Originally published September 1996
When the company's material vendor decided to withdraw its copolyester elastomeric (COPE) material from certain medical applications, Utah Medical Products, Inc., located in Midvale, UT, had to explore alternative materials for the white dual-lumen tubing used in its disposable intrauterine pressure- transducer catheter. Finding another tubing material that met the application's special criteria was not an easy task. Compromising among many material and product requirements was necessary to obtain balanced product performance. This article describes the company's material-selection process, which included tubing extrusion, material testing, prototype evaluation, a cost analysis, and product safety testing to ensure FDA requirements would be met. As a result of this project, a higher quality, higher yielding, and slightly lower cost replacement tubing material was found. In addition, several lessons were learned that may prove useful to others engaged in choosing polymeric materials for medical devices.
A medical-grade, gamma- and ethylene oxidesterilizable elastomeric material and a white color concentrate that together satisfied USP XXII Class VI and Tripartite biocompatibility guideline requirements were needed to replace the COPE being used for the catheter tubing.1 Based on the product's application and assembly requirements,2 the other replacement material criteria were defined, in order of descending importance, as follows:
* Stiffness properties equivalent to the current COPE material under various environmental conditions.
* Heat stability up to 65°C (149°F).
* Good chemical resistance to saline solution and hospital cleaning fluids.
* Smooth surface finish.
* Adequate adhesion to on-line ink-jet printer ink, ultraviolet (UV)- curable adhesives, and a polyvinyl chloride (PVC) overmolding compound.
* Lower cost than COPE, if possible.
Even a material with good chemical resistance can show poor adhesion properties unless its surface is specially treated. Thus, the word adequate was used to define the adhesion requirement since primer would be applied to the tubing during processing to promote adhesion.
Because stiffness is the most important physical characteristic of this tubing, hardness and flexural modulus properties were the measurements used in the initial material screening. The most promising candidate materials for evaluation were selected after a review of the general properties of six different kinds of thermoplastic elastomers: styrene block copolymers, thermoplastic elastomeric olefins, thermoplastic polyurethanes, thermoplastic vulcanizates, COPEs, and copolyamides (COPAs).37 The first four types were eventually dropped from further consideration for various reasons, including inadequate hydrocarbon-fluid chemical resistance, heat stability, and hydrolytic stability, and the unavailability of the proper grades for the application. Primarily because of their low cost, we also examined PVC blends, even though published data did not include all the properties of interest.
We next contacted several vendors of COPE, COPA, and PVC and discussed the product application with them, discovering that several other COPE vendors had adopted the same precautionary policy as our current supplier. Finally, three candidate materials were selected for evaluation. Table I summarizes these candidate materials' typical properties compared with the current (control) material, COPE-1. COPE-1, COPE-2, and COPA are natural in color, while the PVC is a precolored white material. Looking at the table, most people would expect that the COPE-2 material would have the highest probability of success for the project. However, because adding a white color concentrate affects these material properties, it was necessary to undertake a tubing extrusion study.
Tubing Extrusion and Testing
The purpose of the tubing extrusion study was twofold: to observe each material's extrusion characteristics and to determine if it could be used to produce acceptable products, and to create tubing samples for material analyses and other testing. During the extrusion process, problems arose involving the color concentrate supplier, die drawdown ratios, string-up, and the dimensional stability of the new materials. Running conditions had to be adjusted to obtain reasonable tubing for material testing. In general, the COPEs and COPA were run under similar conditions, with only minor process differences. The PVC material, however, was run under different conditions--melt temperature lower by 100°F, higher pull speed--and had a tendency to stick to the uncoated areas of the screw. Although the PVC tubing samples varied greatly in dimensions and surface roughness, it was decided to include them in the material testing program. Ultimately, 2000 pieces of tubing measuring 30.5 in. long by 0.140-in. OD were formed from each of the candidate materials for material testing, prototype building, and product evaluations.
A total of 41 material tests were conducted on the tubing samples, and the results for the candidate materials were compared with those for COPE-1. For each property studied, each new material was assigned a ranking based on its performance compared to this control material, which was ranked 1.00. Table II presents these data for 27 of the most important properties tested. In this table, the higher the number, the better the property is. Although the stiffness of the COPA tubing as extruded ranked higher than the control (1.22 versus 1.00), at 0.96 it was close to the control tubing's stiffness after a 160-minute exposure to body-temperature saline solution. In contrast, the stiffness ranking of the COPE-2 tubing as extruded was 0.87, lower than the control tubing, and it stayed near that level during the saline solution tests.
Based on all of the material test results, it was decided to build 100 prototype product units from COPE-1, COPE-2, and COPA materials for further evaluation. Although the PVC tubing exhibited the best ink adhesion properties, and the vendor stated that the material could be recompounded to match the required stiffness, PVC was dropped from the material candidate list at this time because of its very high stiffness ranking and poor heat stability. After 6 months under room temperature conditions following two gamma sterilization test cycles, the PVC tubing changed from white to yellow.
Prototype Building and Evaluation
To ensure that the materials were nontoxic, the MEM elution cytotoxicity test was performed on the COPE-2 and COPA materials before the prototype units were built. Both materials passed the test.
