Mechanical Properties of Antibacterial Silicone Rubber for Hydrocephalus Shunts R. VAN NOORT, Sheffield University and Area Health Authority (Teaching),Department of Medical Physics, Hallamshire Hospital, Glossop Road, Sheffield SlO 2JF, England, and R. BAYSTON, University Department of Paediatrics, Children’s Hospital, Western Bank, Sheffield SlO 2TH, England
Summary In an effort to find a solution to the serious problem of bacterial colonization of cerebrospinal-fluid shunting devices, room-temperature and heat-vulcanizing silicone rubhers were impregnated with gentamicin sulphate. The effects on the mechanical properties of the two rubbers were studied. Results show that the tensile strength and extensibility of the room-temperature-vulcanizing silicone rubber reduced with increasing concentration of the drug. For the heat-vulcanizing silicone rubber, the tensile strength was also found to decrease with increasing drug content. The extensibility after an initial reduction a t low concentrations was found to increase a t drug concentrations in excess of 10 mg/g. Nevertheless, the changes in mechanical properties measured are considered not to be so great as to preclude the application of drugimpregnated silicone rubbers to cerebrospinal-fluid shunting devices.
INTRODUCTION Bacterial colonization of cerebrospinal-fluid shunting devices for the treatment of hydrocephalus is a serious problem, resulting in removal of the shunt and subsequent replacement. Although the etiology of shunt colonization has been definedl and attempts have been made using various methods to prevent its occurrence, few have shown any S U C C ~ S SBecause . ~ , ~ of this the introduction of gentamicin sulphate, an antibacterial drug, into the rubber so as to prevent colonization is being investigated. The drug is intended to prevent the development of bacterial microcolonies on the surface of the rubber. It is not intended to produce therapeutic concentrations of the drug in the tissues or body fluids which can be dealt with by normal prophylactic methods. Previous studies using room-temperature-vulJournal of Biomedical Materials Research, Vol. 13,623-630 (1979) 0 1979 John Wiley & Sons, Inc. 0021-9304/79/0013-0623$01.00
624
VAN NOORT AND BAYSTON
canizing silicone rubber (Silastic 382, Dow Corning) have been reported* and have shown that the therapeutic concentration of gentamicin sulphate for these purposes is of the order of 25 mg/g of rubber. Heat-vulcanizing silicone rubbers have been used in the manufacture of hydrocephalus shunts since 1957. The materials are, for the time being, the most suitable for the purpose. There are no hinged parts and rapid opening and closing of the slit valves do not occur. The slit valves themselves rarely fail. Failure is usually due to deterioration caused by lipid absorption and calcification, or due to blockage by debris or blood clot. As well as displaying antibacterial activity, the mechanical properties of the impregnated silicone rubber should not deteriorate to a level which will preclude their practical application. The resultant mechanical properties of impregnated silicone rubber are reported here.
MATERIALS AND METHODS Two types of silicone rubber were used for comparison. One was a room-temperature-vulcanizing (RTV) silicone rubber (Silastic 382) and the other a heat-vulcanizing (HV) silicone rubber (Silastic MDX4/4515), both of which are manufactured by Dow Corning Ltd. Gentamicin sulphate (Genticin, Nicholas Laboratories Ltd.) was the antibacterial drug used. The drug was added to the raw stock of the two rubbers in proportions of 5,10,25, and 50 mg/g of rubber. In the case of the RTV rubber, the drug was mixed in manually. Stannous octoate was added as the catalyst and the mixture cast into sheets 1-mm thick. The antibiotic was added to the HV rubber while it was repeatedly passed between two rollers until, by subjective assessment, integration was complete. Sheets 1-mm thick were produced by compression molding at 100°C for 30 min. The sheets were removed from the mold and post-cured for 24 hr a t 150°C. Strips of the two rubbers measuring 50 mm in length and a uniform width of 5 mm were cut from the sheets to test for mechanical properties. Previous studies have shown that strips of these dimensions give accurate and reproducible result^.^.^ Tensile tests were carried out on an Instron (.table model 1026) which had incorporated in it a pogo-extensometer. This extensometer accurately measured increments of strain of 10%. Clamping of
SILICONE RUBBER FOR HYDROCEPHALUS SHUNTS
625
the strips was achieved by the use of pneumatic grips and all tests were carried out at a constant crosshead speed of 20 cm/min. Six strips of each of the rubbers were tested for each of the drug concentrations. The averaged stress-strain curves were computed for strain increments of 20% and 40% for the RTV rubber and HV rubber, respectively. In addition, the means for the ultimate tensile strength and extensibility of the rubbers and their standard deviation were determined.
