ORIGINAL STUDY
Detection of Glaucoma and Its Association With Diabetic Retinopathy in a Diabetic Retinopathy Screening Program Rita A. Gangwani, FRCSEd, FHKAM (Ophthalmology),*w Sarah M. McGhee, PhD, FFPH,z Jimmy S.M. Lai, FRCS, FRCOphth, FHKAM (Ophthalmology), M.Med (Ophthalmology), MD,*w Christina K.W. Chan, MPH, BSc,z and David Wong, FRCS, FRCOphth*w
Purpose: To determine the type of glaucoma in subjects with diabetes mellitus detected during a diabetic retinopathy screening program and to determine any association between diabetic retinopathy (DR) and glaucoma. Materials and Methods: This is a population-based prospective cross-sectional study, in which subjects with diabetes mellitus underwent screening for DR in a primary care outpatient clinic. Digital fundus photographs were taken and graded for presence/ absence and severity of DR. During this grading, those fundus photographs showing increased cup-to-disc ratio (CDR) (Z0.6) were identified and these patients were referred to the specialist ophthalmology clinic for detailed examination. The presence of glaucoma was established based on CDR and abnormal visual field (VF) defects according to Hodapp-Parrish-Anderson’s criteria. An elevation of intraocular pressure was not required for the diagnosis of glaucoma. The patients said to have definite glaucoma were those with vertical CDR >/= 0.6, glaucomatous defects on VF examination, or retinal nerve fiber thinning if VF was unreliable. Results: Of the 2182 subjects who underwent screening, 81 subjects (3.7%) had increased CDR and 40 subjects (1.8%) had confirmed glaucoma. Normal-tension variant of primary open-angle glaucoma was the most prevalent type (1.2%) We did not find any evidence that DR is a risk factor for glaucoma [odds ratio for DR vs. no DR = 1.22 (95% confidence interval, 0.59-2.51)]. Conclusion: The overall prevalence of glaucoma in this diabetic population, based on finding increased cupping of optic disc in a teleretinal screening program was 1.8% (95% confidence interval, 1.0-3.0). Key Words: glaucoma, diabetic retinopathy, screening
(J Glaucoma 2016;25:101–105)
head and corresponding loss of visual field (VF), and increased CDR can be a sign that can alert a clinician in an otherwise asymptomatic patient. In our DR screening program, we screen approximately 70,000 patients a year and the burden of referral is substantial. To exclude glaucoma we will require repeated testing of intraocular pressure (IOP) as well as VF and other tests as appropriate. Raised IOP is the only modifiable risk factor that can slow the progression of the disease. Crucially, the amount of investigation and follow-up depends on the type of glaucoma detected. Diabetes mellitus (DM) causes microvascular damage and affects vascular autoregulation of the optic nerve resulting in optic nerve damage which may exacerbate glaucomatous optic neuropathy.3–6 Diabetes compromises glial and neuronal functions making retinal neurons and ganglion cells more susceptible to glaucomatous damage.7,8 Chronic hyperglycemia is also associated with increased IOP in patients with diabetes.9–12 Despite these links, there is still some debate about the association between glaucoma and DM. Some studies have found no association between DM and open-angle glaucoma,12–17 whereas others have found a positive association between diabetes and openangle glaucoma.18–22 Ong et al1 published a paper on detection of early glaucoma in a UK population. However, the prevalence of glaucoma in a predominantly Chinese diabetic population is unknown. We undertook this study to determine the prevalence and type of glaucoma in subjects with diabetes based on detection during DR screening and to investigate whether there was any association between DR and glaucoma.
MATERIALS AND METHODS
A
diabetic retinopathy (DR) screening program can identify many patients with other eye diseases, one of which—increased cup-to-disc ratio (CDR) is an important finding that requires referral.1 Glaucoma is the second leading cause of blindness globally.2 It is a progressive optic neuropathy with characteristic changes in the optic nerve Received for publication August 8, 2013; accepted August 22, 2014. From the *Department of Ophthalmology; zSchool of Public Health, The University of Hong Kong; and wDepartment of Ophthalmology, Queen Mary Hospital, Hong Kong SAR, China. Presented as a poster at World Ophthalmology Congress 2012. Disclosure: The authors declare no conflict of interest. Reprints: Rita A. Gangwani, FRCSEd, FHKAM (Ophthalmology), Department of Ophthalmology, The University of Hong Kong, Room 301, Level 3, Block B, 100 Cyberport Road, Cyberport 4, Hong Kong SAR, China (e-mail: [email protected]). Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved. DOI: 10.1097/IJG.0000000000000138
J Glaucoma
Volume 25, Number 1, January 2016
The study adhered to the Declaration of Helsinki and ethics approval was obtained from the Institutional Review Board committee. Informed consent according to the Helsinki criteria was obtained. This population-based, prospective, cross-sectional study included subjects with a history of DM who had never undergone examination by an ophthalmologist before. The subjects underwent DR screening in a primary care outpatient clinic in Hong Kong from March 2008 to August 2008. They were assessed for glaucoma based on the identification of increased optic disc cupping on retinal fundus photographs.
