Population-based Rate and Patterns of Diplopia in Giant Cell Arteritis

ABSTRACT

Giant cell arteritis (GCA) is the most common vasculitis in older adults with permanent vision loss as a feared complication. Diplopia has been reported in a small percentage of patients with visual manifestations. The goal of this study was to determine the population-based rates and patterns of binocular diplopia from GCA. The Rochester Epidemiology Project (REP), a medical records linkage system was used to identify all residents of Olmsted County, Minnesota, USA, diagnosed with GCA between January 1, 1950 and December 31, 2019. Medical records were then reviewed to identify patients with binocular diplopia from GCA. There were 301 incident cases of GCA from 1950 to 2019. Fourteen (5%) patients presented with binocular diplopia. Of these 14 patients, nine (3%) had constant diplopia and five (2%) had transient diplopia. Among patients with constant diplopia, cranial nerve VI involvement was suspected in four (44%) cases. Systemic symptoms and inflammatory markers were similar in patients with and without diplopia. There was no difference in the rate of anterior ischaemic optic neuropathy between the two groups (7% vs. 7%, p = 1.00). In conclusion, this population-based study showed that binocular diplopia was present in 5% of patients with GCA, which could either be transient or constant. GCA patients with diplopia had similar systemic manifestations and risk of vision loss as GCA patients without diplopia.

Introduction

Giant cell arteritis (GCA) is a granulomatous vasculitis that targets medium and large sized vessels.¹ It is the most common form of vasculitis affecting older adults with the highest incidence among people of Scandinavian descent.¹ The annual incidence rate of GCA among Americans is 18.8 per 100,000 persons over the age of 50.² Complications from GCA are caused by stenosis and aneurysms of affected vessels.³ Permanent vision loss is one of the most feared complications, ranging from 8.2% to 48.8% of patients with newly diagnosed GCA.⁴

Binocular diplopia as a symptom of GCA has been reported in 1% to 19% of patients.^5–14 It can be caused by ischaemia in any segment of the ocular motor system including the brainstem, ocular motor cranial nerves (CNs), or extraocular muscles.⁶ Diplopia can be constant or transient and associated with other ocular and systemic symptoms, such as blurred vision, eye pain, headaches, jaw claudication, and scalp tenderness.⁷ More importantly, it can accompany transient or permanent vision loss and could be used as a warning sign for other complications.⁷

The majority of prior studies on diplopia in GCA are from tertiary centres and therefore subject to referral and recall bias, which likely contributes to the large range of reported diplopia in GCA.^7,8 The goal of this study was to utilise the Rochester Epidemiology Project (REP), a medical records-linkage system, to determine the population-based rate of diplopia from GCA and to determine the details of patients with diplopia to help guide screening and treatment of this disease and to prevent further complications such as vision loss.

Materials and methods

The REP was used to identify all potential cases of GCA in Olmsted County, Minnesota from January 1, 1950, through December 31, 2019.¹⁵ The cohort of patients with incident GCA has been previously identified.¹⁶ Patients were included either if they met the 1990 American College of Rheumatology criteria (confirmed by a rheumatologist) or if they met the following criteria: age ≥ 50 years with elevated inflammatory markers (i.e., erythrocyte sedimentation rate [ESR] or C-reactive protein [CRP]) and computed tomography/angiography, magnetic resonance arteriography or positron emission tomography evidence of large vessel vasculitis involving the ascending aorta and/or its branches.¹⁷ Institutional review board approval was obtained from Mayo Clinic and Olmsted Medical Center. The study adhered to the tenets of the Declaration of Helsinki and the Health Insurance Portability and Accountability Act.

Medical records were reviewed to identify patients with binocular diplopia from GCA. Binocular diplopia was considered constant if an episode lasted more than 24 hours or transient if an episode lasted less than 24 hours, with or without recurrence. Neuro-ophthalmology reviewed and confirmed all cases of binocular diplopia. Cases of monocular diplopia were excluded. The type and pattern of diplopia, treatment, response to therapy and final outcomes were documented. Other ocular manifestations such as permanent or transient vision loss and systemic manifestations such as headache, jaw claudication, facial pain, scalp tenderness, fever, weakness, weight loss, polymyalgia rheumatica, and fatigue were recorded. Laboratory findings included haemoglobin, ESR, CRP, alkaline phosphatase, and albumin.

Descriptive statistics with raw count and proportions summarised the patient data. Comparisons between GCA patients with and without diplopia were performed using rank sum tests for continuous variables, Fisher’s exact tests for categorical variables with n < 5 in at least one cell, and chi-square tests for the remaining categorical variables were used. The rate of diplopia in GCA was calculated as the number of patients with diplopia divided by the total number of patients with GCA. Analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).

