Acuity Recovery and Cone Pigment Regeneration after a Bleach in Patients with Retinitis Pigmentosa and Rhodopsin Mutations

Abstract

purpose. To assess visual acuity recovery times and cone photopigment regeneration kinetics after a bleach in the fovea of patients with dominant retinitis pigmentosa due to rhodopsin mutations.

methods. The authors measured acuity recovery times by computerized photostress testing in 13 patients with dominant retinitis pigmentosa and one of eight rhodopsin mutations. The authors also measured their time constants of cone photopigment regeneration with a video imaging fundus reflectometer to determine whether acuity recovery time depended on pigment regeneration kinetics. These values were compared with those of normal subjects, by the Mann–Whitney U test. The relationship between acuity recovery time and the time constant of cone photopigment regeneration among the patients was quantified by the Spearman rank correlation.

results. The visual acuity recovery times, which averaged 22.0 seconds for the patients with retinitis pigmentosa and 11.2 seconds for the normal subjects, were significantly slower for the patient group (P < 0.001). The time constants of cone pigment regeneration, which averaged 172 seconds for the patients with retinitis pigmentosa and 118 seconds for the normal subjects, also were significantly slower for the patient group (P = 0.043). The authors also found a significant, positive correlation between the visual acuity recovery time and the time constant of pigment regeneration for the patients with retinitis pigmentosa (r = 0.65, P = 0.017).

conclusions. A slowing of foveal visual acuity recovery and cone pigment regeneration, which are related to each other, can occur in patients with retinitis pigmentosa, due to a rod-specific gene defect.

Patients with retinitis pigmentosa commonly report slowed daytime visual recovery after exposure to high levels of illumination. Clinical testing has demonstrated slow visual acuity recovery in patients with retinitis pigmentosa of unspecified genetic types. ¹ Such delayed recovery may be due to slow foveal cone pigment regeneration, which has been found in a patient with dominant retinitis pigmentosa due to an unknown mutation, ² and in patients with autosomal recessive retinitis pigmentosa, ³ X-linked retinitis pigmentosa, ³ isolate retinitis pigmentosa, ³ and Usher syndrome (type 1). ⁴  

An estimated 10% of patients with retinitis pigmentosa in the United States have rhodopsin mutations as the cause of their disease. ^{5 6} Their decline in rod function is followed by a loss of cone function by a presently unknown mechanism. ⁷ Because some of these patients report impaired daytime visual recovery after exposure to bright lights, it is possible that slowed cone photopigment regeneration may be implicated in their secondary cone degeneration. 

The purposes of the present study were to determine whether patients with retinitis pigmentosa due to a rhodopsin mutation show slow visual acuity recovery after a bleach and whether their recovery times reflect their cone photopigment kinetics. 

Methods

Subjects

Thirteen patients (7 men and 6 women; age, 39.7 ± 3.1 years; mean ± SE) with dominantly inherited retinitis pigmentosa due to a rhodopsin mutation were tested. Table 1 indicates that each of these patients had 1 of 8 rhodopsin mutations and that 6 patients had the Pro23His mutation. Their corrected visual acuities ranged from 20/20 to 20/30, and their central dark-adapted rod thresholds to an 11° test light in the Goldmann–Weekers dark adaptometer ranged from borderline to 2.5 log-units above normal. We excluded patients by direct ophthalmoscopy who had cystoid macular edema because of the positive associations between macular edema and delayed visual acuity recovery after a bleach ⁸ and between macular edema and delayed foveal cone pigment regeneration. ⁹ Also, we excluded patients with cataracts, to minimize light scatter in our fundus reflectometry measurements. Fifteen normal subjects (9 men and 6 women; age, 39.5 ± 4.1 years; mean ± SE) served as controls for the measurements of visual acuity recovery time. Eleven normal subjects (6 men and 5 women; age, 36.9 ± 1.9 years; mean ± SE) served as controls for the measurements of cone pigment regeneration. The differences in mean age between each of these two normal groups and the patient group were not statistically significant by t-test (P = 0.97 and 0.48, respectively). Before inclusion in this study, all subjects signed informed consent approved by the Investigational Review Boards of the Massachusetts Eye and Ear Infirmary and Harvard Medical School. The methods described herein also adhered to the principles of the Declaration of Helsinki. 