Screening Procedure During the DR screening program, the subjects underwent examination of visual acuity, slit-lamp examination, dilatation of pupil with mydriacyl eye drops (Tropicamide 1%), and retinal fundus photography. The digital www.glaucomajournal.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
101
Gangwani et al
retinal color fundus photographs were taken using a Canon CR-DGi retinal camera attached to a Canon EOS 40D digital single-lens reflex camera (Canon, Tokyo, Japan) and graded on the computer screen with spatial resolution of 1024 768 pixels. Two photographic fields were taken of each eye, the first centered on the optic disc and the second centered on the fovea. These nonstereoscopic fundus photographs were graded by a trained optometrist and 2 ophthalmologists for presence/absence and severity of DR according to the English National Screening Programme classification.23 A trained optometrist performed the initial, that is, primary grading of the fundus photograph. If there was presence of any DR or any other eye disease, the fundus photograph was further graded by a secondary grader, an ophthalmologist. The secondary grader was blinded to the primary grader’s results. In case of discrepancy in the grading results between the primary and secondary graders, the photograph was graded by an arbitration grader (another ophthalmologist). Subjects were said to have sight threatening diabetic retinopathy (STDR) if they had any features of maculopathy, preproliferative DR, or proliferative DR. In addition to the grade of DR, the graders noted the presence of other eye diseases which included increased CDR (Z0.6), disc hemorrhage, retinal vein occlusion, macular degeneration, epiretinal membrane, and macular hole. The other eye diseases were referred as appropriate, but this paper deals only with the identification of increased CDR and presence or absence of glaucoma. When one or both of the primary and secondary graders identified increased CDR (Z0.6) and/or features indicative of glaucomatous optic neuropathy such as disc asymmetry (Z0.2), optic disc hemorrhage or notching/thinning of the neuroretinal rim, the photograph was graded by the arbitration grader and, if deemed appropriate, referred for detailed ophthalmological examination to confirm the presence or absence of glaucoma.
Examination at Specialist (Ophthalmology) Clinic At the specialist clinic, these subjects underwent a complete ophthalmologic examination to determine bestcorrected visual acuity, intraocular pressure measurement with Goldmann applanation tonometry, central corneal thickness (CCT) measurement with a pachymeter echograph (Quantel Medical, 63039 Clermonte-Ferrand cedex 2, France), slit-lamp examination, gonioscopy and fundus examination using a 78- or 90-D lens after pupil dilatation. They also underwent automated VF testing using the Swedish Interactive Threshold Algorithm, standard 24-2 method using a Humphrey Visual field Analyzer II (Carl Zeiss Meditec Inc., Dublin, CA) and optical coherence tomography (OCT) for retinal nerve fiber layer (RNFL) thickness using Spectralis OCT (Heidelberg Engineering GmbH, Germany).
Defining Features of Glaucoma The presence of glaucoma was established for those with enlarged vertical CDR and abnormal VF defects according to Hodapp-Parrish-Anderson’s criteria.24–26 An elevation of IOP was not required for the diagnosis of glaucoma. The patients were followed up for a period of 2 years and classified as having definite glaucoma or no glaucoma. The patients said to have definite glaucoma were those with
J Glaucoma
Volume 25, Number 1, January 2016
vertical CDR of Z0.6, glaucomatous defects on VF examination, or RNFL thinning if VF was unreliable. The types of glaucoma were defined as: (i) Primary narrow-angle glaucoma: presence of glaucomatous optic neuropathy plus Z270 degrees peripheral anterior synechiae or appositional closure less than grade 2 on gonioscopy according to the Shaffer grading system. (ii) Primary open-angle glaucoma: presence of glaucomatous optic neuropathy plus Z270 degrees grade 2 or above on gonioscopy according to the Shaffer grading system. (iii) Normal-tension glaucoma: presence of glaucomatous optic neuropathy plus baseline IOPr21 mm Hg on at least 2 initial measurements by Goldman applanation tonometry and open angle. The VF test was performed using the Standard Swedish interactive threshold algorithm 24-2 program. If the glaucoma hemifield test was graded as “outside normal limits” and showed a cluster of 3 contiguous points at 5% level on the pattern SD plot it was considered abnormal. The VF was considered to be reliable if it fulfilled the following criteria: fixation losses
Detection of Glaucoma and Its Association With Diabetic Retinopathy in a Diabetic Retinopathy Screening Program Rita A. Gangwani, FRCSEd, FHKAM (Ophthalmology),*w Sarah M. McGhee, PhD, FFPH,z Jimmy S.M. Lai, FRCS, FRCOphth, FHKAM (Ophthalmology), M.Med (Ophthalmology), MD,*w Christina K.W. Chan, MPH, BSc,z and David Wong, FRCS, FRCOphth*w
