Spectral Transmission of the Human Crystalline Lens in Adult and Elderly Persons: Color and Total Transmission of Visible Light

Abstract

Purpose.: To experimentally measure the spectral transmission of human crystalline lenses belonging to adult and elderly persons, and to determine the color and total transmission of visible light of such crystalline lenses.

Methods.: The spectral transmission curve of 32 human crystalline lenses was measured using a PerkinElmer 800UV/VIS spectrometer. Total transmission of visible light and the chromatic coordinates of these crystalline lenses were determined from these curves for solar illumination.

Results.: The crystalline lens that filters UV and its transmission in the visible spectrum decreases with age; such a decrease is greater for short wavelengths. The total transmission of visible light decreases, especially after the age of 70 years, and the crystalline color becomes yellower and saturated.

Conclusions.: The great variability existing in the spectral transmission of the human crystalline lens is lesser between the ages of 40 and 59 years, but greater from the age of 60 and older. The decrement in transmittance between these two age groups varies from 40% for 420 nm to 18% for 580 nm. Nevertheless, it is proven that age is not the only parameter affecting crystalline transmission. In the range of 40 to 59 years, age does not bear an influence on total transmission of light, but from 60 years and older it does. Moreover, the light transmitted decreases with age. This total transmission of light is similar to or lower than the amount that the different intraocular lenses transmit, even with a yellow or orange filter. The color of the human lens becomes yellowish and saturated with age.

RESUMEN

Introduction

It is a proven fact that the human crystalline lens turns yellow with age; however, there are very few direct experimental measurements that show what this yellowing is like. The clearest way to show this process is to determine the spectral transmission of the crystalline lens. ¹ In the existing literature, references to this are usually indirect, for instance referring to the spectral transmission of ocular media regarding the formation of keratitis or cataracts, ² comparing the color of the crystalline lens with the color of specific IOLs, ³ analyzing changes in the ocular layers brought about by light, ⁴ or measuring the density of the ocular media. ^5,6 Weale ⁷ performed direct measurements of the spectral transmission of the crystalline lens, but only showed the absorbance as a function of age for seven wavelengths and fitted a logarithmic equation. Nonetheless, in most cases, when data on the spectral transmission of the ocular media are required, they are mainly obtained from the classic work by Boettner and Wolter. ¹ These authors experimentally determined the spectral transmission of the ocular media of nine eyes corresponding to the ages of 4 weeks, and 2, 4½, 23, 42, 51, 53, 63, and 75 years, and report from their results that “the number of specimens measured was not sufficient to obtain as good a statistical accuracy as one would like, especially to evaluate more accurately the effect of age on the direct light transmission of the eye.”^{1(pg. 782)}  

Moreover, the controversy as to whether yellow filters should be incorporated in IOLs ^8–10 still continues today, as it is not clear in which cases they can be beneficial. ^11–18 In the case of yellow or orange filters being incorporated in IOLs, it is argued that the crystalline lens of an elderly person is yellowish in order to protect the retina, so the intention is to imitate this yellowing. In fact, current directions for use of intraocular lenses with filters that try to imitate the aged human crystalline lens take the measures of Boettner and Wolter ¹ as a reference, specifying this explicitly in their directions for use. Nonetheless, it should be noted that, regarding aged crystalline lenses, Boettner and Wolter ¹ only show the spectral transmission of the crystalline lens of a child aged 4½-years and two subjects aged 53 and 75 years, who are consequently taken as a reference. This is why the authors believe that more measurements should be available that show what the spectral transmission of the adult and aged human crystalline lens is like. These determinations should be made using more accurate, up-to-date instruments which would, for example, avoid flattening the crystalline lens, and of course, a significant number of crystalline lenses should be determined. ¹  

The main aim of this study was to contribute experimental data on spectral transmission of human crystalline lenses from adult and elderly subjects to gain knowledge into this yellowing characteristic. With the aid of these curves, the percentage of visible light that the crystalline lenses transmit was calculated, as well as their color by calculating their chromatic coordinates. Finally, all these data for a transparent IOL, one with a yellow filter, and another with an orange filter have been calculated. These results were then compared with those obtained from human crystalline lenses. 

