Oscillatory Potentials of the Electroretinogram in Hypertensive Patients
Abstract Because alteration of oscillatory potentials of the electroretinogram has been described in diabetic patients without signs of diabetic retinopathy as an early marker of changes in microcirculation, we studied the behavior of these potentials in patients with early-onset hypertension. Electroretinograms were recorded in 24 subjects with essential hypertension (blood pressure >140/90 mm Hg) and in 9 age-matched normotensive control subjects (blood pressure <140/90 mm Hg). Diabetes and ocular diseases were considered exclusion criteria. Sitting blood pressure was measured by a single investigator with a mercury sphygmomanometer after each subject had been at rest for 10 minutes. Funduscopic changes in all subjects did not exceed stage I World Health Organization classification. The oscillatory index was calculated by adding waves O1, O2, and O3 within the b wave of the electroretinogram. Statistical analysis was performed with Student’s t test for paired and unpaired data and linear regression. The oscillatory index was significantly reduced in hypertensive patients compared with normotensive subjects. An inverse relationship was observed when systolic and diastolic blood pressures were plotted against the oscillatory index. In conclusion, our data demonstrate that the electrical activity of the retina is altered early in the course of hypertension and that the influence of systolic pressure on the oscillatory index is greater than that of diastolic pressure.
Morphological alterations of the retinal vessels in hypertensive patients have been recognized and classified since 1939, when they were reported by Keith, Wagener, and Barker.1 Such alterations can be easily examined by funduscopy, but a strict relationship between retinal features and blood pressure levels can be demonstrated only for the most severe changes.2
The oscillatory potentials (OPs) are a series of three to six wavelets superimposed on the ascending phase of the b wave of the electroretinogram (ERG). Their origin is still not well known, but they seem to be generated in the inner retinal layers, which are supplied by the retinal vascular system.3 4 5
OP reduction has been clearly demonstrated in severe forms of angiopathies such as venous and arterial occlusion. Recently, OP amplitude reduction has also been detected in diabetic patients without signs of diabetic retinopathy, and it has been considered an early marker of microcirculation alterations and a predictor of progression to overt diabetic retinopathy.6 7 8 The ERG has been used as a noninvasive method of studying the retinal microvasculature in humans.9 Although data obtained in patients and in animal models suggest that microvascular alterations are detectable even in early stages of hypertension,10 11 OPs in patients with essential hypertension have been little investigated in this situation.
The aim of the present study was to compare the OPs of subjects with mild or moderate untreated hypertension who do not have characteristic funduscopic alterations of hypertensive retinopathy with the OPs of normotensive age-matched control subjects.
The hypertensive study population was composed of 24 subjects (14 men and 10 women, aged 19 to 57 years [mean age, 41 years]) with mild or moderate untreated hypertension (systolic blood pressure [SBP], 146±15 mm Hg; diastolic blood pressure [DBP], 95±10 mm Hg), recruited from a hospital hypertension clinic. Criteria for entry into the study included mild or moderate essential hypertension, no prior antihypertensive pharmacological treatment, alterations of fundus oculi not exceeding grade 1 according to the World Health Organization classification, normal visual acuity or mild refractive errors, and normal glucose tolerance. Secondary hypertension was excluded by means of clinical history, physical examination, and laboratory and instrumental tests.
The control group consisted of nine healthy age-matched normotensive subjects (five men and four women, aged 28 to 50 years [mean age, 43 years]; blood pressure <140/90 mm Hg).
Sitting blood pressure was measured by a single investigator with a mercury sphygmomanometer after each subject had been at rest for 10 minutes. Neither hypertensive nor normotensive subjects received drugs affecting the autonomic nervous system or peripheral circulation.
Each subject was informed of the objective and the methods of the study and gave written consent. The ethics committee of the hospital gave its approval to the study design.
