Journal of Pediatric Ophthalmology and Strabismus

Original Article 

Feasibility of Retinal Screening in a Pediatric Population With Type 1 Diabetes Mellitus

Anton M. Kolomeyer, MD, PhD; Natasha V. Nayak, MD; Melissa A. Simon, MD; Bernard C. Szirth, PhD; Khadija Shahid, OD, FAAO; Iris Y. Sheng, BS; Tina Xia, BA; Albert S. Khouri, MD

Abstract

Purpose:

To study the feasibility of using a nonmydriatic camera to screen children with type 1 diabetes mellitus (DM1) as young as 2 years for diabetic retinopathy.

Methods:

Prospective pilot imaging study involving children with DM1 aged 2 to 17 years. The screening consisted of: (1) intake form; (2) measurement of blood pressure, pulse, and oximetry; (3) assessment of visual acuity (SIMAV, Padova, Italy); and (4) nonmydriatic color imaging (Canon CX-1 45° 15.1 megapixel camera; Canon Corp., Tokyo, Japan). Images were assessed for signs of diabetic retinopathy and graded for quality on a scale of 1 to 5 by two clinicians. Kappa coefficient was calculated to determine inter-observer agreement.

Results:

One hundred four of 106 (98%) children underwent imaging (mean age: 11.1 years, 51% male, 88% white). One (1%) child had nonproliferative diabetic retinopathy and 2 (1.9%) had incidental findings. Only 62% of children had an eye examination within the past year, with children with DM1 for more than 5 years significantly more likely to have done so (P = .03). Children who had an eye examination within the past year were significantly older than their counterparts (P = .01). Images of high quality (grades 4 and 5) were acquired in 178 (86%) eyes, and images of some clinical value (grades ≥ 2) were obtained in 207 (99.5%) eyes. Inter-observer agreement for image quality was 0.896.

Conclusions:

The feasibility of using a nonmydriatic camera to screen children as young as 2 years for changes related to diabetic eye disease was demonstrated. Nonmydriatic imaging may supplement standard dilated clinical ophthalmology examinations for select patient populations.

[J Pediatr Ophthalmol Strabismus 2014;51(5):299–306.]

From the Department of Ophthalmology, Eye and Ear Institute, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania (AMK); the Institute of Ophthalmology and Visual Science, Rutgers New Jersey Medical School, Newark, New Jersey (NVN, MAS, BCS, IYS, TX, ASK); and the Department of Ophthalmology and Visual Sciences, University of Iowa, Iowa City, Iowa (KS).

The authors have no financial or proprietary interest in the materials presented herein.

Correspondence: Albert S. Khouri, MD, The Institute of Ophthalmology and Visual Science, Rutgers New Jersey Medical School, 90 Bergen Street, Suite 6100, P.O. Box 1709, Newark, NJ 07101-1709. E-mail: albert.khouri@rutgers.edu

Received: September 23, 2013
Accepted: April 28, 2014
Posted Online: July 16, 2014

Abstract

Purpose:

To study the feasibility of using a nonmydriatic camera to screen children with type 1 diabetes mellitus (DM1) as young as 2 years for diabetic retinopathy.

Methods:

Prospective pilot imaging study involving children with DM1 aged 2 to 17 years. The screening consisted of: (1) intake form; (2) measurement of blood pressure, pulse, and oximetry; (3) assessment of visual acuity (SIMAV, Padova, Italy); and (4) nonmydriatic color imaging (Canon CX-1 45° 15.1 megapixel camera; Canon Corp., Tokyo, Japan). Images were assessed for signs of diabetic retinopathy and graded for quality on a scale of 1 to 5 by two clinicians. Kappa coefficient was calculated to determine inter-observer agreement.

Results:

One hundred four of 106 (98%) children underwent imaging (mean age: 11.1 years, 51% male, 88% white). One (1%) child had nonproliferative diabetic retinopathy and 2 (1.9%) had incidental findings. Only 62% of children had an eye examination within the past year, with children with DM1 for more than 5 years significantly more likely to have done so (P = .03). Children who had an eye examination within the past year were significantly older than their counterparts (P = .01). Images of high quality (grades 4 and 5) were acquired in 178 (86%) eyes, and images of some clinical value (grades ≥ 2) were obtained in 207 (99.5%) eyes. Inter-observer agreement for image quality was 0.896.

