Abstract
Purpose:
Chronic kidney disease (CKD) patients often develop cardiovascular disease (CVD) and retinopathy. The purpose of this study was to assess the association between progression of retinopathy and concurrent incidence of CVD events in participants with CKD.
Design:
We assessed 1051 out of 1936 participants in the Chronic Renal Insufficiency Cohort (CRIC) Study that were invited to have fundus photographs obtained at two time points separated by 3.5 years, on average
Methods:
Using standard protocols, presence and severity of retinopathy (diabetic, hypertensive, or other) and vessel diameter caliber were assessed at a retinal image reading center by trained graders masked to study participants’ information. Participants with a self-reported history of CVD were excluded. Incident CVD events were physician-adjudicated using medical records and standardized criteria. Kidney function and proteinuria measurements, along with CVD risk factors were obtained at study visits.
Results:
Worsening of retinopathy by two or more steps in the EDTRS retinopathy grading scale was observed in 9.8% of participants, and was associated with increased risk of incidence of any CVD in analysis adjusting for other CVD and CKD risk factors (odds ratio [OR]: 2.56, 95% confidence interval [CI]: (1.25, 5.22), p<0.01). After imputation of missing data these values were OR=1.66 (0.87, 3.16), p=0.12.
Conclusion:
Progression of retinopathy is associated with higher incidence of CVD events, and that retinal-vascular pathology may be indicative of macrovascular disease even after adjustment for kidney diseases and CVD risk factors. Assessment of retinal morphology may provide important information when assessing CVD in patients with CKD.
Keywords: Retinopathy, chronic kidney disease, cardiovascular disease, myocardial infarct, cerebrovascular accident, cardiac heart failure, cardiovascular risk
Introduction
Chronic kidney disease (CKD) is often linked to high rates of cardiovascular disease (CVD) and retinopathy.(1) We have previously reported the results from the Retinopathy in Chronic Renal Insufficiency (RCRIC) study on the cross-sectional association between retinopathy and cardiovascular disease (CVD) (1) in a group of participants with chronic kidney disease (CKD) enrolled in the Chronic Renal Insufficiency Cohort (CRIC) study.(2, 3) In this cohort, the prevalence of CVD was higher in patients with retinopathy and in patients with retinal venular dilation, and this association remained significant after controlling for traditional risk factors for CVD.(1) We subsequently reported that baseline retinopathy is a risk factor for the development of future CVD events.(4) In this prospective study in which fundus photographs have been obtained at baseline and then a second time about 3.5years later, we investigate the association between progression of retinopathy and concurrent incidence of CVD.
Methods
The CRIC study design (2, 3), along with the methodology used to interpret the retinal images (4–6) have been previously described. Participants for the Retinopathy in Chronic Renal Insufficiency (RCRIC) study, an ancillary investigation of the CRIC study, were recruited at 6 of the 7 CRIC clinical centers. All 2605 CRIC participants from these 6 sites were invited to participate in the RCRIC study. Of these participants 1936 (74.3%) were enrolled in the RCRIC study and baseline fundus photography were obtained. In 1025 (52.9%) participants a second set of photographs were obtained on average 3.5 +/− 0.5 (SD) years (range: 2.6 to 4.9 years) later. Baseline photography was performed between May, 2006 and June, 2008, and follow-up photography between November, 2009 and June, 2011. Our protocol adhered to Helsinki declaration principles and HIPPA, and was approved by an Institutional Review Board. All participants provided written informed consent.
Trained non-ophthalmic personnel obtained photographs. To induce a physiologic, non-pharmacologic dilatation of the pupils, participants were seated in a darkened room for five minutes. Forty-five degree digital, color fundus photographs were obtained using a Canon CR-DGI, Non-Mydriatic Retinal Camera (Canon Inc, Tokyo Japan). Two photographs, one centered on the macula and the other on the optic disc, were obtained from each eye. A participant was included if either the disc or macula photographs of one eye were of sufficient quality for assessment.