The ERG was recorded after a 15-minute mesopic retinal adaptation. The subjects’ pupils were dilated with 1% tropicamide. The ERG stimuli for eliciting OPs were white flashes from a photostimulator installed at a distance of 0.3 m from the examined eye in a Ganzfeld bowl (Lace Electronica Erev 85). The frequency of the stimulus was 0.5 Hz and the intensity was 1 J.
Monopolar contact lens electrodes (Ag/AgCl) were used after local anesthesia (oxibuprocaine 0.4%); the reference electrode was located in the center of the forehead. An additional electrode was located in the mastoid area. The signal was amplified with a bandwidth of 100 to 250 Hz. Analysis time was 0.1 second. Twenty events were recorded from both eyes. The first three components of the ERG were identified and labeled O1, O2, and O3 and their amplitudes were measured. The oscillatory index (OI) was calculated by adding the amplitudes of waves O1, O2, and O3. The OI of each patient was expressed as the mean of the recorded events.
Data were analyzed by use of the computer package spss/pc. All data are reported as mean±SD. The intragroup comparison of OPs recorded from the right eye with those recorded from the left was performed with Student’s t test for paired data. The intergroup comparison was performed with Student’s t test for unpaired data. Blood pressure and OIs were plotted using linear regression.
The intragroup comparison of OIs recorded from the left eye with those recorded from the right showed no difference. Therefore, only the OI recorded from the right eye of each patient was used in further analysis.
OIs were significantly lower in hypertensive subjects than in normotensive control subjects ([78±9] · 10−6 V versus [102±19] · 10−6 V; P=.002) (Fig 1⇓). An inverse linear regression was demonstrated by plotting OI with SBP (r=−.67, P=.0001) and with DBP (r=−.51, P=.002) (Figs 2⇓ and 3⇓).
The results of this study demonstrate that OI is reduced early in subjects with essential hypertension. Previous studies evaluated OPs only in moderately to severely hypertensive subjects with a long history of hypertension in whom funduscopic alterations of the retina (eg, abnormal arteriovenous crossing or narrowing of arterioles) were already detectable.12 13 The present study demonstrates that retinal electric function in hypertensive subjects is also affected before the onset of the morphological changes peculiar to hypertensive retinopathy.
Linear regression analysis indicates that SBP appeared to strongly influence the OI, whereas the effect of DBP was less impressive. It is a matter of fact that SBP is more closely linked to neurological complications of hypertension (eg, atherothrombotic brain infarction) than are other components of blood pressure (DBP, pulse pressure).14 15 Retinal electric potentials are the detectable manifestation of eye neuronal activity. Our study confirms the negative influence of increased SBP on the central nervous system.
Retinal vascular alterations may provide a clue to the status of various systems and organs of the body, especially the central nervous system, the cardiovascular system, and the kidneys. The tissue reactions of the retina and optic nerve to vascular changes are similar to those of the central nervous system. In 1898, Gunn reported retinal vascular changes in patients with cerebrovascular insufficiency and renal diseases. Since then, many physicians, including Keith, Wagener, and Scheie, have examined the retinal vascular changes of systemic hypertension.16 Various forms and stages of hypertensive diseases affect retinal, choroidal, and optic nerve circulations differently. Because the retinal circulatory changes are most readily and easily seen by fundus oculi examination, much attention has been focused on the retinal alterations in hypertension. Although some of the vascular changes in the retina can be found in the normal population, they are more frequent in hypertensive patients; in both populations, changes are age dependent. Furthermore, perception of variations in retinal architecture is somewhat observer dependent. Therefore, not all authors agree on the role of funduscopy in the evaluation of prognosis and target-organ damage in hypertension. A number of classifications of hypertensive retinopathy have been proposed; in the past, classifications such as those of Keith, Wagener, and Barker1 had prognostic significance. The subsequent development of new classes of antihypertensive drugs altered the clinical course of hypertensive retinopathy; many of the advanced funduscopic changes in hypertension are less frequently seen in clinical practice.
