Prevalence of open angle glaucoma in the Faroese population

Abstract

Purpose:

The Faroe Islands are home to 50,000 genetically isolated people in the North Atlantic. The prevalence of open angle glaucoma (OAG) in the Faroese population is unknown. Consequently, we conducted a survey to determine the prevalence of OAG in the Faroese population. We also investigated the role of known glaucoma-causing genes in Faroese OAG.

Methods:

We conducted a prospective survey of known and newly diagnosed glaucoma patients at the Faroese National Hospital, Landssjukrahusid, Tórshavn between Oct. 1^st 2015 to 31th December 2017. In addition we reviewed the only eye care provider in the Faroese Islands by scrutinizing electronic medical records between 2009 and June 15, 20014 Oct. 1^st 2015 and the partly overlapping prescriptions for ocular hypotensive medications in 2016 to identify patients with either a diagnosis of glaucoma, a diagnosis of ocular hypertension or a prescription for ocular hypotensive medications. Next, we prospectively confirmed diagnoses with complete eye examinations. Patient DNA samples were tested for variations in known glaucoma-causing genes (MYOC, OPTN, and TBK1).

Results:

We determined the age-related prevalence of OAG January 1, 2017 in individuals 40 years or older to be 10.7/1,000 (1.07%) and highly age-related. A diagnosis of OAG was present in 264 patients, of whom 211 (79.9%) had primary open angle glaucoma (including normal tension glaucoma), 49(18.6%) had pseudo-exfoliation glaucoma (PXG), and 4 (1.5%) had pigmentary glaucoma (PG). Among patients receiving medications for glaucoma, nearly 50% had POAG, while the majority of the rest had ocular hypertension or secondary glaucoma. No disease-causing variants were detected in MYOC, OPTN, or TBK1.

Conclusion:

The calculated prevalence of OAG in the Faroe Islands was 1.07%. The absence of MYOC, OPTN, or TBK1 disease-causing variants in Faroese POAG patients suggests that a different, potentially unique set of genes may be contributing to the pathogenesis of glaucoma in this population.

Glaucoma is a progressive optic neuropathy that is a major cause of irreversible blindness in the world.^{1, 2}. Glaucoma is diagnosed by structural assessment of the head of the optic nerve (cupping) and functional assessment of the visual fields. The most common form of glaucoma is primary open-angle glaucoma (POAG).^3–5 The prevalence of POAG varies greatly between different racial and ethnic groups⁶, with the highest frequencies in blacks from African and Caribbean nations (2.9 to 8.8%).^7–10 The prevalence of POAG in whites with Northern European ancestry ranges from 1.44 to 3.0%^{7, 11, 12} and is similar to that of Asians from Japan (2.6–3.9%),^{13, 14} Korea 3.5%,¹⁵ and China (0.71–1.0%).^{16, 17}

The Faroe Islands are located in the North Atlantic Ocean, between Iceland, Norway, and the British Isles. Due to the remote location, the approximately 50,000 inhabitants of the Faroe Islands have been genetically isolated for centuries and analysis of this population has led to development of an 8–12 generation genealogical database. These population features, combined with a national pharmacy registry and a centralized Faroese ophthalmological service that are the sole providers of eye care to the Faroese, have facilitated genetic and epidemiological studies of eye disease.^18–22 However, there are no prior studies of the prevalence of primary open angle glaucoma on the Faroe Islands and its genetic basis.

Historical and archeological evidence suggests that the Faroese have mixed heritage, with ancestry from both Norse populations and from the British Isles.^{23, 24} Faroese language, however, has greatest similarities with Icelandic and old Norwegian languages but also has Celtic (Scottish and Irish) influences.^{25, 26} More recently, the ancestry of the Faroese has been explored with genetic analyses. Studies of genetic markers on the Y-chromosome suggested that the Faroese had Norwegian, Swedish, Icelandic, and Celtic male ancestors.²⁷ Similarly mitochondrial markers were used to provide evidence that a large proportion of Faroese female ancestors were from the British Isles.²⁸ The genetic analyses of heredity of the Faroese are consistent with historical data.

