Nonarteritic anterior ischemic optic neuropathy (NAION) is an optic neuropathy believed to be caused by ischemia to the optic nerve head. Its pathophysiology is thought to be related to compromised circulation through the short posterior ciliary arteries due to episodes of hypotension (including nocturnal hypotension) and underlying vascular disease (hypertension, diabetes mellitus dyslipidemia). Eyes with small crowded optic discs with small or nonexistent cupping also seem to be at greater risk, likely due to mechanical obstruction leading to decreased blood flow (1).
Several medications also have been implicated in causing NAION, although evidence of true causation is often lacking (1). Many case reports have suggested a link between phosphodiesterase-5 (PDE-5) inhibitors (sildenafil, tadalafil, and vardenafil) and NAION, largely through an observed temporal relationship between use of these medications and subsequent development of NAION (2–8). These medications act by inhibiting the degradation of phosphodiesterase in the smooth muscle cells lining the corpus cavernosum of the penis, leading to accumulation of cyclic guanosine monophosphate, causing release of nitric oxide and subsequent vasodilation (9). The PDE enzyme is found in smooth muscle of many vascular tissues in the body, but the PDE-5 subtype, on which these drugs preferentially act, is found only in the corpus cavernosum. A different subtype, PDE-6, is found in retinal blood vessels and sildenafil and vardenafil have been shown to have selectivity for these receptors as well; this is likely the mechanism for the transient changes in color perception that are commonly recognized side effects of these medications (9). These medications are also being explored as therapies for pulmonary arterial hypertension, as they have been shown to have some vasodilatory effects on pulmonary vasculature (10). However, there is little evidence that they impact circulation at the optic nerve head or in the short posterior ciliary arteries (11,12).
Epidemiological studies have examined the association of PDE-5 inhibitors with NAION and found no association (13), and there are Phase-III clinical trials examining the ocular safety of sildenafil (14). Limitations exist in these reports such as short follow-up periods and lack of adjustment for confounding variables. Therefore, further study is warranted.
The purpose of this study was to use a health claims database to determine whether patients who have developed NAION had a greater likelihood than the general population of using PDE-5 inhibitors before development NAION.
We used the IMS Lifelink database as the data source for this study. Lifelink is a health claims database that captures prescription drug dispensing, physician visits, hospitalizations, and demographic information for approximately 68 million residents of the United States. The database captures 17% of males aged 45–54 years, 13% of males aged 55–64 years, and 8% of males aged >65 years. Information on all physician visits and hospitalizations are entered as International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. The data for this study were comprised approximately 1 million men randomly selected from the database followed from 2000 to 2011. The database has been subject to routine quality checks and has been used in previous pharmacoepidemiologic research (15,16). Institutional review board approval was obtained for this study.
Case and Control Definition
Cases were defined as those with the first ICD-9 diagnosis of NAION (ICD-9 code 377.41). The date of the first diagnosis of NAION (registration of code 377.41) was deemed the index date. We excluded patients with history or subsequent diagnosis of polymyalgia rheumatica and giant cell arteritis. For each case, all eligible controls with no history of NAION were identified and matched to the cases by age, calendar time, and index date of the case. Since a density-based sampling strategy was used for control selection, each control could have been used more than once. This type of sampling ensures that the odds ratio is a close approximation of the rate ratio (17). Demographic information was examined using descriptive statistics, and a conditional logistic regression model was built to examine rate ratios.
Demographic information was examined using descriptive statistics. Recent users of a PDE-5 inhibitor drug were defined by having 1 prescription of sildenafil, vardenafil, or tadalafil filled within 30 days of the index date. “Any use” was defined as a subject who had filled a prescription for a study anytime in the 1 year before the index date. We further stratified our analysis to the type of a PDE-5 inhibitor where a user was defined as a subject who had received at least 1 prescription of either sildenafil, vardenafil, or tadalafil within 1 year of the index date. A conditional logistic regression model was created to adjust for the following covariates: diabetes, statins, hypertension, myocardial infarction, and stroke. Adjusted rate ratios were computed with non-users of PDE-5 inhibitors (in the year before the index date) as the control group.
Within the initial cohort of 934,283 individuals, a total of 1,109 cases of NAION diagnosis were found and matched to 1,237,290 age-matched controls. The average age of both groups was 69.8 years; cases were more likely to have been taking medication for hyperlipidemia and were more likely to have been diagnosed with diabetes mellitus, hypertension, myocardial infarction, and cerebrovascular accident in the preceding year (P < 0.001) (Table 1).
The crude rate ratio for having a prescription filled of any PDE-5 inhibitor medications in the year before the NAION was 1.0, whereas the adjusted rate ratio was 1.01 (95% confidence interval [CI], 0.79–1.28) (Table 2). When stratified into the rate ratios for individual PDE-5 inhibitor medications, tadalafil showed a trend toward being protective of NAION (adjusted rate 0.74), and vardenafil showed a trend toward a more strong association (adjusted rate ratio 1.47); however, these trends did not reach statistical significance for any of the 3 medications (Table 2). Filling a prescription for any PDE-5 inhibitor in the 30 days before the event also had no significant association, with crude and adjusted rate ratios of 0.95 and 0.96, respectively (95% CI, 0.75–1.23).
