Stroke Rate, Subtype, and Cardiovascular Risk Factors in Nonarteritic Anterior Ischemic Optic Neuropathy: A Population-Based Study

Abstract

Background:

Nonarteritic anterior ischemic optic neuropathy (NAION) is a common cause of acute optic neuropathy in adults and is associated with vascular risk factors. Owing to the overlapping risk factor profiles between NAION and cerebral stroke, previous studies have produced conflicting results with regard to NAION as an independent risk factor for stroke.

Methods:

A retrospective chart review was conducted using the Rochester Epidemiology Project database to identify all cases of NAION occurring among Olmsted County, Minnesota residents from January 1, 1990, through December 31, 2016. Stroke events were characterized using clinical and radiologic data. Comparison was made to an age- and sex-matched control group with similar vascular risk factors.

Results:

One-hundred four patients with NAION and 104 control subjects were analyzed. Median age at diagnosis was 65 years (range, 40–90 years). Thirteen patients (13%) with NAION and 10 controls (10%) had symptomatic strokes after the age of 40 years. Among patients with NAION, 6 (46%) suffered a stroke before the diagnosis of NAION, 5 (39%) at least 5 months after the NAION diagnosis, and 2 patients (15%) suffered strokes both before and after the NAION. The cumulative probability of symptomatic strokes for patients with NAION was not significantly different than for controls (hazard ratio = 1.50, 95% confidence interval: 0.66–3.42; P = 0.34). There were no cardioembolic strokes within 1 month of the NAION diagnosis. The mechanism of symptomatic strokes did not differ between the 2 groups.

Conclusions:

NAION does not confer an increased risk of symptomatic stroke beyond the risk posed by age and existing vascular risk factors.

Introduction:

Nonarteritic anterior ischemic optic neuropathy (NAION) is an incompletely understood, multifactorial disease process thought to be caused by arterial insufficiency to the optic nerve head due to a combination of various systemic and ocular risk factors^{1, 2}. Systemic risk factors for NAION include hypertension, hyperlipidemia, diabetes mellitus, and obstructive sleep apnea^3–5. Unlike the majority of ischemic strokes, NAION is thought to result from hypoperfusion of the optic nerve head as opposed to a thromboembolic process¹. Despite the differing mechanisms, it has been hypothesized that the occurrence of NAION may be associated with a higher risk of insufficiency and infarction of the cerebral vessels producing stroke⁴. Due to the overlapping vascular risk factor profiles between NAION and cerebral stroke, it has proven difficult to determine whether NAION is an independent risk factor for stroke, and prior studies have produced conflicting results^{4, 6–10}. Most recently, a large database study from Taiwan reported an increased risk of stroke in patients with NAION when compared to controls with similar vascular risk factors⁷. However, this study utilized big data and as a result the mechanisms of strokes could not be determined other than differentiating ischemic versus hemorrhagic; additionally, incident cases of NAION could not be confirmed as the study relied upon the use of diagnosis codes.

The primary aim of this study was to examine the population-based occurrence and mechanism of symptomatic strokes in patients with a confirmed diagnosis of NAION to a well-matched control cohort from the Rochester Epidemiology Project (REP) to assess whether NAION is an independent risk factor for stroke.

Methods:

The medical records of all Olmsted County, Minnesota residents diagnosed with optic neuropathies, excluding glaucoma, from January 1, 1990 through December 31, 2016 were reviewed to identify incident cases of NAION. Potential subjects were identified using the REP, a medical records database designed to capture all medical care provided in Olmsted County, Minnesota^{11, 12}. The study was approved by the Institutional Review Boards of the Mayo Clinic and Olmsted Medical Center. It conforms to the requirements of the United States Health Insurance Portability and Accountability Act (HIPPA) and adheres to the tenets of the Declaration of Helsinki.

A total of 1,791 patients with a diagnosis of optic neuropathy were identified and reviewed in detail to identify confirmed incident cases of NAION. A diagnosis of NAION required a new onset of unilateral vision loss and/or optic nerve-type visual field defect with optic disc edema present at initial presentation. Patients were required to have follow-up beyond the initial presentation, and if an alternative cause was found, such as optic neuritis, giant cell arteritis, collagen vascular disease, infection, or chronic uveitis, these patients were excluded.

Data on the patients’ demographics, ocular and medical histories, and clinical course were obtained from their medical records. Diagnosis and imaging codes were used to identify patients who were diagnosed with stroke or transient ischemic attack during the study period, as well as patients who had undergone cranial magnetic resonance imaging (MRI). Records and associated imaging for these patients were then reviewed by a fellowship trained stroke neurologist (JPK) to confirm the occurrence and elucidate the mechanism of stroke. Strokes were classified according to the TOAST criteria^13–15. Strokes identified on imaging were considered to be symptomatic if clinical symptoms correlated temporally and spatially with the infarct documented in the medical record. For patients presenting with multiple strokes on imaging, the appearance and location of infarction, as well as the timing of stroke symptoms, were used to identify the symptomatic stroke(s). Incidentally found asymptomatic strokes and the presence of small vessel disease as noted by the interpreting radiologist were also recorded.