The prevalence of glaucoma in the populations ancestral to the Faroese range from 0.76 – 1.72% in Denmark,^{29, 30} 0.93 – 1.55% in Sweden,^{29, 31} 0.54 – 2.19% in Norway,^{29, 32} 2.03 – 4.0% in Iceland,^{29, 33} 2.16% in Scotland,³⁴ to 2.0% in Ireland.³⁵ The prevalence of glaucoma in the Faroe Islands, however, has not yet been determined.

Heredity has a major role in the pathogenesis of glaucoma.^{36, 37} Most cases of POAG are caused by the combined influence of numerous genetic and environmental influences, however variants in the myocilin (MYOC),³⁸ optineurin (OPTN),³⁹ and TANK binding kinase 1 (TBK1)⁴⁰ genes are responsible for approximately 5% of POAG cases.⁴¹ The role of these three genes in the pathophysiology of POAG in the Faroese is unknown.

This study sought to determine the prevalence of open angle glaucoma in the Faroese and to determine the role of known POAG-causing genes (MYOC, OPTN, and TBK1) in its pathogenesis.

Material and Methods

Faroese patient population.

All study participants provided informed consent and this study adhered to the tenets of the Helsinki Declaration II and was approved by human research / Faroese data protection agency Dátueftirlitið (approval #14-00042-3).

The Department of Ophthalmology at the National Hospital in Tórshavn is the only provider of ophthalmic care on the Faroe Islands. The Faroese National Pharmacy Registry records all prescriptions for glaucoma medications provided to citizens of the Faroe Islands. Consequently, review of medical and pharmacy records from the National Hospital and the National Pharmacy registry is considered to have complete capture of all known patients with glaucoma and patients receiving glaucoma medications.

Faroese citizens over the age of 40 were enrolled in this study between October 1^st, 2015 and December 31^st, 2017 by three means:

• Review of prior medical records. A search for patients aged 40 years or older with either OAG or ocular hypertension and seen at the ophthalmology clinics at the National Hospital in Tórshavn between 2009 and 2014 was made. The data were extracted from a clinical database (https://www.cambiogroup.com) established in 2009 including controls of patients diagnosed before 2009. All prevalent glaucoma patients identified by review of their medical records were invited (by phone and by mail) to return to the eye clinic at the National Hospital in Tórshavn for participation in the study if they had not already undergone a complete examination at this site by one of the authors (E.H., M.H., or K.V.).

• Review of pharmacy records. We screened the Faroese National Pharmacy Registry for active prescriptions for glaucoma medications that were issued between January 1^st, 2016 and December 31^st 2016. Specifically, we searched records for prescription eyedrops containing beta-blocking medications (timolol, betaxolol); prostaglandin analogue medications (latanoprost, bimatoprost, travaprost, tafluprost); carbonic anhydrase-inhibitating medications (dorzolamide, brinzolamide, acetazolamide); alpha-agonist medications (brimonidine, apraclonodine); and cholinergic medications (pilocarpine). Faroese patients receiving glaucoma medications were identified, their medical files consulted and patients with a diagnosis of either OAG or OH were invited to return to the eye clinic at the National Hospital in Tórshavn for participation in the study if they had not already undergone a complete examination at this site by one of the authors (E.H., M.H., or K.V.).

• Clinic population. Additional patients were prospectively evaluated when they presented to the National Hospital in Tórshavn for routine eye care between October 1^st, 2015 to December 31^th, 2017.

Clinical examination.

All patients underwent a standard ophthalmological examination including Snellen vision acuities, slitlamp examination, indirect ophthalmoscopy, and applanation tonometry. When glaucoma was suspected further examinations included gonioscopy, Humphrey 24-2 SITA visual field tests (Carl Zeiss Meditec Inc., Dublin, CA, USA), and OCT analysis of the retinal nerve fiber layer (Topcon 3D OCT-2000 Topcon Corporation, Japan). DNA was prepared from blood samples as we have previously described.⁴⁰

