REFERENCES. Reports 765. Temporal aspects of the dark-adapted cone a-wave in retinitis pigmentosa. MICHAEL

Transcription

1 Volume 21 Reports Reprint requests: Merlyn M. Rodrigues, M.D., Building 10, Room 10N315, National Institutes of Health, Bethesda, Md Key words: Langerhans cells, cornea, conjunctiva, HLA-DR, immunofluorescence, immuno-electron microscopy, ATPase REFERENCES 1. Stingl G, Wolff-Schreiner EC, PichlerWJ, Gschnait F, Knaff VV, and Wolff K: Epidermal Langerhans cells bear Fe and C 3 receptors. Nature 268:245, Rowden G, Lewis MC, and Sullivan AL: la antigen on human epidermal Langerhans cells. Nature 268:247, Stingl G, Tamaki K, and Katz SI: Origin and functions of epidermal Langerhans cells. Immunol Rev 53:149, Rowden G: Expression of la antigens on Langerhans cells in mice, guinea pigs and man. J Invest Dermatol 75:22, Bergstresser PR, Toews GB, and Streilein JW: Natural and perturbed distributions of Langerhans cells: response to ultraviolet light, heterotopic skin grafting and dimetrofluorobenzene sensitization. J Invest Dermatol 75:73, Brown J, Soderstrom CW, and Winkelmann RK: Langerhans cells in guinea pig cornea. Response to chemical injury. INVEST OPHTHALMOL 7:668, Sugiura S and Matsuda H: infrastructures of Langerhans cells in human corneal limbus. Jpn J Ophthalmol 13:197, Bock P and Hanak H: Some remarks on the morphology of the guinea pig conjunctival epithelium. J Submicrosc Cytol 3:1, Klareskog L, Forsum U, Tjernlund UM, and Peterson PA: Expression of la antigen-like molecules on cells of corneal epithelium. INVEST OPHTHALMOL VIS SCI 18:310, MacKenzie IC and Squier CA: Cytochemical identification of ATPase positive Langerhans cells in EDTA-separated sheets of mouse epidermis. Br J Dermatol 92:523, Rowden G: Immunoelectron microscopic studies of surface receptors and antigens of human Langerhans cells. Br J Dermatol 97:593, Rowden G, Phillips TM, and Lewis MG: la antigens on indeterminate cells of the epidermis: immunoelectron microscopic studies of surface antigens. Br J Dermatol 100:531, Streilein JW, Toews GB, and Bergstresser PR: Corneal allografts fail to express la antigens. Nature 282:326, Alerer W, Schuler G, Stingl G, Honigsmann H, and Wolff K: Ultraviolet light depletes surface markers of Langerhans cells. J Invest Dermatol 76:202, Silberg-Sinakin I, Gigli I, Baer RL, and Thorbecke CJ: Langerhans cells: role in contact hypersensitivity and relationship to lymphoid dendritic cells and to macrophages. Immunol Rev 53:203, Rowden G and Lewis MG: Langerhans cells: involvement in the pathogenesis of mycosis fungoides. Br J Dermatol 95:665, Temporal aspects of the dark-adapted cone a-wave in retinitis pigmentosa. MICHAEL A. SANDBERG, PETER L. SULLIVAN, AND ELIOT L. BERSON. Cone electroretinograms were elicited with a full-field red flash after 45 min of dark adaptation from 16 patients with retinitis pigmentosa and from 17 normal subjects. The average latency and implicit time of the cone a-waves recorded from the patients were each significantly delayed (p < 0.005) compared with the average normal values. In addition, the patients' ivaveforms typically showed a loss of two oscillations on the rising slope of the cone b-wave. These delays and waveform changes could be simulated in normal subjects by reducing the luminance, but not the diameter, of the stimulus. These studies in normal subjects suggest that shortening of outer segments of remaining cones and not simply loss of cones could account for the waveforms observed in patients with retinitis pigmentosa. The results also suggest that remaining cones generate an a-wave in the dark that is normal for a reduced cone visual pigment density. (INVEST OPHTHALMOL VIS SCI 21: , 1981.) Ultrastructural studies of postmortem eyes from patients with hereditary retinitis pigmentosa have revealed retinal areas where cones were reduced in number and had shortened outer segments. 1 " 4 Patients with early retinitis pigmentosa have shown reductions in the amplitude of the conedominant early receptor potential 5 "" 7 and elevations of dark-adapted cone thresholds. 8 " 10 These physiologic findings can be explained, at least in part, by reduced numbers of cones or shortened outer segments or both. A reduction solely in the number of cones would be expected to result in a reduction in the amplitude of the cone a-wave without a change in its latency and implicit time." 1 12 If remaining cones had shortened outer segments but were otherwise functioning normally, then cone a-wave responses should be slowed as well as /81/ $00.50/ Assoc. for Res. in Vis. and Ophthal., Inc.