Methods

The lenses were mainly obtained from eyes used for corneal transplants and that, consequently, did not present any pathology. Thirty-two excised lenses in the age range of 41 to 77 years were studied, plus one 30-year-old lens. Earlier studies have shown that if the material was refrigerated, no significant change in either the image-forming properties of the lens or its transmissivity could be detected. ^19,20 In this case, the lens was removed from the eye, carefully cleaned, and immediately frozen at −80°C. Generally, 24 hours were allowed to elapse between this manual procedure and spectral measurement. This was performed immediately after defrosting, so as to minimize the effects of cold cataract, and to allow for the disappearance of effects due to occasional, minor reversible trauma resulting from even the most careful handling. Crystalline lenses without opacities were chosen. They were chosen by taking a photograph of the dry crystalline lens on a grid, and those that allowed the whole grid in the center to be seen were used. 

The transmission curves were obtained by using an 800 UV/VIS spectrometer (Perkin-Elmer Lambda; Perkin-Elmer, Shelton, CT). This apparatus can measure the spectrum from 200 nm onward, which means that spectral transmissions in UV-A, UV-B, and part of UV-C are accurately determined (precision is up to 1 nm). The air was taken as a reference to measure transmittance. The measurements correspond to the total transmission of the crystalline lens, so the apparatus was equipped with an integrating sphere. Figure 1 is a very simplified diagram (for greater clarity) of how to measure with a spectrophotometer. The source is, in fact, two lamps, one halogen and the other deuterium, which can cover both the UV and the visible lights. The apparatus uses two monochromators to select the wavelength (λ) accurately. This radiation passes through the sample (crystalline lens) that is to be measured before entering the integrating sphere. A suitable cuvette is used to place the sample directly in front and covering the complete entrance hole of the integrating sphere. Obviously, the other way for reference, as already stated, is the air (not shown here for clarity's sake). 

Figure 1. 

 

A simplified diagram showing how to measure with a spectrophotometer. The sample is placed in a cuvette directly in front and covering the complete entrance hole of the integrating sphere.

Results

The results are analyzed considering two age groups: the adult group includes crystalline lenses aged between 40 and 59 years, and the elderly group comprises those corresponding to lenses aged 60 years and older. The boundary between adult and elderly age cannot be defined exactly because it does not have the same meaning in all societies. People can be considered elderly because of certain changes in their activities or social roles. In many countries, people are considered elderly when they retire, and this age is generally 65 years. However, the authors are more interested in the age at which physical changes become more apparent, particularly a decrease in vision. This age boundary is very imprecise; thus, it is very difficult to establish a specific age for this division. For this experiment, it was taken into account that at 60 years old presbyopia is fully developed and that in one's 60s the incidence of cataract operations increases considerably. Therefore, this age was established as the point at which the authors consider a person to be elderly from the visual standpoint. 

Figure 2A shows the spectral transmission of the 14 crystalline lenses belonging to the adult group, and Figure 2B shows the spectral transmission of 18 crystalline lenses belonging to the elderly group. The authors have adapted the results of Boettner and Wolter ¹ to include them in these two figures. Nevertheless, it should be considered that the curves pertaining to 53 and 75 years of age that Boettner and Wolter¹show, measure the direct transmission; in other words, they do not add light scattering, so their value is always a little lower than if they measured the total transmission. Figure 2A also shows the transmission of a 30-year-old crystalline lens determined in this study. 

Figure 2. 

 

(A) Total spectral transmission of the 14 crystalline lenses aged between 40 and 59 years, plus that of a 30-year-old crystalline lens. (B) Total spectral transmission of the 18 crystalline lenses aged 60 years and older. The number beside each curve indicates the age of the crystalline lens. In addition, the adapted Boettner and Wolter ¹ curves are shown (indicated as B&W), and the age of the crystalline lens. Those of B&W 53 and B&W 75 refer to direct, not total, transmission.

