|
|
||||||||
1From the The Berman-Gund Laboratory for the Study of Retinal Degenerations and 2The Ocular Molecular Genetics Institute, Harvard Medical School, Massachusetts Eye and Ear Infirmary, Boston, Massachusetts.
| Abstract |
|---|
|
|
|---|
METHODS. Snellen visual acuities, Goldmann visual field areas (V4e white test light), and 30 Hz (cone) full-field ERG amplitudes were recorded for an average of 9.8 years in 113 patients with RPGR mutations. After censoring data to eliminate ceiling and floor effects, we used longitudinal regression to estimate mean rates of change and to compare these rates with those of a previously studied cohort of 134 patients with dominant retinitis pigmentosa due to RHO mutations, who were followed for an average of 8.9 years. Survival analysis was used to compare the age distribution of legal blindness in these two groups. To explain group differences in visual acuity, optical coherence tomograms were recorded in some patients to visualize central retinal structure.
RESULTS. Mean annual exponential rates of decline for the patients with RPGR mutations were 4.0% for visual acuity, 4.7% for visual field area, and 7.1% for ERG amplitude. Each of these rates was significantly different from zero (P < 0.001). The rates of visual acuity and visual field loss were significantly faster than the corresponding rates in the RHO patients (1.6%, P < 0.001 and 2.9%, P = 0.002, respectively), whereas the rate of ERG amplitude loss was comparable to that in the RHO patients (7.7%, P = 0.39). The median age of legal blindness was 32 years younger in the RPGR patients than in the RHO patients, due primarily to loss of visual acuity rather than to loss of visual field. Loss of acuity in RPGR patients appeared to be associated with foveal thinning.
CONCLUSIONS. Patients with X-linked retinitis pigmentosa due to RPGR mutations lose visual acuity and visual field more rapidly than do patients with dominant retinitis pigmentosa due to RHO mutations.
The discovery of the molecular bases for different forms of retinitis pigmentosa has allowed us and others to reclassify patients according to their responsible gene defects. We recently reported mean rates of decline in ocular function for a large cohort of patients with dominant retinitis pigmentosa due to mutations in the rhodopsin (RHO) gene.2 In the present study we report mean rates of decline in a similar-sized cohort of patients with X-linked retinitis pigmentosa due to mutations in the retinitis pigmentosa GTPase regulator (RPGR) gene, the protein of which is expressed in the connecting cilia of cones as well as rods,3 and compare these rates with those of patients with RHO mutations.
| Methods |
|---|
|
|
|---|
Clinical Evaluation
The patients with RPGR mutations and those with RHO mutations underwent identical ocular examinations. We recorded best corrected visual acuity by using a projected Snellen chart and coded them in decimal form (e.g., 20/40 = 0.5). Kinetic visual fields were measured to the V4e white test light and to one or more smaller test lights in the Goldmann perimeter against the standard background of 31.5 apostilbs, bringing the test light from nonseeing to seeing areas. Fields were plotted with a digitizing tablet or scanned by custom software and converted to areas in square degrees. Although a cartographic distortion arises from projecting the curved surface of the perimeter onto the flat visual field chart6 and the projection of the visual field onto the retina is nonlinear based on a schematic eye,7 most longitudinal studies of visual field progression in retinitis pigmentosa have not applied corrections to their chart data,1 2 8 9 10 and we elected to do the same for consistency. We elicited full-field cone ERGs with 10-µs, 30-Hz flashes of white light (0.2 cd-s/m2) after pupillary dilation, 45 minutes of dark adaptation, and having recorded responses in the dark to 0.5 Hz flashes of light. ERGs were monitored with a contact lens electrode on the topically anesthetized cornea and differentially amplified. Consecutive responses to 30-Hz flashes greater than 10 µV in amplitude were photographed from the screen of an oscilloscope or digitized and quantified by computer. Smaller responses were digitized, smoothed with a band-pass filter, and averaged. Waveforms in response to 30-Hz flashes were quantified with respect to trough-to-peak amplitudes, and amplitudes <0.05 µV, considered nondetectable, were recoded as 0.05 µV. In this study, we limited our analyses to the V4e white test light for measuring visual fields and to 30-Hz white flashes for eliciting ERGs, because only these conditions of testing provided us with sufficiently large data sets to estimate rates of change with high precision.