A control group was created with one control subject for every patient diagnosed with NAION. The selected control group was matched for age, sex, and vascular risk factors with the NAION patient cohort. All continuous measures were summarized with median and range, and all categorical measures were summarized with frequency and percent. Proportional differences between patient groups were compared using the Pearson Chi-square test, and continuous differences were compared using the Wilcoxon Rank Sum test.

The occurrence of symptomatic strokes between NAION and control cohort patients was analyzed using single and multivariable Cox’s Proportional Hazards models. The youngest patient diagnosed with NAION in our cohort was 40 years old, although earlier occurrence of NAION has been reported. Accordingly, in our analysis of the risk of symptomatic stroke between NAION and control patients, the time to symptomatic stroke was measured from the age of 40 years until either the first occurrence of a symptomatic stroke or last documented follow-up. To reduce the possibility of confounding, multivariable models were adjusted for age at NAION diagnosis, sex, and presence of at least one vascular risk factor. Hazard ratios and their 95% confidence intervals were estimated. The cumulative probability of stroke between groups was calculated using the Kaplan-Meier method in ten year increments. All tests were two sided and performed at the 0.05 significance level. All statistical analysis was performed using R Statistical Software (version 3.4.2; R Foundation for Statistical Computing, Vienna, Austria).

Results:

From 1990 through 2016, a total of 104 incident cases of NAION were identified. The median age at diagnosis of NAION was 65 years (range, 40–90 years). Patient characteristics of the 104 NAION and 104 control subjects are summarized in Table 1.

Table 1.

Characteristics of NAION and Control Groups

	NAION (N= 104)	Control (N = 104)	Total (N=208)	p-value
Age, yrs	65 (40, 90)	66 (29, 103)	65 (29, 103)	0.96
Sex, male	59 (56.7%)	59 (56.7%)	118 (56.7%)	1
Race				0.39
White	95 (96.9%)	98 (97.0%)	193 (97.0%)
Black	0 (0.0%)	0 (0.0%)	0 (0.0%)
Asian	1 (1.0%)	3 (3.0%)	4 (2.0%)
Other	2 (2.0%)	0 (0.0%)	2 (0.6%)
Hypertension	83 (79.8%)	74 (71.2%)	157 (75.5%)	0.15
Diabetes Mellitus	41 (39.4%)	37 (35.6%)	78 (37.5%)	0.57
Obstructive Sleep Apnea	24 (23.1%)	27 (26.0%)	51 (24.5%)	0.63
Hyperlipidemia	77 (74.0%)	69 (66.3%)	146 (70.2%)	0.23
Coronary Heart Disease	34 (32.7%)	31 (29.8%)	65 (31.2%)	0.65

Median (Minimum, Maximum) were used to summarize age and N (%) were used to summarize sex, race, and vascular risk factors. Ten patients were missing race. Differences in age were compared using the Wilcoxon Rank Sum test, and proportional differences were compared using the Pearson Chi-square test. All tests are two-sided and performed at the 0.05 significance level.

Thirteen (13%) patients with NAION were diagnosed with a symptomatic stroke after the age of forty years; six of these patients (46%) had a stroke prior to the diagnosed NAION, five (39%) occurred at least 5 months after the NAION diagnosis, and two patients (15%) suffered strokes both prior to and following the NAION. One patient (8%) suffered a cryptogenic stroke within one month of NAION diagnosis. The subtypes of strokes diagnosed in each group of patients are shown in Table 2. No difference in the subtypes of symptomatic strokes was found between the NAION and control groups (P=0.68).

Table 2.

Comparison of Symptomatic Stroke Subtypes Between Groups

Mechanism of Stroke	NAION (N= 13 strokes)	Control (N = 10 strokes)	p-value
			0.68
Large Artery Atherosclerosis	0 (0.0%)	0 (0.0%)
Cardioembolic	4 (30.8%)	4 (40.0%)
Lacunar	3 (23.1%)	1 (10.0%)
Cryptogenic	4 (30.8%)	2 (20.0%)
Stroke of other determined etiology	2 (15.4%)	3 (30.0%)

Proportional differences were compared using the Pearson Chi-square test. All tests are two-sided and performed at the 0.05 significance level.

Fifty-three of 104 (51%) patients diagnosed with NAION and 43 of 104 patients (41%) in the control group underwent cranial MRI (P=0.16); asymptomatic strokes were found in 17 of 53 (32%) patients in the NAION group and 7 of 43 (16%) patients in the control group (P=0.075). The majority of asymptomatic strokes in both groups were lacunar in nature. Small vessel disease was noted on MRI in 43 of 49 (88%) patients in the NAION cohort with accessible imaging and 30 of 42 (71%) patients in the control group (P=0.051).

