Hypoxia induced corneal swelling is a well known phenomenon and one of the primary indices of corneal physiological change during contact lens wear. Holden and Mertz1 hypothesized that the minimum oxygen transmissibility (Dk/t) of a lens should be 87±3.3×10−9 (cmmlO2)/(mlsmmHg) in order to prevent overnight lens induced corneal swelling. More recently, a value of 125×10−9 (cmmlO2)/(mlsmmHg) has been proposed as the critical Dk/t of a lens to prevent lens-induced overnight corneal anoxia.2 Studies have shown that silicone hydrogel lenses induce less corneal swelling compared to conventional hydrogel lenses when worn overnight (Bergenske P, Mueller N, Caroline P, Smythe J, Mai-Le K. Uniformity of overnight corneal swelling with extended wear contact lenses. Optom Vis Sci. 2001;78(12S):198; Bullimore MA, Nguyen M, Bozic J, Mitchell GL. Overnight corneal swelling with 7-day continuous wear of soft contact lenses. ARVO Meeting Abstracts. 2002;43:3100; Fonn D, Moezzi A, Simpson T, Situ P. Confirmation of a yoked corneal swelling response between the test and contralateral control eye. Optom Vis Sci. 2004;81(12S):30).3,4 Although all −3.00 D silicone hydrogel contact lenses meet the Holden and Mertz criterion of 87×10−9 (cmml O2)/(mlsmmHg) for the central lens transmissibility, no one has reported the effect of higher powered silicone hydrogel lenses (decreased Dk/t) on central corneal swelling. Previous corneal swelling studies with silicone hydrogel lenses including a previous study by the current authors5 used low powered silicone hydrogel lenses and compared between the lens types. Although Steffen et al.6 studied overnight swelling with silicone hydrogel lenses in a range of powers between −1.00 to −6.00 D to correct 25 adapted daily soft contact lens wearers it is unclear how many subjects wore higher powered lenses in this dispensing study and they did not compare corneal swelling across lens powers.

Previous studies with conventional hydrogel lenses showed greater central corneal swelling with higher minus lens powers than lower minus powers with the same material, central thickness and central oxygen transmissibility (Dk/t).7–9 Tomlinson and Bibby10 showed that the central corneal swelling in minus powered hydrogel lenses was underestimated and in plus powered lenses was overestimated based on the central lens transmissibility. These findings led to the conclusion that the central lens transmissibility is a poor predictor of the magnitude of central corneal swelling and the response is influenced by the averaging of the lens oxygen transmission.7,10–14 To our knowledge no one appears to have investigated the influence of local central compared to average Dk/t of silicone hydrogel lenses to determine the primary driver of overnight central corneal swelling with these lenses.

The main aims of this study were to compare differences in central corneal swelling between different silicone hydrogel lens materials in high and low powered lenses and to determine if at high levels of oxygen transmissibility central corneal swelling with silicone hydrogel lenses can still be differentiated. In addition, this study aimed to investigate whether central corneal swelling is primarily driven by central or average lens transmissibility. Therefore, we compared central overnight corneal swelling induced by four different silicone hydrogel lenses with three different powers and tested the following null hypotheses:

• 1)

  There are no statistically significant differences in central corneal swelling between the lens types for each lens power.

• 2)

  There are no statistically significant differences in overall central corneal swelling between the 4 lens types.

• 3)

  There are no statistically significant differences in overall central corneal swelling between the 3 lens powers.

• 4)

  Average oxygen lens transmissibility is not the main driver of central corneal swelling in silicone hydrogel lenses.

Distribution of central corneal swelling across subjects from this study has been published15 showing that both the lens-wearing and control eyes followed a normal curve. This validates the use of parametric statistics for data analysis and the use of mean values to represent the central tendency of the data in this paper.

Descriptive statistics were generated for all variables. The effects of lens type and lens power were examined. p-Values of less than 0.05 were considered to be statistically significant. Repeated measures analysis of variance (Re-ANOVA) was used to examine the effect of lens type (lotrafilcon A, senofilcon A, balafilcon A and galyfilcon A) and lens power (−3.00, −10.00 and +6.00 D). When appropriate the Huynh–Feldt (HF) correction was applied to adjust the p values and the HF corrected p values are reported in this paper. For each lens power, to compare the effect of lens type, a separate Re-ANOVA was conducted. Tukey HSD Post hoc tests were used to determine the significance of all pair-wise differences.