Results

A total of 301 patients were diagnosed with GCA between 1950 and 2019. There were 232 (77%) females and the mean age was 76.3 (standard deviation [SD] 8.2) years. Among this cohort, a total of 14 (5%) patients had binocular diplopia, which was constant in nine (3%) and transient in five (2%). All cases of diplopia were present at initial GCA diagnosis and were evaluated by an ophthalmologist. There were no differences in the baseline characteristics, systemic manifestations, and laboratory findings of GCA in patients with and without diplopia (Table 1).

Table 1.

Demographic characteristics, systemic manifestations, and laboratory findings in patients with giant cell arteritis

Characteristic	Diplopia (n = 14)	No diplopia (n = 287)	p-value
Sex, n (%)  Female  Male	11 (79%) 3 (21%)	221 (77%) 66 (23%)	1.00
Age in years, mean (± SD)	77.2 (±7.8)	76.3 (±8.3)	0.89
Symptoms, n (%)  Headache  Scalp tenderness  Jaw claudication  Fatigue  Polymyalgia rheumatica  Weight loss  Weakness  Fever  Facial pain	10 (71%) 7 (54%) 7 (50%) 2 (14%) 3 (21%) 2 (14%) 0 (0%) 1 (7%) 1 (7%)	207 (73%) 118 (43%) 128 (45%) 102 (37%) 84 (29%) 75 (27%) 62 (22%) 48 (17%) 33 (12%)	1.00 0.46 0.73 0.15 0.76 0.53 0.08 0.48 1.00
Laboratory values, mean (± SD)  Haemoglobin in g/dL  ESR in mm/hr  CRP in mg/L  Alkaline phosphatase in IU/L  Albumin in g/dL	12.4 (±0.9) 80.5 (±26.2) 109.9 (±117.3) 195.8 (±40.2) 3.1 (±0.4)	11.8 (±1.3) 74.7 (±30.6) 88.2 (±80.2) 195.8 (±115.9) 3.1 (±0.6)	0.10 0.47 0.72 0.57 0.81
TAB positive, n (%) TAB negative or missing, n (%)  Radiologic criteria positive, n (%)  Radiologic criteria negative or missing, n (%)*	13 (93%) 1 (7%) 0 (0%) 1 (7%)	229 (80%) 58 (20%) 15 (5%) 43 (15%)	0.32
Anterior ischaemic optic neuropathy, n (%)	1 (7%)	19 (7%)	1.00

*Clinical features meeting diagnostic criteria for giant cell arteritis

CRP = C reactive protein; ESR = erythrocyte sedimentation rate; SD = standard deviation; TAB = temporal artery biopsy.

Visual manifestations among the 14 GCA patients with diplopia included blurry vision (n = 12, 86%), photopsias (n = 2, 14%), ocular pain (n = 2, 14%), anterior ischaemic optic neuropathy (AION) (n = 1, 7%), nystagmus (n = 1, 7%), and cotton wool spots (n = 1, 7%). One of the patients with transient diplopia had a coincident hemi-retinal vein occlusion, but it was unclear if this was related to GCA. Among the nine patients with constant diplopia, four had horizontal misalignment and an abduction deficit consistent with CN VI involvement, one had vertical misalignment consistent with CN IV palsy, one had motility deficits consistent with a pupil sparing CN III palsy, and three had a vertical deviation that did not follow a CN pattern (Table 2). All nine patients with constant diplopia from GCA were treated with glucocorticoids. The mean interval from the onset of constant diplopia to treatment was 4.4 days (SD 6.9) with complete resolution over a mean of 23.7 (SD 26.0) days among seven of eight patients with follow-up (Table 2).

Table 2.

Details of giant cell arteritis patients with constant diplopia

Patient	Type	Alignment	Cranial nerve involvement	Associated ocular symptoms	Associated ocular signs	Treatment	Onset to treatment in days	Resolution	Treatment to resolution in days
1	Constant	Horizontal	CN VI	Blurry vision	-	GC	3	Yes	68
2	Constant	Vertical	CN IV	Blurry vision	-	GC, heparin	5	Yes	10
3	Constant	Horizontal	CN VI	Blurry vision, photopsias	-	GC	0	Yes	-
4	Constant	Vertical	No CN pattern	Blurry vision	-	GC	0	Yes	14
5	Constant	Vertical	No CN pattern	Blurry vision, ocular pain	Dissociate nystagmus	GC	22	No	-
6	Constant	Vertical	No CN pattern	Blurry vision	-	GC	3	Yes	1
7	Constant	Horizontal	CN VI	Blurry vision	AION, cotton wool spots	GC	0	-	-
8	Constant	Horizontal	CN VI	Blurry vision	-	GC	5	Yes	7
9	Constant	Oblique	CN III	Photophobia	-	GC	2	Yes	42

AION = anterior ischaemic optic neuropathy; CN = cranial nerve; GC = glucocorticoids.

Among the five patients with transient binocular diplopia, there were details in the pattern for three of them: two reported horizontal misalignment and one reported vertical misalignment. The frequency of diplopia ranged between two (n = 3, 60%) and 12 episodes (n = 1, 20%), while duration of diplopia ranged between 5 (n = 2, 40%) and 120 minutes (n = 1, 20%).