Visual Acuity Recovery Time Measurements

Visual acuity recovery time was measured in the fovea by photostress testing ¹⁰ after pupillary dilation. The tested eye was the eye with better visual acuity or, if visual acuities were equal bilaterally, the right eye. Patients first attempted to identify letters of progressively smaller size flashed in random alphabetical order on a computer screen. If a letter was correctly named, the next presentation was a letter that was 0.1 log-unit (21%) smaller; if there were two misses at a given letter size, the previous (larger) size became the minimum angle of resolution. Since any of the 26 letters of the alphabet could be presented, the probability of a correct guess by chance for each trial was less than 4%. 

The 9° diameter white light of a Welch Allyn direct ophthalmoscope was then focused onto the fovea for 10 seconds at its standard retinal illuminance of 7.0 log-td to produce a cone pigment bleach of∼ 99%. ¹¹ Immediately afterward, patients tried to identify letters flashed in random order at twice the minimum angle of resolution. The recovery time was the time between the offset of the bleach and the first correct letter identification. 

Cone Pigment Regeneration Measurements

Cone pigment regeneration was recorded with a video imaging fundus reflectometer, based on a modified Kowa fundus camera; the eye evaluated was the same as that tested for visual acuity recovery. Light from an external halogen lamp, attenuated by an infrared blocking filter and conducted by fiberoptic, was the source for a 15° diameter, yellow-green test field (λ_max = 560 nm; 10 nm half bandwidth) of maximally 5.1 log-td and an 8° diameter yellow bleaching field (λ50% cut-on = 540 nm) of 6.5 log-td. At its maximal retinal illuminance the test field, exposed for 250 msec, would be expected to bleach only ∼1% of the cone visual pigments. ¹¹ The yellow bleaching field, which was presented for 2 minutes, would be expected to bleach ∼99% of the middle- and long-wavelength sensitive cone visual pigments and to a lesser extent rhodopsin and the short-wavelength sensitive cone visual pigment and photoproducts. These fields were centered in the dilated pupil as a 6-mm inner diameter, 7-mm outer diameter annulus in Maxwellian-view. The fundus camera also provided a blue light-emitting diode for fixation in Newtonian-view. 

After 15 minutes of dark adaptation, the stimulus was presented at retinal illuminances between 4.8 and 5.1 log-td in steps of 0.1 log-unit to obtain a series of baseline images for calibration of fundus reflectance. Immediately after the bleach, the stimulus was presented at 4.8 log-td, at increasing intervals for up to 15 minutes, to monitor cone pigment regeneration. After a 10-minute rest period in darkness to prevent cone pigment bleaching, the sequence of calibration, bleach, and testing was then repeated to increase the likelihood of acquiring reliable data (see below). All patients and normal subjects showed full cone pigment regeneration by the end of each sequence, based on the analytical methods described below. 

The fundus was imaged onto the screen of a high-resolution integrating CCD camera (Xybion, Gen-I). The output of this camera was connected to a computer containing a frame-grabber card, frame storage buffer card, and an 8-bit gray scale/color video card. Image frames were analyzed by software (NIH Image 1.49) that converted field brightness to 256 gray levels. Image brightness was evaluated for the central 2° (rod free area) of the fovea to quantify cone pigment regeneration. The differences in pigment double density for each successive postbleach image relative to the initial postbleach image were calculated to obtain the proportion of unbleached pigment as a function of time. An exponential model was used to derive the time constant in seconds (i.e., the time to regenerate ∼63% of visual pigment). We also evaluated fundus image brightness as a function of time for a parafoveal area outside the 8° bleach, to determine whether fundus reflectance was affected by systematic changes in eye position during the recovery phase; if there was a significant trend versus time, then the corresponding foveal pigment regeneration series was discarded. If neither the test nor retest parafoveal series showed a significant trend versus time, then we selected the foveal series with the better fit to the exponential model for comparative statistics. 

Testing of normal subjects revealed a mean time constant for foveal cone pigment regeneration of 118 seconds (see Results section), within the range of mean normative values reported by others, 98 to 120 seconds. ^{3 4 12 13} On the other hand, our mean normative value for double density was 0.19 log-unit, below the range reported by others, 0.21 to 0.30 log-unit. ^{3 4 12 14} This presumably occurred because our annular test beam passed through the margin of the pupil, which results in a lower double density. ¹⁵ Because some patients with retinitis pigmentosa and good visual acuity have foveal cone photoreceptors with reduced directional sensitivity, as measured by the Stiles-Crawford effect, ¹⁶ it is possible that the values for our normal subjects, but not those for some of our patients, would have been higher for a fundus reflectometer with a beam entering the center of the pupil. 