The results of the central corneal swelling for all lens types and powers are shown in Table 5.

Differences in central corneal swelling between the lens types for each lens power

Overnight central corneal swelling for each lens type with each lens power is shown in Fig. 1. For each lens power in Fig. 1, there was a statistically significant effect of lens type (Re-ANOVA, pHF<0.05 for all). The +6.00 D galyfilcon A lens induced greater corneal swelling than lotrafilcon A and senofilcon A (post hoc tests; p<0.05 for both), but was not different than balafilcon A (post hoc test; p>0.05). The −10.00 D lens also induced greater swelling with the galyfilcon A than lotrafilcon A (post hoc test; p<0.05), but not than the other two lenses (post hoc tests; p>0.05 for both). With −3.00 D lens, swelling induced by balafilcon A was greater than lotrafilcon A (post hoc test; p<0.05), but was not different than the other two lenses (post hoc tests; p>0.05 for both).

Figure 1.

Overnight central corneal swelling for each lens power (vertical bars denote 0.95 confidence intervals, the significantly different pairs are indicated by brackets and asterisks).

(0.07MB).

Effect of lens type on central corneal swelling

Averaged for power, there was a significant effect of lens type (Re-ANOVA, pH-F<0.001) lotrafilcon A induced the least (6.2±2.8%) and galyfilcon A the most central corneal swelling (7.6±3.0%) (post hoc tests; p<0.05). There was no difference between galyfilcon A, balafilcon A and senofilcon A, and between lotrafilcon A and senofilcon A (post hoc tests; p>0.05). Immediately after lens removal, all lenses induced significantly more central corneal swelling than their respective controls (all post hoc tests; p<0.05) (Fig. 2).

Figure 2.

Overnight central corneal swelling (lens vs. control) (vertical bars denote 0.95 confidence intervals).

(0.06MB).

Effect of lens power on central corneal swelling

There was a significant effect of lens power on central corneal swelling (Re-ANOVA; pH-F<0.001) as illustrated in Fig. 3. The +6.00 D power, averaged across lens materials, induced significantly greater central swelling than the −10.00 and −3.00 D (post hoc tests; p<0.05 for both), (7.7±2.9 vs. 6.8±2.8 and 6.5±2.5% respectively) but there was no difference between −10.00 and −3.00 D (post hoc test; p>0.05) (Fig. 3).

Figure 3.

Overnight central corneal swelling by lens power (lens vs. control) (vertical bars denote 0.95 confidence intervals).

(0.06MB).

In this study we used 12 different silicone hydrogel lenses with central oxygen transmissibility values ranging from 31 to 211×10−9 (cmml O2)/(mlsmmHg). Central corneal swelling differences between the lens types were particularly pronounced in high powered lenses and are clearly seen in both +6.00 D and −10.00 D lens powers in Fig. 1, suggesting that the corneal swelling response especially in the higher lens powers may be minimized by using silicone hydrogel lenses in the higher transmission end. These results are consistent with the difference in the calculated central oxygen transmission values between the silicone hydrogel lenses shown in Table 3 and show that at high levels of Dk/t, central corneal swelling with silicone hydrogel lenses can still be differentiated based on the lens oxygen transmissibility. The exception to this statement was galyfilcon A which did not induce the most swelling among lenses in −3.00 D power (Fig. 1) as would be predicted from the other lens power results in this study and other published work (Moezzi AM, Fonn D, Simpson TL. Comparison of overnight corneal swelling induced by prototype lotrafilcon A toric versus balafilcon A toric. American Academy of Optometry Meeting Abstract. 2006:060095).17,18 This can be attributed to sampling or perhaps other possible uncontrolled lens specific or ocular surface related factors which may have influenced the corneal swelling response in addition to lens Dk/t warranting further investigation in future studies. The small but significant differences in central corneal swelling between the study lenses in each high powered group (i.e. −10.00 or +6.00) (Fig. 1) are probably not clinically relevant. However, some subjects exhibited high levels of corneal swelling as we reported previously15 and for those it would seem sensible to use the lenses that cause the least amount of swelling.