2 766 Reports Invest. Ophthalmol. Vis. Sci. November 1981 Table I. Dark-adapted full-field cone a-waves in retinitis pigmentosa Patient Age Amplitude (IJLV) Latency (msec) Implicit time (msec) Patients X S.E.M. Normal subjects X S.E.M reduced in amplitude. 13 ' l4 The present study was done to evaluate the temporal characteristics of the cone a-wave in patients with retinitis pigmentosa as a measure of the functional capacity of remaining cones. Methods. Sixteen patients with retinitis pigmentosa (average age 25 years) were selected on the basis of previous electroretinogram (ERG) recordings on file that showed dark-adapted fullfield cone ERG b-wave amplitudes of at least 10 fxv to single flashes of red light. According to pedigree analyses, patients 1 to 8 (Table I) had dominant retinitis pigmentosa with reduced penetrance, 9 to 13 had autosomal recessive retinitis pigmentosa, and 14 to 16 had sex-linked retinitis pigmentosa. Patients had best-corrected visual acuities that ranged from 20/20 to 20/50. All had full visual fields as measured on the Goldmann perimeter with a V-4 white test light except patients 8 and 9, whose fields were constricted to the 40 and 30 isopters, respectively. All had clear media except patient 9, who had a small posterior subcapsular lens opacity. All had elevated rod thresholds and markedly reduced or nondetectable full-field rod ERGs. Cone ERGs to white 30 Hz flashes were normal or reduced in amplitude and were delayed in b-wave implicit time. Seventeen normal observers (average age 23 years) were tested to provide baseline values. The eye to be tested was dilated with 1% cyclopentolate hydrochloride and 10% phenylephrine hydrochloride and was dark-adapted for 45 min. ERGs were monitored with a bipolar Burian-Allen contact lens electrode placed on the topically anesthetized cornea, differentially amplified at a gain of 10,000 (3 db down at 2 Hz and 300 Hz) with respect to a forehead ground, attenuated at 60 Hz with a notch filter (Q = 43), and recorded on magnetic tape with FM electronics. Responses were later replayed from tape into a signal-averaging computer (38 msec analysis time) while monitoring for baseline stability; approximately one fourth of the responses of each patient and each normal subject had a drifting baseline exceeding 2 /xv over the first 5 msec, and these responses were excluded from the computer summation. Since baseline noise for single responses appeared comparable in patients and normal subjects and since with computer averaging the noise is reduced in proportion to the square root of the number of summations, the number of responses summed for a given patient or normal subject was selected to be approximately inversely proportional to the square of his or her a-wave amplitude (baseline to negative peak). This process of equating for signal to noise for all patients and normal subjects was done to help minimize a bias toward overestimating the latency of small amplitude responses,. This bias was also minimized by adjusting the vertical gain of a given summed response to match the a-wave slope of a standard normal waveform.

