Aspirin Use and the Prevention of Acute Ischemic Cranial Nerve Palsy LENWORTH N. JOHNSON, MD, STEVEN W. STETSON, MD, GREGORY B. KROHEL, MD, CHERYL L. CIPOLLO, COA, AND RICHARD W. MADSEN, PHD

● PURPOSE:

To assess the relationship of aspirin use and ischemic cranial nerve palsies among patients with diabetes mellitus and hypertension. ● METHODS: This retrospective case-control study involved 100 patients with ischemic cranial nerve palsies in association with diabetes, hypertension, or both (palsy cases) and 163 age-matched and sex-matched patients with diabetes, hypertension, or both but without ischemic cranial nerve palsies (nonpalsy control subjects). Comparisons were made with respect to duration of diabetes, dose and duration of aspirin use, dose and duration of tobacco use, and presence of cardiac or cerebrovascular disease. ● RESULTS: There were 20 oculomotor, 33 trochlear, 37 abducens, and 10 facial nerve palsy cases. The median duration of diabetes was 6 years for cases and 7 years for control subjects. There were 34 cases (34%) who had used aspirin for a mean duration of 5.5 years before the onset of the cranial nerve palsy and 49 control subjects (30.1%) who had used aspirin for a mean duration of 4.3 years. There were no significant differences between cases and control subjects for duration of diabetes (P ⴝ .94); aspirin use (P ⴝ .51), duration (P ⴝ .50), and dosage (P ⴝ .89); tobacco use (P ⴝ .73) and consumption (P ⴝ .45); and proportion of cardiac disease (P ⴝ .17). Cerebrovascular disease was significantly less common among palsy cases than nonpalsy control subjects (P < .001). There was no significant difference in the odds of a patient having cranial nerve palsy in the aspirin group compared with the nonaspirin group (odds ratio, 1.12; 95% confidence interval, 0.70 –2.04).

Accepted for publication Sept 20, 1999. From the Neuro-ophthalmology Unit, Mason Eye Institute (Dr Johnson), and Biostatistics Unit (Dr Madsen), University of Missouri– Columbia, Columbia, Missouri; and Neuro-Ophthalmology Unit, Lions Eye Institute, Albany Medical Center, Albany, New York (Drs Stetson and Krohel and Ms Cipollo). This study was supported in part by an unrestricted grant to the Department of Ophthalmology at the University of Missouri–Columbia, from Research to Prevent Blindness, Inc, New York, New York. Reprint requests to Lenworth N. Johnson, MD, Neuro-ophthalmology Unit, Mason Eye Institute, University of Missouri–Columbia, Columbia, MO 65212; fax: (573) 884-0346; e-mail: [email protected] 0002-9394/00/$20.00 PII S0002-9394(99)00362-1

©

2000 BY

● CONCLUSION: Aspirin use was not associated with a reduced rate of ischemic third, fourth, sixth, and seventh nerve palsies among patients with diabetes mellitus and hypertension. Aspirin appears to be ineffective in preventing ischemic third, fourth, sixth, and seventh cranial nerve palsies. Patients with ischemic cranial nerve palsy have a significantly lower rate of strokes and transient ischemic attacks than patients who have diabetes or hypertension but who do not have a history of cranial nerve palsy. (Am J Ophthalmol 2000;129:367–371. © 2000 by Elsevier Science Inc. All rights reserved.)

C

EREBROVASCULAR

DISEASE,

WHICH

INCLUDES

transient ischemic attacks and strokes, affects up to 1.0 per 1,000 people each year in the United States.1 As a result, stroke is the third leading cause of death in this country, ranking behind only cardiac disease and cancer. Known risk factors for stroke include hypertension, diabetes mellitus, cigarette smoking, and advancing age.2 Patients with diabetes mellitus, hypertension, or both may develop diplopia from isolated oculomotor (third), trochlear (fourth), or abducens (sixth) nerve palsies, and facial paralysis from facial (seventh) nerve palsy.3– 8 Patients with diabetes have a 10-fold increase in the incidence of ischemic cranial nerve palsies, with an incidence of 1% among diabetics compared with an incidence of 0.1% for the nondiabetic population.9 Although the incidence of ischemic cranial nerve palsies among patients with hypertension is unknown, hypertension is often associated with this disorder.4 – 6 Aspirin’s contribution to the prevention and treatment of stroke both acutely and prophylactically has been recognized in the literature for years.10 –17 Aspirin blocks the conversion of arachidonic acid to prostacyclin and thromboxane A2, thereby inhibiting vasoconstriction and platelet aggregation.16 Antiplatelet effects are evident within 40 minutes, and the platelet inhibitory effect lasts approximately 7 to 9 days, the life of the platelet. The Antiplatelet Trialists’ Collaboration study reported a 30% reduction in stroke risk and a 15% decrease in the risk of cerebrovascular deaths.10 The Swedish Aspirin Low-Dose Trial showed an 18% reduction in stroke or death in the

ELSEVIER SCIENCE INC. ALL

RIGHTS RESERVED.