Patients were judged to have open angle glaucoma if they had optic nerve cupping and/or glaucomatous visual fields as previously described.⁴² Patients who met criteria for open angle glaucoma but were also found to have radial iris transillumination defects, Krukenberg spindles, Scheie stripes, or heavy pigmentation of the trabecular meshwork on gonioscopy were diagnosed with pigmentary glaucoma. Patients who met criteria for open angle glaucoma, but were also found to have exfoliation material on the anterior segment structures (anterior lens capsule, iris, corneal endothelium, or trabecular meshwork) were diagnosed with exfoliation glaucoma.⁴³ Finally, those open angle glaucoma patients with no signs or history of trauma, steroid use, pigment dispersion syndrome, or exfoliation syndrome, were diagnosed with primary open angle glaucoma (POAG). Patients with POAG that had a maximum IOP ≤ 21 mm Hg were further diagnosed with normal tension glaucoma (NTG). Patients with insufficient clinical data (possible or probable glaucoma) were included as OAG patients in this study.

Epidemiological calculations.

The Faroe Islands total population of 24,635 aged 40 years or more in 2019 was derived from Statistics Faroe Islands (www.hagstova.fo). Our approach of reviewing the medical and pharmacy records of the sole healthcare providers to the Faroese had near complete capture of all known glaucoma patients. Consequently, we used the over 40-year population of the Faroe Islands, 24,635, for the denominator in our prevalence calculations. Similarly, we used total population figures from the Faroe Islands in prevalence calculations for subsets of the over 40 year population.

Genetic Testing.

DNA was prepared from blood samples using standard techniques and each patient’s DNA sample was tested for variants in genes previously associated with adult-onset POAG, myocilin (MYOC), optineurin (OPTN), and TANK binding kinase 1 (TBK1). Patients were tested for MYOC variants with Sanger sequencing of exon 3 as we have previously described.⁴⁴ A commercially available real-time PCR assay (TaqMan, Applied Biosystems, Waltham, MA) was used to detect the Glu50Lys OPTN variant and any positive results were confirmed with Sanger sequencing as previously described.⁴⁵ Two commercially-available real-time PCR assays were used to detect copy number variations in the TBK1 gene and positive results were confirmed with microarray analysis as previously described.⁴⁶

Results

The Faroe Islands have a completely centralized healthcare system, where all of the ophthalmologists are at the National Hospital at Tórshavn. Similarly, all medicines prescribed to Faroese are recorded in the National Pharmacy Registry. Consequently, virtually all known cases of glaucoma on the Faroe Islands are documented in the national medical and pharmacy records. We sought to identify the prevalence of glaucoma on the Faroe Islands by reviewing these medical and pharmacy records and confirming diagnoses with prospective eye examinations.

The review of prior medical records identified 467 living patients with a diagnosis of either glaucoma or ocular hypertension. The national pharmacy registry identified 527 living patients with prescription redemption records that suggested a diagnosis of glaucoma or ocular hypertension. A comparison between the two sources showed an overlap of 275 individuals (Table 2). Eye exams were arranged to determine which patients met our criteria for a diagnosis of glaucoma. During the study, we also conducted eye exams on 9,860 residents of the Faroe Islands and identified 43 new glaucoma cases of whom 27 also were recorded in the pharmacy registry of patients receiving glaucoma medications. The remaining 16 cases were identified among patients who did not receive a prescription for glaucoma medications during 2016 (Table 2). After prospective eye exams and applying inclusion criteria for open angle glaucoma (OAG), we identified a total of 264 patients with a confirmed diagnosis of open angle glaucoma (OAG). A total of 180 of the 264 confirmed OAG cases were identified from review of both prior medical and pharmacy records. A smaller number of OAG cases were initially identified only by review of pharmacy records (56 cases), only by review of medical records (12 cases), or were identified as new cases not recorded in the pharmacy registry (16 cases). The majority of these 264 OAG patients had primary glaucoma, 172 (65.2%) had POAG and 39 (14.8%) had normal tension glaucoma (NTG). The remaining patients had secondary open angle glaucoma, 49 (18.6%) had PXG and 4 (1.5%) had pigmentary glaucoma (PG) (Figure 1).

Table 2.