Figure 3 makes it possible to compare the spectral transmission of collateral eyes, as it shows some pairs of crystalline lenses belonging to the same person. 

Figure 3. 

 

Transmission curves of pairs of collateral eyes. The number that appears beside each curve indicates the age of the crystalline lens.

The total amount of visible light that these crystalline lenses transmit (see Appendix) from these curves was calculated, which provides objective information on how opaque they are. This is particularly important because with age, the transmission of the crystalline lens does not only become more selective because it filters short wavelengths more than long wavelengths, but the curves also become lower. Figure 4 shows how the total percentage of visible light that the different crystalline lenses transmit in solar illumination varies. Also included are the results derived from the Boetner and Wolter ¹ curves in Figure 4, as well as the total transmission of three representative IOLs ⁸ : transparent (Alcon SA60AT; Alcon, Fort Worth, TX), yellow (Alcon Acrysof IQ SN60; Alcon), and orange (Ophtec PC440; Optech BV, Groningen, The Netherlands). There are different IOLs that incorporate very similar ⁸ filters, so the authors selected only the three IOLs that show different spectral transmissions clearly, and that are, therefore, considered most representative. The three IOLs selected incorporate a UV filter, but in one case it is transparent, in another yellow, and in the third, orange. Finally, how the color of the crystalline lenses varies by calculating their chromatic coordinates is shown. The color of the crystalline lenses was ascertained by using the chromatic coordinates calculation from the spectral transmission curves and from the characteristic curve of the illuminant D65 (solar illumination), and by applying the equations shown in the Appendix. Figure 5 shows the chromatic coordinates of some of the crystalline lenses (not all of them for clarity's sake) calculated for solar illumination, in a CIE1931 chromatic diagram. This diagram also gives the chromatic coordinates of the three IOLs. 

Figure 4. 

 

Percentage of visible light transmitted by the different crystalline lenses measured under solar illumination, fitted linearly to each age range. The total transmission of a crystalline lens of 30 years (triangle) is given, and also the Boettner and Wolter measurements (crosses). The total transmission of visible light of the three IOLs is also included: Transparent (T), Yellow (Y), and Orange (O).

Figure 5. 

 

Chromatic coordinates of nine of the crystalline lenses under solar illumination (illuminant D65). The coordinates corresponding to the three IOLs, Transparent (T), Yellow (Y), and Orange (O), are also included.

Discussion

Spectral Transmission of the Crystalline Lens

The results show a great variability in the spectral transmissions of crystalline lenses from adult and elderly persons. In general, and as can be seen in Figures 2A and 2B, the curves that represent spectral transmission decrease to a lesser or greater extent with age. They not only filter short wavelengths more, for example, the blue ones, but their maximum transmissions no longer reach 100% in many cases. They transmit, therefore, less visible light in total. However, this decay is not progressive with age, as can be seen, since there are elderly subjects who present much more transparent crystalline lenses than younger persons; moreover, this situation is quite usual. Boettner and Wolter ¹ particularly refer to the case of the lens of a 71-year-old eye that had a better direct transmittance than that of a 53-year-old eye. This is why Boettner and Wolter¹ question the suggestion that age can be determined by transmittance measurements. ²¹ The authors' results confirm that age does not, in fact, exclusively determine crystalline lens transmittance. In Figure 2A, the transmission of a 30-year-old crystalline lens, as well as the results that Boettner and Wolter ¹ show in their article for 4½-year-old and 53-year-old crystalline lenses, have also been adapted. As can be observed, the transmission of the 30-year-old lens reaches 100% in almost the whole visible spectrum, with just a slight decrease in short wavelengths. The results from Boettner and Wolter ¹ of a 75-year-old crystalline lens are included in Figure 2B. Their results coincide entirely with this experiment, allowing for the fact that, as previously stated, those of 53 and 75 years measure the direct transmission. The curve corresponding to a 4½-year-old child ¹ (total transmission) shows that this filter lets 100% of the visible spectrum through after 400 nm. It filters the UV totally, except for the window centered at 320 nm, which gradually disappears with age, although, as can be observed in many cases, it does not disappear completely. 