Statistical Analyses

The visual acuity recovery time, the time constant of foveal cone pigment regeneration, and the foveal cone pigment double density were each compared by diagnosis (retinitis pigmentosa versus normal) using the Mann–Whitney U test, a nonparametric test based on ranks that does not require a normal distribution of the data. For the patients with retinitis pigmentosa, we calculated the correlation between their time constant of foveal cone pigment regeneration and their pigment double density to see whether the former was a reflection of their stage of cone degeneration in the fovea. We also calculated the correlation between their visual acuity recovery time and their time constant of foveal cone pigment regeneration to see whether variation in the psychophysical measure could be accounted for by variation in the physiological measure. These two relationships were quantified by the Spearman rank correlation, which does not require the assumption of a bivariate normal distribution. Calculation of pigment regeneration time constants and the above statistical analyses were done with JMP (version 3.2; SAS Institute, Cary, NC) on a Macintosh computer (Apple Computers, Cupertino, CA). 

Results

Table 1 lists the visual acuity recovery times for the 13 patients with retinitis pigmentosa. Figure 1 plots these recovery times, which averaged 22.0 seconds, against those for the normal subjects, which averaged 11.2 seconds. The values for the patient group are significantly slower than those for the normal group by the Mann–Whitney U test (P < 0.001). Eight of the 13 patients (62%), representing 7 different mutations, had recovery times above the normal range. 

Figure 2 shows foveal cone pigment regeneration data for a normal subject (top) and for a representative patient with retinitis pigmentosa (bottom). Each graph shows the best-fitting exponential function from which a time constant was derived. For these examples, the time constants are 125 and 189 seconds, respectively. 

Table 1 lists the foveal cone pigment regeneration time constants for the patients with retinitis pigmentosa. These time constants, which averaged 172 seconds, and those for the normal subjects, which averaged 118 seconds, are plotted in Figure 3 A. The values for the patient group are significantly slower than those for the normal group (P = 0.043). Two of the 13 patients (15%), with different mutations, had time constants above the normal range. 

Table 1 also lists the cone pigment double-density values for the patients with retinitis pigmentosa. Figure 3B plots these double densities, which averaged 0.15 log-unit, against those for the normal subjects, which averaged 0.19 log-unit. Although these two distributions are not significantly different (P = 0.098), the values for the patients tend to be lower than those for the normal subjects. Four of the 13 patients (31%), each with a different mutation, had double densities below the normal range. For the patients with retinitis pigmentosa, the time constant of pigment regeneration and the pigment double density are inversely related by the Spearman rank correlation (r = −0.26), but this value is not statistically significant (P = 0.40). 

Figure 4 plots the visual acuity recovery time versus the time constant of foveal cone pigment regeneration for the patients with retinitis pigmentosa. The correlation (0.65) is statistically significant (P = 0.017). If we consider only the data for the 6 patients with the Pro23His mutation, this correlation is 0.84 (P = 0.036). 

Discussion

This study shows that patients with dominant retinitis pigmentosa due to rhodopsin mutations, as a group, have visual acuity recovery times and foveal cone pigment regeneration time constants after a bleach that are significantly slower than normal. Although prior studies have reported slow visual acuity recovery or slow foveal cone pigment regeneration in patients with retinitis pigmentosa, ^{1 2 3 4} this is the first demonstration of these abnormalities in patients with retinitis pigmentosa who are known to have rhodopsin gene defects. We also found that the visual acuity recovery times and the time constants of cone pigment regeneration were significantly correlated for our entire group of patients, as well as for the subset of patients with the Pro23His mutation. Just as the rate of foveal cone dark adaptation reflects the rate of foveal cone pigment regeneration among normal subjects of varying age, ¹² the visual acuity recovery time similarly reflects the rate of foveal cone pigment regeneration among patients with retinitis pigmentosa and a rhodopsin mutation. 

