ORIGINAL ARTICLE

Analysis of uveitis in a Canadian aboriginal populationMili Roy, MD

ABSTRACT ● RÉSUMÉ

Objective: To compare patient demographics, uveitis characteristics, and outcomes in aboriginal First Nations (FN) versus non-aboriginal non-FN uveitis patients.

Design: Case–control study.Participants: Forty-three FN patients (80 eyes) and 45 control non-FN uveitis patients (69 eyes).Methods: Retrospective chart review comparing patient demographics (age, sex, residency), disease characteristics (laterality,

anatomic classifications, granulomatous and chronic inflammation, systemic associations, specific uveitis diagnoses), anddisease severity-related outcomes (therapies, visual outcomes, complications) between FN versus control patients.

Results: Mean age at disease onset was significantly younger in FN patients (30.4 years) versus control patients (40.2 years; p o0.0001). Bilateral uveitis was significantly more common in the FN group (86%) versus control patients (51%; p ¼ 0.0005), andgranulomatous uveitis was significantly more common in FN patients (53%) than among control patients (11%; p o 0.0001).There was no significant difference in rates of chronic uveitis between groups, identified in 74% of FN versus 60% of controlsubjects (p ¼ 0.15). Anatomically, panuveitis was the prevalent form of uveitis within the FN group, occurring in 67% (versus 16%in control patients), whereas anterior uveitis was the prevalent anatomic classification in control patients, occurring in 73% (versus26% in FN patients; p o 0.001). Vogt–Koyanagi–Harada (VKH) syndrome was the predominant causative uveitis diagnosis withinthe FN group, identified in 56% of FN patients versus 0% of control subjects (p o 0.001). Idiopathic uveitis was the predominantcausative diagnosis within the control group and was identified in 40% of control patients, compared with 19% of FN uveitispatients (p ¼ 0.002). All therapies except anterior segment surgeries (p ¼ 0.88) were required more often in FN, includingsystemic corticosteroids (p ¼ 0.002), injected corticosteroids (p ¼ 0.042), systemic immunosuppression (p ¼ 0.021), glaucomatherapy (p ¼ 0.005), laser (p o 0.001), and posterior segment surgeries (p ¼ 0.002). Complication rates were higher in FNpatients (p o 0.001). Final visual outcome was worse in FN patients (35% FN r 20/200 versus 9% control subjects; p ¼ 0.001).

Conclusions: Multiple findings differed strikingly between FN and control patients. Uveitis in FN patients was characterized bysignificantly younger age of onset. FN patients were more likely to have bilateral disease, panuveitis, and granulomatous disease.With respect to specific uveitic entities, VKH was more common in FN patients. FN patients also required more aggressivetherapies, and yet had higher complication rates and had poorer visual outcomes.

Objet : Comparaison des démographies, caractéristiques de lʼuvéite et résultats entre patients autochtones des Premières Nations(PN) et les patients uvéitiques non autochtones.

Nature : Étude de cas témoins.Participants : 43 patients PN (80 yeux) et 45 patients témoins non-PN ayant une uvéite (69 yeux).Méthodes : Revue rétrospective des dossiers comparant les démographies (âge, sexe, résidence), les caractéristiques médicales

(latéralité, classifications anatomiques, inflammations granulomateuses et chroniques, associations systémiques, diagnosticsprécis dʼuvéite) et les résultats reliés à la gravité de la maladie (traitements, résultats visuels, complications) entre les patients desPN et les témoins.