Evaluation of retinal abnormalities and vessel calibers
Trained graders and a retinal specialist, without knowledge of the participant’s clinical and demographic information, assessed fundus photographs at the RCRIC Fundus Photograph Reading Center. The Atherosclerosis Risk in Communities (ARIC) (7) and the Early Treatment of Diabetic Retinopathy (ETDRS) protocols (8) were used to grade retinopathy due to diabetes, hypertension, and other conditions. Photographs were evaluated by a single masked reader.(4–6). Reliability assessments have been provided in a previous report (5)
Presence and severity of retinopathy of any cause (diabetic, hypertensive, or other etiology) and diameters of the major retinal vessels were assessed. Because the readers were unaware of the diabetic or hypertensive status of the participants, retinopathy was evaluated without assumption of cause. Fundus pathology including retinopathy was assessed in 1051 participants and the diameters of the major retinal vessels were measured in 959 participants (Figure 1).
Figure 1.
Flow chart describing study participants.
PRP = pan retinal laser photocoagulation.
Retinal abnormalities were graded by referring to standard photographs. From a diverse group of retinal abnormalities, an overall retinopathy severity score was assigned for each eye based on a modification of the ETDRS diabetic retinopathy scale.(6, 8) There were 10 ordinal scores, ranging from no retinopathy (level 10) to advanced proliferative retinopathy (level 80). Scores for an eye were used to determine the eye’s baseline retinopathy classification (4 categories) as normal, very mild non-proliferative (NPR) retinopathy, moderate and severe NPR, or proliferative retinopathy (PR). The eye with more advanced retinopathy was used as the person-level classification at baseline (86% of participants had 2 eyes with good quality photographs at both visits) In the cases when photographs of only one eye were available, the score for that eye was used as the person-level category.
To determine progression of retinopathy, the 10-level ordinal scale scores from each of the two eyes of a participant were combined into a 19-step scale.(9) Progression of retinopathy was defined as an increase of 2 or more steps in the ETDRS person-level score from the first score.(9) An increase of 2 or more steps in the ETDRS person-level scores has been used in previous epidemiologic studies as representing a clinically meaningful progression of retinopathy.(9)
Measurements of vascular arteriolar and venular calibers were performed according to the ARIC Study protocol.(7) The diameter of up to 6 arterioles and 6 venules were averaged (5, 7), providing separate average arteriolar and venular changes. A difference (increase or decrease) of 10 microns or more in vessel diameter between baseline and follow-up was considered a meaningful change based on reproducibility of measurement.(5, 6) For vascular caliber assessment all 161 eyes with laser panretinal photocoagulation at baseline were excluded because this treatment causes large decreases in vascular calibers. Patients that received panretinal photocoagulation during follow up were included in the imputed analysis.
Ascertainment and adjudication of cardiovascular events
At study entry, CRIC participants provided only self-reported information on previous clinical CVD. During follow-up, participants were evaluated at annual in-person visits and interim 6-month phone calls. Electronic medical records were also searched for hospitalizations. For all hospitalizations, International Classification of Diseases codes for any discharges were obtained. If any codes suggestive of a possible cardiovascular event (i.e., heart failure [HF], acute myocardial infraction [MI], stroke, and peripheral artery disease [PAD]) were present, or if participants died during a hospitalization, relevant medical records were retrieved for detailed manual review.(10)
The criteria used for adjudication of HF were based on: clinical symptoms, radiographic evidence, physical examination of the heart and lungs, central venous hemodynamic monitoring data, and echocardiographic imaging that were consistent with Framingham clinical criteria.(11) The criteria for MI were based on symptoms of acute cardiac ischemia, cardiac biomarkers (e.g., creatinine kinase, cardiac troponin), and electrocardiographic data consistent with the third international criteria for MI.(12, 13) Two neurologists reviewed all hospitalizations suggestive of stroke using criteria based on acute neurological symptoms and tests/procedures.(14)
Participant age, gender, race/ethnicity and smoking status were obtained. Hypertension was defined as a systolic BP ≥140 mmHg, diastolic BP ≥90 mmHg, and/or use of antihypertensive medications. Diabetes mellitus was defined as either fasting glucose ≥126 mg/dl, random glucose ≥200 mg/dl, or use of insulin or other anti-diabetic medication (3). Estimated glomerular filtration rate (eGFR) was calculated using the CRIC eGFR equation.(15)
Statistical Methods
We compared participant characteristics using chi-squared tests for categorical variables and t-tests for continuous variables. Data values from the annual visit closest in time to fundus photography were used as baseline characteristics for the RCRIC study analyses.