RESULTS AND DISCUSSION The stress-strain curves for the two rubbers are depicted in Figure 1. These clearly show the reproducibility of the results obtained with standard errors no larger than 5%. Both the ultimate tensile strength and the extensibility of the RTV rubber were found to decrease significantly with increased concentrations of gentamicin sulphate while the stiffness (i.e., the slope of the stress-strain curve) was comparatively little affected. The curves for the tensile strength and extensibility of the RTV rubber, depicted in Figure 2 and Figure 3, respectively, show that small amounts of the drug are relatively more detrimental than large amounts, indicated by the initially rapid reduction in these properties. This pattern of behavior is similar to that suggested by Nielsen when little or no adhesion between the drug and the rubber occurs.6 In such a case, the drug cannot carry any of the load since stress transfer to the drug particles is not possible, so all the load must be carried by the rubber. As shown by the stress-strain curves for the HV rubber [Fig. l(b)], the pattern of behavior for this rubber was found to be considerably different from that observed for the RTV rubber. Whereas the stiffness of the RTV rubber was only slightly affected by impregnation with gentamicin sulphate, the stiffness of the HV rubber was considerably reduced at a concentration of 50 mg/g. Again, small amounts of the drug have a relatively more pronounced effect on the properties. The curve obtained for the tensile strength of the HV rubber, shown in Figure 2, is very similar to that of the RTV rubber except that the strength of the former is far superior to that of the latter. For both rubbers, the impregnation with 50 mg/g of gentamicin sulphate resulted in a 40% reduction in tensile strength compared with controls.
626
VAN NOORT AND BAYSTON
ELONGATION We)
(a) Fig. 1. Averaged stress-strain curves for room-temperature-vulcanizing silicone rubber (a) and heat-vulcanizing silicone rubber (b) with gentamicin sulphate concentrations of 0 ( O ) ,5 ),(. 10 ( A ) , 25 ( A ) , and 50 (0) mg/g. The vertical lines define one standard error mean.
The results of the ultimate extensibility of the HV rubber were found to be strikingly different from the RTV rubber (Fig. 3). A small amount of gentamicin sulphate did cause an initial reduction in the extensibility, but when the concentration of gentamicin was increased beyond 10 mg/g, an increase in extensibility resulted. The authors suggest that this difference in behavior can be explained by the degree to which vacuoles can form around the drug particles. The RTV
SILICONE RUBBER FOR HYDROCEPHALUS SHUNTS
ELONGATION
627
(%I
Fig. 1. (Continued from previous page.)
rubber is considerably less extensible than the HV rubber and failure of the elastomer will result before any reasonably sized vacuoles can form. In the case of the HV rubber, since it is some three times more extensible than the RTV rubber, considerably larger vacuoles can form around the drug particles which can introduce a sizable contribution to the total extensibility of the elastomer. This contribution would depend upon the concentration of the drug in the elastomers and the results suggest that it only becomes pronounced at concentrations in excess of 5 mg/g.
628
VAN NOORT AND BAYSTON
C O N C E N T R A T I O N OF G E N T A M l C l N SULPHATE ( m g l g m )
Fig. 2. The ultimate tensile strength of heat-vulcanizing silicone rubber ( M ) and room-temperature-vulcanizing silicone rubber ( 0 )as a function of the concentration of gentamicin sulphate. Bars denote one standard deviation.
Improvements are being sought in the mechanical properties of impregnated silicone rubbers by enhancing the distribution of gentamicin sulphate in the materials. The narrow scatter of results for mechanical properties, derived from strips of rubber cut from different areas in the same sheet, indicate that a reasonably uniform distribution of the drug has been achieved. The measurement of particle size by optical microscopic examination showed that the drug particles varied in size from less than 1.0to 62.5 pm, with 95%smaller than 2.5 pm. The large agglomerates could act as stress raisers
SILICONE RUBBER FOR HYDROCEPHALUS SHUNTS
CONCENTRATION
OF GENTAMlClN SULPHATE
629
Imglgm)
Fig. 3. The ultimate extensibility of heat-vulcanizing silicone rubber (m) and room-temperature-vulcanizing silicone rubber ( 0 )as a function of the concentration of gentamicin sulphate. Bars denote one standard deviation.
causing premature failure of the silicone rubbers. The breakdown and dispersion of such agglomerates may lead to improved mechanical properties. Some preliminary activity tests carried out on the impregnated HV silicone rubber by the methods described in a previous report4 suggest that the activity of the drug is not significantly affected by the curing procedure adopted and that the required level of drug concentration may only need to be as low as 25 mg/g or less. Should this prove to be the case (work has already been undertaken to assess the required
630
VAN NOORT AND BAYSTON
therapeutic concentration for the HV rubber), then the HV rubber impregnated with 25 mg/g or gentamicin sulphate would retain a tensile strength in excess of 70% compared with the control while increasing the extensibility by only 15%. A change in properties of such a proportion may well prove to be acceptable where such applications as CSF shunts are concerned.