367

group that received 75 mg of aspirin daily as compared with placebo.15 Consequently, the Swedish Aspirin Low-Dose Trial suggested that low-dose aspirin may give adequate antiplatelet activity while minimizing the gastrointestinal side effects often seen at higher doses. Other studies caution against the indiscriminant use of aspirin when the cause of stroke is unknown, citing an increased morbidity and mortality in hemorrhagic stroke.14,15 Although the evidence of aspirin’s benefit in cerebral ischemic stroke prophylaxis is well documented, little is known as to its specific role in the management of ischemic cranial nerve palsies. In the following retrospective case-control study, we assessed the relationship of aspirin use and ischemic cranial nerve palsies among patients with diabetes and hypertension.

smokers, former smokers, and nonsmokers. Medical history of cardiac disease (angina, myocardial infarction, cardiac arrhythmia, coronary artery disease) and cerebrovascular disease (transient ischemic attack or stroke) was also analyzed. Descriptive statistics were used to characterize the palsy cases and nonpalsy control groups. The chi-square test was used to compare the groups on categorical variables. Because the quantitative variables that were compared had skewed distributions, the Wilcoxon rank sum test was used to compare the groups in these cases. To assess the effects of multiple variables, logistic regression was done with palsy (yes/no) as the response variable. Aspirin use (yes/ no), an indicator for cardiac and/or cerebrovascular disease (yes/no), and smoking (current, former, or never) were the predictor variables. For the smoking variable, the response “never” was used as the baseline condition, and indicator variables for being current or former smokers were defined. This allowed an examination of the effect of the variable “aspirin use” on the estimated probability of being in the palsy group while accounting for cardiac and/or cerebrovascular disease and smoking behavior. The HosmerLemeshow test was used to assess the fit of the model. The overall rate of aspirin use was 83 (31.6%) of 263 patients. We expected to find a difference in the rate of aspirin use between cases and control subjects if aspirin use resulted in a decreased frequency of ischemic cranial nerve palsies. In an effort to assess the power to detect a difference with these sample sizes, we considered the following scenario. Assuming a 25% rate of aspirin use for the 100 palsy cases and 40% aspirin use for the 163 nonpalsy control subjects, the power to detect a difference in aspirin use between cases and control subjects was 80% (␤ ⫽ 0.2), with one-sided chi-square test at a significance of P ⬍ .05.

PATIENTS AND METHODS THE RECORDS OF 125 CONSECUTIVE PATIENTS (PALSY

cases) older than 45 years with ischemic cranial nerve palsies associated with diabetes or hypertension were reviewed. All patients had undergone neuro-ophthalmologic examination by one of the authors (L.N.J. or G.B.K.) between 1986 and 1998. The cranial nerve palsies included oculomotor (third), trochlear (fourth), abducens (sixth), and facial (seventh) nerve palsies. All patients had diplopia or unilateral facial paralysis from isolated cranial nerve palsies that spontaneously arose and resolved within 4 months of onset. For each palsy case with diabetes only, we attempted to identify two sex-matched and agematched (within ⫾ 3 years) nonpalsy control subjects with diabetes only who had been evaluated in our practices during the same 12-year time period. Similar matching of cases and control subjects with respect to sex and age was performed for patients with hypertension only and for patients with both hypertension and diabetes. Patients with multiple sclerosis, myasthenia gravis, temporal arteritis, intracranial tumor, pseudotumor cerebri, history of childhood strabismus, clinically significant head trauma, or degenerative neurologic disorders such as progressive supranuclear palsy or chronic progressive external ophthalmoplegia were excluded from both the palsy and nonpalsy groups. Approval from the institutional review boards was exempted, because this was a retrospective study of patients examined by the authors in their clinics. Successful sex and age matches were obtained for 100 palsy cases with 163 nonpalsy control subjects. The cases and control subjects were compared with regard to the duration of diabetes. Because the duration of hypertension was recorded only infrequently in our medical records, this information was not assessed for patients with hypertension. The dose and duration of aspirin use between palsy cases and nonpalsy control subjects were analyzed. Other risk factors analyzed included the dose and duration of tobacco use. Tobacco users were divided into current 368

