Key Points
Question
What is the risk of ocular sequelae with diabetes diagnosed in childhood?
Findings
In this cohort study of 525 youth (aged <22 years), the risk of diabetic retinopathy was 88% greater in those with type 2 diabetes (T2D) compared with those with type 1 diabetes (T1D) within the first 15 years of disease. Similarly, risk of developing proliferative diabetic retinopathy or requiring pars plana vitrectomy was greater in T2D than T1D.
Meaning
These findings suggest that children with T2D have a higher risk of developing retinopathy than those with T1D, potentially requiring earlier surveillance and intervention to prevent vision-threatening complications.
This cohort study describes the risk of developing diabetes-associated ocular complications among children diagnosed with type 1 or type 2 diabetes during a 50-year period.
Abstract
Importance
Despite the increasing prevalence of type 2 diabetes (T2D) diagnosed in childhood, little is known about the natural history of ocular sequelae in youth-onset T2D compared with type 1 diabetes (T1D).
Objective
To assess the risk of developing diabetes-associated ocular complications among youth diagnosed with diabetes.
Design, Setting, and Participants
This retrospective, population-based medical record review included all residents of Olmsted County, Minnesota (95.7% White in 1990), diagnosed with diabetes at younger than 22 years (hereinafter referred to as children) from January 1, 1970, through December 31, 2019.
Main Outcomes and Measures
Risk of developing ocular complications over time.
Results
Among 1362 individuals with a diagnostic code of diabetes, medical record reviews confirmed a diagnosis of T1D or T2D in 606 children, of whom 525 (86.6%) underwent at least 1 eye examination (mean [SD] age at diabetes diagnosis, 12.1 [5.4] years; 264 [50.3%] male). Diabetes-associated ocular complications occurred in 147 of the 461 children (31.2%) with T1D and in 17 of the 64 children (26.6%) with T2D. The hazard ratio illustrating the risk between T2D and T1D rates was 1.88 (95% CI, 1.13-3.12; P = .02) for developing any diabetic retinopathy (nonproliferative or greater), 2.33 (95% CI, 0.99-5.50; P = .048) for proliferative diabetic retinopathy, 1.49 (95% CI, 0.46-4.89; P = .50) for diabetic macular edema, 2.43 (95% CI, 0.54-11.07; P = .24) for a visually significant cataract, and 4.06 (95% CI, 1.34-12.33; P = .007) for requiring pars plana vitrectomy by 15 years after the diagnosis of diabetes.
Conclusions and Relevance
Diabetic retinopathy, proliferative diabetic retinopathy, and the need for pars plana vitrectomy occurred within a shorter diabetes duration for children with T2D compared with T1D in this population-based cohort. Children with T2D had almost twice the risk of developing retinopathy compared with those with T1D. These findings suggest that to prevent serious ocular complications, children with T2D may require ophthalmoscopic evaluations at least as frequently as or more frequently than children with T1D.
Introduction
Diabetes is a common chronic disease of childhood characterized by chronic hyperglycemia with end-organ damage that often results in the microvascular triad of nephropathy, neuropathy, and retinopathy.1,2 Subsequent retinal neurodegeneration and breakdown of the blood-retinal barrier progress to diabetic retinopathy, the leading cause of blindness among working-age and young adults.3,4 Although the ocular sequelae of type 1 diabetes (T1D) and adult-onset type 2 diabetes (T2D) have been well described,1,3,4,5 little is known concerning the progression of diabetic retinopathy among children with T2D, despite its increasing prevalence in recent years.6 Current pediatric diabetic retinopathy screening guidelines from the American Academy of Ophthalmology and the American Academy of Pediatrics have evidence-based guidelines for children diagnosed with T1D, whereas data guiding the management of diabetic retinopathy in childhood-onset T2D are limited.4 The purpose of this study was to assess the risk of developing diabetes-associated ocular complications (DAOC) among a population-based cohort of children diagnosed with either T1D or T2D during a 50-year period.