CONCLUSIONS This work demonstrates that it is possible to impregnate both room-temperature-vulcanizing and heat-vulcanizing silicone rubbers with gentamicin sulphate. Although changes in mechanical properties of the silicone rubbers do occur, these are not so detrimental as to render these materials unsuitable for use. Further investigations of the suitability of silicone rubbers for impregnation are being undertaken and improved methods of impregnation are being tried out in an effort to enhance the mechanical properties of the impregnated silicone rubbers. The authors gratefully acknowledge the financial support of the Science Research Council (grant no. B/RG 77229). the Association for Spina Bifida and Hydrocephalus (grant no. 4018-01) and the Trent Regional Health Authority (grant no. 659).
References 1. R. Bayston and J. Lari, “A Study of the Sources of Infection in Colonized Shunts,” Deu. Med. Child Neurol., Suppl., 32,16 (1974). 2. R. Bayston, “Antibiotic Prophylaxis in Shunt Surgery,” Deu. Med. Child Neurol., 17, (35), 99-103 (1975). 3. K. Welch, “Prevention of shunt infection,” 2. Kinderchir., to appear. 4. R. Bayston, “Preliminary Studies of the Impregnation of Silastic Elastomers with Antimicrobial Substances,” Deu. Med. Child Neurol., Suppl. 37,50 (1976). 5. R. van Noort, Ph.D. Thesis, University of Sussex, 1976. 6. L. E. Nielsen, “Simple Theory of Stress-Strain Properties of Filled Polymers,” J Appl. Polym. Sci., 10,97 (1966). 7. R. van Noort, B. Harris, and M. M. Black, “Elastomeric Materials for Prosthetic Heart Valves,” Plast. Rubber, 7, February (1977).
Received May 10,1978 Revised November 6,1978
Summary In an effort to find a solution to the serious problem of bacterial colonization of cerebrospinal-fluid shunting devices, room-temperature and heat-vulcanizing silicone rubhers were impregnated with gentamicin sulphate. The effects on the mechanical properties of the two rubbers were studied. Results show that the tensile strength and extensibility of the room-temperature-vulcanizing silicone rubber reduced with increasing concentration of the drug. For the heat-vulcanizing silicone rubber, the tensile strength was also found to decrease with increasing drug content. The extensibility after an initial reduction a t low concentrations was found to increase a t drug concentrations in excess of 10 mg/g. Nevertheless, the changes in mechanical properties measured are considered not to be so great as to preclude the application of drugimpregnated silicone rubbers to cerebrospinal-fluid shunting devices.
INTRODUCTION Bacterial colonization of cerebrospinal-fluid shunting devices for the treatment of hydrocephalus is a serious problem, resulting in removal of the shunt and subsequent replacement. Although the etiology of shunt colonization has been definedl and attempts have been made using various methods to prevent its occurrence, few have shown any S U C C ~ S SBecause . ~ , ~ of this the introduction of gentamicin sulphate, an antibacterial drug, into the rubber so as to prevent colonization is being investigated. The drug is intended to prevent the development of bacterial microcolonies on the surface of the rubber. It is not intended to produce therapeutic concentrations of the drug in the tissues or body fluids which can be dealt with by normal prophylactic methods. Previous studies using room-temperature-vulJournal of Biomedical Materials Research, Vol. 13,623-630 (1979) 0 1979 John Wiley & Sons, Inc. 0021-9304/79/0013-0623$01.00
624
VAN NOORT AND BAYSTON
canizing silicone rubber (Silastic 382, Dow Corning) have been reported* and have shown that the therapeutic concentration of gentamicin sulphate for these purposes is of the order of 25 mg/g of rubber. Heat-vulcanizing silicone rubbers have been used in the manufacture of hydrocephalus shunts since 1957. The materials are, for the time being, the most suitable for the purpose. There are no hinged parts and rapid opening and closing of the slit valves do not occur. The slit valves themselves rarely fail. Failure is usually due to deterioration caused by lipid absorption and calcification, or due to blockage by debris or blood clot. As well as displaying antibacterial activity, the mechanical properties of the impregnated silicone rubber should not deteriorate to a level which will preclude their practical application. The resultant mechanical properties of impregnated silicone rubber are reported here.