AMERICAN JOURNAL

RESULTS TABLE 1 SUMMARIZES THE RESULTS. THE MEAN AGE ⫾ SD OF

67.8 ⫾ 8.9 years, median age of 67 years, and age range of 45 to 93 years for palsy cases were similar, as expected, to the mean age of 67.2 ⫾ 8.9 years, median age of 67 years, and age range of 47 to 91 years for nonpalsy control subjects. Of the 100 cases, 23 (15 men, eight women) had diabetes only, 41 (23 men, 18 women) had hypertension only, and 36 (19 men, 17 women) had both diabetes and hypertension. Of the 163 control subjects, 40 (26 men, 14 women) had diabetes only, 66 (33 men, 33 women) had hypertension only, and 57 (30 men, 27 women) had both diabetes and hypertension. Information on the duration of diabetes was available for 56 (95%) of 59 cases and 96 (99%) of 97 control subjects with diabetes. The median duration of diabetes of 6 years (range, 3 days to 30 years) for cases was not different from the median duration of 7 years (range, 4 months to 32 years) for control subjects (P OF

OPHTHALMOLOGY

MARCH 2000

TABLE 1. Comparison of Palsy Cases With Nonpalsy Control Subjects Cases (n ⫽ 100)

Mean age ⫾ SD (range) (yrs) Hypertension only (no.) Diabetes mellitus only (no.) Hypertension and diabetes (no.) Diabetes duration, median (range) Nerve palsy (no. [%]) Oculomotor (3rd) Trochlear (4th) Abducens (6th) Facial (7th) Aspirin use (no. [%]) Duration, median (range) Tobacco use (no. [%]) Current smoker Former smoker Nonsmoker Tobacco use (no. [%]) Cigars, pipe, chew Cigarettes ⬍1 pack/day 1–⬍2 packs/day ⱖ2 packs/day Cardiac disease (no. [%]) Cerebrovascular disease (no. [%])

PREVENTION

OF

P Value

67.8 ⫾ 8.9 41 (23 M, 18 F) 23 (15 M, 8 F) 36 (19 M, 17 F) 6 yrs (3 days–30 yrs)

67.2 ⫾ 8.9 66 (33 M, 33 F) 40 (26 M, 14 F) 57 (30 M, 27 F) 7 yrs (4 mos–32 yrs)

— — — — .94

20 (20) 33 (33) 33 (37) 10 (10) 34 (34) 5 yrs (2 mos–25 yrs)

0 0 0 0 49 (30.1) 4 yrs (3 days–26 yrs)

— — — — .51 .50 .73

14 (14) 12 (12) 74 (74)

25 (15.3) 35 (21.5) 103 (63.2)

1 (4.2)

6 (10.0)

6 (25.0) 9 (37.5) 8 (33.3) 23 (23) 6 (6)

9 (15.0) 25 (41.7) 20 (33.3) 50 (30.7) 39 (23.9)

.45

⫽ .94, Wilcoxon rank sum test). There were 20 oculomotor palsies (20%), 33 trochlear palsies (33%), 37 abducens palsies (37%), and 10 facial nerve palsies (10%) among cases. Of the 100 palsy cases, 34 (34%) were taking aspirin before the onset of the cranial nerve palsy. The mean duration of aspirin use for palsy cases was 5.5 years, with median duration of 5 years and range of 2 months to 25 years. Aspirin use for these 34 cases consisted of 81 mg daily (three cases), 162 mg daily (one case), 325 mg daily (28 cases), 1,300 mg daily (one case), and 3,900 mg daily (one case). Forty-nine control subjects (30.1%) were using aspirin at the following doses: 81 mg daily (four control subjects), 162 mg daily (one control), 325 mg daily (41 control subjects), and 1,300 mg daily (three control subjects). The mean duration of aspirin use for nonpalsy control subjects was 4.3 years, with a median duration of 4 years and a range of 3 days to 26 years. There was no significant difference (P ⫽ .51, chi-square test) between aspirin use before the onset of cranial nerve palsy for palsy cases and aspirin use for nonpalsy control subjects. There also was no significant difference between cases and control subjects as to the duration of aspirin use (P ⫽ .50, Wilcoxon rank sum test) or the daily dose of aspirin consumed (P ⫽ .89, Wilcoxon rank sum test). The odds of a patient having cranial nerve palsy in the aspirin group compared with the odds of having cranial nerve palsy in VOL. 129, NO. 3

