Acritical driving force underpinning the development of silicone hydrogel contact lenses has been their extremely high oxygen performance, whereby hypoxia-related problems have essentially been obviated.1 This superior oxygen performance is due to the incorporation of silicone into the lens material, so that silicone rather than water becomes the main vehicle for the passage of oxygen through the lens.2 Although the manufacturers of silicone hydrogel contact lenses have published values of oxygen permeability (Dk) relating to their products, the precise methodology used in determining these values is not readily available. It is therefore important that, from time to time, independent estimates of the Dk of commercially available products are presented, using a single technique with clear reporting of the precise methodologies adopted.
Three methods have been described for the measurement of contact lens Dk.3 The polarographic (or Fatt) method4 involves placing a contact lens on a polarographic oxygen sensor and measuring the rate at which oxygen passes through the material from the atmosphere to the sensor electrodes. In the coulometric technique,5 an oxygen-free carrier gas passes over one side of the lens and transfers oxygen that has passed through it to an electrolytic fuel cell where a quantitative decomposition occurs. The gas-to-gas (or volumetric) method6 involves mounting a lens material specimen between two fixed-volume chambers, one of which is pressurized with 100% oxygen up to several atmospheres greater than the second chamber, and oxygen passing through the lens in response to this pressure gradient is detected as a pressure increase in the second chamber. Most Dk data reported in the ophthalmic literature has been conducted using the polarographic or coulometric techniques.
As explained by Young and Benjamin,7 the current international standard for Dk measurement (ISO 9913-14) stipulates that the polarographic methodology can only be used for lens materials with a permeability of less than about 100 × 10−11(cm2/s) (ml O2/ml × mm Hg).a The reason for this 100 limit is not given in the international standard, but presumably relates to the notion that the boundary layer correction required for the polarographic methodology becomes nonlinear with high Dk values.7
We believe that the polarographic methodology has utility in measuring the oxygen performance of “hyperpermeable” soft contact lens materials, such as those used to fabricate silicone hydrogel lenses, and think that this can be achieved by using a modified version of the methodology described in the international standard ISO 9913-1.4 The aims of this work, therefore, are to (a) measure and report the Dk and water content of new-generation silicone hydrogel contact lens materials, and (b) suggest a modified version of the methodology described in the international standard ISO 9913-14 for measuring Dk so that it is applicable to values in excess of 100.
Five silicone hydrogel contact lenses on the market at the time of writing, and two conventional hydrogel lenses, were evaluated; these are presented in Table 1 in descending rank order of manufacturer claimed Dk. The conventional lenses were used for reference purposes, in that the Dk of these materials measured using polarography has been well documented.8,9 In essence, this constituted a control procedure. All lenses were −1.00 D in power (such lenses are approximately parallel-sided10), and were obtained through normal commercial channels without reference to the fact that they were to be used for Dk measurement.
In general terms, we adhered to the procedures for measuring Dk as stipulated in ISO 9913-1.4 Before Dk measurement, lenses were removed from their blister packs, placed in glass vials containing phosphate buffered saline solution, and left overnight in a thermostatically controlled water bath at 35 ± 0.5°C. A single lens was placed on a polarographic oxygen sensor (Rehder Development Company, California) comprising a pair of electrodes (gold cathode and silver anode) and a solid state temperature sensor. This assembly was housed in a chamber maintained at 35°C and was connected to an external polarographic amplifier. The electric current passing between the two electrodes was monitored on a digital display on the amplifier unit. The steady-state current was recorded once the current reading had stabilized, which was typically within 20 min of placing the lens on the polarographic oxygen sensor.
This process was repeated using separate stacks of two, three, four, five, and six lenses. Different lenses were used to create each stack, so 21 lenses were used for each lens type. This stacking technique has the advantage of allowing finished lenses to be used (rather than using specially-manufactured lenses) without influencing measurement values.10 In each case, the thickness of the single lens and that of each of the stacks of two, three, four, five, and six lenses was measured using an electromechanical gauge (Rehder Development Company, California).
