Establishing Material Compatibility, Process Conditions, and Bubble Points of Filters

Sterility assurance is critically important for liquid parenteral products. For pharmaceuticals that are not heat-labile, terminal sterilization may be performed. However, in most cases, sterilizing-grade filters must be used. Following their use in manufacturing, these filters are tested to ensure that the membrane remained integral. A method referred to as the bubble-point integrity test is commonly performed. This test may be completed manually or with an automated bubble-point integrity tester. More than 85% of companies use the bubble point as the test for filter integrity when manufacturing sterile products (1). In addition, the US Food and Drug Administration guideline states that the bubble-point test is mandatory to ensure that filter leaks or perforations did not occur during the filtration process (2).

Figure 1: Apparatus used to recirculate the antimicrobial preservative-containing solution through the test filter. (IMAGE IS COURTESY OF MERCK & CO.)

Before using a sterilizing-grade filter to manufacture a parenteral formulation, a bubble-point value is determined with the filter and the product in the laboratory. A procedure for performing and determining the product-specific bubble point has been described in the literature (3). In this procedure, the product is first pumped briefly through tubing connected to a filter. Once the filter has been wetted with the product, it is removed, and the bubble point is determined, typically using an automated bubble-point integrity machine. Three replicates are performed, and the average product-specific bubble point is established. The bubble-point value is then transferred to a batch document that is used during the manufacturing process. If the filter fails the minimum product-specific bubble point following the manufacture of a product, a series of tests may be performed on the filter to determine whether a true leak is present or whether a false failure occurred (3).

Table I: Preservatives used in filter bubble-point studies and typical amounts found in pharmaceutical products. (IMAGE IS COURTESY OF MERCK & CO.)

The authors demonstrated that in some cases the failure of the product-specific bubble point may not result from a leak or loss in membrane integrity, but from factors such as interaction between the product and materials used during the formulation and filling process (4). The selection of tubing was shown to be critically important when filling parenteral liquid products that contain antimicrobial preservatives such as phenol, m-cresol, and benzyl alcohol (4). In these studies, the interaction of these preservatives with the tubing resulted in the release of polydimethyl siloxane (silicone oil), which lowered the product-specific bubble point (4). The lowering of the bubble point resulted from a reduction of the liquid surface tension of the wetted membrane pores and was caused by the adsorption of polydimethyl siloxane to the filter membrane. This article will review these findings and discuss factors to consider when defining a product-specific bubble point.

Table II: Tubing types evaluated in filter bubble-point studies. (IMAGE IS COURTESY OF MERCK & CO.)

Materials and methods

The equipment that was used to evaluate the effect on filters with various types of tubing and antimicrobial solutions is depicted in Figure 1 (4). In this procedure, the solution was recirculated at flow rates of 160–200 mL/min for a minimum of 15 h. The bubble point was measured using an automatic integrity tester (Sartocheck 4, Sartorius, New York).

Table III: Corrected product-wetted bubble-point (CPBP)–preservative solutions. (IMAGE IS COURTESY OF MERCK & CO.)

A series of tests was performed using various types of tubing and antimicrobial preservative-containing solutions. Table I lists some of the antimicrobial preservatives that are used in parenteral products and the typical amounts found in these products (5). The levels of antimicrobial preservatives ranged from 0.0715% to 1.9% (5). Solutions containing 0.25% of each preservative in saline solution were used in previous filtration studies (4). Each of these preservative-containing solutions was recirculated through several types of tubing (listed in Table II) with a sterilizing-grade, 0.22-µm filter (Millipak 20, Millipore, Bedford, MA).

Table IV: Effect of preservative-containing solutions and platinum-cured silicone tubing on the bubble point. (IMAGE IS COURTESY OF MERCK & CO.)

The experiments were performed by first determining the corrected product-wetted bubble point (CPBP) for the preservative solutions tested. Each filter was flushed with water, and the bubble point was determined. These tests were performed with replicates according to the procedures outlined by the Parenteral Drug Association (3). The results of these tests are summarized in Table III.