As part of a separate program,11 we had recorded optical coherence tomograms (OCTs) from 5 of the patients with RPGR mutations (age range, 1947 years) and from 10 of the patients with RHO mutations (age range, 2052 years). The RPGR patients had visual acuities of 20/30 to 20/200, and the RHO patients had visual acuities of 20/20 to 20/60 (excluding an eye of one patient with a history of deep amblyopia). We evaluated these tomograms to search for a structural basis for visual acuity differences in these two groups.
Statistical Analyses
For estimating mean rates of change, we censored visual acuities of 20/20, except those that followed a lower value, to minimize a ceiling effect, because on our coding sheet we had constrained Snellen visual acuities to be
20/20. To minimize floor effects, we also censored patients with baseline visual acuities <20/100 and follow-up data after visual acuity declined to <20/100. For patients who became aphakic or pseudophakic in either eye at follow-up, those follow-up visits were excluded from visual acuity analyses. We also censored baseline visual field areas <78 deg2 (i.e., equivalent to a diameter of 10°) and follow-up data after the first occurrence of an area <78 deg2 to minimize floor effects. To minimize floor effects, we censored baseline ERG amplitudes <0.68 µV and follow-up data after the amplitude decreased to <0.34 µV. The censoring criteria were those applied in a previous study of patients with dominant RHO mutations.2 After applying these criteria, we eliminated patients from a given analysis if their residual follow-up was <3 years.
We converted all measures of ocular function to natural logarithms, because an exponential model has been shown to be optimal for evaluating cell loss over time in animal models of retinitis pigmentosa,12 provides a good fit for describing short-term disease progression in patients with retinitis pigmentosa,2 and has been used in several longitudinal studies of retinitis pigmentosa.1 2 8 9 10 13 14 Repeated-measures longitudinal regression (performed with PROC MIXED of SAS, ver. 9; SAS Institute, Cary, NC) was used to estimate the mean rate of change for each outcome measure, based on the average loge value for both eyes at each visit (when data for both eyes were available). By including terms for genotype (i.e., RPGR versus RHO mutation) and the cross-product of time x genotype, we compared mean slopes in patients with RPGR mutations versus mean slopes in patients with RHO mutations. We also used longitudinal regression to compare the mean rates of progression in patients with RPGR mutations in exons 1 to 14 (n = 33) with the mean rates in patients who had RPGR mutations in open reading frame (ORF) 15 (n = 80), because a previous analysis had suggested differences in ocular function between these two groups based on single visits.5
We used the commercial software (PROC LIFEREG of SAS) to fit a Weibull function to survival data and compare the age distribution of legal blindness in patients with RPGR mutations to the corresponding distribution in patients with RHO mutations. These plots provide a visualization of the long-term course of disease, and the model allows inclusion of left-censored data (i.e., a patient failing at baseline) and right-censored data (i.e., a patient not failing during follow-up) as well as interval-censored data (i.e., a patient failing between exams occurring at ages x1 and x2). For this purpose, we applied failure criteria (i.e., a visual acuity
20/200 or a visual field area
314 deg2 in one eye and a visual acuity
20/200 or a visual field area
314 deg2 in the fellow eye) to the entire dataset. The area of 314 deg2 corresponds to an equivalent diameter of 20° (i.e., a criterion for legal blindness) and was used in lieu of measuring the linear extent of each remaining visual field directly from charts. We also used a visual acuity
20/200 alone and a visual field area
314 deg2 alone as failure criteria, to determine which was the critical factor that led to legal blindness in each group.
| Results |
|---|
|
|
|---|
|
|
|
Median Age to Reach Legal Blindness
We found a significant effect of genotype on the age distribution for legal blindness (P < 0.001). Figure 1 shows that our patients with RPGR mutations reached legal blindness, based on loss of acuity and/or field, at a median age (45 years) that was 32 years younger than that of our patients with RHO mutations (77 years). Figure 2 shows that the development of legal blindness was driven primarily by visual acuity loss in the patients with RPGR mutations and by visual field loss in the patients with RHO mutations. That is, in the patients with RPGR mutations, the survival curve based on a visual acuity of 20/200 or less is shifted to younger ages compared with the survival curve based on a visual field area of 314 deg2 or less. In contrast, the visual field survival curve is shifted to younger ages compared with the visual acuity survival curve of the patients with RHO mutations.