Building the prototypes at the manufacturing site enabled us to find out what problems would be encountered for each material during the assembly process. Eleven manufacturing steps are involved to complete the product. Table III shows comparative data for four of those steps, three of which were difficult to perform using the COPE-2 tubing. Since there were so many problems with this material during insertion operations, the vendor was contacted for advice on processing. The vendor indicated that blocking and tackiness were normal for the material, and that at high humidity the blocking would be even more pronounced. Nevertheless, it was decided to conduct product functional evaluations on the prototype units made from all three materials.
Eleven product functional property tests were conducted on the prototype units. The final testing involved four electrical properties: sensitivity, balance, plug, and leak. Table IV presents comparative data for six important tests. These results show that the COPA prototypes had a lower aminoinfusion rate than the COPE units. This variation was caused by the smaller cross-sectional area of the COPA tubing's lumen, mentioned earlier in Table II, and the problem can be corrected during the tubing extrusion process. Because of the relatively low volume of tubing material used for the product, the COPE-2 vendor was not willing to modify the material to reduce or eliminate its blocking and tackiness. Therefore, the COPE-2 material was withdrawn from the candidate list and the COPA material was selected as the replacement for the current COPE-1 material.
According to the initial material cost information given in terms of price per pound in Table I, the COPA material is more expensive than COPE materials. However, those data assumed a constant cost for the color concentrate, which turned out to be a poor assumption. After a thorough cost analysis was performed using information submitted by the polymer and color concentrate vendors, we were surprised to find out that the COPA's unit cost is actually slightly lower than that of COPE-1, which depends on the amount of material ordered at one time. Thus, switching to a new material will result in cost savings.
Product Safety Testing
After the COPA material was selected as the replacement for COPE-1, safety testing was done on the COPA prototypes only. In order to meet the requirements set in the United States Pharmacopeia XXII for Class VI materials as well as those of the Tripartite biocompatibility guidelines for medical devices in the "Internal devices, tissue and tissue fluid, short-term duration" category, devices built with COPA tubing were sent to an outside laboratory for toxicity testing for risk-assessment determination. The following nine tests were performed: acute systemic toxicity, cytotoxicity, hemolysis, implantation, irritation, mutagenicity, pyrogenicity (LAL), sensitization, and subchronic toxicity. The COPA material was found to satisfy all requirements.
Products built at our manufacturing facility were also sent to an outside lab for gamma sterilization qualification tests. This evaluation found that the gamma cycle used for the COPE-1 product would also work well for the device made with the new tubing material. The results of these two types of testing should be sufficient for the modified product to meet all FDA product safety requirements.
Based on the material and product evaluation program described above, a higher quality, slightly lower cost COPA material with a potentially shorter extrusion cycle was found to replace the COPE-1 tubing material in Utah Medical Product's disposable catheter product, without requiring any modifications of the existing assembly procedures and equipment. The COPE-2 material could also have been used, but the assembly methods would have had to be modified to account for the tubing's tackiness.
Some of the lessons learned during this project may also be helpful to others who engage in polymeric material selection for medical devices. Although the initial material screening saved both time and money, we found that some of the material properties data used to select the candidate materials--such as that for hardness and flexural modulus--were not accurate for amorphous materials like PVC. In addition, the fine differences in stiffness among the materials tested could not be determined from the vendors' material data sheets, but could only be detected by testing the actual extruded tubing.
Polymeric materials in the same family may not perform the same way in an application, and such differences are also not apparent on material data sheets. For example, the COPE-1 and COPE-2 materials investigated in this project had different surface properties, which caused assembly problems with COPE-2 prototypes. In sum, testing of actual processes and prototype products is necessary to determine a material's performance in a specific application.
It is also important to observe the material color changes following gamma radiation over an extended period in ambient conditions. The PVC tubing material studied in this project changed from white to yellow 6 months after the second gamma radiation test, although it ranked higher than the control in other color tests.
Finally, it is always a wise practice to keep good records of testing results and the methods used for a product material evaluation. Subsequent material selection projects will be easier and faster if the base material comparison data collected during the original evaluation is readily available.
1. Rakita PE, and Foure M, "Gamma Radiation Stabilization of PVC," J Vinyl Tech, 6(2):7376, 1984.
2. Woo L, Westphal S, and Ling M, "Consideration for Medical Plastics Selection and Substitution," in Society of Plastics Engineers Technical Papers, vol XXXX, (ANTEC 94), Brookfield, CT, Society of Plastics Engineers (SPE), pp 26432646, 1994.
3. Eller R, "Thermoplastic Elastomers," in Modern Plastics Encyclopedia, Hights-town, NJ, McGraw-Hill, pp B56B61, 1995.
4. Lebel MA, "Thermoplastic Elastomers," Plastics Engineering, January, pp 2326, 1994.
5. School RJ, "Markets for Thermoplastic Elastomers," in Society of Plastics Engineers Technical Papers, vol XXXX, (ANTEC 94), Brookfield, CT, SPE, pp 33883392, 1994.
6. Donohue J, "Metallocene Catalysts: Driving the Polyolefin Revolution," Med Plast Biomat, 1(2):2024, 1994.
7. Clark RJ, and Henkee CS, "Poly(Ether Amide) Triblock and Star Block Copolymers," in Society of Plastics Engineers Technical Papers, vol XXXX, (ANTEC 94), Brookfield, CT, SPE, pp 34083412, 1994.
George Chang is a manufacturing engineer at Utah Medical Products, Inc. (Midvale, UT), where he specializes in the injection molding and extrusion of medical polymers.