Discussion

This population-based study demonstrated that the rate of binocular diplopia among a cohort of 301 GCA patients was 5% (3% constant and 2% transient). This rate is at the lower end of other single- and multi-centre non-population-based studies where rates of diplopia in GCA ranged between 1% and 19%.^{5,6,8–10,18} Given the risk of referral bias with tertiary centre-based studies, it is not surprising to find higher rates throughout the literature. Furthermore, many of these studies did not offer a clear description of diplopia, which further complicates rate comparisons. In our study, medical records were carefully reviewed to exclude cases of monocular diplopia, and only cases of constant and definite transient binocular diplopia were included. Our rate was similar to a study in the Lothian Region of Scotland where six (4.4%) patients among a GCA cohort of 136 presented with diplopia; two had CN IV and two had CN III paresis, which completely resolved in corticosteroid treatment.¹³ Similar low rates were published in a population-based study from Iceland where two (1.5%) out of 133 patients with GCA reported diplopia but without further details being provided.^11,19

Several mechanisms for diplopia in GCA have been proposed including brainstem, ocular motor CN, and extraocular muscle ischaemia.^1,20 Of the patients with constant binocular diplopia, most had horizontal misalignment suggesting involvement in the CN VI distribution. While further deductions on the exact mechanism of diplopia could not be made, many studies support CN ischaemia and specifically cranial VI involvement as the most likely aetiology.^6,21 Three of our patients did not follow a clear ocular motor CN pattern of misalignment; however, not enough information was available to determine if brainstem (i.e. skew deviation) or extraocular muscle ischaemia had occurred.

All patients were treated with glucocorticoids. Seven out of the eight patients with follow-up had a resolution of diplopia between 1 day to 3 months of treatment. These results are in contrast to a prospective study conducted at the University Hospital of Basel where complete resolution was reached in eight out of nine patients with diplopia from GCA just a few days after the start of glucocorticoid therapy.¹⁴

In our study, patients who presented with diplopia from GCA had other systemic symptoms and laboratory markers suggestive of a GCA diagnosis but were not different from the rest of the GCA cohort. Such similar findings have been previously reported.^7,10,14 From a multi-centre, retrospective study by Ross et al., patients with diplopia from GCA were compared with patients with diplopia from other causes. They determined the presence of jaw claudication to be highly indicative of a GCA diagnosis as it was found in 80% of patients with diplopia, higher than the 50% reported in the current study.⁷ These results were consistent with Singh et al. who found more jaw claudication in patients with visual manifestations than without (55% vs. 38%, p = .03) in a cohort of 204 patients with GCA.¹¹ While systemic symptoms can help diagnose GCA in patients with diplopia, they are not always present so providers must also rely on other clinical and laboratory findings.¹² We noted similar levels of haemoglobin, ESR, CRP, alkaline phosphatase, and albumin in GCA patients with and without diplopia. Other studies have shown higher levels of ESR and CRP in patients with diplopia from GCA compared with patients with diplopia from other causes, further reinforcing their value as supportive tests.^7,10

Vision loss is a feared complication of GCA. In our cohort, AION was present in 7% of all diplopia cases, lower than the 33% and 16% reported in prior non-population-based studies of diplopia from GCA.^7,14 Given the differences in study design and concerns about referral bias from prior studies, we suspect the true AION rate in GCA patients presenting with diplopia to be lower than those previously reported and closer to our 7%. Interestingly, we did not find a difference in the rate of AION in GCA patients presenting with or without diplopia (7% vs. 7%, p = 1.00). These results are reassuring as there does not appear to be an increased risk of ischaemic ocular complications in patients presenting with diplopia from GCA.

This study has several limitations. Its retrospective design is limited by variability of documentation in medical records and non-standardised assessments. Patients were assigned CN palsies based on the pattern of misalignment; however, ischaemia to the muscle or brainstem could also mimic an ocular motor CN palsy. In patients with baseline or new severe vision loss, diplopia may be under-reported. The number of GCA patients with diplopia was relatively small because this was focused on a population-based cohort within the REP. Finally, because Olmsted County is predominantly white Caucasian, the results may not be applicable to other patient populations.

In conclusion, this study provides an accurate population-based rate of binocular diplopia from GCA, which occurs in 5% of GCA patients. Patients with diplopia from GCA had similar characteristics to GCA patients without diplopia, including systemic manifestations of GCA, risk of vision loss, and laboratory findings. Because GCA is a potentially blinding disease, a high index of suspicion is required. Elderly patients presenting with binocular diplopia should undergo further workup for systemic symptoms and inflammatory markers suggestive of GCA.

Funding Statement

This study was made possible using the resources of the Rochester Epidemiology Project, which is supported by the National Institute on Aging of the National Institutes of Health under Award Number R01AG034676, and Grant Number UL1TR002377 from the National Center for Advancing Translational Sciences (NCATS), a component of the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Declaration of interest statement

The authors report no conflict of interest.