Moreover, it can be seen that in the adult group (Fig. 2A) dispersion of the curves is less than in the elderly group (Fig. 2B). In order to evaluate this dispersion, and to facilitate the comparison between these two groups, Table 1 shows the mean spectral transmission, the SD, and the confidence interval (CI) at 95% of confidence for five selected wavelengths and the two age ranges. The five wavelengths considered are 420, 460, 500, 540, and 580 nm, corresponding to the most representative areas of the visible spectrum. Table 1 shows that the mean values for the adult group are higher than those of the elderly group; that is, there is a decrement in mean transmittance with aging. The value of the decrement, calculated in relative terms (percentage), is 40%, 29%, 26%, 24%, and 18% for wavelengths 420, 460, 500, 540, and 580 nm, respectively. This means that the decrement in transmittance due to aging, between the two age ranges studied, linearly depends on the light wavelength value (R ² = 0.91); the shorter wavelength values (blue light) are higher. The SD for 420 nm and 460 nm is similar for both age groups, but for 500, 540, and 580 nm this SD remains similar for the adult group, but it undergoes a sharp increase in the elderly group. This result was expected, because as age advances, each individual's experience of life (exposure to the sun, diet, diseases, and so forth) differs, therefore, the crystalline lens develops differently; thus, its spectral transmission has much more variability. 

Table 1. 

 

Mean Value, SD, and Confidence Interval (CI) of Spectral Transmission

λ (nm)*	Adults†			Elderly‡
	Mean Value	SD	CI§	Mean Value	SD	CI§
420	14.0	8.9	[9.3–18.7]	8.4	5.7	[5.4–11.4]
460	39.9	15.6	[31.7–48.1]	28.3	15.0	[13.3–43.3]
500	72.0	12.5	[65.5–78.5]	53.0	20.9	[42.0–64.0]
540	89.3	6.4	[85.9–92.7]	67.9	21.1	[46.8–89.0]
580	93.8	7.9	[89.7–97.9]	77.1	18.5	[67.4–86.8]

*  Differences between both groups are significant for all wavelengths (P < 0.01). Only for the wavelength 420 nm, the significance degree is P = 0.059.

†  Adults (40–59 years old).

‡  Elderly (60–80 years old).

§  [CI, with 95% of confidence].

Figure 3 shows the spectral curves of collateral eyes. As can be observed, the eyes of one subject usually evolve, with regard to spectral transmission, in the same way, presenting practically the same curves in both eyes. We only found differences between the transmissions of the two eyes in subjects of over 70 years old, as can be seen in the cases of the 75-, 76-, and 77-year-olds. 

Total Transmission of Visible Light by the Crystalline

An aspect that is just as important to know as the crystalline lens spectral transmission curve, is knowing the whole percentage of visible light the lens can transmit. This is important, as the total amount of visible light that reaches the retina is usually what mainly allows a subject to have better or worse vision, regardless of whether the lens transmits long wavelengths more than short wavelengths. The Appendix provides the equations used to calculate the total transmission of the crystalline lenses. Figure 4 shows the percentage of transmission of visible light that these lenses let through. It can be seen that in general, as age increases, the total transmission of light decreases. That is clearly evident after the age of 60, but the considerable data dispersion may occasionally result in an increased transparency of an individual older lens with respect to a younger one (see Fig. 4). If the crystalline lenses are now grouped into the two above-mentioned age groups, it shows that for adult persons (40–59 years) the mean transmission is 88%; the maximum in our study was 96%, and the minimum 76%. In the elderly persons group (60 years and older), the mean transmission was 70%, but the dispersion was much greater, as the maximum transmission was 89%, and the minimum 21%. Moreover, this dispersion increases as age increases. Nevertheless, some of the results obtained in very old lenses might have been influenced by the existence of very mild age-related lens opalescence, which could not be detected by the photographic procedure used to assess lens transparency. 