The entire procedure described earlier was conducted a second time, after allowing sufficient time for the lenses to fully rehydrate. If the results of the two procedures generated data that was generally concordant, the resultant current for each stack was taken as the average of the two determinations. If data from the two determinations was discordant, a third procedure was undertaken (again after allowing the lenses to fully rehydrate) and the average of the two data sets with the closest matching currents was taken to be the resultant current.
Typically, to acquire a full dataset for each lens type using the procedure described earlier took about 5 working days. The order in which the lens types were measured was randomized and the investigator undertaking these procedures was masked with respect to the lens type under evaluation. Masking was achieved by removing the lenses from the blister packs in which they were supplied and placing them in coded glass vials containing 0.9% phosphate buffered saline.
The oxygen transmissibility for each lens stack was calculated by multiplying the mean of the current in microamperes by 2.854 × 10−9. This value is based on Faraday's constant, the partial pressure of oxygen in the atmosphere and the surface area of the electrode used in this work. A correction was made for the edge effect by using the following formula:
where t is lens center thickness in millimetres and D is the diameter of the cathode in millimetres4 (it was 4 mm in this work).
The values for the inverse of the calculated oxygen transmissibility (resistance) were plotted graphically against stack thickness; the inverse of the gradient of the line of best fit through the data points for this graph represents the Dk of the lens material. This process eliminates error due to the boundary effect.4 For each lens type, linear regression analysis was undertaken to determine goodness of fit.
The water content of all lenses was measured gravimetrically, using the procedure stipulated in international standard ISO 10339.11 In essence, each lens in its fully hydrated state was weighed on an analytical balance. Before weighing, excess water on the surface of the lens was removed by placing it on a clean dry lint-free cloth, folding the cloth over the lens, and lightly pressing three times with a fingertip. The lens was dried in an oven, and then re-weighed. The difference in mass before and after oven drying represents the mass of water present in the fully hydrated lens. The procedure was repeated using 10 new lenses for six of the seven lens types evaluated (measurements were not completed for 1.Day Acuvue), with fully hydrated lenses weighed at both room temperature (22 ± 1°C) and after overnight storage in a water bath at 35 ± 0.5°C.
The relation between the inverse of the calculated transmissibility (resistance) vs. thickness for the five silicone hydrogel lenses is plotted in Fig. 1. The goodness of fit to a linear relationship is evident by inspection; this is verified by the fact that all r2 values for these lenses were 0.98 or greater. The center thickness values of each lens and lens stack for all lenses measured are tabulated in the Appendix which is available online at www.optvissci.com.
The Dk values measured in this study compared with those claimed by the manufacturers are given in Table 2. The claimed Dk values fall within the 95% confidence interval of our measured values for both reference lens types –1.DAY ACUVUE and Seequence. Our measured values for these reference lens types are also in good agreement with previously published measures.8,9
The claimed Dk value for Acuvue Oasys also fell within the 95% confidence interval of our measured value. Focus Night & Day and Acuvue Advance had claimed values of Dk below the lower 95% confidence limit of our measured value, and O2 Optix and PureVision had claimed values of Dk above the upper 95% confidence limits of our measured values.
Table 2 displays the water content values measured in this study compared with those claimed by the manufacturers. It is clear from this table that there is very good agreement between our water content values measured at room temperature vs. those claimed by the manufacturers, the only discrepancy being the claimed value for Acuvue Oasys (38%) falling just outside the 95% confidence intervals of the mean value of our measurement (36.9 ± 1.0%).