|
|
Optical Coherence Tomography
Five of the 10 RHO patients with available OCTs had reduced visual acuity associated with macular cysts and were not considered further. Of the remaining patients with tomograms, the five with RHO mutations had a mean visual acuity of 20/22 and the five with RPGR mutations had a mean visual acuity of 20/53 (Table 4) . This difference was significant (P = 0.002). Their mean retinal thicknesses at the foveal center were 171 and 104 µm, respectively. The thickness in the RHO patients is similar to the normal mean thickness for our test system (167 µm),11 and that in the RPGR patients is significantly smaller than that in the RHO patients (t-test for unequal variances, P = 0.03). Figure 3 shows a tomogram from a 40-year-old RHO patient with a visual acuity of 20/25 and from a 38-year-old RPGR patient with a visual acuity of 20/100. The patient with the RHO mutation had a normal retinal thickness profile, whereas the patient with the RPGR mutation had a broad foveal depression with attenuation of the outer nuclear layer centrally, indicating loss of central foveal cones.
|
|
| Discussion |
|---|
|
|
|---|
OCT recordings revealed that the difference in mean visual acuity between these two groups may be attributable to photoreceptor loss. The tomograms of one RPGR patient showed a broad thinning of the fovea resembling that in the tomograms of patients with Stargardts disease17 or occult macular dystrophy.18 When visual acuity was reduced in patients with RHO mutations, it tended to be associated with macular cysts. Study of OCTs from additional RHO patients and RPGR patients who have reduced visual acuity will reveal whether these features are characteristic of these two groups.
We also found that patients with RPGR mutations lost visual field area to the V4e stimulus at a mean rate that was approximately 50% faster than that in patients with RHO mutations and were left with a central island of vision
20° at a younger median age than the RHO patients. This result suggests that the faster loss of visual acuity by the RPGR patients may be a consequence of their faster loss of central field, consistent with a significant correlation between visual acuity and central visual field diameter in retinitis pigmentosa.19 However, the two groups had similar age distributions for retaining 20° of total visual field and had nearly identical mean rates of progression of the full-field cone ERG, which derives mostly from the peripheral retina.
In a previous study based on single visits with adjustment for differences in age, we reported that patients with RPGR mutations in exons 1 to 14 had a borderline smaller mean visual field area (P = 0.04) and mean cone ERG amplitude (P = 0.06) than did patients with RPGR mutations in ORF15,5 suggesting that the former group had more severe disease at a given age than did the latter group. In the present study we evaluated whether mean rates of disease progression were different in these two groups. We found that patients with exon 1 to 14 mutations lost ERG amplitude 50% faster than did patients with ORF15 mutations, whereas visual acuity and visual field area declined comparably in the two groups. We, therefore, conclude that RPGR mutations in exons 1 to 14 tend to cause a more rapid loss of peripheral cone retinal function than do mutations in ORF15.
| Acknowledgements |
|---|
| Footnotes |
|---|
Submitted for publication August 16, 2006; revised October 31, 2006; accepted January 23, 2007.
Disclosure: M.A. Sandberg, None; B. Rosner, None; C. Weigel-DiFranco, None; T.P. Dryja, None; E.L. Berson, None
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked "advertisement" in accordance with 18 U.S.C.
1734 solely to indicate this fact.
Corresponding author: Michael A. Sandberg, Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, MA 02114; masandberg{at}aol.com.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. A. Sandberg, B. Rosner, C. Weigel-DiFranco, T. L. McGee, T. P. Dryja, and E. L. Berson Disease Course in Patients with Autosomal Recessive Retinitis Pigmentosa due to the USH2A Gene Invest. Ophthalmol. Vis. Sci., December 1, 2008; 49(12): 5532 - 5539. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Walia, G. A. Fishman, A. Swaroop, K. E. H. Branham, M. Lindeman, M. Othman, and R. G. Weleber Discordant Phenotypes in Fraternal Twins Having an Identical Mutation in Exon ORF15 of the RPGR Gene Arch Ophthalmol, March 1, 2008; 126(3): 379 - 384. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. S. Aleman, A. V. Cideciyan, A. Sumaroka, S. B. Schwartz, A. J. Roman, E. A. M. Windsor, J. D. Steinberg, K. Branham, M. Othman, A. Swaroop, et al. Inner Retinal Abnormalities in X-linked Retinitis Pigmentosa with RPGR Mutations Invest. Ophthalmol. Vis. Sci., October 1, 2007; 48(10): 4759 - 4765. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |