rod increment thresholds in cone-rod dystrophy

Rod increment thresholds in cone-rod dystrophy

David W. Yates,Fishman

Deborah J. Derlacki, David R. Pepperberg, Kenneth R. Alexander, and Gerald A.

Rod system increment threshold functions (ITFs) were studied in patients with cone-rod dystrophy (CRD).Rod thresholds (It) for a 104-min, 500-nm test stimulus (TS), superimposed on an 110 long wavelengthbackground (luminance Ib), were measured in eleven CRD patients and fourteen normal subjects. Thresh-olds in normals were also measured using a 7-min TS. A modified version of Weber's law [logIt = logK +log(IP + c)] was fitted to the data from each subject and test condition to yield a description of the ITF in terms

of the free parameters K, c, and n. Four of the CRD patients exhibited an abnormally high absolutethreshold; of these, two showed abnormalities in K and c consistent with a reduced efficiency of quantumcapture by the rods. Abnormalities in the ITFs of CRD patients did not resemble the effect of reducing thediameter of the TS from 104 to 7 min in normals. This suggests that threshold abnormalities in the CRDpatients did not result from altered spatial summation. The results illustrate use of a parametric representa-tion of the ITF to evaluate the loss of sensitivity in visual disorders.

1. Introduction

Cone-rod dystrophy (CRD) is a group of inherited,progressive retinal disorders involving dysfunction ofthe cone and, to a lesser degree, the rod photoreceptorsystems.1-8 As patients diagnosed with CRD repre-sent a genetically and clinically heterogeneous group,6

the mechanisms responsible for CRD are likely to bediverse. A reduction in amplitude of the rod-mediat-ed electroretinogram (ERG), often accompanied byperipheral retinal pigmentary changes, distinguishesCRD from progressive cone dystrophy. 7 9 10 ERG evi-dence for impairment of the rod system in CRD raisesinterest in further characterizing the loss of rod sensi-tivity in this group of disorders.

The increment threshold function [here abbreviatedITF; also referred to as a threshold vs intensity (TVI)function] represents the variation of threshold (It)with background luminance (Ib). The ITF has been

When this work was done all authors were with University ofIllinois at Chicago, Lions of Illinois Eye Research Institute, Ophthal-mology Department, Chicago, Illinois 60612; D. W. Yates is now withSmith Kline & French Laboratories, King of Prussia, Pennsylvania19406.

Received 2 June 1988.0003-6935/89/061115-07$02.00/0.© 1989 Optical Society of America.

widely used to describe properties of the rod system,both in normals and in patients with retinal disease"1 -16; also see Refs. 17-19. Previous investigators haveproposed that variations in the rod system ITF ob-served among subjects and in different test conditionsmay be governed by, and thus may characterize, specif-ic physiological processes, e.g., a gain-control mecha-nism in the retina. 1 4 1 5

We have analyzed properties of the ITF to studyabnormalities of the rod system in CRD patients.Rod-mediated psychophysical thresholds were mea-sured over a range of background luminance and ana-lyzed in terms of Eq. (1), a modified form of Weber'slaw:

logI, = logK + log(Ib + c). (1)

By fitting Eq. (1) to the threshold data, we have char-acterized the ITFs of CRD patients and normal sub-jects in terms of the free parameters K, c, and n; such aparametric representation facilitates intersubjectcomparisons of the ITF. Although we have found thatEq. (1) provides a close fit to the experimental data,there is no conclusive evidence that this test functionuniquely reflects the action of specific threshold-regu-lating processes. Nevertheless, an evaluation of ab-normalities in ITF parameters obtained with the use ofEq. (1) can potentially constrain hypotheses regardingthe underlying pathophysiology of CRD. The presentpaper compares relationships among the ITF parame-ters to the predictions of several such hypotheses.Some of our results were presented at the 1987 meetingof the Association for Research in Vision and Ophthal-mology. 20

15 March 1989 / Vol. 28, No. 6 / APPLIED OPTICS 1115

2

I

0:

5uz

)

-1

0 -2J_

-3

7 * 0104 '

*/ X~~~~-A v

-./-oo -4 -3 -2 -1 0 1

Log Retinal lluminance (scot td)

Fig. 1. ITFs of a normal subject (number 14) to a 104-min TS(circles) and a 7-min TS (triangles). Each data point represents theaverage of three trials. Solid curves plot the least-squares fit of Eq.

(1) to each set of measured thresholds.

II. Methods

A. Subjects

Fourteen normal subjects, ages 24-43 years, servedas controls. Each had a best-corrected Snellen visualacuity of 20/20 or better, was normal on ophthalmo-scopic examination, and had normal color vision asdetermined by testing on a Nagel anomaloscope.

The diagnosis of CRD for the eleven patients includ-ed in this investigation (ages 16-56 years; Table I) wasbased on patient history, electroretinographic find-ings, and clinical examination. The chief complaint ofeach patient was blurred or reduced central vision.Most of these patients also complained of difficultywith color vision; none indicated poor night vision asthe initial or most significant ocular symptom. Onexamination of the retina, none of these patientsshowed characteristic bone-spiculelike clumping ofpigment within the midperipheral retina, as seen inpatients with retinitis pigmentosa. Only nonspecific

pigment mottling was occasionally apparent. Bilater-al atrophic-appearing foveal lesions were also found tovarying degrees in seven of the patients. On slit-lampexamination, none showed posterior subcapsular lensopacification, or other than minimal yellowing of thelens nucleus.

Electroretinograms (ERGs) were recorded in bothdark- and light-adapted conditions using a protocoldescribed previously.2' Responses obtained from thepatients (Table I) were compared with the 90% toler-ance limits for a population of normal subjects testedin the same laboratory (normative values in Table I).ERGs were recorded in conditions that isolate the conesystem response (30-Hz long wavelength flicker pre-sented to the dark-adapted eye, and white flash pre-sented to the light-adapted eye), and in conditions forwhich the rod system response is isolated or likely topredominate (short wavelength or white flash present-ed to the dark-adapted eye). Relative to the responsesof normal subjects, cone responses of the CRD patientswere in general more severely affected than the rod-isolated or rod-dominated responses (Table J).22

The absence of a specific complaint of night blind-ness, a primary complaint of poor or blurred centralvision with impaired color vision, the relatively greaterreduction of cone (vs rod) function evident from theERG testing, and the absence of peripheral bone-spi-culelike pigment together indicated a diagnosis ofCRD. By contrast, patients with a rod-cone dystro-phy (i.e., retinitis pigmentosa) typically complain pri-marily of poor night vision, exhibit greater impairmentof rod (vs cone) function on ERG testing, and generallyhave peripheral pigmentary degenerative changes, of-ten in the form of bone-spiculelike pigment clumping.Based on family history, one of the present patients(number 3) showed the dominant form of CRD. Theother ten patients described here were isolated cases,having no other known affected family member.

Table 1. Electroretinographic Data Obtained from CRD Patients

ERG amplitudes (V)aDark-adaptedb Light-adaptedb

Patient Snellen Short wavelength 30-Hz longnumber Age acuity flash White flash wavelength flicker White flash

1 24 20/70 170 320 7 502 42 20/20 100 300 6 603 18 20/60 170 320 - 504 23 20/70 150 280 2.5 nd5 41 20/40 100 170 2.5 106 53 20/70 160 320 nd nd7 16 20/100 40 120 nd nd8 31 20/100 nd 110 3 509 56 20/200 100 280 7.5 40

10 18 20/50 nd 130 nd nd11 45 20/30 nd 40 nd 10

Normative valuesc 190-300 400-610 13.5-31 100-180

a nd = nondetectable response.b Testing conditions described in Ref. 21.c 90% tolerance limits with 90% coverage proportion (see text).

1116 APPLIED OPTICS / Vol. 28, No. 6 / 15 March 1989

B. Procedures

The pupil of the test eye was dilated by instilling2.5% phenylephrine hydrochloride and 1% tropica-mide drops. Each subject was dark-adapted for 30min prior to testing. The nontest eye was occludedduring the test period.

Increment thresholds were measured with a Tu-binger perimeter. Trial lenses were used to correctrefractive error and compensate for testing distance.The standard test condition, based on the rod isolationtechnique of Aguilar and Stiles, 2 employed a 500-nm,104-min, 500-ms test stimulus (TS). ITFs for a 7-minTS of identical duration and wavelength were alsomeasured in all normal subjects. Comparison of thedata obtained with the 7- vs 104-min TS afforded a testof the possibility that in normals, reduction of the TSsize mimics threshold abnormalities exhibited in CRD.Test stimuli were presented in the center of an 11° longwavelength background at 15° in the nasal field.

Subjects fixated on a dim, long wavelength cross.As fixation by CRD patients often is not precisely.foveal, it is uncertain whether the TS fell at 150 in thetemporal retina of the patients. To characterize thetest area in the CRD patients, absolute threshold pro-files to both a 500-nm and a 655-nm TS were measuredalong the horizontal meridian, from 200 nasal to 200temporal.232 4 For patients 1-10, this examinationshowed the test area to be representative of parafovealrod sensitivity. Thus, small instabilities in fixationthat may have occurred during testing would not havesignificantly affected the ITF. Patient 11 had a lessuniform threshold profile. However, repeat testing ofincrement thresholds in this patient (separate session)yielded results identical to those obtained initially,indicating stable fixation.

To determine whether increment thresholds for the500-nm TS were rod-mediated, ITFs were additionallyobtained with a 655-nm TS. Thresholds for the twowavelengths were measured alternately at each back-ground luminance; values obtained at each wavelengthwere expressed in units of log scotopic trolands. Eval-uation of the difference in the log threshold values forthe two wavelengths afforded identification of themechanism mediating detection at threshold.23 A dif-ference of 0 log scot td between the two values ofthreshold indicates rod-mediated detection at bothwavelengths; a difference of 1.9 log scot td indicatescone-mediated detection at both wavelengths. Inter-mediate values of the threshold difference indicaterod-mediated detection of the 500-nm TS and cone-mediated detection of the 655-nm TS.2 5 Analysis ofthe threshold differences indicated, for all subjects,rod-mediated detection of the 500-nm TS throughoutthe range of background luminance employed. Forexample, at the highest background luminance, thecondition in which thresholds would be most likely tobe cone-mediated, the range of threshold differencesamong the normals was 0.56-0.97 log unit. AmongCRD patients 1-10, the range was -0.16-0.90 log unit.Patient 11 was unable to detect the long wavelength

TS at any background luminance, a result consistentwith rod-mediated detection for the 500-nm TS.

Thresholds were measured using an ascendingmethod of limits. The luminance of the TS was ini-tially set at a subthreshold value and then increased bysteps of 0.1 log unit until the subject signaled detec-tion. Plotted values of threshold represent the meansof three consecutive determinations at each back-ground luminance. Backgrounds were presented inorder of increasing luminance; subjects adapted toeach background for 1 min prior to presentation of theTS. Through an iterative least-squares procedure,Eq. (1) was fitted to each set of data to obtain best-fitvalues for K, c, and n. Because the slope of the ITF isknown to vary with stimulus conditions, 13 we allowedfree variation of the slope parameter (n) in the fittingprocedure, rather than fixing n at some arbitrary val-ue.

Ill. Results

Figure 1 shows ITFs measured in a representativenormal subject. Circles and triangles indicate resultsobtained with the 104- and 7-min TS, respectively.Solid curves indicate the least-squares fit of Eq. (1) tothe measured thresholds (rms error = 0.09 for eachdata set). For all sets of data obtained from normalsubjects, the fitting of Eq. (1) yielded an rms error<0.13.

CRD patients 1-7 exhibited absolute thresholdswithin the normal range. With the exception of pa-tient 5 (see below), these patients exhibited normalincrement thresholds at all background luminancestested. Absolute thresholds for CRD patients 8-11were above the normal range. Figures 2(a)-(d) showITFs measured in patients 8-11 and the range of ITFsmeasured in normal subjects. As in normals, dataobtained from the CRD patients were well describedby Eq. (1) (rms error <0.15).

Figures 3(a)-(c) show intersubject distributions forlogK, logc, and n, respectively, determined by least-squares fit of Eq. (1) to the threshold data. Datadisplayed within each panel indicate results obtainedfrom normals, 7-min TS (left column); normals, 104-min TS (middle column); and CRD patients, 104-minTS (right column). Figure 3(d) shows the distributionof absolute thresholds for the respective subjects andtest conditions. Shown above the parametric data arepairs of curves illustrating how a change in value of asingle parameter affects the position and shape of theITF. These pairs of curves describe the influence of,respectively, an increase in logK [Fig. 3(a)], an increasein logc [Fig. 3(b)], and a decrease in n [Fig. 3(c)].

We first consider relationships among parametricdata obtained with the 104-min TS. As noted above,CRD patients 8-11 exhibited absolute thresholdshigher than the normal range. Figure 3 shows thatvalues of logK for these patients, as well as for patient5, were above the range of logK measured in normals.Patients 8, 10, and 11 also exhibited abnormally highvalues of logc. With the exception of patients 6 and 8,


2 * Patient 8

1

(a)/_ /

-0a -4 -3 -2 -1 0 1Log Retinal liluminance (scot td)

0 Ptient 10

..0/

-c~c (C)

r_--[4 -3 -2 -1 0

Log Retinal Illuminance (scot td)

2

I.01

o -1

_11° -2

-4 -3 -2 - 1 0Log Retinal Illuminance (scot td)

* Patient 11

- -W / *

0 -

-1 -

-2: -

_3

v

a)

I_-

Fo

-4 -3 -2 -1 0Log Retinal Illuminance (scot td)

Fig.2. ITFs of CRD patients 8-11 [(a)-(d), respectively]. Solid curves plotthe least-squares solution of Eq. (1). The shaded area indicatesthe range of thresholds measured in normal subjects.

all the CRD patients exhibited values of n within thenormal range.

Figure 3 also compares ITF parameters for the CRDpatients (104-min TS) with those of normals for the 7-min TS. In the normal subjects, reducing the TSdiameter from 104 to 7 min significantly increased thelog absolute threshold [mean increase: 2.26 log units(t 13 = 49.93, p < 0.01)]. Accompanying the thresholdelevations were changes in the ITF parameters; reduc-ing the TS diameter increased logK [mean increase:1.86 log units (t 1 3 = 54.12, p < 0.01)], increased logc[mean increase: 0.49 log unit (t 13 = 9.07, p < 0.01)],and decreased n [mean decrease: 0.12 (t13 = 9.62, p <0.01)]. The relative values of ITF parameters for CRDpatients 8-11 (104-min TS) differed from those fornormals using the 7-min TS. That is, for the normals,a decrease in the TS diameter led to an elevation ofabsolute threshold and a decrease in n. By contrast,CRD patients 8-11, who had elevated absolute thresh-olds (104-min TS), exhibited values of n that werewithin or above the range shown by normals (104-minTS). These results argue against an alteration of spa-tial summation properties in CRD patients 8-11 as thebasis of the observed ITF abnormalities.

We considered the alternative possibility that thethreshold abnormalities exhibited by CRD patients 8-11 reflect a reduced efficiency of quantum capture bythe rod photoreceptors. In its simplest form, the hy-pothesis of reduced quantum catch assumes equal at-

tenuation of the effective luminances of a backgroundlight and a stimulating flash. The resulting transla-tion of the ITF is a 450 shift in a logIt vs loglb plot, i.e.,equal displacements along the horizontal and verticalaxes. On the assumption of a constant ITF exponent(n), we may characterize the resulting changes in logKand loge as follows. We assume an attenuation of Iband It to the physiologically effective levels Ib and tywhere

(2)

Defining K' and c' as the altered values of the ITFparameters, we may write [through Eqs. (1) and (2)]

I = a'-'K I + aKc = K'I + K'c', (3)

where K' = a1 nK, and K'c' = aKc. We may addition-ally define

A logK = logK'- logK; A loge = logc' - log. (4)

A variation of a thus gives rise to a family of ITFssatisfying [from Eqs. (3) and (4)]

A logK = [(1 - n)/n]A logc. (5)

In a plot of logK vs loge, the resulting family of ITFs isthus represented by a line of slope (1 - n)/n.

The possibility of reduced quantum capture effi-ciency is evaluated in Fig. 4, which plots logK vs logcfor each CRD patient and normal subject (filled andopen circles, respectively). Dashed lines in the figure


0

a1.

II

o-1

-3

2

0

a-

3 (d)|

/

Is

-

151

PIFII-

-2

l

I'b = Ibla; 1' , = Ila; a > .

-IJ -6-0 -6 -5 -4 -3 -2 -1 0

2

- 10

-ELO 0a,

1

-2

, - 3-J- 4

Log Retinal Illuminance

0.0

-0.5

-o

u

01al

0-J

-1 '0

-1.5

-2.0

-2.5

-3.0N: 7' N: 104' CRD: 104'.

(b)

-1/

-u-F

a,

0__J

-6 -5 -4 -3 -2 -1 0


N: 7' N: 104' CRD: 104'

a,

0to

xCa

2- ---- - - - ---- -- - __

1 (C)

0-

-1

-2 --3 -I /_____ -4 /'I ' ' l ' I '

-00 -6 -5 -4 -3 -2 -1 0 1


1.28

0.95

0.90

0.85

0.80

0.75

0.70

0.65

0.60 N: 7' N: 104' CRD: 104

0-o

a,

0__J

1.o (d)

0.5

0.0

-0.5

-1.0

-1.5

-2.0

-2.5

-3.0

11-

0 10 0

08° a

0 1,3,4 21l ,

N: 7' N: 104' CRD: 104'

identify the normal ranges of logK and loge. Fromeach of the fourteen data points representing normalsubjects, we extended a line of slope (1 - n)/n, with thevalue of n determined by the ITF of that subject.These lines describe the changes in logK and logeresulting from a change in a. For clarity, Fig. 4 illus-trates only the two lines representing the extremes ofthis family of lines. Data from two CRD patients(numbers 8 and 10) fall outside- the normal ranges oflogK and loge, as indicated by the dashed lines in Fig. 4,but within the region enclosed by the solid lines. Thisresult is consistent with the possibility that the ITFabnormality in these two patients results from a re-duced efficiency of quantum capture by individual rodphotoreceptors.

Fig. 3. Parametric representation of ITF data. (a)-(c) Values of logK, loge, and n resultingfrom least-squares fit of Eq. (1) to the measured thresholds. (d) Absolute thresholds. Datawithin each panel show results obtained from normals, 7-min TS (left column); normals, 104-min TS (center column); and CRD patients, 104-min TS (right column). Numbers in theright-hand columns identify CRD patients. Pairs of curves above the parametric dataillustrate the dependence of the shape and position of the ITF on variation of a singleparameter. Relative to a reference (solid curve of each pair), the dashed curves show the

change due to (a) an increase in logK; (b) an increase in loge; and (c) a decrease in n.

IV. Discussion

We have employed a three-parameter test equationto evaluate rod function in CRD, a group of visualdisorders that has not, to our knowledge, previouslybeen described in terms of the ITF. Of the CRDpatients exhibiting absolute thresholds within the nor-mal range (numbers 1-7), numbers 2, 3, 4, and 7showed normal values for all three test parameters.This finding is noteworthy, as it was possible, a priori,that CRD patients with normal absolute thresholdsmight exhibit abnormal ITFs. Patient 5, character-ized by abnormally high logK and relatively low loge,showed an absolute threshold (= Kc) in the normalrange. The ITFs of patients 1 and 6 were within thenormal range at all values of background luminance


-o

I)

-oHa'

0-J

2

0

- 1

-2

-3

-4

- / - --

1.5

1.0

-o8

0

_J

0.5

0.0

I 110

0

9 @

10 5

S7i8Ok 2lb 1

-0.5

11

10

8 8

4 30 05

-1.0

8-

0

0

0 03So 2,11 5

10:0 101S 9

6

I. . . . . . .I l l i.-orl l

1 5lie

1.0 -

0 5 *9

, 0 -

-10- 0~~~~~61

0 -l 5 -1.5

-3 0 -2 -2 0 -1 5 -1 o -0.5 0.0

Log c (scot td)

Fig. 4. Plot of logK vs logc (104-min TS) for CRD patients andnormals (filled and open circles, respectively). Numbers identifyCRD patients 8-11. Dashed lines identify the normal ranges of logcand logK. Solid lines enclose a region describing the variation of theITFs of normals with change in the parameter a. See text for

further details.

but were characterized by abnormally low values oflogK and n, respectively. Patients 8-11 exhibited ab-normally high absolute thresholds and abnormal val-ues of at least one ITF parameter.

Previous studies, both psychophysical and electro-physiological, have discussed implications of the shapeand position of the ITF for mechanisms that regulatevisual sensitivity.13 -1526-31 It is of interest to considerwhether properties of the ITFs of CRD patients pro-vide information on the mechanism underlying theabnormality in CRD. We address this question in thefollowing paragraphs.

A. Knee Point Value of lb

The knee point of the ITF may be defined as thepoint at which It equals twice the value of absolutethreshold; let bO represent the background luminanceat this point. Two distinct hypotheses have been pro-posed to account for the observed value of bo. Thefirst of these postulates an intrinsic dark light thatlimits visual sensitivity and that is exceeded when Ib >ib (reviewed in Ref. 13). The second supposes theexistence of a gain-control mechanism that begins todepress sensitivity at b - bo (reviewed in Ref. 30).Simply interpreted in terms of Eq. (1), an elevateddark light should increase c and preserve the value of Kin the normal range. An abnormality in the gain-regulating mechanism might be expected to preserveabsolute threshold (= Kc) within the normal range,i.e., to increase c and decrease K. However, as indicat-ed by Figs. 3 and 4, in none of the CRD patients was theITF abnormality characterized by an elevated value oflogc and by a value of logK within or below the normalrange. The present data thus argue against both anelevated dark light and an altered gain-control mecha-nism as a principal basis of rod threshold abnormali-ties in CRD.

B. Vertical Position of the ITF

Alteration of a mechanism operating at a level proxi-mal to the site of visual adaptation might be expected

to shift the ITF vertically.'4"18 Such a vertical scalingof the ITF would be reflected as an increase in thevalue of K. ITF data obtained from one patient (num-ber 9) were characterized by an abnormally high valueof K, consistent with dysfunction of a postadaptationprocess. An abnormally high criterion for thresholddetection by patient number 9 could also account forthe elevation of K. However, for two reasons, thisalternative seems unlikely. First, the value of K ex-hibited by patient 9 differs substantially from the nor-mal range. Second, the notion of a criterion changeassumes a normally functioning retina, which is clearlynot the case for this patient (see ERG data, Table I).

C. Efficiency of Quantum Capture

A reduction in the efficiency of quantum capture bythe rods has been hypothesized as a possible basis forthe sensitivity loss in photoreceptor disease such asretinitis pigmentosa32-34 and certain types of progres-sive cone dystrophy. 3 5 A reduction in quantum catchcould arise, e.g., from shortened outer segments, a lossof photoreceptors, or an altered orientation of the out-er segment disk membranes. In the context of Eq. (1),the hypothesis of reduced quantum catch predicts theinterrelationship between K and c given by Eq. (5).Two of the CRD patients (numbers 8 and 10) exhibitedthreshold abnormalities consistent with this predictedrelationship.

D. Spatial Summation

There is evidence that threshold abnormalities invisual disorders may result from altered spatial sum-mation. Sandberg and Berson36 found that the ITFsof patients with retinitis pigmentosa could be mim-icked by a reduction in TS diameter. On this basisthey concluded that certain retinitis pigmentosa pa-tients have a reduced capacity for spatial summationfor the middle wavelength cone mechanism. In thepresent study we found that, among normals, a reduc-tion of the TS diameter elevates logK and loge anddecreases the exponent n. Similar changes are evidentin the data of Barlow.13 However, CRD patients 8-11showed a different pattern of values for the ITF pa-rameters (Fig. 3), which argues against reduced spatialsummation as the mechanism underlying thresholdchanges in CRD.

V. Conclusion

We have used the ITF to characterize rod systemdysfunction in patients with CRD. We have foundthat CRD patients with elevated absolute thresholdsdo not exhibit a single pattern of abnormality withrespect to ITF parameters. This is perhaps not sur-prising, in view of clinically observed heterogeneitiesknown to exist among CRD patients. Conceivably,the phenotypic expression of CRD could involve multi-ple mechanisms of the types considered above. Theobserved differences in the ITF properties of thesepatients indeed emphasize the value of parametricrepresentation of the ITF for intersubject comparisonsand for the evaluation of mechanisms of visual loss.