In the retinal vasculature, the vascular tone is autoregulated by pacemaker cells of the vessel wall; the muscular tone varies in response to changes in perfusion pressure and intraocular pressure to provide a constant blood flow. Metabolic changes, the partial pressure of oxygen, and blood pH also influence vascular reactivity; aging and hyperglycemia also impair vascular reactivity.16
Although the exact site of origin of OPs in the retina is still a matter of debate, their relationship with retinal circulation is clearly demonstrated.3 4 5 Their amplitude is reduced in all severe forms of angiopathy, for example in venous and arterial occlusion of the retina and in pulseless disease (Takayasu’s disease)17 ; also, OPs appear modified in less severe forms of retinal angiopathy such as the early stages of diabetic retinopathy.7 8 Not only is the amplitude of single waves or their sum (OI) reduced, but also latency is increased.
OPs seem to be generated in an area between the inner nuclear layer and the amacrine-ganglion cell layer; because these retinal structures are supplied by the retinal vascular system, they are possibly affected in hypertension even in its early stages, considering that microvascular dynamics is soon altered in hypertension.11 Some studies in spontaneously hypertensive rats clearly demonstrated that two mechanisms produce an increase in microvascular resistance: a decreased internal diameter of the arterioles and the rarefaction of arterioles and capillaries.10 Few studies have been performed in the microcirculation of humans with essential hypertension, and most of these were done by noninvasive methods. Therefore, these studies have been restricted to the microvasculature of the skin and of the bulbar conjunctiva.18 19 Some morphological studies on muscle microcirculation have been performed by use of muscle biopsy.20 These studies confirmed that in subjects with mild or borderline hypertension, similar microvascular abnormalities can be found.
The OI may be a useful alternative to funduscopy in evaluating changes in small vessels of the central nervous system. This method permits a quantitative and objective evaluation of retinal electrical activity; funduscopic evaluation, by contrast, is qualitative and subjective, with possible interobserver variations.
Moreover, it would be interesting to evaluate with a long-term study whether these functional changes can be reversed by antihypertensive treatment in the absence of morphological alterations. Thus, the OI could be used as a marker of the effect of treatment on target-organ function.
Reprint requests to Prof G. Bellini, MD, Istituto di Medicina Clinica dell’Università di Trieste, c/o Ospedale di Cattinara, Strada di Fiume 447, 34149 Trieste, Italy.
Simonsen SE. Prognostic value of ERG (oscillatory potential) in juvenile diabetics. Acta Ophthalmol. 1975;123(suppl):223-224.
Zweifach BW, Kolvacheck S, De Lano FA, Chen PCY. Micro- pressure-flow relationships in a skeletal muscle of spontaneously hypertensive rats. Hypertension. 1981;2:601-614.
Sullivan JM, Prewitt RL, Josephs BS. Attenuation of the microcirculation in young patients with borderline hypertension. Hypertension. 1983;5:844-851.
Eichler J, Stave J, Boehm J. Oscillatory potentials in hypertensive retinopathy. Doc Ophthalmol. 1984;40:161-165.
Mueller W, Gauss J, Spittel U, Dueck H. Oscillatory potentials in cases of systemic hypertension. Doc Ophthalmol. 1984;40:167-171.
Kannel WB, Dawber TR, McGee DL. Perspectives on systolic hypertension: the Framingham study. Circulation. 1980;61:1179-1182.
Tso MOM, Jampol LM. Hypertensive retinopathy, choroidopathy, and optic neuropathy of hypertensive disease. In: Laragh JH, Brenner BM, eds. Hypertension: Pathophysiology, Diagnosis, and Management. New York, NY: Raven Press Publishers; 1990:433-465.
Draaijer P, Cosenzi A, de Leeuw P, Leunissen K. Nailfold capillary density in sodium sensitive and sodium resistant borderline hypertensive patients. Am J Hypertens. 1992;5:10a. Abstract.
Heinrich HA, Romen W, Heimgartner W, Hartung E, Baumer F. Capillary rarefaction characteristic of the skeletal muscle of hypertensive patients. Clin Wochenschr. 1988;66:54-60.