3 Volume 21 Dark-Adapted Cone a-waves in RetinHIs Pigmentosa Reports 767 Effect of Stimulus Diameter on the Normal Cone a-wave 120' r Normal Dom. RP 40' Auto. Rec RP X-Linked RP Fig. 1. Representative computer-averaged responses from a normal subject (n = 16) and patients 4 (n = 48), 13 (n = 96), and 14 (n = 80) from top to bottom, Tracings begin at flash onset. For each pair of horizontal lines through a given tracing, the upper line defines the average baseline and the lower line defines the criterion displacement for measuring a-wave latency at the intersection with the tracing. Vertical lines are drawn through normal a-wave latency (left line) and implicit time (right line). A-wave latency was then measured on a photographic print of the display from stimulus flash (i.e., trace onset) to the first point of the descending slope that was 5 mm below a horizontal line bisecting the baseline noise; this distance exceeded the remaining baseline variation of all tracings and intersected the a-wave well before the initiation of the b-wave. A-wave implicit time (sometimes designated as b-wave latency) was measured from stimulus flash to the first major Fig. 2. Computer-averaged responses from a normal subject to the 120 (full-field) flash (n = 16) and a 40 flash (n = 120). Horizontal and vertical lines are as described in Fig. 1. negative peak preceding the cornea positive b-wave. Cone a-waves were elicited every 2 sec by a full-field 10 yusec xenon flash attenuated with a Kodak Wratten 26 (red) filter; this stimulus had a luminance of 3.4 log foot-lamberts. Isolation of cone function was confirmed by the observation that two patients with congenital rod monochromatism tested with this stimulus and seven normal subjects tested with a scotopically matched blue light that isolated rod function gave no response within the 38 msec analysis time. Testing of a patient with congenital protanopia and a patient with congenital deuteranopia showed the cone a-wave elicited by this red flash to be about 30% generated by the green cones and 70% by the red cones. Some normal observers were seated at different distances from the dome so that the stimulus flash subtended visual angles of 120 (fullfield), 50, 40, 30, and 20 at the cornea; this manipulation stimulated a reduction in the number of functional cones. For other normal observers, neutral density filters were inserted in front of the red flash to simulate the effects of shortened outer segments. Differences in the

4 768 Invest, Ophthalmol. Vis. Sci. November 1981 Reports Effect of Stimulus Luminance on the Normal Cone a-wave o.o N O 0.6 NO Fig. 3. Computer-averaged responses from a normal subject at standard luminance (n = 23) and at reduced luminance (n = 113). Horizontal and vertical lines are as described in Fig. 1. amplitude, latency, and implicit time of the cone a-wave between the group of retinitis pigmentosa patients and the group of controls were assessed statistically with Student's t test for means from independent samples for normally distributed data or the Mann-Whitney U test for skewed data. Results. Cone a-wave amplitudes, latencies, and implicit times for the 16 patients with retinitis pigmentosa are listed in Table I; means and standard errors for the patient and normal groups are given below each column. As a group the patients with retinitis pigmentosa had cone a-wave amplitudes that were significantly reduced below normal (p < ) based on the Mann-Whitney U test. Their average latency and implicit time were significantly delayed with respect to the normal mean values (p < 0.005) based on t tests. Neither latency nor implicit time correlated significantly with amplitude for the patients. Representative responses from a normal subject and three patients with different genetic types of retinitis pigmentosa are illustrated in Fig. 1; vertical lines designating normal a-wave latency and implicit time have been extended through wave- forms of the patients to demonstrate these delays. The delays in a-wave implicit time seen in these three patients' recordings were accompanied by a loss of two oscillations on the rising phase of the b-wave (Fig. 1). Twelve patients showed this b-wave defect; patients 3 ; 8, and 9 showed reduced oscillations; and patient 15 showed normal oscillations. Reducing the angular subtense of the stimulus flash had no effect on the temporal aspects of the normal a-wave under these test conditions (Fig. 2). In contrast, reducing the luminance of the stimulus flash slowed both latency and implicit time of the normal cone a-wave. With 0.6 neutral density attenuation (i,e., simulating approximately a 75% reduction in cone outer segment length), the timing of the a-wave and the loss ot two oscillations on the b-wave approximated a waveform representative of those seen in our patients with retinitis pigmentosa tested without neutral density filters (Fig. 3). Discussion. The delays in