Conclusions:

The feasibility of using a nonmydriatic camera to screen children as young as 2 years for changes related to diabetic eye disease was demonstrated. Nonmydriatic imaging may supplement standard dilated clinical ophthalmology examinations for select patient populations.

[J Pediatr Ophthalmol Strabismus 2014;51(5):299–306.]

From the Department of Ophthalmology, Eye and Ear Institute, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania (AMK); the Institute of Ophthalmology and Visual Science, Rutgers New Jersey Medical School, Newark, New Jersey (NVN, MAS, BCS, IYS, TX, ASK); and the Department of Ophthalmology and Visual Sciences, University of Iowa, Iowa City, Iowa (KS).

The authors have no financial or proprietary interest in the materials presented herein.

Correspondence: Albert S. Khouri, MD, The Institute of Ophthalmology and Visual Science, Rutgers New Jersey Medical School, 90 Bergen Street, Suite 6100, P.O. Box 1709, Newark, NJ 07101-1709. E-mail: albert.khouri@rutgers.edu

Received: September 23, 2013
Accepted: April 28, 2014
Posted Online: July 16, 2014

Introduction

Diabetes mellitus type 1 (DM1) affects school-aged children in the United States with a prevalence of approximately 2 per 1,000, and its incidence may be increasing.1 Diabetic retinopathy is the leading cause of blindness in young adults, with prevalence ranging from 10% to 30%, depending on the study.2 Current guidelines from the American Academy of Pediatrics recommend annual screening beginning 3 to 5 years after diagnosis of DM1 in children older than 9 years.1 The American Academy of Ophthalmology recommends beginning annual screening of DM1 3 to 5 years after disease onset in patients younger than 30 years.3 The American Diabetes Association recommends screening patients with DM1 10 years or older within 5 years of disease onset.4

Although there have been innumerable studies describing the benefits of telemedicine screenings focused on detection of diabetic retinopathy in the adult population, few efforts have included the pediatric population.5–7 The purpose of our study was to demonstrate the feasibility of screening for diabetic retinopathy in children with DM1 as young as 2 years using nonmydriatic high-resolution imaging at the Children with Diabetes convention ( http://www.childrenwithdiabetes.com).

Patients and Methods

This was a prospective pilot imaging study of children with DM1 that was approved by the Rutgers University Internal Review Board and was fully HIPAA compliant. The imaging was conducted at an annual “Friends for Life” convention organized by Children with Diabetes, Inc., in Orlando, Florida, between June 29 and July 4, 2010. The purpose of this conference is to provide education and support for children with DM1 and their families. Inclusion criteria for imaging were age of 17 years or younger and a self-reported diagnosis of DM1. The screening program was promoted on the Children with Diabetes web site. At the convention, the children’s families made appointments for 15-minute time slots. The day was divided into a morning and afternoon screening session, whereas the children were divided into the following age groups: 2 to 4 years; 5 to 7 years; 8 to 11 years; 12 to 14 years; and older than 15 years, with previously identified ocular pathology. We witnessed improved cooperation by separating children into age brackets, thereby streamlining the screening process. The screening protocol was explained in full detail to the parents/guardians, who signed general consent and release forms during registration (these forms remained with conference organizers) and were present throughout the examination.

A detailed description of our adult screening protocol has been published previously.8 In brief, the screening team consisted of: (1) senior medical students who completed an intake form (including age, gender, ethnicity, time since last eye examination, duration of DM1, insulin pump use, personal and family history of medical and ocular problems, etc.), measured blood pressure, pulse, oxygen saturation, and assessed visual acuity (SIMAV, Padova, Italy); (2) an imaging professional who performed nonmydriatic color imaging (Canon CX-1 45° 15.1 megapixel camera with CMOS sensor; Canon Corp., Tokyo, Japan); and (3) an on-site medical director who analyzed the complete screening data, reviewed color images, counseled participants and guardians regarding results, and made referrals for follow-up examination as needed. Intraocular pressure was not measured in this study because it was found to be stable over the previous 4 years of follow-up in children with DM1 (unpublished observation). All data were entered on a computer. For the analysis, Snellen visual acuity was converted into logarithm of the minimum angle of resolution (logMAR) scale, averaged, and then re-converted back into Snellen visual acuity.