Ninety six participants who had advanced retinopathy at the time of the baseline photography were excluded from our analyses as they could not worsen by ≥2 steps (Figure 1). We also excluded participants who had had history of a specific CVD event at the time of baseline retinal imaging, in order to focus only on incident CVD events. This resulted in the exclusion of 102 participants with any CVD events, 30 with MI, 10 with stroke, 9 with PAD, and 65 with HF.
We evaluated the association of worsening of retinopathy by two or more steps between initial and follow-up photography with development of CVA, MI, PAD, CHF, or any of those four (any CVD) during the time elapsed between these two photography sessions, by using univariate logistic regression models, and multivariable models adjusted for age, sex, baseline smoking status, low density lipoprotein, high density lipoprotein, triglycerides, diabetes, hypertension, hemoglobin A1c, systolic blood pressure, initial eGFR, 24-hour urine protein, and initial retinopathy level. Similar analyses were performed for the association between change of retinal vascular calibers between two photograph sessions (categorized into decreased ≥10 um, increased≥10 um, within 10um change) and incidence of any CVD (the composite cardiovascular outcome).
We first performed these analyses using only patients with both initial and follow-up photographs (two imaging session cohort, primary analysis). Due to the large number of participants without follow-up photographs (n=587, 34% for retinopathy analyses, and n=526, 34% for retinal vascular caliber analysis; Figure 1), that may have biased the results of the two imaging sessions cohort, we performed multiple imputation (11) of missing data for the change retinopathy score (secondary analysis). In this type of analysis we estimated the missing data based on the data available from the participants that had both baseline and follow up photography. Among those without follow-up photographs, we first selected the time interval between initial and follow-up photograph by random sampling time intervals from patients who had both initial and follow-up photographs. If the selected time was after the patient had died or been censored, the time of death or censoring measurement was used as the imputed time. Next, we imputed whether ≥2-step retinal level worsening had occurred by using a logistic regression model that included as covariates previously identified predictors of change in retinopathy (4, 5) measured at the time of the initial photographs and the imputed length of follow-up. We added the imputed data to the data from the cohort patients with photographs at both times (“imputation added cohort”) and repeated the regression analyses as described previously. We repeated these steps 50 times (16), and combined the regression analysis results using SAS 9.4 (SAS Institute Inc, Cary, NC).
Results
Among 1936 participants we excluded 11 with no CRIC follow up. For the analysis of retinopathy progression, from the 1925 remaining patients, we excluded 116 with baseline photos ungradable for retinopathy assessment and 96 with advanced retinopathy level that could not increase by two steps. Out of these remaining 1713 participants 587 did not have follow up assessment for retinopathy (76 died before the beginning of the follow up imaging period and 68 died during the imaging follow up period) and 75 had ungradable retinopathy at follow-up. Therefore, there were 1051 participants that had retinopathy analysis, Figure 1.
For vessel diameter analysis, we excluded 161 patients with panretinal photocoagulation at baseline. From the remaining 1764 participants, we excluded 203 with baseline photos ungradable for vascular caliber assessment. From the remaining 1561 cases, 526 were excluded because of no follow up photographs (64 died before the beginning of the follow up period and 65 died during the follow up period), and 76 had ungradable retinal vascular calibers at follow-up. Therefore, we had 959 participants included in the vascular diameter analysis, Figure 1.
Characteristics of the participants have been described.(1,5) All 1713 participants who had gradable baseline retinal photographs and retinopathy that could progress by two steps were included in the study. Median age at the time of first photography was 59.7 years (range: 23 to 77 years), 48% were white, 42% were black, 55% were men, and 750 (44%) had diabetes. This age, sex, ethnicity/race and presence of diabetes distribution is similar to the main CRIC cohort (3). At baseline, 1242 (73%) participants had no retinopathy, 142 (8%) had mild non-proliferative retinopathy, 240(14%) had non-proliferative retinopathy, 89 (5%) had proliferative retinopathy (Table 1).