Symptomatic Stroke

Of the 208 total patients analyzed, 23 experienced a symptomatic stroke within the surveillance period; 13 of 104 (13%) patients in the NAION cohort and 10 of 104 (10%) in the control group. The median follow-up time was 73 years (range, 43 to 103 years). The cumulative probability of stroke for patients with NAION was not significantly different than for controls (HR=1.50, 95% CI: 0.66–3.42; p=0.34; Figure 1). Adjusting for age at NAION diagnosis, gender, or the presence of a vascular risk factor did not reveal a significant risk of stroke in patients with NAION when compared to controls (Table 3). The cumulative probability of stroke at 80 years of age for NAION and control patients was 16.6% (95% CI: 5.1%, 26.7%) and 13.0% (95% CI: 3.6%, 21.6%), respectively (Table 4).

Figure 1.

Kaplan-Meier cumulative probability of stroke between NAION and control groups

Table 3.

NAION and the cumulative probability of stroke

	HR (95% CI)	p value
Unadjusted	1.50 (0.66, 3.42)	0.34
Age at NAION diagnosis	1.51 (0.66, 3.46)	0.33
Adjusted for gender	1.50 (0.66, 3.43)	0.34
Hypertension	1.51 (0.66, 3.45)	0.33
Diabetes Mellitus	1.45 (0.63, 3.33)	0.38
Obstructive Sleep Apnea	1.48 (0.65, 3.39)	0.35
Hyperlipidemia	1.49 (0.65, 3.41)	0.35
Coronary Heart Disease	1.51 (0.66, 3.45)	0.33

Table 4.

Kaplan-Meier cumulative probability rates of stroke

Group	50 years old	60 years old	70 years old	80 years old
Control	0.0% (0.0%, 0.0%)	2.1% (0.0%, 4.9%)	3.4% (0.0%, 7.1%)	13.0% (3.6%, 21.6%)
NAION	0.0% (0.0%, 0.0%)	1.1% (0.0%, 3.1%)	6.4% (0.8%, 11.7%)	16.6% (5.1%, 26.7%)

Discussion:

Many of the vascular risk factors predisposing patients to ischemic strokes are also risk factors for NAION. In this population-based retrospective study, we found a similar lifetime incidence of symptomatic strokes among patients with NAION compared to those in our sex, age, and vascular risk matched control cohort. Thus, our data indicate that NAION does not independently contribute to the overall risk of symptomatic stroke. Prior studies have reported conflicting results regarding the association between NAION and cerebrovascular disease^{4, 6–10}. The strengths of this study include the ability to confirm the diagnoses of NAION and stroke by manual review of the medical record, the ability to sub-classify strokes, and the population-based setting allowing for incidence calculation with avoidance of referral bias.

A previous Taiwanese study from Lee et al. reported an increased incidence of stroke for patients with NAION and vascular risk factors when compared to controls also having at least one vascular risk factor⁷. However, as this was a large healthcare database study, the diagnoses of NAION and stroke were reliant on diagnosis codes; in our experience, diagnosis codes for ischemic optic neuropathy and stroke are often inaccurate. Because big data studies are dependent on diagnosis codes, it is possible that the study by Lee et al. captured a large number of incidentally found asymptomatic strokes in patients with NAION, as our study has demonstrated a trend toward a higher rate of MRI’s performed in NAION patients compared to control patients with an equal number of vascular risk factors⁷. Although our study population was primarily Caucasian with very few Asian subjects, we did not find a difference when we compared the likelihood of symptomatic strokes between patients with NAION and the control cohort.

The most common etiologies of symptomatic stroke in our NAION cohort were lacunar, cardioembolic, and cryptogenic; this finding is consistent with published incidence rates for stroke subtypes in the normal population, which matched our control cohort¹⁶. Notably, none of the patients in our NAION cohort experienced a cardioembolic stroke around the time of their NAION diagnosis. This finding is consistent with previous assertions that NAION is very rarely caused by a thromboembolic mechanism¹ and suggests that a cardioembolic workup is unnecessary for patients with NAION in the absence of embolic retinal vascular disease.

Limitations of this study include the racial homogeneity of the Olmsted county population, which is primarily Caucasian; thus, the applicability of our findings to other ethnicities is uncertain. The retrospective nature of our study inevitably resulted in some patients having limited follow-up data. The limited power of this study may have impacted our ability to identify an association between diagnosis of NAION and symptomatic stroke (Type II error). Additionally, our determination of vascular risk factors relied on diagnosis codes, and our categorization of vascular risk factors as discrete variables did not allow for consideration of the severity of these risk factors in individual patients. Finally, our determination of symptomatic versus asymptomatic strokes was reliant on the quality and completeness of the patient history and the medical record.

In summary, we demonstrate that NAION does not confer an increased risk of symptomatic stroke beyond the risk posed by age and existing vascular risk factors. Given that the systemic risk factors for NAION largely overlap with those of stroke, a thorough assessment of modifiable risk factors remains crucial to preventing future ocular and systemic morbidity.