Both eyes were imaged without the use of mydriatic agents. We attempted to obtain only one image per eye, but, if the quality was poor and if permissible by the child, a repeat image was acquired. The color images were captured at the following flash settings: light iris in age 2 to 6 years (35 watt seconds [Ws]); dark iris in age 2 to 6 years and light iris, age older than 6 years (50 Ws); and dark iris in age older than 6 years (75 Ws).9 Canon proprietary software EyeScape (Canon, Los Angeles, CA) was used to manage all Digital Imaging and Communications in Medicine based images. Color images were evaluated for potential changes in the optic nerve (ie, cup-to-disc ratio, asymmetry, neuroretinal rim color, and nerve fiber layer integrity), blood vessels (ie, artery-to-vein ratio and nicking, microaneurysm formation, venous beading, and neovascularization), and macula and posterior pole (ie, hemorrhage, exudates, ischemia, retinal detachment, and pigmentary changes). If media opacity was suspected (eg, due to acquisition of a blurry image), anterior segment settings on the fundus camera were used to evaluate the eye for cataract formation. In addition, although not part of the image-grading rubric, a digital red filter (610-nm wavelength, focusing on retinal pigment epithelium/choroid) was used by the imaging specialist and medical director to improve evaluation of the posterior pole.10 Our primary variable of interest was findings suggestive of diabetic retinopathy, whereas the secondary outcome variable was compliance with an annual eye examination, if applicable.

To assess the quality of nonmydriatic images in this pediatric population, two ophthalmology residents (AMK and NVN) masked to each other’s assessment reviewed the fundus images in a random order and graded them for general quality on a scale from 1 to 5 (Figure 1).11 The images were graded in the following manner: grade 1 (of no clinical value, fundus barely visible), grade 2 (small portion of fundus visible, not able to evaluate all findings), grade 3 (majority of fundus visible, able to evaluate abnormalities), grade 4 (near perfect image, whole fundus visible, able to evaluate most subtle findings), and grade 5 (perfect image, whole fundus visible, able to evaluate all subtle findings). If more than one image was obtained per eye, the “grade” included was that of the best image quality. Images of “some clinical value” were defined as grades 2 and above, whereas those of “high quality” included grades 4 and 5.

Examples of images obtained during the screening process. (A) Grade 1: Optic nerve is barely visible. Photograph not of clinical value. (B) Grade 2: Optic nerve, superior arcade, and small portion of inferior arcade are seen. Approximately 50% of the fundus is obscured. (C) Grade 3: Optic nerve, macula, and superior arcade are completely visible; majority of the inferior arcade can be evaluated. Approximately 20% of the fundus is obscured. (D) Grade 4: Optic nerve, macula, and superior and inferior arcades are completely visible. Fundus photograph shows minor peripheral image blurring. (E) Grade 5: Optic nerve, macula, and superior and inferior arcades are completely visible. There is no image distortion of any kind.

Figure 1.

Examples of images obtained during the screening process. (A) Grade 1: Optic nerve is barely visible. Photograph not of clinical value. (B) Grade 2: Optic nerve, superior arcade, and small portion of inferior arcade are seen. Approximately 50% of the fundus is obscured. (C) Grade 3: Optic nerve, macula, and superior arcade are completely visible; majority of the inferior arcade can be evaluated. Approximately 20% of the fundus is obscured. (D) Grade 4: Optic nerve, macula, and superior and inferior arcades are completely visible. Fundus photograph shows minor peripheral image blurring. (E) Grade 5: Optic nerve, macula, and superior and inferior arcades are completely visible. There is no image distortion of any kind.