Résultats : La moyenne dʼâge au début de la maladie était significativement plus jeune chez les patients PN (30,4 ans) versus celledes témoins (40,2 ans; p o 0,0001). Lʼuvéite bilatérale était significativement plus répandue dans le groupe PN (86 %) versus lestémoins (51 %; p ¼ 0,0005) et lʼuvéite granulomateuse était significativement plus répandue chez les patients PN (53 %) versusles témoins (11 %; p o 0,0001). Il nʼy avait pas de différence significative dans les taux dʼuvéite chronique entre les groupes,relevés chez 74 % des PN versus 60 % des témoins (p = 0,15). Anatomiquement, la panuvéite était la forme prédominantedʼuvéite dans le groupe PN avec une proportion de 67 % (contre 16 % chez les témoins), alors que la classification anatomiquede lʼuvéite antérieure prédominait chez les témoins avec un taux de 73 % (contre 26 % chez les PN; 0o0,001). Le syndrome deVogt Koyanagi Harada (VKH) était le diagnostic étiologique de lʼuvéite prédominant chez le groupe PN, soit 56 % des patients PNversus 0 % chez les témoins (p o 0,001). Lʼuvéite idiopathique a été le diagnostic étiologique prédominant chez le groupetémoin, soit 40 % comparativement à 19 % chez les PN (p ¼ 0,002). Toutes les thérapies, sauf les chirurgies du segmentantérieur (p ¼ 0,88), ont été requises plus souvent chez les PN, y compris les corticostéroïdes systémiques p ¼ 0,21), lescorticostéroïdes injectés (p ¼ 0,042), lʼimmunosuppression systémique (p ¼ 0,002), la thérapie du glaucome (p ¼ 0,005), leschirurgies au laser (p o 0,001) et du segment postérieur (p = 0,002). Les taux de complication étaient plus élevés chez les PN(p o 0,001) et le résultat visuel final était pire chez les PN (35 % PN ≤ 20/200 versus 9 % chez les témoins p ¼ 0,001).

Conclusions : Beaucoup de données diffèrent considérablement entre les patients des Premières Nations et les témoins. Lʼuvéitedes PN se caractérisait par lʼâge significativement plus jeune au début. Ces patients étaient plus sujets à avoir des maladiesbilatérales, des panuvéites et des maladies granulomateuses. Quant aux entités uvéitiques spécifiques, le VKH était plus répanduchez les patients des Premières Nations. Ceux-ci avaient requis des thérapies plus agressives, pour des taux plus élevés decomplications et de plus faibles résultats visuels.

From the Department of Ophthalmology, University of Manitoba,Winnipeg, Canada and the Department of Ophthalmology and VisualSciences, University of Toronto, Toronto, Canada.

Originally received Jan. 1, 2013. Final revision Sep. 18, 2013. AcceptedSep. 22, 2013

Correspondence to Mili Roy, MD, 3200 Erin Mills Parkway, MississaugaON L5L 1W8; roym@cc.umanitoba.ca

Can J Ophthalmol 2014;49:128–1340008-4182/14/$-see front matter & 2014 Canadian OphthalmologicalSociety. Published by Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.jcjo.2013.09.020

128 CAN J OPHTHALMOL—VOL. 49, NO. 2, APRIL 2014

Uveitis in aboriginal Canadians—Roy

In addition to inherent historic precedent, the Canadianaboriginal population today, composed of the First Nations(FN), Metis, and Inuit, represents a rapidly expandinggroup within the overall population, with a significantlyhigher growth rate than non-aboriginal Canadians.1 Healthcare provision for aboriginal peoples may be impacted byspecial considerations including access, disease epidemiol-ogy, and cultural and linguistic perspectives.

The limited ophthalmologic literature to date specific tothe North American aboriginal population has focusedlargely on issues such as diabetic retinopathy,2 age-relatedmacular degeneration,3 and refractive issues.4,5 Despitewide recognition of the relevance of race and genetics inuveitis, literature addressing uveitis in the North Americanaboriginal population is sparse. A few studies regardingspecific single uveitis entities in FN or aboriginal popula-tions have been published.6–8 A few studies have alsotabulated the range of eye diseases encountered withincertain American Indian groups but did not focus onuveitis.9–11 The author was unable to document anyNorth American studies analyzing the overall spectrumof uveitic disease encountered in aboriginal persons.

Mounting evidence in the rheumatologic literatureprovides a strong rationale to study uveitis in an aboriginalcontext. Associations between intraocular inflammation andrheumatologic conditions such as the seronegative spondy-loarthropathies and human leukocyte antigen (HLA) B27positivity, rheumatoid arthritis, and lupus are well-known.In turn, numerous rheumatologic studies have established adistinct predisposition for certain autoimmune conditionswithin some aboriginal populations. Adult rheumatic dis-eases have been documented at 5 to 7 times higher rates inthe aboriginal than the Caucasian population.12 Reactivearthritis, juvenile idiopathic arthritis, and vasculitis aresignificantly more prevalent in the aboriginal than non-aboriginal population.13 Studies of the Haida First Nationdocumented one of the highest prevalence rates worldwideof both HLA B27 positivity and ankylosing spondylitis.14

The largely Cree First Nation population of central Canadahas one of the highest known prevalence rates globally ofrheumatoid arthritis with more severe disease phenotypesand earlier onset.15

This study analyzes the spectrum of uveitic diseaseencountered in a Canadian FN population with compar-ison against an appropriate non-aboriginal control popu-lation. The author was unable to identify any other suchstudies published to date.