Fig. 2 demonstrates an inverse relationship between oxygen transmissibility and mean overnight central corneal swelling induced by the silicone hydrogel study lenses. Also, this figure illustrates that with all study lenses, the lens wearing eye showed significantly more swelling than in the contralateral non-lens wearing eye as shown previously,4,5 (Fonn D, Moezzi A, Simpson T, Situ P. Confirmation of a yoked corneal swelling response between the test and contralateral control eye. Optom Vis Sci. 2004;81(12S):30) suggesting that even lenses that exceed the Holden and Mertz 87×10−9 (cmml O2)/(mlsmmHg)1 will not avoid overnight lens induced edema (Table 5). The additional swelling produced by the lens compared to the same closed eye condition without the lens has been partially attributed to lens-related corneal swelling factors other than hypoxia.19–25

Despite differences in thickness of the central and peripheral cornea there are similar oxygen demands across the cornea, independent of corneal location.26 Even after blinking, the effect of tear mixing to equilibrate the oxygen tension under a soft lens is insignificant.27 To maintain normal corneal physiology and health, it is important that high powered soft contact lenses provide sufficient local oxygen transmissibility through the thickest part of the lens.2,28–30 Bruce29 compared local Dk/t measurements of spherical lotrafilcon A and balafilcon A lenses of various powers and found that lotrafilcon A exceeded the Holden and Mertz criterion of 87×10−9(cmml O2)/(mlsmmHg) at all lens locations in the range of +3.00 to −6.00 powers, however only balafilcon A of +1.00 D power met this criterion across the lens. In our study none of the plus powered lenses met the minimum Dk/t of 87×10−9 (cmml O2)/(mlsmmHg) at the center as the highest central Dk/t was 70×10−9 (cmml O2)/(mlsmmHg) with +6.00 D lotrafilcon A (Table 3). All minus powered lenses in this study meet or exceed the Holden and Mertz criterion for extended wear at the center (Table 3).

Previous studies with conventional hydrogel lenses under closed eye conditions showed greater central corneal swelling with higher minus lens power but the same center thickness as the lower power lenses, and similar levels of central corneal swelling with minus compared to plus hydrogel lenses. This was despite greater central thickness and therefore lower central Dk/t of plus lens powers. These findings were explained by assuming that average of the central area of the lens instead of local central lens transmissibility was responsible.7,10–14 However, findings from these previous studies with hydrogel lenses are at odds with results from our study with silicone hydrogel lenses (Fig. 3).

For the ease of illustration, the mean values of central and average lens transmissibility for each lens power are shown in the right side of the last rows in Tables 3 and 4 respectively. Central corneal swelling with +6.00 D was significantly greater than −10.00 D lens power (Fig. 3) despite similar levels of average lens transmissibility of these two lens powers (Table 4, last row). However, central oxygen transmissibility in +6.00 D was significantly lower than −10.00 D lens power as shown in Table 3 and this is in line with the higher central corneal swelling induced by +6.00 D lenses in this study. Therefore, the greater central swelling induced by +6.00 D compared to the −10.00 D and −3.00 D in Fig. 3 can be explained by the lower central oxygen transmission of the plus lens power. Also central corneal swelling induced by −10.00 and −3.00 D lens powers in Fig. 3 were not significantly different despite obvious differences in average lens transmissibility between high and low minus lens powers (Table 4, last row). In contrast, the similar level of central corneal swelling induced by −10.00 D and −3.00 D lens powers (Fig. 3) can be easily predicted from the similarity in central lens transmissibility between these two lens powers as shown in Table 3 (last row). A correlation analysis (that was not statistically significant as only three pairs of data were compared) showed a very strong linear association between central corneal swelling and central lens Dk/t (r2=0.94) and a weaker association between the central corneal swelling and average lens transmissibility (r2=0.68). Therefore, these findings from our study suggest that in silicone hydrogel lenses central corneal swelling is mainly driven by central lens oxygen transmissibility. Average oxygen transmissibility is less likely to primarily affect the outcome here and a likely reason was that these lenses were in the high transmissibility range. These results are in agreement with a suggestion from at least one previous study which predicted a more prominent role for the effect of local oxygen transmissibility, rather than the averaging effect in higher transmissible hydrogel lenses.14

This study was supported by a grant from CIBA Vision USA, Duluth, GA, and in part by a COETF grant.

Professor Desmond Fonn is a consultant for Alcon Laboratories and CooperVision Inc. None of the other authors of this manuscript have any potential financial interest in a product reported in the manuscript.