Control Subjects (n ⫽ 163)

.17 ⬍.001

the nonaspirin group also showed no significant difference (odds ratio, 1.12; 95% confidence interval, 0.70 to 2.04). In the logistic regression analysis, we found that the estimated regression coefficient for aspirin use (⫹0.5618) did not differ significantly from zero (P ⫽ .065). Furthermore, the sign of the regression coefficient indicated that although the relationship was not statistically significant, the aspirin users had slightly higher odds of being in the palsy group than the nonusers did. The estimated odds ratio was 1.75 (95% confidence interval estimate, 0.97 to 3.18). This result is consistent with what was observed in the univariate analyses in which 34% of the palsy group used aspirin, while 30% of the nonpalsy control subjects used aspirin. The Hosmer-Lemeshow test for goodness of fit of the logistic regression model had a P value of .63, indicating a satisfactory fit. Among palsy cases, 14 (14%) were current smokers and 12 (12%) were former smokers. Former or current tobacco use consisted of cigars and pipe (one case), less than one pack of cigarettes per day (six cases), at least one but less than two packs of cigarettes per day (nine cases), and two or more packs of cigarettes per day (eight cases). For two cases the dose of cigarette use was not recorded. Among 18 cases, there was a median of 35 pack-years (range, 0.08 to 90 pack-years) of total cigarette consumption for cases. Among nonpalsy control subjects, 25 (15.3%) were current smokers and 35 (21.5%) were former smokers. Former

ACUTE ISCHEMIC CRANIAL NERVE PALSY

369

or current tobacco use consisted of cigars, pipe, or chewing tobacco (six control subjects), less than one pack of cigarettes per day (nine control subjects), at least one but less than two packs of cigarettes per day (25 control subjects), and two or more packs of cigarettes per day (20 control subjects). Among 49 control subjects, there was a median of 40 pack-years (range, 0.17 to 138 pack-years) of total cigarette consumption for control subjects. There were no significant differences between cases and control subjects as to tobacco use (P ⫽ .73, chi-square test) and total consumption of cigarettes (P ⫽ .45, Wilcoxon rank sum test). A total of 27 palsy cases (27%) had cardiac or cerebrovascular disease, of whom 21 (21%) had cardiac disease only, four (4%) had cerebrovascular disease only, and two (2%) had both cardiac and cerebrovascular disease. A total of 68 nonpalsy control subjects (41.7%) had cardiac or cerebrovascular disease, of whom 29 (17.8%) had cardiac disease only, 18 (11%) had cerebrovascular disease only, and 21 (12.9%) had both cardiac and cerebrovascular disease. The proportion of palsy cases and nonpalsy control subjects with cardiac disease did not differ (P ⫽ .17, chi-square test). We did find that cerebrovascular disease was significantly less common among palsy cases than nonpalsy control subjects (P ⬍ .001, chi-square test). Of the 100 palsy cases, a total of six (23.1%) of the 26 former or current smokers and 21 (28.4%) of 74 nonsmokers had cardiac or cerebrovascular disease. Among nonpalsy control subjects, cardiac or cerebrovascular disease was present in 34 (56.7%) of the 60 former or current cigarette smokers and 34 (33.0%) of 103 nonsmokers. Former or current tobacco use was significantly related to the presence of cardiac or cerebrovascular disease (P ⫽ .02, chi-square test). A significant relationship was observed between aspirin use and the presence of cardiac or cerebrovascular disease (P ⬍ .001, chi-square test), reflecting practice guidelines recommending aspirin for patients with cardiac or cerebrovascular disease. Because the onset, clinical manifestations, and clinical course of seventh nerve palsies resulting from viralassociated Bell’s palsy and those due to presumed ischemia are generally indistinguishable, we performed a separate analysis of third, fourth, and sixth nerve palsies (while excluding the 10 seventh nerve palsy cases).18,19 The findings remained similar. Aspirin use was not beneficial in preventing acute ischemic cranial nerve palsies. However, whereas the previous logistic regression analysis was marginally insignificant, reanalysis without the seventh nerve palsy cases now showed the estimated regression coefficient for aspirin use (⫹0.7193) to be marginally significantly different from zero (P ⫽ .02). This finding suggested that aspirin users had a slightly higher odds of being in the palsy group than patients who did not use aspirin. 370