METHODS

Approvals from the Assembly of Chiefs representing thelocal FN and from the local university ethics review boardwere obtained. Retrospective chart review was conductedof 43 consecutive FN uveitis patients (80 uveitic eyes) and45 consecutive non-FN control patients (69 uveitic eyes)

from a single geographic catchment area and presenting toa single uveitis subspecialist from 2006 to 2010. Metis andInuit patients comprising aboriginal but non-FN patientswere excluded from both the FN and the control groups.These exclusions were made in respect of the directconsultative process with the Assembly of Chiefs specifi-cally and solely representing the FN population, and tostrengthen methodology by defining the study populationby more stringent criteria to avoid consideration of3 different and distinct aboriginal groups as one.

All patients underwent a standardized intake review ofsystems relevant to uveitis, including documentation ofracial ancestry, and thorough eye examination includingSnellen visual acuity, tonometry, pupillary, anterior seg-ment, and fundus examination. Tailored individualizedsystemic investigations were performed, guided by history,findings, and review of systems. Follow-up and therapywere also individualized as determined by a single practi-tioner, reducing interobserver variability in evaluation andmanagement.

The outcome measures studied encompassed patientdemographics, disease characteristics, and outcomesreflecting disease severity. Every outcome was measuredin both groups and then statistically compared betweenthe control versus FN groups.

Patient demographicsOutcomes measured included patient age at uveitis

onset, age at presentation, sex, follow-up duration, andurban versus rural residency.

Disease characteristicsOne outcome measured included quantification of the

various specific uveitides identified, whether purely ocularsyndromes or systemic associations. In addition, uveitisclassification was documented as acute versus chronic,granulomatous versus nongranulomatous, unilateral versusbilateral, and anatomic classification as per the Stand-ardized Uveitis Nomenclature (SUN).16

Disease severityOutcomes measured to reflect disease severity included

visual outcome analyzed by 2 methods. Final absolute bestcorrected Snellen visual acuities were grouped categoricallyas mild, moderate, or severe visual loss, and the distribu-tions were compared between groups. The net change invision over study duration was also analyzed. Otheroutcomes measured included the need for specific medicaland surgical therapies, and uveitic complications. Medicaltherapies evaluated included systemic and injected cortico-steroids, glaucoma therapy, and systemic immunosuppres-sion. Therapies not directly associated with uveitis, such asglaucoma therapy for pre-existing or nonuveitic glaucoma,were excluded. Surgical therapies studied included laser,anterior segment, and posterior segment surgeries requiredbecause of uveitis. Surgeries required for nonuveitic

CAN J OPHTHALMOL—VOL. 49, NO. 2, APRIL 2014 129

Table 1—Self-reported patient racial ancestries

Ojibway FrenchCree FinnishCree RussianFN United KingdomCree Northern EuropeanOjibway CaucasianOjibway United Kingdom/UkrainianCree UkrainianCree NigerianFN RussianSaulteaux Scottish/GermanDene Russian/FrenchSaulteaux Danish/UkrainianFN GreekSaulteaux UkrainianOji-Cree FilipinoFN Northwest EuropeanCree Mexican/CaucasianDene United Kingdom/French/SwissCree CaucasianFN United Kingdom/ItalianFN HungarianOji-Cree Jamaican/AfricanFN CaucasianCree ScottishOji-Cree United KingdomOji-Cree United KingdomOji-Cree Argentinian/Spanish/PolishFN NigerianCree CaucasianFN Swedish/UkrainianFN United KingdomFN English/ItalianFN English/GermanCree AfricanCree FilipinoCree IrishOjibway East IndianOji-Cree European (Romanian)Cree Icelandic/Hungarian/ScottishOji/Carrier ChineseOjibway EuropeanOjibway Belgian

ChineseCaucasian

FN, First Nations.