Methods
The medical records of all patients younger than 22 years (hereinafter referred to as children) who were newly diagnosed with diabetes from January 1, 1970, through December 31, 2019, in Olmsted County, Minnesota, were retrospectively reviewed. Approval for the study was obtained from the institutional review boards of the Mayo Clinic, Rochester, Minnesota, and Olmsted Medical Center. The study adhered to the ethical principles described in the Declaration of Helsinki.7 Given the retrospective nature, no informed consent was required. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
An extensive search of all Berkson hospital adaptations of the International Classification of Diseases, and International Classification of Diseases, Ninth Revision, codes diagnosing diabetes was performed using the Rochester Epidemiology Project (REP) to identify all potential patients within the age and diagnosis date criteria.8 The REP links all medical care delivered to residents of Olmsted County, Minnesota, using diagnostic codes.9 Given the relative isolation of residents in Olmsted County from other urban centers, the REP captures virtually all medical care provided to residents by the Mayo Clinic, Olmsted Medical Group, and their affiliated hospitals.10
A diagnosis of T1D was confirmed for inclusion if the medical record reported a diagnosis of insulin-dependent diabetes mellitus or type 1 diabetes with subsequent treatment consistent with management of T1D, whereas a diagnosis of T2D was confirmed as non–insulin-dependent diabetes mellitus or type 2 diabetes.3 Exclusion criteria included patients living outside Olmsted County at the time of their diabetes diagnosis, those diagnosed after 21 years of age or outside the study period, or if the patient or guardian denied research authorization. All medical records identified via the diagnostic code search were individually reviewed to assess for diagnostic accuracy and to extract relevant demographic data such as self-reported sex, race, and ethnicity. Hemoglobin A1c (HbA1c) values were collected according to the laboratory value that was closest, within a maximum of 6 months, to the date of diabetes diagnosis and at each follow-up eye examination. Blood pressure and antihypertensive medications were also documented at each recorded visit.
Among 1362 unique individuals with a diagnostic code of diabetes before 22 years of age, 606 met inclusion criteria, of whom 525 (86.6%) underwent at least 1 eye examination after their diagnosis of T1D or T2D, excluding 81 (13.4%) without an examination. Follow-up eye examinations were recorded for any change in retinopathy status through 40 years of age. Diabetes-associated ocular complications assessed in this study included nonproliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), diabetic macular edema (DME), a visually significant cataract (VSC), and the need for pars plana vitrectomy (PPV). Diabetic retinopathy was determined by the clinical documentation of NPDR (including background diabetic retinopathy) or of PDR. If fundus imaging was present without a corresponding clinical examination, a retinal specialist (A.J.B.) reviewed the images to assess for evidence of retinopathy. In addition, DME was noted if physician documentation included the terms diabetic macular edema or clinically significant macular edema. Surgical records were reviewed to determine whether any of the study participants received a PPV due to a DAOC. The presence of a VSC was defined as a cataract requiring extraction or identified on slitlamp examination with a concurrent visual acuity of less than 20/40, without confounding factors such as a vitreous hemorrhage or macular edema. The observation of a cataract predating the diagnosis of diabetes or resulting from ocular trauma was excluded.
An analysis of the differences in continuous variables between groups was completed using a 2-sample t test. Categorical differences were compared between groups using the Fisher exact test. The main outcome measures of developing diabetic retinopathy (NPDR or greater), PDR, DME, a VSC, or the need for a PPV were estimated using the Kaplan-Meier survival method,11 with additional calculation of hazard ratios (HRs) and 95% CIs for comparison. Potential baseline risk factors for these outcomes such as HbA1c levels or a recorded elevation in blood pressure were evaluated using Cox proportional hazards models. Data analysis was performed using SAS, version 9.4 (SAS Institute Inc).
Results
A total of 606 children were diagnosed with diabetes during the 50-year period, yielding an incidence of T1D of 26 per 100 000 children per year, whereas T2D occurred in 5 per 100 000 per year. Five hundred twenty-five participants (86.6%) underwent at least 1 eye examination and sufficiently met diagnostic criteria to distinguish between T1D (461 [87.8%]) and T2D (64 [12.2%]), after which they were followed up for a mean (SD) of 13.6 (9.4) years (range, 1 day to 37.1 years), and 8.6 (6.9) years (range, 17 days to 23.9 years), respectively. The mean (SD) age at diabetes diagnosis was 12.1 (5.4) years (range, 73 days to 21.8 years); 261 participants (49.7%) were female and 264 participants (50.3%) were male. Additional self-reported demographic and initial clinical characteristics are shown in Table 1. White children constituted a significantly larger percentage of the T1D cohort compared with the T2D cohort (384 [83.3%] vs 35 [54.7%]; P < .001), whereas Asian (9 [14.1%] vs 3 [0.7%]; P < .001) and Black (12 [18.7%] vs 21 [4.5%]; P < .001) children constituted a significantly larger percentage of the T2D cohort compared with the T1D cohort. The proportion of female participants in the T2D cohort was significantly higher than that in the T1D cohort (46 [71.9%] vs 215 [46.6%]; P < .001). There was no significant difference in HbA1c values at the diagnosis of diabetes (mean [SD], 13.4% [14.2%] for T1D vs 10.0% [3.6%] for T2D; P = .07) or at the initial diagnosis of diabetic retinopathy (mean [SD] 10.6% [2.7%] for T1D vs 10.6% [2.5%] for T2D; P = .96) between patients with T1D compared with T2D. Those with T2D were more likely to have a blood pressure of greater than 130/80 mm Hg or be using antihypertensive medications at the initial diagnosis of diabetic retinopathy compared with patients with T1D (11 [17.2%] vs 38 [8.2%]; P = .02).