cone a-wave latency and implicit time found in the present group of retinitis pigmentosa patients and the simulation of these delays in normal subjects tested with a dimmer light flash are both consistent with the idea that the patients have shortened cone outer segments. The delays in the cone a-wave in the patients cannot be attributed to light attenuation by lens opacities, since only one patient had a minimal cataract. Moreover, the waveform simulation in normal subjects would suggest that remaining cone photoreceptors in patients with retinitis pigmentosa can generate an a-wave under darkadapted conditions that is normal for a reduced cone visual pigment density. The finding that delays were not present in normal subjects tested with stimuli reduced in visual angle supports the idea that the delays in patients cannot be caused simply by a reduction in the number of functional cones in the retina. This latter conclusion is reinforced by a previous report that localized retinal destruction secondary to photocoagulation was associated with a decline in cone a-wave amplitudes but no change in cone a-wave implicit times under dark-adapted conditions.l2 Furthermore, for the present group of patients cone a-wave latency and implicit time were not correlated with cone a-wave amplitude which should reflect in part the number of remaining cones. Absence of two oscillations on the rising phase of the cone b-wave was evident in most patients'

5 Volume 21 Reports 769 recordings and was likewise simulated in normal subjects with a neutral density filter. The four patients who retained these oscillations also had the shortest cone a-wave implicit times despite having reduced amplitudes (Table I). The possibility must be raised that implicit time of the cone a-wave in retinitis pigmentosa patients varies with the presence or absence of these oscillations. This inference may also be drawn from previously published full-field cone ERGs of patients with dominant retinitis pigmentosa with reduced penetrance 15 ; those patients with oscillations had faster cone a-wave implicit times than those without oscillations. The mechanisn that leads to loss of these oscillations remains to be defined. From the Berman-Gund Laboratory for the Study of Retinal Degenerations, Harvard Medical School, Massachusetts Eye and Ear Infirmary, Boston, Mass. This work was supported in part by National Eye Institute Project Grant EY00169 and in part by the National Retinitis Pigmentosa Foundation, Baltimore, Md., and the George Gund Foundation, Cleveland, Ohio. Submitted for publication May 21, Reprint requests: Dr. Michael A. Sandberg, Berman-Gund Laboratory, Massachusetts Eye and Ear Infirmary, 243 Charles St., Boston, Mass Key words: cone, a-wave, electroretinogram, latency, retinitis pigmentosa, outer segment. REFERENCES 1. Mizuno K and Nishida S: Electron microscopic studies of human retinitis pigmentosa. I. Two cases of advanced retinitis pigmentosa. Am J Ophthalmol 63:791, Kolb H and Gouras P: Electron microscopic observations of human retinitis pigmentosa, dominantly inherited. INVEST OPHTHALMOL 13:487, Szamier RB and Berson EL: Retinal ultrastructure in advanced retinitis pigmentosa. INVEST OPHTHAL- MOL Vis Sci 16:447, Szamier RB, Berson EL, Klein R, and Meyers S: Sex-linked retinitis pigmentosa: ultrastructure of the photoreceptors and pigment epithelium. INVEST OPHTHALMOL VIS SCI 18:145, Berson EL and Goldstein EB: Early receptor potential in dominantly inherited retinitis pigmentosa. Arch Ophthalmol 83:412, Berson EL and Goldstein EB: The early receptor potential in sex-linked retinitis pigmentosa. INVEST OPHTHALMOL 9:58, Tamai A: Studies on the early receptor potential in the human eye. III. ERP in primary retinitis pigmentosa. Yonago Acta Med 18:18, Haig C and Saltzman SL: Correlation of visual acuity and absolute luminance threshold in retinitis pigmentosa. Arch Ophthalmol 53:109, Berson EL, Gouras P, Gunkel RD, and Myrianthopoulos NC: Rod and cone responses in sex-linked retinitis pigmentosa. Arch Ophthalmol 81:215, Massof RW and Finkelstein D: Rod sensitivity relative to cone sensitivity in retinitis pigmentosa. IN- VEST OPHTHALMOL VIS SCI 18:263, Berson EL and Howard J: Temporal aspects of the electroretinogram in sector retinitis pigmentosa. Arch Ophthalmol 86:653, Francois J and de Rouck A: Behavior of ERG and EOG in localized retinal destruction by photocoagulation. In Clinical Electroretinography: Proceedings of the Third International Symposium, Burian HM and Jacobson JH, editors. Oxford, 1966, Pergamon Press, Inc., pp Berson EL, Gouras P, and Hoff M: Temporal aspects of the electroretinogram. Arch Ophthalmol 81:207, Weleber RG: The effect of age on human cone and rod ganzfeld electroretinograms. INVEST OPHTHAL- MOL VIS SCI 20:392, Berson EL, Gouras P, Gunkel RD, and Myrianthopoulos NC: Dominant retinitis pigmentosa with reduced penetrance. Arch Ophthalmol 81:226, 1969.