86%(37)

74%(32)

53%(23)

51%(23)

60%(27)

11%(5)

Bilateral uvei�s(p=0.0005

odds ra�o 5.90)

Chronic uvei�s(p=0.1527

odds ra�o 1.94)

Granulomatousuvei�s (p<0.0001odds ra�o 9.20)

Percen

tages

First Na�onsControls

( ) = absolute number of cases

Fig. 1—Uveitis characteristics. Dark grey, First Nations; lightgrey, controls. Numbers in parentheses are absolute numberof cases.

Uveitis in aboriginal Canadians—Roy

complications, such as cataract predating uveitis, wereexcluded. Complications cited were rigorously defined.Cataracts were cited as uveitic complications only if gradedas minimum 2þ posterior subcapsular cataract or mini-mum 3þ nuclear sclerosis and arising because of uveitis. Ifmultifactorial such as age-related or predating uveitis,those cataracts were excluded. Uveitic posterior synechiaewere included in analysis if involving a minimum of3 quadrants or requiring peripheral iridotomy (PI).Hypotony was defined as an intraocular pressure of r5mm Hg at a minimum of 2 consecutive visits. Glaucomawas not analyzed as a specific complication except theparticular instance of absolute glaucoma defined as nolight perception vision because of uveitic glaucoma.Otherwise, uveitic glaucoma was reflected in the analysesof medical and surgical interventions required for glau-coma. Multiple other uveitic sequelae, such as macularscarring, edema, (pseudo)holes, epiretinal membranes(ERMs), and others cited, were included as complicationsif identifiable to any extent.

130 CAN J OPHTHALMOL—VOL. 49, NO. 2, APRIL 2014

All outcomes were measured in both groups andcompared statistically between groups. Comparisons ofsex, urban versus rural residency, follow-up duration, aswell as disease characteristics including laterality, granu-lomatous disease, and medical and surgical therapies wereanalyzed with t tests, χ2 tests, and odds ratio analyses. Thedistribution of various specific uveitis diagnoses and typesencountered was compared between groups using non-parametric binomial tests to compare observed incidencerates. Absolute visual outcome and anatomic classificationcomparisons were performed with χ2 testing. An analysisof covariance model was used to compare complications,and an analysis of variance (ANOVA) procedure was usedto compare group mean ages and mean change in visualacuity.

RESULTS

Patient demographicsThe FN group consisted of 43 patients, self-reported as

13 Cree (30%), 6 Ojibway (14%), 6 Oji-Cree (14%),3 Saulteaux (7%), 2 Dene (5%), 1 Ojibway-Carrier (2%),and 12 (28%) of unspecified FN ancestry. Self-reportedracial ancestries of all patients in both the FN and controlgroups are summarized in Table 1. The FN groupincluded 80 uveitic eyes of 43 patients, whereas thecontrol group included 69 uveitic eyes of 45 patients.Mean age at uveitis onset in the FN group at 30.4 � 14.6years (range 4–61 years) was significantly younger than thecontrol group at 40.2 � 18.9 years (range 2.5–88 years; po 0.0001). Age at presentation was also significantlyyounger in the FN group at an average of 32.1 � 15.4years (range 6–62 years) compared with control patients atan average of 44.5 � 18.0 years (range 5–88 years; p o0.0001). Follow-up duration was comparable betweengroups with a mean of 4.4 � 6.2 years in FN versus 4.9� 9.4 years in control patients (p ¼ 0.879).

26%(11)

0%(0)

7%(3)

67%(29)

73%(33)

9%(4) 2%

(1)

16%(7)

anterior intermediate posterior panuvei�s

Percen

tage

First Na�onsControl

( ) = absolute number of cases

P<0.001, comparison of distribu�on of anatomic classifica�on between groups

Fig. 2—Anatomic uveitis classifications. p o 0.001, compar-ison of distribution of anatomic classification betweengroups. Dark grey, First Nations; light grey, controls. Numbersin parentheses are absolute number of cases.

Uveitis in aboriginal Canadians—Roy

Sex distributions were female predominant in bothgroups without significant differences between groups,measuring 34 female (79%) and 9 male (21%) patientsin FN group compared with 29 female (64%) and 16 male(36%) patients in the control group (p ¼ 0.130). Of theFN group, 32 (74%) lived in nonurban settings versus 11(26%) in urban settings. A significantly higher proportionof control patients (32 patients [71%]) resided in urbansettings than FN patients (p o 0.0001, odds ratio 7.16),with 13 (29%) control patients living in nonurbansettings.

