A difference of opinion
Retinal side-effects of sildenafil (Lancet Vol 353 1999)
Michael A Vobig, Theo Klotz, Michael Staak, Karl Ulrich Bartz-Schmidt, Udo Engelmann, Peter Walter
See Commentary page 340
Sildenafil is an effective oral treatment for erectile dysfunction.1 It acts as a relatively specific inhibitor of the type 5 phosphodiesterase isozyme (PDE5) in the corpus cavernosum.2,3 In-vitro studies have shown that PDE6, found predominantly in retinal photoreceptors,4 is a key enzyme in the phototransduction cascade,
since it is also inhibited by sildenafil. The inhibition efficacy is about 1/10 of that for PDE5.2 Although patients in studies have occasionally complained about visual disturbances after sildenafil,1 no data are available on visual or retinal function after sildenafil.
We studied the effect of a single oral dose of 100 mg sildenafil (Viagra) in five healthy volunteers on visual acuity, visual field, colour vision, intraocular pressure, anterior segment, fundus appearance, and electrophysiological tests (electroretinogram [ERG], visual evoked potentials [VEP]), which are well
established as objective and sensitive tests of the visual system.5 All tests were done immediately before medication, and 1 and 6 hours after. Sildenafil plasma concentrations were measured 1, 3, and 6 hours after
the dose.
Time after medication (hours)
0
1
6
Visual acuity (ETDRS-charts)
1·4 (0·3)
1·5 (0·3)
1·5 (0·3)
Intraocular pressure (IOP) (mm Hg)
17·4 (1·4)
15·8 (1·5)
16·2 (1·5)
Visual field, mean defect
4·5 (2·1)
4·3 (1·9)
3·5 (3·6)
VEP, amplitude P100-N80 (µV)
4·2 (1·0)
3·3 (1·8)
3·1 (0·3)
ERG, saturated a-wave
Amplitude (µV)
-110·4 (28·9)
-69·6 (20·6)
-109·6 (17·2)
Implicit time (ms)
22·4 (0·5)
21·8 (3·3)
22·8 (0·6)
ERG, saturated b-wave
Amplitude (µV)
210·6 (68·6)
161·6 (57·1)
226·2 (48·7)
Implicit time (ms)
42·8 (0·7)
44·0 (1·2)
44·2 (1·3)
Mean (SD) n=5.
Effects of sildenafil on retinal function
There was no significant effect of sildenafil on visual acuity, visual field, colour vision, intraocular pressure, or visual evoked potentials (table). None of the volunteers complained about visual disturbances after sildenafil. The anterior and posterior segments of the eye showed no abnormalities after drug administration.
ERG measurements showed significant changes: the a-wave amplitude of the saturated response was reduced to 63% (p=0·0015, t-test) 1 hour after sildenafil and was completely recovered 6 hours after. The
b-wave amplitude of the saturated response showed a reduction to 77% (p=0·04, t-test) 1 hour after drug administration and a complete recovery 6 hours after. Scotopic and photopic b-wave amplitudes, and
amplitudes of the oscillatory potential OP2 showed a small, insignificant reduction after 1 hour and complete recovery after 6 hours. The ERG implicit times remained unaffected. The ERG reduction correlated well with the sildenafil plasma concentration which showed its peak 1 hour after administration.
Though visual acuity, visual field, and the VEP remained unchanged, a single oral dose of 100 mg sildenafil induced an a-wave and b-wave reduction in the saturated response of the ERG. PDE6 inhibition in rod
photoreceptors seemed to be the most likely mechanism of sildenafil-associated retinal dysfunction.
Fortunately, the ERG changes showed a complete recovery. However questions remain: will recovery also be seen in older patients or those with pre-existing abnormalities and retinal disease, such as in patients with PDE mutations leading to receptor degeneration, and do repeated applications of sildenafil cause prolonged
or further retinal dysfunction? From the data presented here, it seems that retinal function in patients receiving sildenafil should be monitored by clinical and electrophysiological tests.