Purpose: To determine the type of glaucoma in subjects with diabetes mellitus detected during a diabetic retinopathy screening program and to determine any association between diabetic retinopathy (DR) and glaucoma. Materials and Methods: This is a population-based prospective cross-sectional study, in which subjects with diabetes mellitus underwent screening for DR in a primary care outpatient clinic. Digital fundus photographs were taken and graded for presence/ absence and severity of DR. During this grading, those fundus photographs showing increased cup-to-disc ratio (CDR) (Z0.6) were identified and these patients were referred to the specialist ophthalmology clinic for detailed examination. The presence of glaucoma was established based on CDR and abnormal visual field (VF) defects according to Hodapp-Parrish-Anderson’s criteria. An elevation of intraocular pressure was not required for the diagnosis of glaucoma. The patients said to have definite glaucoma were those with vertical CDR >/= 0.6, glaucomatous defects on VF examination, or retinal nerve fiber thinning if VF was unreliable. Results: Of the 2182 subjects who underwent screening, 81 subjects (3.7%) had increased CDR and 40 subjects (1.8%) had confirmed glaucoma. Normal-tension variant of primary open-angle glaucoma was the most prevalent type (1.2%) We did not find any evidence that DR is a risk factor for glaucoma [odds ratio for DR vs. no DR = 1.22 (95% confidence interval, 0.59-2.51)]. Conclusion: The overall prevalence of glaucoma in this diabetic population, based on finding increased cupping of optic disc in a teleretinal screening program was 1.8% (95% confidence interval, 1.0-3.0). Key Words: glaucoma, diabetic retinopathy, screening
(J Glaucoma 2016;25:101–105)
head and corresponding loss of visual field (VF), and increased CDR can be a sign that can alert a clinician in an otherwise asymptomatic patient. In our DR screening program, we screen approximately 70,000 patients a year and the burden of referral is substantial. To exclude glaucoma we will require repeated testing of intraocular pressure (IOP) as well as VF and other tests as appropriate. Raised IOP is the only modifiable risk factor that can slow the progression of the disease. Crucially, the amount of investigation and follow-up depends on the type of glaucoma detected. Diabetes mellitus (DM) causes microvascular damage and affects vascular autoregulation of the optic nerve resulting in optic nerve damage which may exacerbate glaucomatous optic neuropathy.3–6 Diabetes compromises glial and neuronal functions making retinal neurons and ganglion cells more susceptible to glaucomatous damage.7,8 Chronic hyperglycemia is also associated with increased IOP in patients with diabetes.9–12 Despite these links, there is still some debate about the association between glaucoma and DM. Some studies have found no association between DM and open-angle glaucoma,12–17 whereas others have found a positive association between diabetes and openangle glaucoma.18–22 Ong et al1 published a paper on detection of early glaucoma in a UK population. However, the prevalence of glaucoma in a predominantly Chinese diabetic population is unknown. We undertook this study to determine the prevalence and type of glaucoma in subjects with diabetes based on detection during DR screening and to investigate whether there was any association between DR and glaucoma.
MATERIALS AND METHODS
A
diabetic retinopathy (DR) screening program can identify many patients with other eye diseases, one of which—increased cup-to-disc ratio (CDR) is an important finding that requires referral.1 Glaucoma is the second leading cause of blindness globally.2 It is a progressive optic neuropathy with characteristic changes in the optic nerve Received for publication August 8, 2013; accepted August 22, 2014. From the *Department of Ophthalmology; zSchool of Public Health, The University of Hong Kong; and wDepartment of Ophthalmology, Queen Mary Hospital, Hong Kong SAR, China. Presented as a poster at World Ophthalmology Congress 2012. Disclosure: The authors declare no conflict of interest. Reprints: Rita A. Gangwani, FRCSEd, FHKAM (Ophthalmology), Department of Ophthalmology, The University of Hong Kong, Room 301, Level 3, Block B, 100 Cyberport Road, Cyberport 4, Hong Kong SAR, China (e-mail: [email protected]). Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved. DOI: 10.1097/IJG.0000000000000138
J Glaucoma
Volume 25, Number 1, January 2016
The study adhered to the Declaration of Helsinki and ethics approval was obtained from the Institutional Review Board committee. Informed consent according to the Helsinki criteria was obtained. This population-based, prospective, cross-sectional study included subjects with a history of DM who had never undergone examination by an ophthalmologist before. The subjects underwent DR screening in a primary care outpatient clinic in Hong Kong from March 2008 to August 2008. They were assessed for glaucoma based on the identification of increased optic disc cupping on retinal fundus photographs.