Table 1:
Comparison of baseline characteristics of CRIC participants with and without follow-up photography among eligible participants with no prior known cardiovascular disease.
| Has follow-up photograph |
|||||
|---|---|---|---|---|---|
| Total n=1713 | No n=587 (34%) | Yes n=1126 (66%) | p-value* | ||
| Baseline Retinal | No NPR | 1242 (73%) | 407 (69%) | 835 (74%) | 0.009 |
| Level Findings | Mild NPR | 142 (8%) | 46 (8%) | 96 (9%) | |
| NPR | 240 (14%) | 97 (17%) | 143 (13%) | ||
| PR | 89 (5%) | 37 (6%) | 52 (5%) | ||
| Arteriole diameter (^m) | mean (SD) | 149 (14.5) | 149 (15.3) | 149 (14.1) | 0.73 |
| Venule diameter (^m) | mean (SD) | 220 (23.5) | 220 (24.1) | 220 (23.2) | 0.95 |
| eGFR (ml/min/1.73m2) | mean (SD) | 45.4 (18.4) | 41.5 (18.3) | 47.4 (18.0) | <0.001 |
| Urine Protein g/24 hours | <0.10 | 623 (36%) | 183 (31%) | 440 (39%) | <0.001 |
| 0.10 - <0.50 | 482 (28%) | 159 (27%) | 323 (29%) | ||
| 0.50 - <1.50 | 236 (14%) | 86 (15%) | 150 (13%) | ||
| 1.50+ | 183 (11%) | 72 (12%) | 111 (10%) | ||
| Missing | 189 (11%) | 87 (15%) | 102 (9%) | ||
| Age | mean (SD) | 59.7 (10.8) | 60.1 (11.3) | 59.4 (10.6) | 0.21 |
| Race/ethnicity | Hispanic | 80 (5%) | 27 (5%) | 53 (5%) | 0.45 |
| Non-Hispanic Black | 727 (42%) | 263 (45%) | 464 (41%) | ||
| Non-Hispanic White | 824 (48%) | 267 (45%) | 557 (49%) | ||
| Other | 82 (5%) | 30 (5%) | 52 (5%) | ||
| Gender | Women | 777 (45%) | 275 (47%) | 502 (45%) | 0.37 |
| Men | 936 (55%) | 312 (53%) | 624 (55%) | ||
| Diabetes mellitus | Yes | 750 (44%) | 285 (49%) | 465 (41%) | 0.004 |
| Hypertension | Yes | 1503 (88%) | 538 (92%) | 965 (86%) | <0.001 |
| Smoking status | Current smoker | 205 (12%) | 77 (13%) | 128 (11%) | 0.07 |
| Former smoker | 731 (43%) | 261 (44%) | 470 (42%) | ||
| Non-smoker | 777 (45%) | 249 (42%) | 528 (47%) | ||
| Body mass index (kg/m^2) | mean (SD) | 31.5 (7.6) | 31.9 (8.0) | 31.3 (7.4) | 0.12 |
| Normal (<25 kg/m2) | 310 (18%) | 103 (18%) | 207 (18%) | 0.49 | |
| Obese (≥30 kg/m2) | 887 (52%) | 312 (54%) | 575 (51%) | ||
| Overweight (25-<30 kg/m2) | 503 (30%) | 164 (28%) | 339 (30%) | ||
| Systolic BP (mmHg) | mean (SD) | 126.0 (21.2) | 129.3 (22.0) | 124.3 (20.6) | <0.001 |
p-values are trend tests for retinal level, smoking, and BMI, chi-square tests for other categorical variables, and t-tests for continuous variables
(N)PR=(Non) Proliferative Retinopathy, eGFR=Estimated Glomerular Filtration Rate, BP=Blood Pressure
In comparison to the 1126 participants who had follow up photography, the 587 participants without follow up photography had more advanced baseline retinopathy level (p=0.009), worse baseline eGFR (p<0.001), greater urine protein excretion(p<0.001), higher prevalence of diabetes mellitus (p=0.004), higher prevalence of systemic hypertension (p<0.001), and higher mean systolic blood pressure (p<0.001, Table 1).