Overview of the three partial overlapping data sources

	Medical record review	Overlap between medical and pharmacy records	Pharmacy record review	Overlap between pharmacy record review and new cases	New OAG diagnosis	Total confirmed OAG cases
	Jan 1, 2009 to July 15, 2014		Jan 1, 2016 to Dec 31, 2016		Oct 1, 2015 to Dec 31, 2017
Glaucoma suspects identified	604		558		43
Glaucoma suspects excluded
Deceased prior to exam	106
Diagnosis retracted in records	31		31
Glaucoma suspects evaluated for POAG	467	275	527	27	43
Confirmed OAG diagnosis (10/1/2015 – 12/31/2017)	192	180	236	27	43	264

Figure 1. Open angle glaucoma cases detected in the Faroe Islands.

A total of 264 cases of open angle glaucoma were detected including 172 (65%) primary open angle glaucoma cases, 39 (15%) normal tension glaucoma cases, 49 (19%) exfoliation glaucoma cases, and 4 (1.5%) pigmentary glaucoma cases.

The age and gender-specific distribution of OAG is presented in Table 1 together with Faroese population aged 40 years or more as of January 1^st 2018. The prevalence of OAG rose exponentially from 0.03% in the age group 40–49 years old to 5.49% in the age group 85–89 years old. The average prevalence rates for OAG in individuals aged 40 years or more was 1.07% with a higher frequency in males (1.24%) than females (0.89%).

Table 1.

Five-year age- and gender specific prevalence rates of open angle glaucoma in the Faroese population January 1, 2018

	Males			Females			Males and Females
Age (years)	Cases of OAG	Population	Prevalence (rate/1,000)	Cases of OAG	Population	Prevalence (rate/1,000)	Cases of OAG	Population	Prevalence (rate/1,000)
40–44	0	1,701	0	0	1,480	0	0	3,181	0
45–49	1	1,668	0.6	1	1,539	0.6	2	3,207	0.6
50–54	1	1,867	0.5	1	1,616	0.6	2	3,483	0.6
55–59	11	1,572	7.0	2	1,564	1.3	13	3,136	4.1
60–64	11	1,472	7.5	3	1,383	2.2	14	2,855	4.9
65–69	20	1,404	14.2	10	1,343	7.4	30	2,747	10.9
70–74	33	1,176	28.1	18	1,051	17.1	51	2,227	22.9
75–79	25	791	31.6	23	801	28.7	48	1,592	30.2
80–84	29	486	59.7	16	576	27.8	45	1,062	42.4
85–89	19	297	64.0	22	450	48.9	41	747	54.9
90–94	6	97	61.9	11	238	46.2	17	335	50.7
95–99	0	15	0	1	48	20.8	1	63	15.9
40–99	156	12,546	12.4	108	12,089	8.9	264	24,635	10.7

Our review of the national pharmacy records identified 527 Faroese residents over 40 years old that were prescribed glaucoma medications including 24 who were registered citizens elsewhere. Among the remaining 503 Faroese citizens receiving anti-glaucomatous therapy the 236 (46.9%) met our diagnostic criteria for OAG, while 140 (27.8%) had ocular hypertension and 8 (1.6%) had chronic narrow angle glaucoma. In an additional 22 patients (4.4%) the suspicion for glaucoma was not confirmed by prospective examination. A diagnosis of glaucoma could not be confirmed for other reasons in another 16 patients (3.2%). Finally, 81 individuals (16.1%) were treated for secondary glaucoma. Among the 81 cases, glaucoma was secondary to anterior uveitis in 20 (24.7%), to retinal detachment in 18 (22.2%), to vascular disease in 14 (17.3%), to trauma in 8 (9.9%), to a surgical procedure in 8 (9.9%), and to congenital anomaly in 6 (7.4%). In seven cases (8.6%) the primary eye disease was unknown.

Data on intra-ocular pressure (IOP) range and mean, inter-ocular differences range and mean as well as ocular and systemic morbidity were calculated for each of the four OAG types and presented in Table 3.

Table 3.

Clinical data from the Faroese glaucoma population.