Table 1. Historical and Clinical Characteristics of 525 Patients Younger Than 22 Years Diagnosed With T1D or T2D Who Underwent at Least 1 Eye Examination, 1970-2019.
| Characteristic | Patient groupa | Difference (T1D − T2D) % (95% CI) | P value | |
|---|---|---|---|---|
| T1D (n = 461) | T2D (n = 64) | |||
| Age at diabetes diagnosis, mean (SD), yb | 10.8 (5.1) | 17.3 (3.4) | –6.5 (–7.8 to –5.2) | <.001 |
| Sex | ||||
| Female | 215 (46.6) | 46 (71.9) | –25.3 (–37.2 to –13.3) | <.001 |
| Male | 246 (53.4) | 18 (28.1) | 25.3 (13.3 to 37.2) | <.001 |
| Race and ethnicity | ||||
| American Indian/Alaska Native | 0 | 1 (1.6) | –1.6 (–4.6 to 1.5) | .12 |
| Asian | 3 (0.7) | 9 (14.1) | –13.4 (–22.0 to –4.9) | <.001 |
| Black | 21 (4.5) | 12 (18.7) | –14.2 (–23.9 to –4.4) | <.001 |
| Hispanic | 3 (0.7) | 4 (6.3) | –5.6 (–11.6 to 0) | .005 |
| White | 384 (83.3) | 35 (54.7) | 28.6 (–39.7 to –14.4) | .001 |
| Mixed | 0 | 1 (1.6) | –1.6 (–4.6 to 1.5) | .12 |
| Otherc | 8 (1.7) | 1 (1.6) | 0.2 (–3.1 to 3.4) | >.99 |
| Unknown | 42 (9.1) | 1 (1.6) | 7.5 (3.5 to 11.6) | .047 |
Abbreviations: T1D, type 1 diabetes; T2D, type 2 diabetes.
Unless otherwise indicated, data are expressed as number (%) of patients.
Combined mean (SD) age: 12.1 (5.4) years (range, 73 days to 21.8 years).
Includes patients who self-identified as non-Hispanic or had an unknown Hispanic ethnicity and self-reported their race as other.
Diabetes-associated ocular complications occurred in 147 of the 461 children with T1D (31.9%) during a mean follow-up of 14 years (range, 1 day to 37 years) and in 17 of the 64 children with T2D (26.6%) during a mean follow-up of 9 years (range, 18 days to 24 years). The HRs illustrating the risk between T2D and T1D rates of developing complications were 1.88 (95% CI, 1.13-3.12; P = .02) for any diabetic retinopathy (NPDR or greater), 2.33 (95% CI, 0.99-5.50; P = .048) for PDR, 1.49 (95% CI, 0.46-4.89; P = .50) for DME, 2.43 (95% CI, 0.54-11.07; P = .24) for a VSC, and 4.06 (95% CI, 1.34-12.33; P = .007) for requiring PPV by 15 years after the diagnosis of diabetes. The Kaplan-Meier rates in T1D vs T2D for developing any retinopathy were 31% vs 53%; PDR, 5% vs 8%; DME, 5% vs 4%; a VSC, 1% vs 9%; or requiring PPV by 15 years after diagnosis of DM, 2% vs 8% (Table 2). The sex-adjusted HR for developing any retinopathy in T2D vs T1D was 1.88 (95% CI, 1.13-3.11; P = .02), whereas the sex-adjusted HR of developing PDR was 2.35 (95% CI, 0.99-5.56; P = .05). After adjusting for race using self-identified categories of White or not White, the adjusted HR of developing any retinopathy was 1.63 (95% CI, 0.96-2.79; P = .07), and the adjusted HR of developing PDR was 2.02 (95% CI, 0.79-5.16; P = .14). Given the shorter available follow-up in the T2D cohort, the Kaplan-Meier risks for developing DAOC in T2D beyond 15 years after diagnosis of diabetes were too unstable to calculate. There was no significant difference in the rate of PPV performed across each decade of diabetes diagnoses (rate per decade from 1970 to 2019, 6.4%, 0, 0, 3.7%, and 0 at 15 years; P = .11). Although 24 patients with PDR required PPV, of the 59 total patients with PDR, 57 also received anti–vascular endothelial growth factor, panretinal photocoagulation, or both. Of the 2 patients with PDR without documented treatment, one was lost to follow-up shortly after diagnosis, and the other had PDR without high-risk characteristics diagnosed near the end of the study period and was closely observed.