Statistical analysis was performed using Microsoft Excel 12.2.7 (Microsoft Corporation, Redmond, WA) and JMP 10.0 (SAS Institute, Cary, NC) software. Descriptive analyses included mean, standard deviation, and percentage of total. The t test was used to compare means and Fisher’s exact test to evaluate percentages. Pearson’s chi-square test was used to compare categorical variables. Kappa co-efficient was calculated to determine inter-observer agreement in grading image quality. A P value less than .05 was considered statistically significant.

Results

Of the 92 children with self-reported duration since the most recent eye examination, 38% were not evaluated by their ophthalmologist/optometrist in more than 1 year prior to the screening (Table 1). Children who had DM1 for at least 5 years were significantly more likely to have had an examination within the past year (Table 2). Children who had an eye examination within the past year were also significantly older than their counterparts (12.2 ± 3.4 vs 10.2 ± 3.8 years; P = .01). Gender (P = .39), body mass index (BMI) (P = .25), ethnicity (P = .56), visual acuity (P = .37), insulin pump use (P = .32), and duration of diabetes mellitus (P = .09) were not significantly associated with undergoing an eye examination by their ophthalmologist/optometrist within the past year (Table 3).

Screening Participants’ Demographics and Clinical Characteristics

Table 1:

Screening Participants’ Demographics and Clinical Characteristics

Relationship Between Duration of Diabetes Mellitus Type 1 and Annual Eye Examination

Table 2:

Relationship Between Duration of Diabetes Mellitus Type 1 and Annual Eye Examination

Relationship of Population Characteristics and Undergoing an Eye Examination Within the Past Year

Table 3:

Relationship of Population Characteristics and Undergoing an Eye Examination Within the Past Year

One hundred four of 106 (98%) children who presented to our screening location and met inclusion criteria were included in the analysis. Two children younger than 3 years were unable to remain still for imaging. The parents of these children were advised to have them examined by an ophthalmologist/optometrist. Table 1 lists demographic and clinical characteristics of our study population. Both eyes were imaged in each participant. A total of 218 images were acquired (ie, 10 [4.8%] eyes required a second image to improve quality). One (1%) child was found to have non-proliferative diabetic retinopathy at the screening (age: 15 years, disease duration: 10 years) and two (1.9%) had incidental findings (one with a choroidal nevus and cup-to-disc asymmetry and one with a myopic fundus). None of the children were found to have strabismus, cataracts, corneal or other retinal disease, or nystagmus. Based on quality of acquired images (as agreed upon by the two graders), 1 (0.5%) was assessed as grade 1, 5 (2.4%) as grade 2, 24 (11%) as grade 3, 103 (50%) as grade 4, and 75 (36%) as grade 5. Photographs of some clinical value (grades 2 and above) were obtained in 207 (99.5%) eyes and high quality (grades 4 and 5) images were acquired in 178 (86%) eyes. Inter-observer agreement for overall fundus image quality (“grade”) was 0.896 (95% confidence interval: 0.857 to 0.935).

Discussion

Diabetic retinopathy is a common microvascular manifestation of diabetes mellitus, with its prevalence directly related to the duration of diabetes mellitus. Although vision is spared in patients with retinopathy not affecting the macula, a small proportion of these eyes (with proliferative diabetic retinopathy) may eventually progress to irreversible blindness.12 The importance of screening and early diagnosis of diabetic retinopathy is supported in part by the Early Treatment of Diabetic Retinopathy Study, which showed that the risk of moderate vision loss from diabetic macular edema and severe vision loss from proliferative diabetic retinopathy are decreased with focal and panretinal laser photocoagulation, respectively.13 We performed a thorough screening with a 15.1 megapixel nonmydriatic fundus camera in addition to determination of vision, blood pressure, pulse, oximetry, and BMI in children as young as 2 years of age.

The general consensus on the recommended screening regimen by the American Academies of Pediatrics and Ophthalmology and the American Diabetes Association suggests beginning 3 to 5 years after diagnosis of DM1 in children older than 9 years (based on whichever occurs sooner). However, based on our below outlined observations, we believe that retinal screening may have a significant role in reducing DM1 morbidity in much younger children. First, many of the children in our cohort have been diagnosed as having DM1 since 18 months of age (and some as early as 3 months). Second, a significant proportion of the children use latest generation insulin pumps, which were not available at the time of the American Diabetes Association position statement. Third, improvement in retinal camera resolution, use of imaging processing software technologies (eg, red-free filters), and development of novel imaging modalities (eg, fast-scan optical coherence tomography) allow for earlier detection of diabetes mellitus-related eye pathology. Finally, refinements in our understanding of diabetic disease pathophysiology may influence future recommendations regarding optimal age and frequency of retinal screening programs.