Delayed visual acuity recovery and/or cone pigment regeneration was found in more than half of our patients with disease due to several different dominant rhodopsin mutations. Earlier studies had shown these abnormalities in patients with diverse genetic types of retinitis pigmentosa. ^{2 3 4} Therefore, this abnormality is not restricted to patients with a particular rhodopsin mutation or even with dominant disease. 

The physiologic basis for impaired cone pigment regeneration in retinitis pigmentosa is not known. Although not statistically significant, our finding of an inverse relationship between the time constant of cone pigment regeneration and the cone pigment optical density could mean that a prolonged time constant is a sign of abnormal cone outer segment morphology. Alternatively, the slowing of cone pigment regeneration could be the direct result of rod photoreceptor degeneration independent of the presence or stage of cone photoreceptor degeneration. It is possible, for example, that loss of rods results in less 11-cis-retinal in the retinal pigment epithelium after a bleach. Although cones are about fivefold more efficient than neighboring rods in competing for 11-cis-retinal after a bleach, ¹⁷ its shortfall in the retinal pigment epithelium might slow cone pigment regeneration, even in otherwise healthy cones. For example, it has been shown that the kinetics of cone dark adaptation are slowed in patients with systemic vitamin A deficiency and are normalized in these patients after vitamin A supplementation. ^{18 19} It remains to be established whether some critical delay in cone pigment regeneration or some critical slowing of visual acuity recovery time is a predictor of impending cone photoreceptor cell death in patients with retinitis pigmentosa and rhodopsin mutations. 

 Supported by National Eye Institute Grant EY00169; the Foundation Fighting Blindness, Hunt Valley, Maryland; and the Massachusetts Lions Eye Research Foundation, Inc.

 Submitted for publication December 29, 1998; revised April 5, 1999; accepted April 28, 1999.

 Proprietary interest category: N.

 Corresponding author: Michael A. Sandberg, The Berman-Gund Laboratory, The Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, MA 02114. E-mail: [email protected]

 

Table 1.

 

Patient Demographics and Ocular Findings by Age

Age/Sex	Rhodopsin mutation	Visual acuity	Dark-adapted threshold*	Acuity recovery time (sec), †	Photopigment double density, ‡	Photopigment regeneration time constant (sec), §
19/M	Pro347Leu	20 /20	4.30	44	0.14	365
24/M	Cys110Tyr	20 /20	2.18	21	0.17	178
33/F	Pro23His	20 /20	1.18	10	0.21	109
33/M	Pro23His	20 /20	2.04	14	0.16	97
36/F	Pro23His	20 /20	2.18	16	0.14	144
36/M	Phe45Leu	20 /20	1.90	11	0.15	94
39/F	Pro171Leu	20 /30	4.48	21	0.05	275
40/F	Pro23His	20 /25	1.90	17	0.08	162
46/M	Pro23His	20 /20	1.90	22	0.19	156
50/F	Pro23His	20 /20	2.00	22	0.20	189
51/F	Gly114Asp	20 /30	4.18	37.5	0.06	162
52/M	Gly89Asp	20 /30	3.78	28.5	0.24	159
57/M	Thr17Met	20 /30	3.78	22	0.10	143

*  Central 11° white light, in log microapostilbs (normal < 1.7).

†  Foveal; upper range of normal is 20 seconds (n = 15).

‡  Central 2°; lower range of normal is 0.13 (n = 11).

§  Central 2°; upper range of normal is 194 seconds (n = 11).

Figure 1.

 

Visual acuity recovery times in normal subjects and patients with retinitis pigmentosa. Filled circles are means; error bars, ±SE. Points with the same (or similar) value have been shifted horizontally to facilitate their identification.

Figure 2.

 

Data and fitted exponential curves for representative foveal cone pigment regeneration from a normal subject (top) and a retinitis pigmentosa patient with a rhodopsin mutation (bottom).

Figure 3.

 

Time constants (A) and double densities (B) from measurements of foveal cone pigment regeneration in normal subjects and patients with retinitis pigmentosa. Filled circles are means; error bars, ±SE. Points with the same (or similar) value have been shifted horizontally to facilitate their identification.

Figure 4.

 

Scatter plot of the visual acuity recovery time versus the time-constant of foveal cone pigment regeneration for the patients with retinitis pigmentosa. The straight line represents the best least-squares fit to the entire data set.