Table 2. Kaplan-Meier Rate of Developing Ocular Sequelae in Patients Younger Than 22 Years Diagnosed With Diabetes .
| DAOC, time after diabetes diagnosis, y | Patients with T1D (n = 461) | Patients with T2D (n = 64)a | HR (95% CI)b | P value | Overall HR (95% CI) | ||||
|---|---|---|---|---|---|---|---|---|---|
| No. at risk | No. of cumulative events | Kaplan-Meier rate, % of patients | No. at risk | No. of cumulative events | Kaplan-Meier rate, % of patients | ||||
| Any DR (NPDR or greater) (n = 163) | .02 | ||||||||
| 5 | 363 | 2 | 1 | 37 | 1 | 2 | NA | .20 | 1.88 (1.13-3.12) |
| 10 | 243 | 30 | 10 | 19 | 8 | 27 | NA | .005 | |
| 15 | 133 | 78 | 31 | 10 | 14 | 53 | NA | .004 | |
| 20 | 56 | 122 | 57 | NA | NA | NA | NA | NA | |
| 25 | 25 | 137 | 72 | NA | NA | NA | NA | NA | |
| 30 | 5 | 145 | 84 | NA | NA | NA | NA | NA | |
| PDR (n = 59) | .048 | ||||||||
| 5 | 364 | 0 | 0 | 37 | 0 | 0 | NA | c | 2.33 (0.99-5.50) |
| 10 | 261 | 1 | 0.4 | 23 | 2 | 8 | NA | <.001 | |
| 15 | 177 | 10 | 5 | 15 | 2 | 8 | NA | .24 | |
| 20 | 102 | 32 | 19 | NA | NA | NA | NA | NA | |
| 25 | 47 | 45 | 33 | NA | NA | NA | NA | NA | |
| 30 | 11 | 51 | 45 | NA | NA | NA | NA | NA | |
| DME (n = 40) | .50 | ||||||||
| 5 | 364 | 0 | 0 | 37 | 0 | 0 | NA | c | 1.49 (0.46-4.89) |
| 10 | 262 | 0 | 0 | 24 | 1 | 4 | NA | c | |
| 15 | 180 | 10 | 5 | 15 | 1 | 4 | NA | .84 | |
| 20 | 111 | 25 | 15 | NA | NA | NA | NA | NA | |
| 25 | 58 | 34 | 23 | NA | NA | NA | NA | NA | |
| 30 | 20 | 37 | 28 | NA | NA | NA | NA | NA | |
| VSC (n = 22) | .24 | ||||||||
| 5 | 362 | 2 | 1 | 37 | 0 | 0 | NA | c | 2.43 (0.54-11.07) |
| 10 | 260 | 2 | 1 | 23 | 1 | 4 | NA | .19 | |
| 15 | 181 | 4 | 1 | 15 | 2 | 9 | NA | .04 | |
| 20 | 120 | 10 | 6 | NA | NA | NA | NA | NA | |
| 25 | 68 | 13 | 9 | NA | NA | NA | NA | NA | |
| 30 | 21 | 17 | 19 | NA | NA | NA | NA | NA | |
| PPV (n = 24) | .007 | ||||||||
| 5 | 364 | 0 | 0 | 37 | 0 | 0 | NA | c | 4.06 (1.34-12.33) |
| 10 | 262 | 0 | 0 | 23 | 2 | 8 | NA | c | |
| 15 | 180 | 5 | 2 | 3 | 4 | 8 | NA | .04 | |
| 20 | 117 | 13 | 8 | NA | NA | NA | NA | NA | |
| 25 | 65 | 17 | 12 | NA | NA | NA | NA | NA | |
| 30 | 21 | 20 | 18 | NA | NA | NA | NA | NA | |
Abbreviations: DAOC, diabetes-associated ocular complications; DR, diabetic retinopathy; DME, diabetic macular edema; HR, hazard ratio; NPDR, nonproliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy; PPV, pars plana vitrectomy; T1D, type 1 diabetes; T2D, type 2 diabetes; VSC, visually significant cataract.