	IOP Right Range (Median)	IOP Left Range (Median)	R-L Difference Range (Median)	Diabetes N (%)	Vascular Disease N (%)	Cataract Extraction (%)
POAG	12–55 (26)	13–56 (26)	0–39 (3)	27 (15.7)	104 (60.5)	76 (44.2)
NTG	8–21 (19)	11–21 (18)	0–9 (1)	10 (25.6)	28 (71.8)	13 (33.3)
PXG	14–45 (25)	23–50 (26)	0–32 (6)	7 (14.3)	38 (77.6)	26 (53.1)
PG*	16	34	18	1 (25)	2 (50)	1 (25)

Intraocular pressure (IOP) mmHg at diagnosis; R-L difference (inter-ocular difference); N, number; primary open angle glaucoma (POAG); normal tension glaucoma (NTG); exfoliation glaucoma (XFG); pigmentary glaucoma (PG).

^*IOP at diagnosis was available in 25% of patients.

DNA samples were available from 194 (62%) of the 264 OAG patients (130 POAG, 30 NTG, 27 PXG, and 7 PG) and 89 patients with ocular hypertension. Familial clustering was seen among 61.6% of patients with OAG. All samples were tested for glaucoma-causing variants in the MYOC, OPTN, and TBK1 genes. No plausible disease-causing variants were detected in these genes. A single synonymous coding-sequence variant in the MYOC gene, however, was detected in 6.0% of the Faroese cohort. This silent mutation (Tyr347Tyr) was previously reported at similar frequency in control populations and has been judged to be a benign MYOC polymorphism.⁴⁷

Discussion

The Faroe Islands is an archipelago consisting of 18 islands situated in the North Atlantic between UK and Iceland. The Islands were inhabited about 1,000 years ago and today the population comprises 50,000 individuals of whom the majority descends from a very limited number of ancestors. It is therefore possible that the occurrence of glaucoma in the Faroe Islands may be strongly influenced by this population’s isolated geography, unique history, and genetic architecture.

We performed the first formal epidemiological study of glaucoma on the Faroe Islands and we determined that the prevalence of OAG in this population is 1.07% in adults over 40 years old, a prevalence that is lower than what has been observed in Norse, Irish, and Scottish populations from which the Faroese were derived. The relatively low prevalence of glaucoma detected on the Faroe Islands may be due in part to differences in study designs.

We used a stepwise search for glaucoma patients with an initial identification of subjects likely to have glaucoma (based on medical and pharmacy records review (Table 2)), followed by a confirmatory eye exam. When these patients were examined, many did not meet criteria for a diagnosis of glaucoma. We also screened for glaucoma during examinations of 9,860 Faroese citizens referred to the ophthalmological department at the National Hospital during the period of inclusion. The studies of glaucoma prevalence in Norse and European populations used different methods including prospective eye exams, reviews ophthalmic medical records in a tertiary care center, and reviews of population-based glaucoma medication usage.^{29, 31–34} Founder effects and genetic isolation may be alternative explanation for the lower frequency of glaucoma detected on the Faroe Islands. It is possible that the Faroe Islands population was established by a relatively small number of founders that by chance had a low prevalence of glaucoma and they produced descendants, the current population of the Faroe Islands, that also have a low prevalence of glaucoma.

Our study of the prevalence of glaucoma on the Faroe Islands confirmed a strong association between OAG and increasing age. We detected a log-linear increase in glaucoma prevalence rates of 0.59% per decade with the highest rate in the 85–89 year-interval after which a decline was seen (Table 1). Most of the OAG cases 211 (80%) of 264 were primary glaucoma (Figure 1), either primary open angle glaucoma, 172 cases (65%), or normal tension glaucoma, 39 cases (15%). The proportion of normal tension glaucoma cases detected in the Faroe Islands was lower than the 30 to 40% proportion of OAG cases detected in other Caucasian populations.^48–50 One explanation for this difference may be ascertainment bias. A comparison between screening examinations and routine clinical examinations in Malmo, Sweden detected a higher proportion of normal tension glaucoma patients (52.9%) in a screened population than in an ophthalmological clinical practice (13%).⁵¹ This study suggests that normal tension glaucoma may be overlooked in routine ophthalmological examinations. In the same study, the opposite finding was found for exfoliation glaucoma, which was detected at a lower prevalence with screening examinations (16%) than in clinical practice (44%).⁵¹ Alternatively, normal tension glaucoma may have been rare in the founders of the Faroe Islands population, which resulted in a lower prevalence in the current population.