Screening for diabetic retinopathy has been extensively described in the adult literature and few previous studies demonstrated feasibility of non-mydriatic fundus imaging for diabetic retinopathy in children 6 to 19 years of age.14–17 Raman et al. evaluated children aged 8 to 19 years for diabetic retinopathy using a nonmydriatic retinal camera in one location and a slit lamp with mydriasis in another.16 Stillman et al. were the first to demonstrate the feasibility of “telepediatric” nonmydriatic technology to screen 83 children (age: 6.1 to 18 years) with diabetes mellitus.15 Massin et al. evaluated 504 summer camp attendees (age: 10 to 18 years) for the presence of diabetic retinopathy using nonmydriatic fundus photography.14 Toffoli et al. showed that adequate quality nonmydriatic images could be obtained in children older than 3 years and in some as young as 22 months.11 Mayer-Davis et al. determined the overall prevalence of diabetic retinopathy in patients younger than 20 years with nonmydriatic fundus photography.17 The nonmydriatic technologies used in the abovementioned findings compared to ours are summarized in Table 4. In comparison to previous studies specifically focusing on diabetic retinopathy,14–16 our study: (1) included children 2 to 6 years of age (previous youngest age: 6.1 years); (2) used a camera with the ability to acquire 15.1 megapixel images with a resolution of 1,598 × 1,062 pixels (previous highest: 6 megapixel and 1,490 × 960 pixels); (3) employed an imaging specialist to acquire the images (previously acquired by medical students, orthoptists, nurses, and physicians); and (4) included a more comprehensive assessment of overall health in addition to visual acuity (ie, blood pressure, pulse, oximetry, and BMI). Previous studies did not include or selectively included these measures.

Comparison of Nonmydriatic Imaging Characteristics

Table 4:

Comparison of Nonmydriatic Imaging Characteristics

Image acquisition was successful in 104 of 106 (98%) children, although in only 1 of 3 (33%) children younger than 3 years. The difficulty in imaging children younger than 3 years is consistent with the findings of Toffoli et al.11 The overall ability to acquire images was higher in our study than previously reported (98% vs 85.3% in Toffoli et al.). In addition, a second image was acquired to improve its quality in only 10 (4.8%) children (1.05 images/eye) versus 1 to 6 images per eye in the Toffoli et al. study (2.07 images/eye). The ability to obtain nonmydriatic images of quality sufficiently high to ensure a thorough image analysis may reduce the frequency with which dilating drops may be needed in children for comprehensive ocular examinations.

In our study, 99.5% of photographs were of some clinical value (grades 2 and above) and 86% were of high quality (grades 4 and 5). Inter-observer agreement for image quality was 0.896. Stillman et al. reported that 87% of images were “diagnostic” and 80% were “good” or “excellent.”15 Although not explicitly stated in their study, “diagnostic” likely corresponds to grades 2 and above and “good” or “excellent” to grades 4 and 5, respectively. Soto-Pedre et al. reported that photograph quality was “adequate” (likely corresponding to grades 2 and above) in 97.7% of patients.18 Massin et al. graded their images from 1 (“excellent”) to 4 (“not gradable”) and, although they did not specify exact percentages, mentioned that “all fundus photographs were of very good quality and gradable.”14 In a study by Toffoli et al., images of some clinical value (grades 2 and above) and high quality images (grades 4 and 5) were obtained in 95% and 65%, respectively.11 Inter-observer agreement in their study for overall image quality was identical to our study at 0.89.