The results of this large retrospective case–control study support the lack of association between NAION and PDE-5 inhibitor use. Similarly, Margo and French (13) conducted a large case–control study in a large cohort of 4 million veteran patients of 50 years or older in the United States from 2004 to 2005 and did not find an increase in the risk of NAION (rate ratio = 1.02; 95% CI, 0.92–1.12) with sildenafil users. The limitations of this study were its relatively short follow-up and lack of adjustment for potential confounding variables, which were addressed by our study with a longer follow-up period (12 years) and adjustment for potential confounders.
A collated review of over 14,000 cases of sildenafil demonstrated no increased incidence of NAION over the general population (18), which is in agreement with our findings. Similarly, Wirostko et al (14) examined the ocular safety of chronic sildenafil use in a 12-week, double-blind, randomized, placebo-controlled, Phase-III trial where 277 men with idiopathic pulmonary arterial hypertension were prescribed oral sildenafil (20, 40, or 80 mg) or placebo and followed for 18 months. Ocular adverse events were examined using ophthalmic examinations, visual function tests, patients' reports of adverse events, and adverse events reported by investigators. At the end of 18 months, none of the groups taking sildenafil were found to have NAION.
In contrast, a retrospective study of men with NAION was conducted in which men with NAION were asked about their PDE-5 inhibitor history; 671 men were found to have NAION without a history of PDE-5 inhibitor use, whereas 43 did have definite NAION and PDE-5 inhibitor use in the 1-day preceding it. This study found an odds ratio of 2.15 for PDE-5 inhibitor use in NAION (95% CI, 1.06–4.34; P = 0.033) (19).
Although our data show a trend toward varying relationships of each PDE-5 inhibitor medication with NAION (adjusted rate ratio for tadalafil of 0.74, for vardenafil of 1.47), none of these associations reached statistical significance.
There are 2 principle mechanisms proposed by which PDE-5 inhibitors may cause NAION: first, they cause systemic hypotension and second, they cause impaired local autoregulation at the optic nerve head in the short posterior ciliary arteries (20). However, neither of these have been conclusively proven or even supported by animal models and human trials (11,12,20).
The only studies present in the literature that have suggested a causal relationship are case reports, with no other clinical studies that have confirmed this apart from the retrospective study noted above. To the contrary, clinical studies have examined the effect of PDE-5 inhibitor use on ocular blood flow, including flow at the short posterior ciliary arteries to the optic nerve head, and these have found either no change in blood flow or the opposite of increased flow (11,12).
The most convincing causal relationship established in the literature is a rechallenge in 1 patient who reported transient monocular visual field defects on 4 occasions followed by a persistent monocular visual field defect, each occurring within hours of taking tadalafil; the patient was diagnosed with NAION when he sought medical attention for the persistent defect (3). In a series of 10 patients who developed sequential NAION while on PDE-5 inhibitors, Galvez-Ruiz and Arishi (21) were not able to confirm or exclude whether it was the PDE-5 inhibitor, or other risk factors, that most likely contributed to sequential NAION. No other evidence in the literature supports the association so strongly, and most patients reported in the other case reports had 1 or more established risk factors for NAION (2–6,8).
As with all observational studies, our report is subject to limitations. As with all studies using administrative data, we could only assess PDE-5 inhibitor use through prescription drug dispensation in the database, which does not necessarily correlate with use and certainly does not indicate the timing of use, given that PDE-5 inhibitors are not chronic daily-use medications. Nevertheless, with such a large number of patients included, it can be inferred that filling a prescription in the 1 month or 1 year before the event should correlate with use of the medication in most patients. In addition, although our study was adequately powered to show no association with PDE-5 inhibitors and NAION, it was underpowered for the association with specific PDE-5 inhibitor medications. Finally, due to the nature of the data collected through the database, medications purchased out of insurance plans were not captured in the analysis, either for the case or control groups.
In summary, we found no association between having a prescription filled for a PDE-5 inhibitor medication and receiving a diagnosis code of NAION. Although PDE-5 inhibitors are known to affect retinal circulation and thereby cause transient changes in color perception, there is no epidemiological association with NAION, nor is there any support in the literature for the proposed mechanisms of their adverse effect on the circulation of the optic nerve head.
STATEMENT OF AUTHORSHIP
Category 1: a. Conception and design (N. A. Nathoo, F. S. Mikelberg, M. Etminan); b. Acquisition of data (M. Etminan, F. S. Mikelberg); c. Analysis and interpretation of data (M. Etminan, F. S. Mikelberg, N. A. Nathoo). Category 2: a. Drafting the article (M. Etminan, F. S. Mikelberg, N. A. Nathoo); b. Revising it for intellectual content (M. Etminan, F. S. Mikelberg, N. A. Nathoo). Category 3: a. Final approval of the completed article (M. Etminan, F. S. Mikelberg, N. A. Nathoo).