We thank Neal S. Peachey and Harris Ripps forhelpful discussions. This research was supported bygrants EY-07038, EY-05494, EY-01792, and EY-04848from the National Eye Institute and by the NationalRetinitis Pigmentosa Foundation Fighting Blindness.D.R.P. is a Robert E. McCormick Scholar of Researchto Prevent Blindness, Inc.

References

1. G. Goodman, H. Ripps, and I. M. Siegel, "Cone DysfunctionSyndromes," Arch. Ophthalmol. 70, 214 (1963).

2. E. L. Berson, P. Gouras, and R. D. Gunkel, "Progressive Cone-Rod Degeneration," Arch. Ophthalmol. 80, 68 (1968).

3. A. E. Krill, A. F. Deutman, and M. Fishman, "The Cone Degen-erations," Doc. Ophthalmol. 35, 1 (1973).

4. R. E. Carr and I. M. Siegel, "Cone Dysfunctions in Man," Trans.Am. Acad. Ophthalmol. Otolaryngol. 81, 653 (1976).

5. A. E. Krill, "Cone Degenerations," in Krill's Hereditary Retinaland Choroidal Diseases. Vol. II. Clinical Characteristics, A.E. Krill, Ed. (Harper & Row, Hagerstown, MD, 1977), p. 421.

6. G. A. Fishman, "Progressive Human Cone-Rod Dysfunction(Dystrophy)," Trans. Am. Acad. Ophthalmol. Otolaryngol. 81,716 (1976).

7. G. A. Fishman, "Electroretinography and Inherited MacularDystrophies," Retina 5, 172 (1985).

8. R. G. Weleber and A. Eisner, "Cone Degenerations (Bull's-EyeDystrophies) and Color Vision Defects," in Retinal Dystrophiesand Degenerations, D. A. Newsome, Ed. (Raven Press, NewYork, 1988), p. 223.

9. J. Babel and N. Stangos, "Progressive Degeneration of the Pho-topic System," Am. J. Ophthalmol. 75, 511 (1973).

10. D. G. Birch and G. E. Fish, "Rod ERGs in Retinitis Pigmentosaand Cone-Rod Degeneration," Invest. Ophthalmol. Visual Sci.28, 140 (1987).

11. S. Hecht, "Rods, Cones, and the Chemical Basis of Vision,"Physiol. Rev. 17, 239 (1937).

12. M. Aguilar and W. S. Stiles, "Saturation of the Rod Mechanismof the Retina at High Levels of Stimulation," Opt. Acta 1, 59(1954).

13. H. B Barlow, "Dark and Light Adaptation: Psychophysics," inHandbook of Sensory Physiology. Vol. VII/4, D. Jameson andL. M. Hurvich, Eds. (Springer-Verlag, Berlin, 1972), p. 1.

14. D. C. Hood and V. C. Greenstein, "Increment Threshold (tvi)Data and the Site of Disease Action," in Technical Digest ofTopical Meeting on Noninvasive Assessment of the VisualSystem (Optical Society of America, Washington, DC, 1988),Vol.3, p. 2.

15. R. S. L. Young, J. Price, and J. Harrison, "Psychophysical Studyof Rod Adaptation in Patients with Congenital Stationary NightBlindness," Clin. Vis. Sci. 1, 137 (1986).

16. B. Stabell, K. Nordby, and U. Stabell, "Light-Adaptation of theHuman Rod System," Clin. Vis. Sci. 2, 83 (1987).

17. D. C. Hood and V. C. Greenstein, "An Approach to TestingAlternative Hypotheses of Changes in Visual Sensitivity Due toRetinal Disease," Invest. Ophthalmol. Visual Sci. 23, 96 (1982).

18. W. S. Geisler, "Mechanisms of Visual Sensitivity: Backgroundsand Early Dark Adaptation," Vision Res. 12, 1423 (1983).

19. R. W. Massof, S. Marcus, G. Dagnelie, D. Choy, J. Sunness, andM. Albert, "Theoretical Interpretation and Derivation of Flash-on-Flash Threshold Parameters in Visual System Diseases,"Appl. Opt. 27, 1014 (1988).