AMERICAN JOURNAL

DISCUSSION OUR FINDINGS INDICATE THAT ASPIRIN MAY BE INEFFEC-

tive in preventing ischemic third, fourth, sixth, and seventh cranial nerve palsies. We found no significant difference in either the proportion of patients using aspirin or the dose and duration of aspirin use between cranial nerve palsy cases and nonpalsy control subjects. Additionally, there was no significant difference in the odds of a patient with cranial nerve palsy being in the aspirin group compared with the odds in the nonaspirin group. Our study sample size was sufficient to detect a difference in aspirin use between cranial nerve palsy cases and nonpalsy control subjects at a modest power of 80% when the actual proportions of use are 25% and 40%, respectively. Because the study was not prospectively designed with randomized assignment to treatment with aspirin, we were unable to assess the change in incidence rate of ischemic cranial nerve palsies resulting from aspirin use. Consequently, we are unable to conclusively determine whether aspirin is ineffective as prophylaxis for preventing ischemic cranial nerve palsies. Interestingly, although aspirin use has been shown to reduce the risk of myocardial infarction and strokes in middle-aged persons, aspirin users had a 52% higher risk of stroke among individuals aged 65 years and older.14 Additionally, aspirin was shown in one study to be ineffective in reducing the severity of visual loss in another cranial neuropathy, notably ischemic optic neuropathy.20 Likewise, controversy exists as to whether aspirin is effective in preventing the overall occurrence of ischemic optic neuropathy.21–23 Because aspirin use is so common, it is possible that some of our patients could have inadvertently omitted informing us of their intermittent or regular use of aspirin. Such omissions should have been equally prevalent among palsy cases and nonpalsy control subjects because both groups were randomly identified from the same patient population base. Tobacco use among individuals with diabetes or hypertension poses an increased risk for the development of strokes, transient ischemic attacks, and cardiac disease. Cigarette smoking produces vasoconstriction and an increase in platelet aggregation, erythrocyte count, leukocyte count, and fibrinogen, as well as a decrease in high-density lipoprotein, all of which may increase the risk of vascularrelated diseases. Despite the potential adverse effects of tobacco, our study did not find tobacco use to be an independent risk factor for ischemic cranial nerve palsies. In a study by Jacobson and associates,3 tobacco use also was not found to be a significant independent risk factor for ischemic ocular motor nerve palsies. Only previously diagnosed diabetes mellitus, left ventricular hypertrophy, and elevated hematocrit were found to pose significant independent risks. In our study, patients with ischemic cranial nerve palsies had a significantly lower rate of strokes and transient OF

OPHTHALMOLOGY

MARCH 2000

ischemic attacks than nonpalsy control subjects. This could suggest that there may be two populations of patients with diabetes mellitus and hypertension: one population that is more prone to develop small-vessel ischemic cranial nerve palsies and another population with a tendency to develop strokes and transient ischemic attacks. A more plausible explanation is that patients in the palsy group could have initiated aspirin use prophylactically because of knowledge that aspirin reduces the incidence of cardiac disease and strokes. In contrast, nonpalsy control subjects could have initiated aspirin after having been affected by cardiac or cerebrovascular disease. In such a scenario, the palsy cases could have a decreased rate of stroke because of stroke prophylaxis. That these patients developed cranial nerve palsies despite stroke prophylaxis with aspirin—and despite both cases and control subjects having similar dose and duration of aspirin use—would again indicate that aspirin may be ineffective in preventing ischemic cranial nerve palsies. Because of the retrospective nature of this study, we are unable to assess whether either of these hypotheses is correct. A prospectively designed, randomized, placebo-controlled study would be helpful in addressing these issues. In our 100 cases, hypertension alone was more commonly associated with ischemic cranial nerve palsies than diabetes alone. Although our study is susceptible to physician referral bias, it paralleled several other studies showing the abducens nerve to be the most commonly affected cranial nerve in ischemic cranial palsies.4 – 6 Whereas our study and others focused on mononeuropathies, it should be noted that simultaneous bilateral diabetic ophthalmoplegia has been reported.24 As clinicians, it is important that we exclude other possible causes of ocular motor nerve palsies before rendering the diagnosis of ischemic cranial nerve palsy.25,26 Although our study suggests that aspirin may be ineffective in preventing ischemic cranial nerve palsies, the possible benefit of aspirin on the recurrence of ischemic cranial nerve palsies or on recovery time was not evaluated.