Apart from the total transmission of the lenses, Figure 4 also shows the total transmittance of the 30-year-old lens which is 95%. In addition, it shows the data of Boettner and Wolter ¹ calculated from their transmission curves, which, as can be seen, are within the ranges obtained. Their 53-year-old lens transmitted 75% and their 75-year-old lens transmitted 31%, both measured by direct transmission, therefore, slightly undervalued regarding this experiment's determinations which refer to total transmission. The crystalline lens of the 4½-year-old child transmitted 96% of the visible light (total transmission). 

If fitting a straight line to the points in Figure 4 is attempted, it will show that the correlation is practically zero in the age interval from 40 to 59 years (R ² = 0.057). This indicates that, in this range, age has a minimal influence on the variation between the total light transmission values, as only 5.7% of such variation can be explained by the age factor. The cause for the greater or lesser transmission in this range should be found mainly in other factors, such as exposure to the sun, metabolism, diseases, and so forth. On the other hand, the fit of the straight line is better in the 60 year and older range with a correlation of R = 0.686 (R ² = 0.470), which indicates that age explains 47% of the variation in crystalline lens transmission and the rest must be ascribed to other factors. 

All these calculations have been made using solar illumination (illuminant D65). The authors have also made the calculations (not shown) with incandescent illumination (illuminant A). In this case, the percentage of light transmitted is somewhat greater than in the case of solar light. This is consistent with the fact that the lens yellows and therefore filters short wavelengths, but as incandescent light is predominantly yellow, its light is almost entirely transmitted by the yellow lens. Nevertheless, as the intensity of solar light is much greater than that of any incandescent source, the crystalline lens will always transmit more light in absolute values with solar illumination. 

Color of Crystalline Lenses

The chromatic coordinates of several (not all, for clarity's sake) crystalline lenses determined in this study are shown in Figure 5 in a CIE1931 diagram. The point labeled as D65 marks the initial position (white or transparent). This point has the coordinates of the illuminant (solar illumination) and it is to be noted that the point labeled as T, which is the point of an IOL with a transparent filter, is very close to point D65. We can see that the color of crystalline lenses develops from white to orange and approaches the spectral locus in more advanced ages, which means that the color becomes much more saturated. Nonetheless, as in the previous analyses, there are older eyes with more transparent (less yellow) crystalline lenses than other younger eyes. The IOLs with yellow or orange filters (labeled as Y and O, respectively, in Fig. 5) are in intermediate positions. Their color corresponds with that of adult or not very old human lenses. 

As in the previous section, the same calculations (not shown) were performed with incandescent illumination (illuminant A), and all the coordinates were much more grouped in the yellow area and around the illuminant A, which is also yellowish. 

IOLs and Spectral Transmission of the Crystalline Lens

Although in this study the main aim was to contribute experimental data on the spectral transmission of human crystalline lenses from adult and elderly persons, the authors believe it would be helpful to make a small comparison between these transmissions and the spectral transmissions of the three representative IOLs ⁸ selected. As previously stated, when manufacturers decide to include this filter in their IOLs, they usually say that it imitates the spectral transmission of the human crystalline lens of an advanced age. In view of the great variability of the curves shown in Figures 2A and 2B, it is difficult, if not impossible, to choose a curve that clearly represents an adult or elderly crystalline lens. However, if the curves are grouped into four age ranges (40–49, 50–59, 60–69, and 70–79 years old) instead of just two, and the mean transmission curve for each range is calculated, then the curves shown in Figure 6 will be obtained. In this figure, the authors have included the curves of the spectral transmission of the three IOLs ⁸ analyzed, and it can be deduced from this figure that the mean spectral transmission between 40 and 69 years is practically the same. This is true, but once again the great dispersion these curves present should be remembered. For example, in this age range (40–69 years), for a wavelength of 470 nm, there are transmissions that go from 36% to 75%. What can be stated, in view of Figure 6, is that after the age of 70 there is usually a sharp decrease in the spectral transmission. 