MATERIALS AND METHODS Two types of silicone rubber were used for comparison. One was a room-temperature-vulcanizing (RTV) silicone rubber (Silastic 382) and the other a heat-vulcanizing (HV) silicone rubber (Silastic MDX4/4515), both of which are manufactured by Dow Corning Ltd. Gentamicin sulphate (Genticin, Nicholas Laboratories Ltd.) was the antibacterial drug used. The drug was added to the raw stock of the two rubbers in proportions of 5,10,25, and 50 mg/g of rubber. In the case of the RTV rubber, the drug was mixed in manually. Stannous octoate was added as the catalyst and the mixture cast into sheets 1-mm thick. The antibiotic was added to the HV rubber while it was repeatedly passed between two rollers until, by subjective assessment, integration was complete. Sheets 1-mm thick were produced by compression molding at 100°C for 30 min. The sheets were removed from the mold and post-cured for 24 hr a t 150°C. Strips of the two rubbers measuring 50 mm in length and a uniform width of 5 mm were cut from the sheets to test for mechanical properties. Previous studies have shown that strips of these dimensions give accurate and reproducible result^.^.^ Tensile tests were carried out on an Instron (.table model 1026) which had incorporated in it a pogo-extensometer. This extensometer accurately measured increments of strain of 10%. Clamping of
SILICONE RUBBER FOR HYDROCEPHALUS SHUNTS
625
the strips was achieved by the use of pneumatic grips and all tests were carried out at a constant crosshead speed of 20 cm/min. Six strips of each of the rubbers were tested for each of the drug concentrations. The averaged stress-strain curves were computed for strain increments of 20% and 40% for the RTV rubber and HV rubber, respectively. In addition, the means for the ultimate tensile strength and extensibility of the rubbers and their standard deviation were determined.
RESULTS AND DISCUSSION The stress-strain curves for the two rubbers are depicted in Figure 1. These clearly show the reproducibility of the results obtained with standard errors no larger than 5%. Both the ultimate tensile strength and the extensibility of the RTV rubber were found to decrease significantly with increased concentrations of gentamicin sulphate while the stiffness (i.e., the slope of the stress-strain curve) was comparatively little affected. The curves for the tensile strength and extensibility of the RTV rubber, depicted in Figure 2 and Figure 3, respectively, show that small amounts of the drug are relatively more detrimental than large amounts, indicated by the initially rapid reduction in these properties. This pattern of behavior is similar to that suggested by Nielsen when little or no adhesion between the drug and the rubber occurs.6 In such a case, the drug cannot carry any of the load since stress transfer to the drug particles is not possible, so all the load must be carried by the rubber. As shown by the stress-strain curves for the HV rubber [Fig. l(b)], the pattern of behavior for this rubber was found to be considerably different from that observed for the RTV rubber. Whereas the stiffness of the RTV rubber was only slightly affected by impregnation with gentamicin sulphate, the stiffness of the HV rubber was considerably reduced at a concentration of 50 mg/g. Again, small amounts of the drug have a relatively more pronounced effect on the properties. The curve obtained for the tensile strength of the HV rubber, shown in Figure 2, is very similar to that of the RTV rubber except that the strength of the former is far superior to that of the latter. For both rubbers, the impregnation with 50 mg/g of gentamicin sulphate resulted in a 40% reduction in tensile strength compared with controls.
626
VAN NOORT AND BAYSTON
ELONGATION We)
(a) Fig. 1. Averaged stress-strain curves for room-temperature-vulcanizing silicone rubber (a) and heat-vulcanizing silicone rubber (b) with gentamicin sulphate concentrations of 0 ( O ) ,5 ),(. 10 ( A ) , 25 ( A ) , and 50 (0) mg/g. The vertical lines define one standard error mean.
The results of the ultimate extensibility of the HV rubber were found to be strikingly different from the RTV rubber (Fig. 3). A small amount of gentamicin sulphate did cause an initial reduction in the extensibility, but when the concentration of gentamicin was increased beyond 10 mg/g, an increase in extensibility resulted. The authors suggest that this difference in behavior can be explained by the degree to which vacuoles can form around the drug particles. The RTV
SILICONE RUBBER FOR HYDROCEPHALUS SHUNTS
ELONGATION
627
(%I
Fig. 1. (Continued from previous page.)
rubber is considerably less extensible than the HV rubber and failure of the elastomer will result before any reasonably sized vacuoles can form. In the case of the HV rubber, since it is some three times more extensible than the RTV rubber, considerably larger vacuoles can form around the drug particles which can introduce a sizable contribution to the total extensibility of the elastomer. This contribution would depend upon the concentration of the drug in the elastomers and the results suggest that it only becomes pronounced at concentrations in excess of 5 mg/g.
628
VAN NOORT AND BAYSTON
C O N C E N T R A T I O N OF G E N T A M l C l N SULPHATE ( m g l g m )
Fig. 2. The ultimate tensile strength of heat-vulcanizing silicone rubber ( M ) and room-temperature-vulcanizing silicone rubber ( 0 )as a function of the concentration of gentamicin sulphate. Bars denote one standard deviation.
Improvements are being sought in the mechanical properties of impregnated silicone rubbers by enhancing the distribution of gentamicin sulphate in the materials. The narrow scatter of results for mechanical properties, derived from strips of rubber cut from different areas in the same sheet, indicate that a reasonably uniform distribution of the drug has been achieved. The measurement of particle size by optical microscopic examination showed that the drug particles varied in size from less than 1.0to 62.5 pm, with 95%smaller than 2.5 pm. The large agglomerates could act as stress raisers
SILICONE RUBBER FOR HYDROCEPHALUS SHUNTS
CONCENTRATION
OF GENTAMlClN SULPHATE
629
Imglgm)
Fig. 3. The ultimate extensibility of heat-vulcanizing silicone rubber (m) and room-temperature-vulcanizing silicone rubber ( 0 )as a function of the concentration of gentamicin sulphate. Bars denote one standard deviation.
causing premature failure of the silicone rubbers. The breakdown and dispersion of such agglomerates may lead to improved mechanical properties. Some preliminary activity tests carried out on the impregnated HV silicone rubber by the methods described in a previous report4 suggest that the activity of the drug is not significantly affected by the curing procedure adopted and that the required level of drug concentration may only need to be as low as 25 mg/g or less. Should this prove to be the case (work has already been undertaken to assess the required
630
VAN NOORT AND BAYSTON
therapeutic concentration for the HV rubber), then the HV rubber impregnated with 25 mg/g or gentamicin sulphate would retain a tensile strength in excess of 70% compared with the control while increasing the extensibility by only 15%. A change in properties of such a proportion may well prove to be acceptable where such applications as CSF shunts are concerned.
CONCLUSIONS This work demonstrates that it is possible to impregnate both room-temperature-vulcanizing and heat-vulcanizing silicone rubbers with gentamicin sulphate. Although changes in mechanical properties of the silicone rubbers do occur, these are not so detrimental as to render these materials unsuitable for use. Further investigations of the suitability of silicone rubbers for impregnation are being undertaken and improved methods of impregnation are being tried out in an effort to enhance the mechanical properties of the impregnated silicone rubbers. The authors gratefully acknowledge the financial support of the Science Research Council (grant no. B/RG 77229). the Association for Spina Bifida and Hydrocephalus (grant no. 4018-01) and the Trent Regional Health Authority (grant no. 659).
References 1. R. Bayston and J. Lari, “A Study of the Sources of Infection in Colonized Shunts,” Deu. Med. Child Neurol., Suppl., 32,16 (1974). 2. R. Bayston, “Antibiotic Prophylaxis in Shunt Surgery,” Deu. Med. Child Neurol., 17, (35), 99-103 (1975). 3. K. Welch, “Prevention of shunt infection,” 2. Kinderchir., to appear. 4. R. Bayston, “Preliminary Studies of the Impregnation of Silastic Elastomers with Antimicrobial Substances,” Deu. Med. Child Neurol., Suppl. 37,50 (1976). 5. R. van Noort, Ph.D. Thesis, University of Sussex, 1976. 6. L. E. Nielsen, “Simple Theory of Stress-Strain Properties of Filled Polymers,” J Appl. Polym. Sci., 10,97 (1966). 7. R. van Noort, B. Harris, and M. M. Black, “Elastomeric Materials for Prosthetic Heart Valves,” Plast. Rubber, 7, February (1977).
Received May 10,1978 Revised November 6,1978