Departments of Ophthalmology (M A Vobig e-mail: vobig@uni-koeln.de), Urology, and Forensic
Medicine, University of Cologne, D50931 Köln, Germany
And the commentary:
No cause for alarm over retinal side-effects of
sildenafil
Are the visual symptoms of which some patients taking sildenafil complain an alarming side-effect of the drug? In this issue of The Lancet, Michael Vobig and colleagues report a decrease in the a-wave and b-wave amplitude in the electroretinogram (ERG) of five healthy men 1 h after they took 100 mg sildenafil citrate (Viagra) orally. The changes had completely disappeared 5 h later. Results of all other electrophysiological and clinical tests (visual acuity, visual field, colour vision, intraocular pressure) were normal.
Phosphodiesterase type 6 is required for the transformation of light into electrical signals; therefore, retinal effects of hosphodiesterase inhibitors are not unexpected and are well known from studies in cats1,2 and in human beings.2-4 Since sildenafil, which is mainly an inhibitor of phosphodiesterase type 5, also weakly
inhibits phosphodiesterase type 6, high plasma concentrations of sildenafil can be expected to inhibit the phototransduction process in retinal photoreceptors, if the drug passes the blood-retina barrier. Tests by Pfizer in anaesthetised dogs5 have shown that sildenafil transiently increases the implicit times of the
a-waves and b-waves and reduces the amplitude of the a-wave at threshold plasma concen-trations that were four times higher than those achieved after the recommended maximum therapeutic dose in human
beings. These effects were reversible and declined in a manner consistent with the plasma half-life (4 h) of sildenafil. McNemar's test of the results of a pilot randomised double-blind placebo-controlled crossover study (protocol 148-232, report of Nov 25, 1997, Pfizer) in eight volunteers who took 200 mg sildenafil did not reveal statistically significant effects of sildenafil on cone-mediated or rod-mediated ERG responses.
Can there be an explanation for the discrepancy between these findings and those that Vobig and colleagues found with only 100 mg of the drug? First, there are methodological questions. Vobig and colleagues' report does not make clear whether the protocol satisfied every important specification made by the International
Standard for Clinical Electrophysiology (ISCEV)6 for ERGs--eg, fixed duration of dark adaptation (at least 25 min), full dilatation of the pupil by mydriatic agents, and stimulation by standardised light flashes--because although their test light produced a fully developed ("saturated") a-wave, the amplitudes of the a-wave show less than half the amplitude of scotopic a-waves recorded according to the ISCEV standard.7
Second, mydriatic agents generally used for ERG have a short duration of action. If the investigators had assumed that the drops applied before the recording of the control ERG were still active and did not reapply them before the recording 1 h after ingestion of sildenafil, the ERG amplitude would be reduced, as observed.
However, the reduction would be due to a narrowing of the pupil,8 not to sildenafil. The drops would certainly have been reapplied for the final test 6 h after ingestion of sildenafil, so the response then would be the same as that obtained before sildenafil.
Nevertheless, even if the ERG changes were confirmed to be due to sildenafil, they are not alarming. The drop in b-wave amplitude correlates with only a very weak loss in light sensitivity of less than 0·2 log units, which is similar to the light-absorbing effect of a car windshield. No sensitivity decrease was observed in the
visual field data and all effects were fully reversible. Such an effect would not be deemed toxic (as long as the individual has a normal healthy retina) but would be taken to be the expected reversible effect of retinal phosphodiesterase, probably in principle not different from the effect in sexual organs. Studies in dogs that
received 65 times the maximum recommended dose for human beings daily for 12 months indicate that repeated exposure of human beings to therapeutic doses of sildenafil are unlikely to impair retinal function or alter retinal morphology .
Thus, Vobig and colleagues' findings do not imply that sildenafil taken in therapeutic doses and not repeated excessively increases the risk of visual impairment for healthy individuals with normal retinas. Nor do they imply that patients taking sildenafil should have their retinal function monitored. People with genetic phospho-diesterase defects, which are very rare, will know from an early age of their retinal impairment, and they should be advised not to take sildenafil.9 More extensive and carefully controlled studies are necessary to clarify the mode of action of sildenafil on retinal function and to study the long-term effects of the drug in
patients with retinal diseases.
Eberhart Zrenner
University Eye Hospital Tübingen, Department of Pathophysiology of Vision and Neuro-Ophthalmology,
D-72076 Tübingen, Germany |