It appears that the rate of obtaining high quality photographs in our study was somewhat higher than that previously described.11,19 This finding may, in part, be due to differences in equipment, imaging set-up, children’s comfort level and experience, and level of expertise of the clinicians obtaining the images. For example, in the Toffoli et al. study, an ophthalmologist and a medical student acquired the photographs after less than 30 minutes of training on the camera, whereas in our study, an imaging specialist with experience in pediatric retinal imaging obtained all images.11 In addition, because imaging time slots were grouped based on age, many children considered this process as a friendly “competition” with their peers, which served as motivation to sit still.

The rate of nonproliferative diabetic retinopathy in our cohort was 1%, which, although lower than the 5% to 30% previously reported in the literature, was similar to Stillman et al. (1.2%).15 This was likely multifactorial and related to diabetes mellitus control and duration, and social and dietary factors, among others. In our cohort, the average duration of DM1 was 5.0 ± 3.5 years. Although the mean duration of diabetes mellitus in the Massin et al. study was 4.8 years, diabetic retinopathy was identified in 4.6% of children.14 In a different study by Lueder et al., no diabetic retinopathy was detected in children with DM1 diagnosed before 2 years of age and with at least 5 years of disease duration.20

Of the 92 participants who reported duration since last eye examination, the likelihood of being evaluated by an ophthalmologist/optometrist within the past year was significantly associated with older age and increased duration of diabetes mellitus (Table 2). Previously reported ethnicity differences in undergoing routine eye examination were not found in our study, most likely because only 12% of our population was non-white.21 Thirteen (27%) of 48 children with diabetes mellitus for 5 years or more (mean ± standard deviation age: 11.6 ± 2.8 years) and 23 (36%) of 64 children with diabetes mellitus for 3 years or more (mean ± standard deviation age: 11.0 ± 3.6 years) did not see their ophthalmologist/optometrist within the past year. Based on current recommendations, it is advisable for these children to undergo more frequent examinations. When prompted, parents mentioned the following reasons for less frequent visits to their ophthalmologist/optometrist: (1) unchanged vision; (2) endocrinologist did not recommend checking vision; (3) vision examination performed by a school nurse; and (4) returning to Children with Diabetes conference for an eye examination (even though we ask every family to visit their own ophthalmologist/optometrist yearly). Telemedicine fundus imaging may serve as a useful supplement for physicians in meeting screening demands of a rapidly growing patient population with childhood diabetes mellitus and may improve patient adherence to regular dilated eye examinations.22

Some of the limitations of this study include: (1) a self-selected study population of likely higher socioeconomic status, which may have affected the rate of diabetic retinopathy and the rate of undergoing an annual eye examination in our cohort; (2) relatively homogenous patient population (88% white); and (3) socioeconomic data were not collected, which may limit our understanding of accessibility to care for children in the general population. Subsequent studies in varied settings with diverse pediatric populations should be pursued to determine generalizability of our findings.

We demonstrated the feasibility of using a non-mydriatic camera to screen children as young as 2 years for changes related to diabetic eye disease. Photographs of “some clinical value” were obtained in 99.5% of eyes. These findings are consistent with those previously reported9 and support the assertion that nonmydriatic imaging may have a role in supplementing or providing an alternative means by which clinical ophthalmology examinations are conducted for select patient populations. Subsequent larger studies should aim to precisely define the specific patient populations and ocular diseases for which the above statements may hold true.