A strength of our study is the centralized ophthalmologic service at the National Hospital in Tórshavn that provides a high proportion of the citizens of The Faroe Islands with eye examinations. The centralized Faroe Island’s healthcare system facilitates an effective capture of the vast majority of the country’s glaucoma patients. Moreover, a significant proportion, almost 20%, of the Faroese population was examined, which suggests the data we collected is representative of the group as a whole. Given these strengths, we suggest that our estimate of glaucoma prevalence has similar validity to prevalence calculations from population-based studies. It is known that the prescription rates of glaucoma medications overestimate the prevalence of glaucoma. Many patients taking glaucoma medications, such as patients with ocular hypertension, do not meet criteria for a diagnosis of glaucoma. We used pharmacy records as a first step in identifying glaucoma patients on the Faroe Islands, and we similarly found that only a fraction of those individuals actually had primary glaucoma when they had an eye examination. Among 528 Faroese who filled prescriptions for glaucoma medicines in 2016, only 45% had glaucoma, confirming that estimates of open angle glaucoma prevalence using this method may result in considerable overestimates of disease.

Our study had weaknesses as well. Our review of medical and pharmacy records focused enrollment on patients that had received eye examinations and a diagnosis or prescription for glaucoma or ocular hypertension, which might produce an enrollment bias. However, this potential bias is mitigated somewhat by our study’s large sampling (20%) of the Faroese population. The glaucoma population included mainly elderly people among whom some had disabilities, which influenced their ability to participate in the follow-up examinations or take part in the visual field measurements. These cases were included despite clinical evidence that only qualified for a classification as possible or probable glaucoma. The decline in prevalence rates after 90 years of age suggests that the detection of glaucoma in this part of the population may be poor, which led to a low estimated prevalence of glaucoma. However, due to the small size of this age group the influence of the overall glaucoma prevalence is likely to be minimal.

A large proportion (61.6%) of Faroese OAG patients have reported a family history of glaucoma. The Baltimore Eye Survey and the Blue Mountains Eye study reported much lower rates of positive family history in patients with glaucoma, 16.1% and 15.7% respectively.^{52, 53} The more frequent positive family history of glaucoma in the Faroese glaucoma patients might be due to differences in methods, study size, and/or unique features of the isolated population of the Faroe Islands. A higher frequency of positive family history might suggest an especially strong genetic basis for glaucoma on the Faroe Islands.

In numerous studies, approximately 3 to 4% of cases of primary open angle glaucoma cases are caused primarily by MYOC variants^{38, 44} and 1% of normal tension glaucoma cases are caused by MYOC,⁵⁴ OPTN,^{39, 45} or TBK1 variants.^{40, 55, 56} However, no defects in these genes were detected among the tested study participants. The small sample size of patients may explain the absence of relatively rare OPTN and TBK1 variants. Our genetic tests were designed to detect mutations that had been most strongly associated with glaucoma in prior studies (mutations in the third exon of MYOC, the Glu50Lys mutation in OPTN, and TBK1 gene duplications). Consequently, it is possible that other mutations in these genes were not detected by our assays. It is also possible that the genetic basis of glaucoma in the Faroe Islands is different than in other populations derived from Europe, perhaps due to geographical isolation and a limited number of population founders. These same factors may also suggest that glaucoma on the Faroe Islands is due to a unique set of disease-causing genes (or possibly to undetected MYOC, OPTN, or TBK1 mutations) and that further study to identify such causes is warranted.

In summary, we have conducted the first formal study of glaucoma prevalence in the Faroe Islands and we estimate that 1.07% of inhabitants over 40 years old have open angle glaucoma, with the majority of them having primary open angle glaucoma (65%) or normal tension glaucoma (15%). Secondary open angle glaucomas accounted for 20% of cases, which were attributable to exfoliation syndrome (18.6%) or pigment dispersion syndrome (1.5%). Furthermore, our studies showed that glaucoma on the Faroe Islands is not due to mutations in the known single-gene causes of disease (MYOC, OPTN, or TBK1), suggesting that this population may harbor mutations in a unique set of glaucoma-causing genes.