The T2D cohort did not have enough years of follow-up to calculate a stable Kaplan-Meier estimate beyond 15 years.
Indicates T2D vs T1D.
No events occurred during this time in 1 or both groups, so P value comparisons cannot be made.
The prevalence of diabetic retinopathy, including both NPDR and PDR, was 32.6% at 15 years among all patients with documentation of eye examinations during a mean (SD) follow-up of 13.0 (9.3) years, with 18 individuals dying before 40 years of age. Among patients with T1D, 30.6% developed either form of retinopathy by Kaplan-Meier estimates at 15 years of diabetes duration compared with 52.7% of patients with T2D. The relative risk of developing retinopathy in patients with T2D vs T1D was 1.88 (95% CI, 1.33-3.12; P = .02) by 15 years of diabetes duration. The youngest ages at diagnosis were 12.6 years for NPDR, 18.4 years for PDR, and 19.9 years for DME among patients with T1D; among patients with T2D, the youngest ages at diagnosis were 21.4 years for NPDR, 23.5 years for PDR, and 23.8 years for DME. No PDR was detected within 5 years after diabetes diagnosis.
Discussion
In this 50-year population-based cohort, children diagnosed with T2D had a higher risk of developing diabetic retinopathy, developing PDR, and requiring PPV compared with those diagnosed with T1D. The duration between the diagnosis of diabetes and the development of diabetic retinopathy was shorter in the T2D cohort compared with the T1D cohort, and patients with T2D developed vision-threatening retinopathy at a higher rate than those with T1D. This suggests that the natural history of retinopathy development among youth diagnosed with T2D may differ from that in youth diagnosed with T1D, where patients with T2D may be more susceptible to developing retinopathy than those with T1D despite controlling for diabetes disease duration.
Because the occurrence of youth-onset T2D was relatively infrequent before the 21st century,1,12 publications reporting on the development of diabetic retinopathy in children consisted of those diagnosed with T1D alone. From 2001 to 2009, however, the SEARCH for Diabetes in Youth Study6 reported a 30.5% increase in T2D, with estimates projecting a quadrupled prevalence to affect 84 131 US children by 2050.13 Since 2006, several studies14,15,16,17,18 have reported a 4% to 44.7% prevalence of all forms of diabetic retinopathy in youth-onset T2D with a mean or median follow-up since diabetes diagnosis ranging from 1.1 years to 8.1 years, compared with 3.4% to 20.1% in T1D with a mean or median follow-up since diabetes diagnosis ranging from 3.2 years to 6.8 years.14,15,16,17 The prevalence of retinopathy in the present study (30.6% in T1D and 52.7% in T2D by 15 years of diabetes duration) was higher than that reported in most prior studies, likely owing in part to the larger cohort size and longer mean follow-up.
A recently published observational study of diabetic complications in participants with youth-onset T2D 7 years after initial enrollment in the Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY) clinical trial18 found that 50.0% developed diabetic retinopathy, 3.2% developed PDR, and 3.5% developed DME, further emphasizing the development of potentially serious retinopathy in youth-onset T2D. The study also demonstrated that hyperglycemia and hypertension led to an association with an increase in risk for T2D-associated microvascular complications, including retinopathy.18 In the present study, hyperglycemia does not sufficiently explain the increased development of vision-threatening retinopathy in the T2D cohort compared with the T1D cohort, because there was no significant difference in HbA1c values between the 2 groups; there was, however, a higher percentage of patients with T2D who met hypertensive criteria at the initial diagnosis of diabetic retinopathy compared with patients with T1D. Adjusting for sex, the differences between Kaplan-Meier curves for developing any diabetic retinopathy remained statistically significant (P = .02) and PDR became nonsignificant (P = .052), suggesting that sex differences may influence some of the differences observed in the development of vision-threatening retinopathy between the T2D and T1D cohorts. Adjusting for sex in the rate of requiring PPV was not performed owing to limitations in the sample size.