1. Kerr NM, Chew SS, Danesh-Meyer HV. Non-arteritic anterior ischaemic optic neuropathy: a review and update. J Clin Neurosci. 2009;16:994–1000.
2. Akash R, Hrishikesh D, Amith P, Sabah S. Case report: association of combined nonarteritic anterior ischemic optic neuropathy (NAION) and obstruction of cilioretinal artery with overdose of Viagra. J Ocul Pharmacol Ther. 2005;21:315–317.
3. Bollinger K, Lee MS. Recurrent visual field defect and ischemic optic neuropathy associated with tadalafil rechallenge. Arch Ophthalmol. 2005;123:400–401.
4. Boshier A, Pambakian N, Shakir SA. A case of nonarteritic ischemic optic neuropathy (NAION) in a male patient taking sildenafil. Int J Clin Pharmacol Ther. 2002;40:422–423.
5. Escaravage GK Jr, Wright JD Jr, Givre SJ. Tadalafil associated with anterior ischemic optic neuropathy. Arch Ophthalmol. 2005;123:399–400.
6. Pomeranz H, Bhavsar A. Nonarteritic ischemic optic neuropathy developing soon after use of sildenafil (Viagra): a report of seven new cases. J Neuroophthalmol. 2005;25:9–13.
7. Wooltorton E. Visual loss with erectile dysfunction medications. CMAJ. 2006;175:355.
8. Pomeranz HD, Smith KH, Hart WM Jr, Egan RA. Sildenafil-associated nonarteritic anterior ischemic optic neuropathy. Ophthalmology. 2002;109:584–587.
9. Wright PJ. Comparison of phosphodiesterase type 5 (PDE5) inhibitors. Int J Clin Pract. 2006;60:967–975.
10. Ghofrani HA, Voswinckel R, Reichenberger F, Olschewski H, Haredza P, Karadas B, Schermuly RT, Weissmann N, Saegar W, Grimmigel F. Differences in hemodynamic and oxygenation responses to three different phosphodiesterase-5 inhibitors in patients with pulmonary arterial hypertension: a randomized prospective study. J Am Coll Cardiol. 2004;44:1488–1496.
11. Grunwald JE, Siu KK, Jacob SS, Dupont J. Effect of sildenafil citrate (Viagra) on the ocular circulation. Am J Ophthalmol. 2001;131:751–755.
12. Harris A, Kagemann L, Ehrlich R, Ehrlich Y, Lopez CR, Purvin VA. The effect of sildenafil on ocular blood flow. Br J Ophthalmol. 2008;92:469–473.
13. Margo CE, French DD. Ischemic optic neuropathy in male veterans prescribed phosphodiesterase-5 inhibitors. Am J Ophthalmol. 2007;143:538–539.
14. Wirostko BM, Tressler C, Hwang LJ, Burgess G, Laites AM. Ocular safety of sildenafil citrate when administered chronically for pulmonary arterial hypertension: results from phase III, randomised, double masked, placebo controlled trial and open label extension. BMJ. 2012;344:e554.
15. Bird S, Etminan M, Brophy J, Hartzema AG, Delaney JA. Risk of acute kidney injury associated with use of fluoroquinolones. CMAJ. 2013;185:E475–482.
16. Bird ST, Brophy JM, Hartzema AG, Delaney JA, Eitminan M. Male breast cancer and 5-alpha reductase inhibitors, finasteride and dutasteride. J Urol. 2013;190:1811–1814.
17. Rodrigues L, Kirkwood BR. Case-control designs in the study of common diseases: updates on the demise of the rare disease assumption and the choice of sampling scheme for controls. Int J Epidemiol. 1990;19:205–213.
18. Giuliano F, Jackson G, Montorsi F, Martin-Morales A, Raillard P. Safety of sildenafil citrate: review of 67 double-blind placebo-controlled trials and the postmarketing safety database. J Sex Med. 2014;11:885–887.
19. Pfizer. Case Crossover Study of PDE5 Inhibitor Exposure as a Potential “Trigger Factor” for Acute NAION. In: ClinicalTrials.gov [Internet]. Bethesda, MD: National Library of Medicine (US), 2000. Available at http://www.clinicaltrials.gov/ct2/show/NCT00759174
. NLM Identifier: NCT00759174. Accessed May 5, 2014.
20. Fraunfelder FW, Pomeranz HD, Egan RA. Nonarteritic anterior ischemic optic neuropathy and sildenafil. Arch Ophthalmol. 2006;124:733–734.
21. Galvez-Ruiz A, Arishi N. Sequential, non-arteritic anterior ischemic optic neuropathy in patients taking sildenafil: a report of ten cases. Saudi J Ophthalmol. 2013;27:241–246.