Screening Procedure During the DR screening program, the subjects underwent examination of visual acuity, slit-lamp examination, dilatation of pupil with mydriacyl eye drops (Tropicamide 1%), and retinal fundus photography. The digital www.glaucomajournal.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
101
Gangwani et al
retinal color fundus photographs were taken using a Canon CR-DGi retinal camera attached to a Canon EOS 40D digital single-lens reflex camera (Canon, Tokyo, Japan) and graded on the computer screen with spatial resolution of 1024 768 pixels. Two photographic fields were taken of each eye, the first centered on the optic disc and the second centered on the fovea. These nonstereoscopic fundus photographs were graded by a trained optometrist and 2 ophthalmologists for presence/absence and severity of DR according to the English National Screening Programme classification.23 A trained optometrist performed the initial, that is, primary grading of the fundus photograph. If there was presence of any DR or any other eye disease, the fundus photograph was further graded by a secondary grader, an ophthalmologist. The secondary grader was blinded to the primary grader’s results. In case of discrepancy in the grading results between the primary and secondary graders, the photograph was graded by an arbitration grader (another ophthalmologist). Subjects were said to have sight threatening diabetic retinopathy (STDR) if they had any features of maculopathy, preproliferative DR, or proliferative DR. In addition to the grade of DR, the graders noted the presence of other eye diseases which included increased CDR (Z0.6), disc hemorrhage, retinal vein occlusion, macular degeneration, epiretinal membrane, and macular hole. The other eye diseases were referred as appropriate, but this paper deals only with the identification of increased CDR and presence or absence of glaucoma. When one or both of the primary and secondary graders identified increased CDR (Z0.6) and/or features indicative of glaucomatous optic neuropathy such as disc asymmetry (Z0.2), optic disc hemorrhage or notching/thinning of the neuroretinal rim, the photograph was graded by the arbitration grader and, if deemed appropriate, referred for detailed ophthalmological examination to confirm the presence or absence of glaucoma.
Examination at Specialist (Ophthalmology) Clinic At the specialist clinic, these subjects underwent a complete ophthalmologic examination to determine bestcorrected visual acuity, intraocular pressure measurement with Goldmann applanation tonometry, central corneal thickness (CCT) measurement with a pachymeter echograph (Quantel Medical, 63039 Clermonte-Ferrand cedex 2, France), slit-lamp examination, gonioscopy and fundus examination using a 78- or 90-D lens after pupil dilatation. They also underwent automated VF testing using the Swedish Interactive Threshold Algorithm, standard 24-2 method using a Humphrey Visual field Analyzer II (Carl Zeiss Meditec Inc., Dublin, CA) and optical coherence tomography (OCT) for retinal nerve fiber layer (RNFL) thickness using Spectralis OCT (Heidelberg Engineering GmbH, Germany).
Defining Features of Glaucoma The presence of glaucoma was established for those with enlarged vertical CDR and abnormal VF defects according to Hodapp-Parrish-Anderson’s criteria.24–26 An elevation of IOP was not required for the diagnosis of glaucoma. The patients were followed up for a period of 2 years and classified as having definite glaucoma or no glaucoma. The patients said to have definite glaucoma were those with
J Glaucoma
Volume 25, Number 1, January 2016
vertical CDR of Z0.6, glaucomatous defects on VF examination, or RNFL thinning if VF was unreliable. The types of glaucoma were defined as: (i) Primary narrow-angle glaucoma: presence of glaucomatous optic neuropathy plus Z270 degrees peripheral anterior synechiae or appositional closure less than grade 2 on gonioscopy according to the Shaffer grading system. (ii) Primary open-angle glaucoma: presence of glaucomatous optic neuropathy plus Z270 degrees grade 2 or above on gonioscopy according to the Shaffer grading system. (iii) Normal-tension glaucoma: presence of glaucomatous optic neuropathy plus baseline IOPr21 mm Hg on at least 2 initial measurements by Goldman applanation tonometry and open angle. The VF test was performed using the Standard Swedish interactive threshold algorithm 24-2 program. If the glaucoma hemifield test was graded as “outside normal limits” and showed a cluster of 3 contiguous points at 5% level on the pattern SD plot it was considered abnormal. The VF was considered to be reliable if it fulfilled the following criteria: fixation losses