More importantly, PDR developed after a shorter diabetes duration among children with T2D compared with T1D in this cohort, which is consistent with prior studies that suggest retinopathy occurs sooner and visual acuity is worse in youth with T2D.15,19,20 In contrast, a retrospective review within a managed care network16 reported that youth with T1D developed all forms of retinopathy earlier than those with T2D, citing a 6-year risk for developing any retinopathy of 27.6% in T1D compared with 8.6% in T2D. However, the investigators obtained the diagnoses from billing codes, which requires a more cautious interpretation given the known limitations of misdiagnosis, underdiagnosis, overdiagnosis, and improper coding, whereas using both diagnostic codes and medical record review has been demonstrated to have a higher sensitivity of determining true diagnoses.21,22 In addition to a difference in the rate of developing retinopathy between children diagnosed with T1D and T2D, studies have suggested that there may be a difference between youth-onset and adult-onset T2D, in which the development of insulin resistance and beta-cell deterioration progresses faster in the former, emphasizing the insufficiency of extrapolating the management of adult-onset T2D to youth-onset T2D.18
The elevated risk of retinopathy among children with T2D compared with T1D has been shown to correlate with demographic factors such as race.17,18 The investigators of the SEARCH pilot study17 hypothesized that the higher rate of retinopathy in their T2D cohort may correlate with the increased racial and ethnic minority group representation among their patients with T2D compared with those with T1D. In the present study, the T1D and T2D cohorts were mostly White, although the T2D cohort included a higher percentage of Asian and Black youth relative to the T1D cohort. After adjusting for race using self-identified categories of White or not White, the risk of developing any retinopathy between the T2D and T1D cohorts (P = .07), and the risk of developing PDR became nonsignificant (P = .14). Race has been suggested to be a surrogate for factors such as lower rates of optimal follow-up care received by racial and ethnic minority populations, which may influence the increased rates of retinopathy observed.6 The racial differences observed between this study’s T1D and T2D cohorts are additionally consistent with prior reports of youth-onset T2D occurring at a higher rate among American Indian, Asian, Black, and Hispanic youth than in White youth, whereas T1D occurs at a higher rate among White youth.23,24
Diabetic macular edema in the context of any level of diabetic retinopathy can significantly affect visual acuity. The Wisconsin Epidemiologic Study of Diabetic Retinopathy XXIII25 reported a 14-year cumulative incidence for DME of 26.1% in T1D, whereas patients with T1D in our study had a 5% risk of developing DME within 15 years of diabetes diagnosis, potentially reflecting improvements in diabetes management and control in recent decades. A similar 15-year risk of 4% for developing DME was observed in our study’s T2D cohort, although the data from previous studies on DME incidence in T2D26 are limited and predominantly focus on adult populations instead of those with childhood-onset T2D.
In addition to vision-threatening retinopathy, patients with diabetes are at increased risk of developing other ocular complications such as early-onset cataracts. Patients with diabetes are reported to be as much as 5 times more likely to develop cataracts at an earlier age than those without diabetes.27 However, few studies have described the risk of cataract development between early-onset T1D and T2D. Although the Wisconsin Epidemiologic Study of Diabetic Retinopathy cataract study did not include patients with T2D,28 it found that those with early-onset T1D (aged <30 years) had a lower 10-year cumulative incidence of requiring cataract extraction than those with older-onset diabetes. In the present study, we report a 1% risk of developing a VSC for patients with T1D compared with a 9% risk for those with T2D within 15 years of diabetes diagnosis, although this difference did not reach statistical significance.
Limitations
There are several limitations to the findings of this study. Its retrospective nature is limited by incomplete data and irregular follow-up. Some children with milder forms of diabetes may have eluded detection, a limitation that is more likely to affect T2D, which may exist undetected for years before a diagnosis.3 Thus, the observed time from diabetes diagnosis to the onset of PDR and other DAOC could be artificially shortened. In addition, given the rare but increasing prevalence of youth-onset T2D, the number of children diagnosed with T2D was relatively small, although larger compared with the previously discussed population-based studies, with 64 children (12.2%) with diabetes diagnosed with T2D. This relatively small number of children with ocular sequelae could potentially exaggerate the rates of DAOC in T2D compared with the larger T1D cohort. This study was also limited in that 13.4% of patients did not have documentation of an ocular evaluation at an REP-affiliated institution. It is possible that the patients without documentation of eye examinations were seen at local optometry clinics or outside Olmsted County. However, given the chronic nature of diabetes management, established referral patterns, and the significant distance to the nearest non–REP-affiliated specialist managing PDR and DME, it is unlikely that a significant number of patients reaching these primary end points would not be captured in this study. Moreover, many without documented eye examinations had only recently received a diagnosis of diabetes and may have not yet been due for an eye examination. Even for the 86.6% of patients with eye examinations, the rates of DAOC could have been underestimated if patients moved out of state or became lost to follow-up. In addition, given the changes in diagnostics and management of diabetes during the 50-year period, there are limitations in comparing patients across decades. Last, Olmsted County has a predominately White population; therefore, the findings from this study may not be generalizable to more racially diverse populations. Because racial and ethnic minority populations have experienced higher rates of diabetes and retinopathy in other studies,6 in a more racially diverse population, rates of diabetes and DAOC may be higher than what was observed in this population-based cohort. This study cohort, however, contains, to our knowledge, the largest sampling and longest follow-up of diabetes-associated ocular sequelae occurring among patients diagnosed with diabetes mellitus in childhood.