Figure 6. 

 

The curves show the spectral transmission of the crystalline lenses averaged for each to the four age ranges. The spectral transmission of the three IOLs is included.

Moreover, in the light of these curves, it seems that yellow and especially orange IOLs reproduce the spectral transmission of the average crystalline lens reasonably well for the ages between 40 and 69 years. Nonetheless, it should again be remembered that the dispersion of the curves is very great. Hence, the assertion that a specific curve represents the adult crystalline lens should be viewed with caution. 

Concerning the total transmission of light, Figure 4 includes (for some random ages) the total transmission of visible light that these three IOLs transmit. This shows that in this case their total transmission is excellent, as both the IOL with a yellow filter and the IOL with an orange filter transmit 89% of visible light and the transparent IOL transmits 99%. This is because although these (yellow and orange) IOLs filter part of the short (blue) wavelengths, they have a transmission of almost 90% and 100% for yellow and orange, respectively, in the zone of the visible spectrum they let through. 

Regarding color, the chromatic coordinates of the IOLs can be observed in Figure 5 (T, Y, and O). Our results confirm those of Romano et al. ³ as the maximum difference in color is indeed found between the transparent IOL and the older (76 years) human crystalline lens. The minimum difference is found between the IOLs that filter blue (yellow and orange IOLs) and the aged crystalline lenses. Romano et al ³ also state that these yellow and orange IOLs reproduce the human crystalline lens aged between 40 and 50 years. In view of the authors' results, the transmission of these IOLs could also represent crystalline lenses of other ages, given the great dispersion there is in their spectral transmission.  

To conclude, the spectral transmission of the human crystalline lens of adult and elderly humans is very variable, and it is impossible to establish a good correlation between age and spectral transmission because other parameters are involved in the spectral transmission variations. This variability is smaller in the adult group, ages 40 to 59 years, and greater for the elderly group, ages 60 and older. If the measurements are grouped and averaged by decades, it can be said that the averaged curve stays more or less the same until 70 years, and that after this age there is a severe decrease in spectral transmission. 

The total amount of visible light transmitted by human adult and elderly lenses is also very variable. In the age range of 40 to 59 years, the total transmission of light is practically independent of age, since it only accounts for about 6% of the variation; nonetheless, in the range of 60 years and older, age bears a greater influence on total light transmission, accounting for almost 50% of the variation. 

With age, the color of the human crystalline lens generally yellows and becomes saturated, but this yellowing is not progressive as the crystalline lenses of older persons are often less yellow than those of younger subjects. 

Supplementary Materials

pdf S1 

Footnotes

 Supported by grants from Catedra Alcon-Universitat de Valencia.

Footnotes

 Disclosure: J.M. Artigas, None; A. Felipe, None; A. Navea, None; A. Fandiño, None; C. Artigas, None

Appendix

To determine the total transmission of the visible spectrum of a crystalline lens, first the tristumulus values (X, Y, and Z) were calculated in accordance with the following formulas:  

 

 

 

 

 

where x¯(λ), y¯(λ), z¯(λ) are the color-matching functions (CMFs) of the standard observer, S(λ) is the spectral distribution of the source (in this case illuminant D65 or solar illumination) and τ(λ) is the spectral transmittance of the crystalline lens. The value of the tristumulus Y indicates the lightness or luminance in the case of a surface color and the transmittance of the filter in this case (crystalline lens). Then, the total transmission (T) in the visible spectrum was calculated following the formula:  

 

To calculate the chromaticity coordinates (x,y) in the CIE1931 chromaticity diagram we used the formula:  

 

 

 

where X, Y, and Z are the tristumulus values.