Table V: Effect of preservative-containing solutions and C-Flex tubing on the bubble point. (IMAGE IS COURTESY OF MERCK & CO.)

Before the 15-h recirculation experiments were initiated, the filters were wetted with the preservative-containing solutions (see Tables IV–VII), and the bubble point at time zero (i.e., preuse bubble point) was measured (4). The solution was then recirculated for 15 h, then a postuse bubble-point test of the filter was performed (see Tables IV–VII). The results indicated that the bubble point of the filter was affected the most when using platinum-cured silicone tubing (see Table IV). The smallest effects on the bubble point occurred when using C-Flex, PharMed, and Biopharm tubing (Cole-Parmer, Vernon Hills, IL) (see Tables V–VII).

Table VI: Effect of preservative-containing solutions and PharMed tubing on the bubble point. (IMAGE IS COURTESY OF MERCK & CO.)

Conclusion

These studies demonstrated that the type of tubing selected for filter-sterilizing an antimicrobial preservative-containing formulation may affect the filter's postuse product bubble point. In the case of platinum-cured silicone tubing, bubble-point failures were caused by the presence of poly(dimethyl) siloxane, or silicone oil, trapped on the filter membrane (4). The presence of poly(dimethyl) siloxane was detected using Fourier transform infrared spectroscopy (4). The lowering of the bubble point was confirmed by readdition of poly(dimethyl) siloxane in a separate recirculation study (4). The lowering of the bubble points resulted from a reduction in the surface tension of the liquid that was wetting the membrane pores. For example, it has been previously demonstrated that silicone oils have low surface tensions (6). The surface-tension reduction was caused by the adsorption of polydimethyl siloxane to the filter membrane. The lower surface tension caused the gas to pass through the pores of the filter more easily, resulting in lower bubble-point values. The following items must be considered when determining the product-specific bubble point of a filter for sterilizing parenteral drugs:

• Materials that have product contact, specifically those upstream of the filter, should be tested to determine if they affect the bubble point. If the materials are found to suppress the bubble point, a material screening-procedure should be used to identify those that have the least effect on the bubble point.

• When selecting tubing material that will be in contact with product that contains preservatives, consider the length of time that the product will be in contact with the different materials, including the filter, during the manufacturing process. Incorporate the processing time into laboratory studies for determining the product-specific bubble point (7).

Table VII: Effect of preservative-containing solutions and Biopharm tubing on the bubble point. (IMAGE IS COURTESY OF MERCK & CO.)

Brian K. Meyer* is a research fellow, and Diego Vargas is a senior engineer, both at Merck & Co., 770 Sumneytown Pike, West Point, PA 19486, tel. 215.652.3992, fax 215.652.5299, brian_meyer@merck.com

*To whom all correspondence should be addressed.

References

1. M. Russell and T.H. Meltzer, "An Interpretation of the Pharmaceutical Industry Survey of Current Sterile Filtration Practices," PDA J. Pharm. Sci. Technol.52 (6), 337–339 (1998).

2. FDA, Guideline on Sterile Drug Products Produced by Aseptic Processing (Rockville, MD, 1987).

3. Parenteral Drug Association, "Technical Report No. 26: Sterilizing Filtration of Liquids," PDA J. Pharm. Sci. Technol.52 (S-1) (1998).

4. B.K. Meyer and D. Vargas, "Impact of Tubing Material on the Failure of Product-Specific Bubble Points of Sterilizing-Grade Filters," PDA J. Pharm. Sci. Technol.60 (4), 1–6 (2006).

5. B.K. Meyer et al., "Antimicrobial Preservative Use in Parenteral Products: Past and Present," J. Pharm. Sci.96 (12), 3155–3167 (2007).

6. T. Svitova et al., "Wetting Behavior of Silicone Oils on Solid Substrates Immersed in Aqueous Electrolyte Solutions," Langmuir18 (18), 6821–6829 (2002).

7. B.K. Meyer and D. Vargas, "Impact of Tubing Material on the Failure of Product-Specific Bubble Points of Sterilizing-Grade Filters," presented at the National PDA Conference, Colorado Springs, CO, April 2008.