Uveitis characteristicsThe comparison of uveitis characteristics including

unilateral versus bilateral, chronic versus acute, andgranulomatous versus nongranulomatous disease areshown in Figure 1. Bilateral and granulomatous uveitiswere both significantly more common in the FN group(p ¼ 0.0005, odds ratio 5.90 and p o 0.0001, odds ratio9.20, respectively). Chronic uveitis was more common

Fig. 3—Specific uveitis diagnoses: First Nations group. Numberetinal necrosis; CMV, cytomegalovirus; HLA, human leukocyte

than acute in both groups without a significant differencebetween groups (p ¼ 0.1527, odds ratio 1.94).

The distribution of anatomic uveitis classification wascompared between groups as shown in Figure 2. Pan-uveitis was the most prevalent anatomic classification inthe FN group, whereas anterior uveitis was prevalent incontrol patients. There was a statistically significant differ-ence between groups in the distribution of uveitis byanatomic classification (p o 0.001).

Each case of uveitis was also tabulated by specificdiagnosis as shown in Figures 3 and 4. Statistical compar-ison of observed incidence rates for specific uveitis typesdemonstrating greater than 5% incidence rate revealedmultiple significant differences in the distribution ofspecific uveitis diagnoses between groups. Vogt–Koya-nagi–Harada (VKH) syndrome was exclusive to the FNgroup and also the most commonly identified causativeuveitis diagnosis within the FN group, occurring in 56%of FN patients versus 0% of control patients (p o0.0001). Cases of VKH syndrome were diagnosed inaccordance with current revised diagnostic criteria(RDC).17 Idiopathic uveitis was the predominant causa-tive diagnosis within the control group and was identifiedin 40% of control patients compared with 19% of FNpatients with uveitis (p ¼ 0.002). Among other specificuveitis types analyzed, HLA B27-associated uveitis andsarcoidosis-associated uveitis were significantly more com-mon in the control group than the FN group (p ¼ 0.003and p ¼ 0.007, respectively).

Disease severityVisual outcomes were analyzed both as final visual

outcome and as net change in vision over study duration.

( ) = absolutenumber of cases

ARN = acute re�nalnecrosis

CMV = cytomegalo-virus

HLA = human leukocytean�gen

rs in parentheses are absolute number of cases. ARN, acuteantigen.

CAN J OPHTHALMOL—VOL. 49, NO. 2, APRIL 2014 131

idiopathic (18)44%

HLAB27/seronega�ve spondyloarthropathy (11)

27%

sarcoidosis (3)8%

mul�focal panuvei�s choroidi�s (2)

5%

herpe�c (1)2%

birdshot re�nochoroidopathy

(1)2%

APMPPE (1)2%

Posner-Schlossman (1)2%

mul�ple sclerosis

associated (1)2% celiac disease

associated (1)2%

electrocu�on induced (1)2%

Fuch's intermediate uvei�s (1)

2%

( ) = absolute number of cases

APMPPE = acute posterior mul�focal placoid pigment epitheliopathy

HLA = human leukocyte an�gen

Fig. 4—Specific uveitis diagnoses: control group. Numbers in parentheses are absolute number of cases. APMPPE, acuteposterior multifocal placoid pigment epitheliopathy; HLA, human leukocyte antigen.

Uveitis in aboriginal Canadians—Roy

Final visual outcomes were classified as mild visual loss(Z20/40), moderate visual loss (20/160 to 20/50), orsevere visual loss (r20/200) for statistical comparison, asshown in Figure 5. The distribution of visual outcomesdiffered significantly between groups (p ¼ 0.001).

The mean change in visual acuity from baseline at studyentry until final visit at study exit was also comparedbetween groups by ANOVA design. The FN groupexperienced a mean net loss of 0.42 line (SD 4.88) ofSnellen acuity, whereas the control group experienced amean net gain of 1.33 lines (SD 3.44). The large SDsreflected multiple extreme outliers particularly in the FNgroup. Box plots for the mean change in visual acuityrevealed there was only 1 extreme outlier in the controlgroup. In the FN group, there were 4 patients less than themean and 3 more than the mean. The difference in changein acuity was significant between groups (p ¼ 0.05) withall patients included, but not statistically significant (p ¼0.14) if extreme outliers (4 controls and 7 FN) wereexcluded, resulting in a recalculation of a net mean gain of