Conclusions
In this population-based cohort of children diagnosed with diabetes during a 50-year period, patients with T2D developed vision-threatening retinopathy after a shorter diabetes duration and at a higher rate than children with T1D. These findings suggest that to prevent serious ocular complications, children with T2D may require ophthalmoscopic evaluations at least as frequently as or more frequently than children with T1D.
References
- 1.Forlenza GP, Stewart MW. Diabetic retinopathy in children. Pediatr Endocrinol Rev. 2012;10(2):217-226. [PubMed] [Google Scholar]
- 2.Vithian K, Hurel S. Microvascular complications: pathophysiology and management. Clin Med (Lond). 2010;10(5):505-509. doi: 10.7861/clinmedicine.10-5-505 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Flaxel CJ, Adelman RA, Bailey ST, et al. Diabetic Retinopathy Preferred Practice Pattern®. Ophthalmology. 2020;127(1):P66-P145. doi: 10.1016/j.ophtha.2019.09.025 [DOI] [PubMed] [Google Scholar]
- 4.Lueder GT, Silverstein J; American Academy of Pediatrics Section on Ophthalmology and Section on Endocrinology . Screening for retinopathy in the pediatric patient with type 1 diabetes mellitus. Pediatrics. 2005;116(1):270-273. doi: 10.1542/peds.2005-0875 [DOI] [PubMed] [Google Scholar]
- 5.Klein R, Klein BE, Moss SE, Davis MD, DeMets DL. The Wisconsin Epidemiologic Study of Diabetic Retinopathy, II: prevalence and risk of diabetic retinopathy when age at diagnosis is less than 30 years. Arch Ophthalmol. 1984;102(4):520-526. doi: 10.1001/archopht.1984.01040030398010 [DOI] [PubMed] [Google Scholar]
- 6.Hamman RF, Bell RA, Dabelea D, et al. ; SEARCH for Diabetes in Youth Study Group . The SEARCH for Diabetes in Youth study: rationale, findings, and future directions. Diabetes Care. 2014;37(12):3336-3344. doi: 10.2337/dc14-0574 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.World Medical Association . World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191-2194. doi: 10.1001/jama.2013.281053 [DOI] [PubMed] [Google Scholar]
- 8.St Sauver JL, Grossardt BR, Yawn BP, et al. Data resource profile: the Rochester Epidemiology Project (REP) medical records-linkage system. Int J Epidemiol. 2012;41(6):1614-1624. doi: 10.1093/ije/dys195 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Melton LJ III. History of the Rochester Epidemiology Project. Mayo Clin Proc. 1996;71(3):266-274. doi: 10.4065/71.3.266 [DOI] [PubMed] [Google Scholar]
- 10.Rocca WA, Yawn BP, St Sauver JL, Grossardt BR, Melton LJ III. History of the Rochester Epidemiology Project: half a century of medical records linkage in a US population. Mayo Clin Proc. 2012;87(12):1202-1213. doi: 10.1016/j.mayocp.2012.08.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc . 1958;53(282):457-481. doi: 10.1080/01621459.1958.10501452 [DOI] [Google Scholar]
- 12.Group SS; SEARCH Study Group . SEARCH for Diabetes in Youth: a multicenter study of the prevalence, incidence and classification of diabetes mellitus in youth. Control Clin Trials. 2004;25(5):458-471. doi: 10.1016/j.cct.2004.08.002 [DOI] [PubMed] [Google Scholar]
- 13.Imperatore G, Boyle JP, Thompson TJ, et al. ; SEARCH for Diabetes in Youth Study Group . Projections of type 1 and type 2 diabetes burden in the US population aged <20 years through 2050: dynamic modeling of incidence, mortality, and population growth. Diabetes Care. 2012;35(12):2515-2520. doi: 10.2337/dc12-0669 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Eppens MC, Craig ME, Cusumano J, et al. Prevalence of diabetes complications in adolescents with type 2 compared with type 1 diabetes. Diabetes Care. 2006;29(6):1300-1306. doi: 10.2337/dc05-2470 [DOI] [PubMed] [Google Scholar]