References

  1. Lueder GT, Silverstein JAmerican Academy of Pediatrics Section on O, Section on E. Screening for retinopathy in the pediatric patient with type 1 diabetes mellitus. Pediatrics. 2005;116:270–273. doi:10.1542/peds.2005-0875 [CrossRef]
  2. Sivaprasad S, Gupta B, Crosby-Nwaobi R, Evans J. Prevalence of diabetic retinopathy in various ethnic groups: a worldwide perspective. Surv Ophthalmol. 2012;57:347–370. doi:10.1016/j.survophthal.2012.01.004 [CrossRef]
  3. Preferred Practice Pattern: Diabetic Retinopathy 2008. Available at: http://one.aao.org/preferred-practice-pattern/diabetic-retinopathy-ppp--september-2008-4th-print. Accessed June 26, 2014.
  4. American Diabetes Association. Standards of medical care in diabetes—2013. Diabetes Care. 2013;36(suppl 1):S11–S66. doi:10.2337/dc13-S011 [CrossRef] Available at: http://care.diabetesjournals.org/content/36/Supplement_1/S11.full; accessed 9/17/2013).
  5. Lamminen H, Ruohonen K. Fundus imaging and the telemedical management of diabetes. J Telemed Telecare. 2002;8:255–258. doi:10.1258/135763302760314207 [CrossRef]
  6. Zimmer-Galler IE, Zeimer R. Telemedicine in diabetic retinopathy screening. Int Ophthalmol Clin. 2009;49:75–86. doi:10.1097/IIO.0b013e31819fd60f [CrossRef]
  7. Li HK, Horton M, Bursell SE, et al. Telehealth practice recommendations for diabetic retinopathy, second edition. Telemed J E Health. 2011;17:814–837. doi:10.1089/tmj.2011.0075 [CrossRef]
  8. Shahid K, Kolomeyer AM, Nayak NV, et al. Ocular telehealth screenings in an urban community. Telemed J E Health. 2012;18:95–100. doi:10.1089/tmj.2011.0067 [CrossRef]
  9. Szirth BC. Light toxicity symposium, ophthalmic photography equipment evaluation. J Ophthalmic Photogr. 1988;10:18–20.
  10. Kolomeyer AM, Szirth BC, Shahid KS, Pelaez G, Nayak NV, Khouri AS. Software-assisted analysis during ocular health screening. Telemed J E Health. 2013;19:2–6. doi:10.1089/tmj.2012.0070 [CrossRef]
  11. Toffoli D, Bruce BB, Lamirel C, Henderson AD, Newman NJ, Biousse V. Feasibility and quality of nonmydriatic fundus photography in children. J AAPOS. 2011;15:567–572. doi:10.1016/j.jaapos.2011.07.010 [CrossRef]
  12. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329:977–986. doi:10.1056/NEJM199309303291401 [CrossRef]
  13. Cantrill HL. The Diabetic Retinopathy Study and the Early Treatment Diabetic Retinopathy Study. Int Ophthalmol Clin. 1984;24:13–29. doi:10.1097/00004397-198402440-00004 [CrossRef]
  14. Massin P, Erginay A, Mercat-Caudal I, et al. Prevalence of diabetic retinopathy in children and adolescents with type-1 diabetes attending summer camps in France. Diabetes Metab. 2007;33:284–289. doi:10.1016/j.diabet.2007.03.004 [CrossRef]
  15. Stillman JK, Gole GA, Wootton R, et al. Telepaediatrics and diabetic retinopathy screening of young people with diabetes in Queensland. J Telemed Telecare. 2004;10:92–94. doi:10.1258/1357633042614203 [CrossRef]
  16. Raman V, Campbell F, Holland P, et al. Retinopathy screening in children and adolescents with diabetes. Ann N Y Acad Sci. 2002;958:387–389. doi:10.1111/j.1749-6632.2002.tb03009.x [CrossRef]
  17. Mayer-Davis EJ, Davis C, Saadine J, et al. Diabetic retinopathy in the SEARCH for Diabetes in Youth Cohort: a pilot study. Diabet Med. 2012;29:1148–1152. doi:10.1111/j.1464-5491.2012.03591.x [CrossRef]
  18. Soto-Pedre E, Hernaez-Ortega MC, Pinies JA. Duration of diabetes and screening coverage for retinopathy among patients with type 2 diabetes. Ophthalmic Epidemiol. 2007;14:76–79. doi:10.1080/09286580600879032 [CrossRef]
  19. Bruce BB, Lamirel C, Wright DW, et al. Nonmydriatic ocular fundus photography in the emergency department. N Engl J Med. 2011;364:387–389. doi:10.1056/NEJMc1009733 [CrossRef]
  20. Lueder GT, Pradhan S, White NH. Risk of retinopathy in children with type 1 diabetes mellitus before 2 years of age. Am J Ophthalmol. 2005;140:930–931. doi:10.1016/j.ajo.2005.05.023 [CrossRef]
  21. Nsiah-Kumi P, Ortmeier SR, Brown AE. Disparities in diabetic retinopathy screening and disease for racial and ethnic minority populations: a literature review. J Natl Med Assoc. 2009;101:430–437.
  22. Conlin PR, Fisch BM, Cavallerano AA, Cavallerano JD, Bursell SE, Aiello LM. Nonmydriatic teleretinal imaging improves adherence to annual eye examinations in patients with diabetes. J Rehabil Res Dev. 2006;43:733–740. doi:10.1682/JRRD.2005.07.0117 [CrossRef]