44%(35)

21%(17)

35%(28)

78%(54)

13%(9)

9%(6)

Mild Visual Loss(>/=20/40)

Moderate Visual Loss(20/160-20/50)

Severe Visual Loss(</=20/200)

Percen

tage

s First Na�ons

Controls

( ) = absolute number of cases

P=0.001, comparison of distribu�on of visual outcomes between groups

Fig. 5—Visual outcomes: final absolute visual acuities. p ¼0.001, comparison of distribution of visual outcomes betweengroups. Dark grey, First Nations; light grey, controls. Numbersin parentheses are absolute number of cases.

132 CAN J OPHTHALMOL—VOL. 49, NO. 2, APRIL 2014

0.37 line (SD 2.44) in the FN group versus 1.17 lines gain(SD 2.29) in control patients (p ¼ 0.14).

Additional outcomes measured as markers of diseaseseverity included the need for various therapies andcomplications. Analysis of required medical and surgicalinterventions are summarized in Figures 6 and 7, respec-tively. Laser interventions were required significantly moreoften in the FN group (36%) versus control group (4%;p o 0.001). Twenty-five YAG PIs and 4 argon retinalphotocoagulations were performed in the FN group,whereas all 3 control eyes required YAG PI only. Anteriorsegment procedures were required approximately equallyoften in both groups (39% of FN patients versus 38% ofcontrol patients; p ¼ 0.88) and included cataract surgery,trabeculectomy, Ahmed valve surgery, goniotomy, andband keratopathy chelation. In contrast, posterior segmentinterventions were required significantly more often in FNpatients (12.5%) versus control patients (1.4%; p ¼0.002). These included lensectomy, vitrectomy, ERMpeeling, and retinal detachment repair with scleral buckleand silicone oil.

77%(33)

37%(16) 28%

(12)

51%44%(20)

18%(8) 9%

(4)

(22)

22%(10)

systemiccor�costeroids

(p=0.002)

injectedcor�costeroids

(p=0.04)

systemicimmunosuppression

(p=0.02)

medical glaucomatherapy

(p=0.005)

Percen

tages

First Na�ons

Control

( ) = absolute number of cases

Fig. 6—Medical therapeutic interventions required. Dark grey,First Nations; light grey, controls. Numbers in parentheses areabsolute number of cases.

36%(29)

39%(31)

12.5%(10)

4%(3)

38%(26)

1.4%(1)

Laser surgeries(p<0.001)

Anterior segmentsurgeries (p=0.88)

Posterior segmentsurgeries (p=0.002)

Percen

tages First Na�ons

Control

( ) = absolute number of cases

Fig. 7—Surgical therapeutic interventions required. Dark grey,First Nations; light grey, controls. Numbers in parentheses areabsolute number of cases.

Uveitis in aboriginal Canadians—Roy

The mean number of uveitic complications as definedin the Methods was significantly higher in the FN group at4.6 complications per eye versus the control group at1.5 complications per eye (p o 0.001). Complicationsencountered included cataract, posterior synechiae, periph-eral anterior synechiae, pupillary membranes, anteriorhyaloid membranes, hypotony, phthisis bulbi, cystoidmacular edema, ERMs, retinal detachment, retinal and/or macular scarring, and optic neuropathy.

DISCUSSION

Strikingly different spectra of uveitides and patientdemographics were documented between the FN andcontrol non-FN groups. Uveitis onset occurred at asignificantly younger age, one decade earlier on average,in FN patients. Age at first presentation correspondinglyoccurred significantly earlier by approximately 12.5 years,on average, in the FN group compared with the controlgroup. Sex was comparably skewed toward female pre-dominance in both groups without significant differences.Non-FN patients resided in urban centres in significantlygreater numbers than FN patients.

Comparison of the various uveitic entities encounteredrevealed significant differences in uveitis causative factorsoccurring between the 2 groups. VKH syndrome was thesingle most common uveitis diagnosis in the FN group(56% of all cases) but was not found at all in the non-FNgroup (0%). In contrast, idiopathic uveitis was the mostcommon diagnosis in the non-FN group, representing40% of cases, but the second most common diagnosis inthe FN group occurring at less than half the frequency,representing only 19% of FN cases. The distribution ofanatomic uveitis classification also differed significantlybetween groups, with panuveitis being the prevalentanatomic classification in the FN group versus anterioruveitis being most prevalent in the non-FN group.Bilateral uveitis and granulomatous disease occurred sig-nificantly more often in the FN group than in the controlgroup. Chronic uveitis was more common than acuteuveitis in both groups without significant differences.