- 15.Porter M, Channa R, Wagner J, Prichett L, Liu TYA, Wolf RM. Prevalence of diabetic retinopathy in children and adolescents at an urban tertiary eye care center. Pediatr Diabetes. 2020;21(5):856-862. doi: 10.1111/pedi.13037 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wang SY, Andrews CA, Herman WH, Gardner TW, Stein JD. Incidence and risk factors for developing diabetic retinopathy among youths with type 1 or type 2 diabetes throughout the United States. Ophthalmology. 2017;124(4):424-430. doi: 10.1016/j.ophtha.2016.10.031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mayer-Davis EJ, Davis C, Saadine J, et al. ; SEARCH for Diabetes in Youth Study Group . Diabetic retinopathy in the SEARCH for Diabetes in Youth Cohort: a pilot study. Diabet Med. 2012;29(9):1148-1152. doi: 10.1111/j.1464-5491.2012.03591.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bjornstad P, Drews KL, Caprio S, et al. ; TODAY Study Group . Long-term complications in youth-onset type 2 diabetes. N Engl J Med. 2021;385(5):416-426. doi: 10.1056/NEJMoa2100165 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tapley JL, McGwin G Jr, Ashraf AP, et al. Feasibility and efficacy of diabetic retinopathy screening among youth with diabetes in a pediatric endocrinology clinic: a cross-sectional study. Diabetol Metab Syndr. 2015;7:56. doi: 10.1186/s13098-015-0054-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Okudaira M, Yokoyama H, Otani T, Uchigata Y, Iwamoto Y. Slightly elevated blood pressure as well as poor metabolic control are risk factors for the progression of retinopathy in early-onset Japanese type 2 diabetes. J Diabetes Complications. 2000;14(5):281-287. doi: 10.1016/S1056-8727(00)00114-8 [DOI] [PubMed] [Google Scholar]
- 21.Rhodes ET, Laffel LM, Gonzalez TV, Ludwig DS. Accuracy of administrative coding for type 2 diabetes in children, adolescents, and young adults. Diabetes Care. 2007;30(1):141-143. doi: 10.2337/dc06-1142 [DOI] [PubMed] [Google Scholar]
- 22.Wei WQ, Teixeira PL, Mo H, Cronin RM, Warner JL, Denny JC. Combining billing codes, clinical notes, and medications from electronic health records provides superior phenotyping performance. J Am Med Inform Assoc. 2016;23(e1):e20-e27. doi: 10.1093/jamia/ocv130 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Nadeau KJ, Anderson BJ, Berg EG, et al. Youth-onset type 2 diabetes consensus report: current status, challenges, and priorities. Diabetes Care. 2016;39(9):1635-1642. doi: 10.2337/dc16-1066 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Mayer-Davis EJ, Bell RA, Dabelea D, et al. ; SEARCH for Diabetes in Youth Study Group . The many faces of diabetes in American youth: type 1 and type 2 diabetes in five race and ethnic populations: the SEARCH for Diabetes in Youth Study. Diabetes Care. 2009;32(suppl 2):S99-S101. doi: 10.2337/dc09-S201 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Klein R, Knudtson MD, Lee KE, Gangnon R, Klein BE. The Wisconsin Epidemiologic Study of Diabetic Retinopathy XXIII: the twenty-five-year incidence of macular edema in persons with type 1 diabetes. Ophthalmology. 2009;116(3):497-503. doi: 10.1016/j.ophtha.2008.10.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Lee R, Wong TY, Sabanayagam C. Epidemiology of diabetic retinopathy, diabetic macular edema and related vision loss. Eye Vis (Lond). 2015;2:17. doi: 10.1186/s40662-015-0026-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kiziltoprak H, Tekin K, Inanc M, Goker YS. Cataract in diabetes mellitus. World J Diabetes. 2019;10(3):140-153. doi: 10.4239/wjd.v10.i3.140 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Klein BE, Klein R, Moss SE. Incidence of cataract surgery in the Wisconsin Epidemiologic Study of Diabetic Retinopathy. Am J Ophthalmol. 1995;119(3):295-300. doi: 10.1016/S0002-9394(14)71170-5 [DOI] [PubMed] [Google Scholar]