC) at Cardiff University, UK. The micromachining centre that was used (Kern,
www.rainfordprecision.com) was employed extensively to develop micromachining strategies suitable for machining micro structured mould tools. Much research work has been conducted on machining microfluidic mould tools, creating micro walls and other features in brass and steel that replicate as channels in polymer. This has been driven by funded research projects such as the European Commission’s (EC) Framework Programme (FP) 6, Surface Enhanced Micro Optical Fluidic Systems (
www.semofs.com) project, which is a multi-national collaboration for the development of an all-polymer microfluidic system for protein sensing, as shown in Figure 5. This device incorporated a fully integrated thermo-
plastic microfluidic substrate complete with active gel-electrolysis fluid pumps, electrical control circuit, controlled surface modification and fluid sealing surfaces suitable for protein analysis applications.
Simulation capability
| Figure 5. Brass microfluidic mould tool cavity showing microwall structures, moulding of polymer microfluidic substrate and final microfluidic device. |
 |
Because of mould flow simulation research performed by MEC, Starbridge Systems was confident in being able to mould such a thin, flat component. In part of this research, a model for simulating the flow behaviour of polymer melts in micro cavities was investigated; this focused in particular on the factors affecting shear rate, pressure and temperature. The model was developed within Moldflow Plastics Insight 5.1 software (www.moldflow.com) and then validated against the experimental results. The model, shown in Figure 6, was similar in size and thickness to the thin wafer part moulded for the micro litre pump. This simulation work, supported by experiment, showed that the critical factors affecting mould filling were injection speed and polymer melt temperature at the start of the injection process. Other factors such as mould and runner surface finish and mould tool temperature also had an effect, but to a lesser degree.
Next steps
| Figure 6. Simulation of melt temperature during a short shot (a) and simulation during a full shot (b). |
 |
The development of the micro litre drug pump is an important product for its manufacturer and access to micro moulding and tool making facilities that are closeby has been an advantage during what has been an involved and complex development process. The company is now planning the next stage of its route to market and this will involve sourcing manufacturing capability in the millions over the next five years together with mould tooling to match. The development of micro featured mould tool manufacture and micromoulding expertise is critical to this plan.
Acknowledgements
This article has been prepared with the kind permission of Starbridge Systems Ltd. In mid 2009 Starbridge changed its name to Cellnovo Ltd (
www.cellnovo.com). Some of the research that drove the process capability developed at MEC was funded by the Engineering and Physical Sciences Research Council Programme, The Cardiff Innovative Manufacturing Research Centre (
www.cuimrc.cf.ac.uk) and the EC FP 6 Project Surface Enhanced Micro Optical Fluidic Systems (
www.semofs.com). In addition, work was performed within the framework of the EU FP6 Network of Excellence Programme Multi-Material Micro Manufacture (4M): Technologies and Applications (
www.4M-association.org).
Dr Robert Hoyle
is Business Development Manager at
MicroBridge Services Ltd, Manufacturing
Engineering Centre, The Parade, Newport Road, Cardiff University, Cardiff CF24 3AA, UK
MicroBridge Services Ltd is wholly owned by Cardiff University and has been set up to exploit the commercial potential of the micro and nano engineering capability developed at MEC.
Login or
register to post comments