Multiple outcomes reflecting disease severity evidencedsignificantly greater uveitis severity in the FN than thecontrol group. The final absolute visual acuity wassignificantly lower in the FN group. Visual change frombaseline declined overall in the FN group but improvedoverall in the control group. This was significant when allpatients were included in the analysis but was notsignificant if extreme outliers were excluded. The overallcomplication rates and need for all medical and surgicaltherapies studied except anterior segment surgeries weresignificantly higher in the FN group than the controlgroup. Therapies required with greater frequency in theFN group included systemic and injected corticosteroids,systemic immunosuppressives and medical glaucoma ther-apy, laser, and posterior segment surgical interventions.Anterior segment surgeries were needed at approximatelyequal rates in both groups, possibly reflecting the higherfrequency of anterior uveitis noted in the control groupcompared with the FN group.

Limitations of this study include relatively small num-bers of patients studied. This reflects the low frequency ofthe disease under scrutiny, particularly when selecting foroccurrence specifically in FN patients. Notwithstanding,this study brings to light significant new findings regardingthe range of uveitic disease in North American FN. Aninherent selection bias exists in the series because patientswere recruited from a subspecialty uveitis practice, con-ceivably skewing toward more complex and chronic cases.However, this bias applies equally to both the case (FN)and control (non-FN) groups and, therefore, does notweaken the strength of comparative analyses betweengroups.

To the best of the authorʼs knowledge, this is the firstand largest study examining a representative spectrum ofuveitis in a North American aboriginal FN population.Significant differences in patient demographics, diseasecharacteristics, severity, and outcomes in FN patientscompared with non-FN patients with uveitis are docu-mented. Conclusions cannot be directly extrapolated toother FN or aboriginal populations but aid in expandingthe hitherto limited understanding of the issues. Pertinentophthalmologic literature has been lacking to date andrequires further elucidation.

This study complements emerging rheumatologic liter-ature in a North American aboriginal context. Multiplestudies have documented a distinct predisposition forcertain rheumatologic conditions among some FN pop-ulations including very high prevalence and more severedisease phenotypes. A recent ophthalmologic studyshowed approximately comparable relative prevalence ratesof uveitis within this Canadian FN study population ascompared with the general non-FN population.18 Incontradistinction to the similar prevalence rates betweenthe 2 groups shown in the former study, this studydocuments a very different spectrum of uveitis occurringin this FN population as compared with the non-FN

CAN J OPHTHALMOL—VOL. 49, NO. 2, APRIL 2014 133

Uveitis in aboriginal Canadians—Roy

population. The FN group demonstrates significantlyearlier uveitis onset, different prevalent causative factorsand anatomic involvement, as well as greater diseaseseverity, higher complication rates, and greater need fortherapies, whereas experiencing significantly poorer out-comes compared with the non-FN group.

In managing uveitis care in aboriginal populations, thepresent findings are compounded by myriad cultural andlinguistic issues. Access to care is impacted by theprimarily nonurban residence base of FN patients. Beyondthe scope of this report, comorbidities from other healthconcerns such as diabetes and endemic tuberculosis alsoimpact uveitis care in FN patients. It is understood thatthe impact of uveitis including disease duration, lostproductivity, reproductive implications when systemicimmunosuppression is required,19 and need for compen-satory supports is amplified by the relatively youngpopulation largely affected, often of working and repro-ductive age.20 This underlines the importance of specialconsiderations in furthering our understanding of uveitisin the unique FN population and tailoring future healthcare delivery strategies accordingly.

Disclosure: The authors have no proprietary or commercialinterest in any materials discussed in this article.

Acknowledgements: The author acknowledges M.L. Schmuck(The Programme for Educational Research and Development,McMaster University, Hamilton, Ont.) and Dr. R. Tate (Depart-ment of Community Health Sciences, Faculty of Medicine,University of Manitoba, Winnipeg, Man.), who contributedsignificantly to analyses performed.