20. D. W. Yates, D. J. Derlacki, D. R. Pepperberg, G. A. Fishman,and K. R. Alexander, "Rod Increment Threshold Function inCone-Rod Dystrophy," Invest. Ophthalmol. Visual Sci. (ARVOSuppl.) 28, 112 (1987).

21. G. A. Fishman, G. Buckman, and T. van Every, "Fundus Flavi-maculatus: a Clinical Classification," Doc. Opthalmol. 13, 213(1977).

22. The ERG responses of patients 8, 10, and 11 to a single shortwavelength flash were nondetectable. However, in patients 8and 10, markedly reduced but detectable responses were appar-ent to a 10-Hz short wavelength flickering stimulus when severalresponses were averaged. Furthermore, the responses of thesetwo patients to a white flash was appreciably greater in dark-adapted (vs light-adapted) conditions, strongly suggesting thatthe dark-adapted response was rod-mediated. Patient 11 wassomewhat atypical, in that her response to even a bright whiteflash was <50 ,uV. Nevertheless, this patient did not indicatenight blindness as her initial or major complaint and did notmanifest any clinical evidence of peripheral retinal degenera-tion. The visual acuity of 20/30 in this patient is consistent withthe diagnosis of CRD, as was her impaired color vision.

23. R. W. Massof and D. Finkelstein, "Two Forms of AutosomalDominant Primary Retinitis Pigmentosa," Doc. Ophthalmol.51, 289 (1981).

24. K. R. Alexander and G. A. Fishman, "Prolonged Rod DarkAdaptation in Retinitis Pigmentosa," Br. J. Ophthalmol. 68,561(1984).

25. S. L. Guth, R. W. Massof, and T. Benzschawel, "Vector Modelfor Normal and Dichromatic Color Vision," J. Opt. Soc. Am. 70,197 (1980).

26. C. B. Blakemore and W. A. H. Rushton, "Dark Adaptation andIncrement Threshold in a Rod Monochromat," J. Physiol. (Lon-don) 181, 612 (1965).

27. D. G. Green, J. E. Dowling, I. M. Siegel, and H. Ripps, "RetinalMechanisms of Visual Adaptation in the Skate," J. Gen. Physiol.65, 483 (1975).

28. J. W. Clack and D. R. Pepperberg, "Desensitization of SkatePhotoreceptors by Bleaching and Background Light," J. Gen.Physiol. 80, 863 (1982).

29. D. A. Baylor, B. J. Nunn, and J. L. Schnapf, "The Photocurrent,Noise and Spectral Sensitivity of Rods of the Monkey, MacacaFascicularis," J. Physiol. (London) 357, 575 (1984).

30. R. Shapley and C. Enroth-Cugell, "Visual Adaptation and Reti-nal Gain Controls," Prog. Retinal Res. 3, 263 (1984).

31. H. Ripps and D. R. Pepperberg, "Photoreceptor Processes inVisual Adaptation," Neurosci. Res. Suppl. 6, S87 (1987).

32. V. N. Highman and R. A. Weale, "Rhodopsin Density and VisualThreshold in Retinitis Pigmentosa," Am. J. Ophthalmol.75,822(1973).

33. H. Ripps, K. P. Brin, and R. A. Weale, "Rhodopsin and VisualThreshold in Retinitis Pigmentosa," Invest. Ophthalmol. VisualSci. 17, 735 (1978).

34. I. Perlman and E. Auerbach, "The Relationship Between VisualSensitivity and Rhodopsin Density in Retinitis Pigmentosa,"Invest. Ophthalmol. Visual Sci. 20, 758 (1981).

35. H. Ripps, K. G. Noble, V. C. Greenstein, I. M. Siegel, and R. E.Carr, "Progressive Cone Dystrophy," Ophthalmology 94, 1401(1987).

36. M. A. Sandberg and E. L. Berson, "Blue and Green Cone Mecha-nisms in Retinitis Pigmentosa," Invest. Ophthalmol. Visual Sci.16, 149 (1977).

0


rod increment thresholds in cone-rod dystrophy

Documents