Figure 2 plots Dk vs. water content at 35°C for the silicone hydrogel lenses investigated in this study. Also shown in this figure is data relating to hydrogel lenses, obtained from one of our previous Dk studies.8 It is evident from Fig. 2 that, for silicone hydrogel lenses, a general relationship between Dk and water content can be inferred, whereby increasing water content is associated with a decrease in Dk. This contrasts with the situation with respect to hydrogel lenses, whereby there is a highly correlated (r2 = 0.96) positive exponential relationship between water content and Dk.8
The Dk values we obtained for our two reference hydrogel materials—1.DAY ACUVUE and Seequence—were in precise agreement with previously-published values,8,9 thereby validating the methodology employed in this work. This was to be expected because, as described in the Methods, we have followed the general protocol outlined in ISO 9913 to 14 for the measurement of contact lens Dk.
The Dk value for Focus Night & Day of 140 claimed by the manufacturer is in agreement with the value of 141 determined by Compan et al.,12 using the polarographic technique. The reason why these values are substantially lower than our Dk value of 162 for Focus Night & Day is unclear; however, our higher value is in agreement with other estimates of the Dk of Focus Night & Day published by employees of CIBA Vision,13,14 the manufacturer of this lens. Alvord et al.13 reported values of 140, 150, and 170; these differences related to the various techniques employed as part of their coulometric methodology. Morgan et al.14 reported values of 155 and 169 when applying coulometric and polarographic methodologies, respectively. Young and Benjamin7 made two determinations and obtained values of 176 and 190 using the polarographic technique. The evidence of ourselves and other workers7,13,14 therefore seems to suggest that the Dk value claimed by CIBA Vision underestimates the true Dk of Focus Night & Day.
The value of 91 claimed by Bausch & Lomb for the Dk of PureVision is somewhat higher than our measured value of 76; again, the reason for this is unclear. Conversely, the manufacturer-claimed Dk value of 91 is somewhat lower than the values of 102 and 111 reported by Young and Benjamin7 and 107 reported by Compan et al.12 for PureVision. That is, Bausch & Lomb's claimed Dk of 91 is about “half-way” between our estimates and those of Young and Benjamin7 and Compan et al.12 At the time of writing, independent estimates of the Dk of O2 Optix, Acuvue Oasys and Acuvue Advance were not available in the refereed literature.
The general inverse relationship between Dk vs. water content for silicone hydrogel lenses is to be expected because for this category of material, water is the limiting factor in oxygen permeation.3 That is, the higher the water content, the greater the difficulty for oxygen to move through the material. The silicone component has an extremely high permeability to oxygen, so lowering the water content of a silicone hydrogel material allows the silicone component to become the primary factor governing oxygen permeation,2 resulting in high Dk values.
The tight relationship between Dk and water content for hydrogel lenses can be attributed to the fact that all hydrogel lenses can be considered, from a material chemistry standpoint, as a single “family” of polymers, albeit with different chemical compositions.2 Conversely, silicone hydrogels can be fabricated as different “families” of polymers,2 with each family employing different structures to enhance oxygen transport. The ramification of this is that, although there will always be an inverse relation between Dk and water content for silicone hydrogel materials, the precise nature of this relation can differ between different silicone hydrogel polymer families. Thus, for example, silicone hydrogel lenses manufactured by different companies might belong to different polymer families. This may explain why the relationship between Dk and water content for silicone hydrogel lenses is not as tight as that for hydrogel lenses, as is evident in Fig. 2.
International standard ISO 9913-14 stipulates that Dk can be determined by measuring the current in a polarographic oxygen sensor when lens samples of various thickness are placed upon the sensor, as long as (a) the test samples have parallel or near parallel anterior and posterior surfaces, (b) the thickest lens sample does not exceed 0.40 mm, and (c) the resultant Dk is <100. Previous authors have found it necessary to adapt the polarographic technique in different ways to measure the Dk of silicone hydrogel lenses, and in all cases ISO 9913-1 has been technically violated. For example, Young and Benjamin7 and Compan et al.12 used powered lenses of nonuniform thickness. Morgan et al.14 described 4 exceptions to the procedures described in ISO 9913-1 that they implemented when measuring Dk, and suggested that their modified technique could be used to measure soft lenses with Dk values up to 150. These authors criticized ISO 9913-1, stating that it “… did not contain sufficient, appropriate detail to unambiguously implement … [the procedures described therein].”14
In the present study, instead of using lens samples of different thickness as prescribed by ISO 9913-1, we used stacks of 1 to 6 parallel-sided lenses, all −1.00 DS, according to the methodology described by Weissman and Fatt.10 Note 7 of ISO 9913-1 states: “The near parallel condition would correspond to dioptric powers within the range +0.50 to −0.50”; however, Weissman and Fatt10 have demonstrated that the best approximation to parallel-sided soft lenses is achieved by using lenses of −1.00 D.
The thickest stack of 6 silicone hydrogel lenses reached 0.53 mm for Focus Night & Day. Also, the resultant Dk values exceeded 100 for two of the five silicone hydrogel lenses tested. Notwithstanding these technical violations of ISO 9913-1, we have demonstrated that using stacks of up to six lenses, when applying the polarographic methodology to the measurement of the Dk of silicone hydrogel contact lens materials, results in robust data acquisition. Indeed, when plotting resistance (t/Dk) vs. stack thickness (t), an r2 value of >0.98 was obtained for all five silicone hydrogel lenses tested, and the 95% confidence limits of all Dk estimates were within 8 to 20% of the mean.
This stacking methodology has been previously validated by Weissman and Fatt10; however, a possible disadvantage of the stacking technique is the potential for additional fine layers of saline to form between the lenses in the stack, cumulatively adding to the resistance to oxygen flow as the stack become larger, resulting in a potential underestimation of the true Dk value. As is clear from the online Appendix, as the lens stack increases in thickness, the actual thickness of the stack tends to be greater than that which would be expected based upon the sum of the individual lens thicknesses. This discrepancy, which occurred for all lens types assessed in this work, is probably due to nonperfect alignment of the lens surfaces, and any gaps that form between stacked lenses can be assumed to be filled with saline.
It should be noted, nevertheless, that the Dk values reported here for the two reference materials (1.Day Acuvue and Seequence)—also measured using the stacking technique and subject to the same potentially confounding effects as described earlier— yielded virtually identical Dk values with those that have been reported extensively in the literature8,9 and claimed by the manufacturers. Further research is required to develop a paradigm for correcting for any errors induced by additional layers of saline within lens stacks should they be significant, and we shall be addressing this issue in a subsequent article.
We conclude that the polarographic technique can be used for measuring the Dk of silicone hydrogel lenses, with suitable modifications to the procedures laid out in ISO 9913-1 and with due consideration to the potentially confounding influences of additional resistance due to fine layers of saline in lens stacks, as we have described. We join with Young and Benjamin7 and Morgan et al.14 in calling for ISO 9913 to 1 to be updated so as to accommodate the measurement of contact lenses with Dk values in excess of 100. Specifically, we think that a revised standard should sanction (a) the use of the lens stacking technique as validated in this and previous studies,8,10 and (b) the use of data obtained from the measurement of lens stack thicknesses >0.40 mm.
At the time of conducting this study, Nathan Efron, Philip Morgan, Ian Cameron and Marie Goodwin worked for Eurolens Research, and Noel Brennan worked for Brennan Consultants. Eurolens Research and Brennan Consultants are contact lens consultancy units, both of which have a direct commercial association with Bausch & Lomb and Johnson & Johnson Vision Care. This association relates to the clinical evaluation of near-market or commercially available contact lenses, including those referred to in the article. In addition, Eurolens Research receives untied research funding from Bausch & Lomb, CIBA Vision and Johnson & Johnson Vision Care.
Institute of Health and Biomedical Innovation, and School of Optometry
Queensland University of Technology
Kelvin Grove, Queensland 4059, Australia