REFERENCES

1. Statistics Canada 2006 Census [Internet]. Aboriginal peoples: releaseno. 5. 2008 Jan 15 [updated 2011 May 25; cited 2011 Oct 2].Statistic Canada: Ottawa, Ont. http://www12.statcan.ca/census-recensement/2006/rt-td/ap-pa-eng.cfm.

2. Silver K, Williams M, Macario E. The National Eye HealthEducation Program: increasing awareness of diabetic eye diseaseamong American Indians and Alaska Natives. Ethn Dis. 2006;16:920-5.

134 CAN J OPHTHALMOL—VOL. 49, NO. 2, APRIL 2014

3. Butt AL, Lee ET, Klein R, et al. Prevalence and risk factors of age-related macular degeneration in Oklahoma Indians: the VisionKeepers Study. Ophthalmology. 2011;118:1380-5.

4. McClure TM, Choi D, Wooten K, et al. The impact of eyeglasseson vision-related quality of life in American Indian/Alaska Natives.Am J Ophthalmol. 2011;151:175-182.e2.

5. Harvey EM, Dobson V, Miller JM. Prevalence of high astigmatism,eyeglass wear, and poor visual acuity among Native American gradeschool children. Optom Vis Sci. 2006;83:206-12.

6. Nussenblatt RB. Clinical studies of Vogt-Koyanagi-Harada’s diseaseat the National Eye Institute, NIH, USA. Jpn J Ophthalmol.1988;32:330-3.

7. Davis JL, Kamal KM, Freidlin V, et al. HLA associations andancestry in Vogt-Koyanagi-Harada disease and sympathetic oph-thalmia. Ophthalmology. 1990;97:1137-42.

8. Martinez JA, Lopez PF, Sternberg P Jr, et al. Vogt-Koyanagi-Haradasyndrome in patients with Cherokee Indian ancestry. Am JOphthalmol. 1992;114:615-20.

9. Bettman JW Jr. Eye disease among American Indians of theSouthwest. I. Overall analysis. Arch Ophthalmol. 1972;88:263-8.

10. Lee ET, Russell D, Morris T, et al. Visual impairment and eyeabnormalities in Oklahoma Indians. Arch Ophthalmol. 2005;123:1699-704.

11. Mansberger SL, Romero FC, Smith NH, et al. Causes of visualimpairment and common eye problems in Northwest AmericanIndians and Alaska Natives. Am J Public Health. 2005;95:881-6.

12. Jarvis JN, Cleland SY. Rheumatic disease in Native American children:opportunities and challenge. Curr Rheumatol Rep. 2003;5:471-6.

13. Barnabe C, Elias B, Bartlett J, et al. Arthritis in AboriginalManitobans: evidence for a high burden of disease. J Rheumatol.2008;35:1145-50.

14. Gofton JP, Chalmers A, Price GE, et al. HL-A 27 and ankylosingspondylitis in B.C. Indians. J Rheumatol. 1984;11:572-3.

15. El-Gabalawy HS, Robinson DB, Daha NA, et al. Non-HLA genesmodulate the risk of rheumatoid arthritis associated with HLA-DRB1 in a susceptible North American Native population. GenesImmun. 2011;12:568-74. http://dx.doi.org/10.1038/gene.2011.30(Epub 2011 May 26).

16. Jabs DA, Nussenblatt RB, Rosenbaum JT. Standardization ofUveitis Nomenclature (SUN) Working Group Standardization ofuveitis nomenclature for reporting clinical data. Results of theFirst International Workshop. Am J Ophthalmol. 2005;140:509-16.

17. Read RW, Holland GN, Rao NA, et al. Revised diagnostic criteriafor Vogt-Koyanagi-Harada disease: report of an internationalcommittee on nomenclature. Am J Ophthalmol. 2001;131:647-52.

18. Roy M. Relative prevalence of Aboriginal patients in a Canadianuveitic population. Can J Ophthalmol. 2012;47:185-8.

19. Wakefield D, Abu El-Asrar A, McCluskey P. Treatment of severeinflammatory eye disease in patients of reproductive age and duringpregnancy. Ocul Immunol Inflamm. 2012;20:277-87.

20. De Smet MD, Taylor SR, Bodaghi B, et al. Understanding uveitis:the impact of research on visual outcomes. Prog Retin Eye Res.2011;30:452-70.