REFERENCES 1. Pulsinelli WA. Cerebrovascular diseases. In Bennett JC, Plum F, editors. Cecil Textbook of Medicine (20th ed). Volume 2. Philadelphia: WB Saunders Co, 1996:2057. 2. Wolf PA. Risk factors for stroke. Stroke 1985;16:359 –360. 3. Jacobson DM, McCanna TD, Layde PM. Risk factors for ischemic ocular motor nerve palsies. Arch Ophthalmol 1994;112:961–966. 4. Richards BW, Jones FR, Younge BR. Causes and prognosis in 4,278 cases of paralysis of the oculomotor, trochlear, and abducens cranial nerves. Am J Ophthalmol 1992;113:489 – 496. 5. Rush JA, Younge BR. Paralysis of cranial nerves III, IV, and VI: cause and prognosis in 1,000 cases. Arch Ophthalmol 1981;99:76 –79.

VOL. 129, NO. 3

PREVENTION

OF

6. Rucker CW. Paralysis of the third, fourth, and sixth cranial nerves. Am J Ophthalmol 1958;46:787–794. 7. Saito O, Aoyagi M, Tojima H, Koike Y. Diagnosis and treatment for Bell’s palsy associated with diabetes mellitus. Acta Otolaryngol 1994;511:153–155. 8. Paolino E, Granieri E, Tola MR, et al. Predisposing factors in Bell’s palsy: a case-control study. J Neurol 1985;232:363– 365. 9. Watanabe K, Hagura R, Akanuma Y, et al. Characteristics of cranial nerve palsies in diabetic patients. Diabetes Res Clin Pract 1990;10:19 –27. 10. Antiplatelet Trialists’ Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. BMJ 1988;296:320 –331. 11. Stroke Prevention in Atrial Fibrillation Investigators. Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study. Lancet 1994;343:687– 691. 12. UK-TIA Study Group. United Kingdom Transient Ischemic Attack Aspirin Trial: Interim results. BMJ 1988;296:316 –320. 13. The Canadian Cooperative Study Group. A randomized trial of aspirin and sulfinpyrazone in threatened stroke. N Engl J Med 1978;299:53–59. 14. Manolio TA, Kronmal RA, Burke GL, et al. Short-term predictors of incident stroke in older adults: the Cardiovascular Health Study. Stroke 1996;27:1479 –1486. 15. The SALT Collaborative Group. Swedish aspirin low-dose trial (SALT) of 75 mg aspirin as secondary prophylaxis after cerebrovascular ischaemic events. Lancet 1991;338:1345– 1349. 16. Roth GJ, Calverley DC. Aspirin, platelets, and thrombosis: theory and practice. Blood 1994;83:885– 898. 17. Dennis M, Bamford J, Sandercock P, Warlow C. Prognosis of transient ischemic attacks in the Oxfordshire Community Stroke Project. Stroke 1990;21:848 – 853. 18. Baringer JR. Herpes simplex virus and Bell palsy. Ann Intern Med 1996;124:63– 64. 19. Murakani S, Mizobuchi M, Makashrio Y, et al. Bell palsy and herpes simplex virus identification of viral DNA in endoneurial fluid and muscle. Ann Intern Med 1996;124(1, part 1):27–30. 20. Botelho PJ, Johnson LN, Arnold AC. The effect of aspirin on the visual outcome of nonarteritic anterior ischemic optic neuropathy. Am J Ophthalmol 1996;121:450 – 451. 21. Beck RW, Hayreh SS, Podhajsky PA, et al. Aspirin therapy in nonarteritic anterior ischemic optic neuropathy. Am J Ophthalmol 1997;123:212–217. 22. Kupersmith MJ, Frohman L, Sanderson M, et al. Aspirin reduces the incidence of second eye NAION: a retrospective study. J Neuro Ophthalmol 1997;17:250 –253. 23. Johnson LN. Nonarteritic anterior ischemic optic neuropathy. J Am Optom Assoc 1998;69:422– 423. 24. Sergott RC, Glaser JS, Berger LJ. Simultaneous, bilateral diabetic ophthalmoplegia: report of two cases and discussion of differential diagnosis. Ophthalmology 1984;91:18 –22. 25. Johnson LN, Hepler RS, Yee RD, Batzdorf U. Sphenoid sinus mucocele (anterior clinoid variant) mimicking diabetic ophthalmoplegia and retrobulbar neuritis. Am J Ophthalmol 1986;102:111–115. 26. Johnson LN, Hepler RS. Isolated abducens nerve paresis from intrapontine, fascicular abducens nerve injury. Am J Ophthalmol 1989;108:459 – 461.

ACUTE ISCHEMIC CRANIAL NERVE PALSY

371