Screening Participants’ Demographics and Clinical Characteristics

CharacteristicNo.
Age (y)
  Mean ± SD11.1 ± 3.7
  Range2 to 17
Gender
  Male/female53 (51%)/51 (49%)
Ethnicity
  White91 (88%)
  Hispanic6 (5.8%)
  African American1 (1.0%)
  Other6 (5.8%)
Body mass index
  Mean ± SD19.8 ± 3.4
  Range12.8 to 32.0
Family history
  Diabetes32 (31%)
  Hypertension22 (21%)
  Glaucoma12 (12%)
  Age-related macular degeneration10 (9.6%)
Duration of diabetes mellitus type I (y)
  Mean ± SD5.0 ± 3.5
  Range0.33 to 14
Insulin pump use
  No.22 (21%)
  Mean ± SD (y)3.5 ± 2.5
  Range (y)0.5 to 10
Last eye examination (y)
  ≤ 157 (62%)
  1 to 227 (29%)
  > 28 (8.7%)
Visual acuity
  Right
    Mean ± SD20/28.4 ± 11.1
    Range20/20 to 20/60
  Left
    Mean ± SD20/27.5 ± 9.7
    Range20/20 to 20/50

Relationship Between Duration of Diabetes Mellitus Type 1 and Annual Eye Examination

Duration (y)> 1 Year Since Last Eye ExaminationPa
< 3 (n = 27)11 (41%).66
≥ 3 (n = 64)23 (36%)
< 5 (n = 43)21 (49%).03
≥ 5 (n = 48)13 (27%)

Relationship of Population Characteristics and Undergoing an Eye Examination Within the Past Year

CharacteristicEye Examination Within Past Year
P
YesNo
Age (y), mean ± SD12.2 ± 3.410.2 ± 3.8.01a
Male28 (49%)21 (60%).39b
Body mass index, mean ± SD20.3 ± 2.819.4 ± 3.7.25a
White51 (89%)34 (97%).56b
Average logMAR best-corrected visual acuity, mean ± SD0.13 ± 0.160.10 ± 0.13.37a
Insulin pump11 (20%)10 (29%).32b
Duration of diabetes mellitus (y), mean ± SD5.81 ± 3.614.53 ± 3.22.09a

Comparison of Nonmydriatic Imaging Characteristics

StudyNo. of ParticipantsAge Range (y)Type of Nonmydriatic CameraDegree of FieldMPNo. of Images Obtained/EyeObtained Image CharacteristicsImaging Performed ByAdditional Assessments
Raman et al., 200216738 to 19N/AaN/AN/AN/AN/AN/AN/A
Stillman et al., 200415836.1 to 18.4Nidek Handy NM-200301.54 x 30°Digital, 24-bit color, 72 pixels/inchRNs, pediatric and adolescent endocrine specialistVisual acuity
Massin et al., 20071450410 to 18Topcon TRC-NW64565 x 45° (total area 120°)Digital (Fuji, S2-PRO), 24-bit color, 1,490 x 960 pixelsOrthoptistBlood pressure, blood sugar, BMI
Toffoli et al., 2011112121 to 18Kowa alpha-D4551 x 45°DigitalOphthalmologist and medical studentNone
Mayer-Davis et al., 201217222< 20bVisucam Pro N4551 x 45°DigitalImaging study staffPhysical examination, HgA1c, blood pressure, BMI, blood work
Current, 20131042 to 17Canon CX-14515.11 x 45°Digital, 1598 x 1062 pixels, 24-bit colorImaging specialistVisual acuity, blood pressure, pulse, oximetry, BMI

10.3928/01913913-20140709-01

Sign up to receive

Journal E-contents