Among 1051 patients who had both gradable baseline and follow-up photographs, 885 (84.2%) patients had no retinal level change, while 63 (6.0%) participants had one-step worsening, 28 (2.7%) had 2-step worsening, 34 (3.2%) had 3-step worsening, and 41 (3.9%) had more than 3-step worsening, with a total of 103 (9.8%) participants having ≥2-step worsening after the initial assessment. Worsening of ≥2-step retinal level from baseline occurred in 32 (5.0%) of 639 participants without diabetes, and 71 (17.2%) of 412 participants with diabetes. Among 1004 patients who had no prior CVD and had both baseline and follow-up photographs, 911 had no progression of retinopathy and 58 (6.3%) of them developed any CVD. Out of 93 participants with progression of retinopathy 18 (19.4%) developed any CVD (Table 2).
Table 2:
Association between 2-step retinal level step change and incidence of each cardiovascular outcome.
| Retinal Level Step Change | Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|---|
| <2 step worsening | ≥2 step worsening | OR (95% CI) | p | OR (95% CI) | p | ||
| Any cardiovascular event ** (n=1004) | No | 853 (93.6%) | 75 (80.6%) | Ref | <0.001* | Ref | 0.01* |
| Yes | 58 (6.4%) | 18 (19.4%) | 3.53 (1.98, 6.30) | 2.56 (1.25, 5.22) | |||
| Atherosclerotic event ** (n=1025) | No | 887 (95.7%) | 86 (87.8%) | Ref | 0.001 | Ref | 0.04 |
| Yes | 40 (4.3%) | 12 (12.2%) | 3.09 (1.56, 6.12) | 2.44 (1.05, 5.67) | |||
| Stroke (n=1045) | No | 934 (98.8%) | 98 (98.0%) | Ref | 0.48 | Ref | 0.60 |
| Yes | 11 (1.2%) | 2 (2.0%) | 1.73 (0.38, 7.93) | 1.56 (0.29, 8.42) | |||
| Acute myocardial infarction (n=1037) | No | 910(97.3%) | 91 (89.2%) | Ref | <0.001 | Ref | 0.002 |
| Yes | 25 (2.7%) | 11 (10.8%) | 4.40 (2.10, 9.23) | 4.28 (1.67, 10.95) | |||
| Peripheral artery disease (n=1045) | No | 933 (98.9%) | 98 (96.1%) | Ref | 0.03 | Ref | 0.17 |
| Yes | 10 (1.1%) | 4 (3.9%) | 3.81 (1.17, 12.37) | 3.00 (0.62, 14.66) | |||
| Heart failure (n=1026) | No | 896 (96.4%) | 86 (88.7%) | Ref | <0.001 | Ref | 0.0497 |
| Yes | 33 (3.6%) | 11 (11.3%) | 3.47 (1.69, 7.12) | 2.43 (1.00, 5.90) | |||
Adjusted for retinal level at first photo, age, sex, smoking, LDL, HDL, triglyceride level, systolic BP, hypertension, diabetes, hemoglobin A1C, eGFR, and 24-hour urine protein
The p-values for the interaction between step change and diabetes status are 0.49 unadjusted, and 0.95 adjusted
Atherosclerotic events include stroke, acute myocardial infarction, and peripheral artery disease.
Any cardiovascular events include all atherosclerotic events and heart failure.
Progression of retinopathy by two or more steps was associated with increased risk of concurrent incidence of any CVD event in univariate analysis (odds ratio [OR] 3.53, 95% Confidence Interval [CI] (1.98, 6.30), p<0.001, Table 2). This association remained statistically significant after adjustment for other CVD and CKD risk factors such as age, sex, smoking, LDL, HDL, triglycerides, systolic pressure, hypertension, diabetes, hemoglobin A1c, eGFR, and urinary protein (multivariate analysis, OR 2.56 (1.25, 5.22), p=0.01, Table 2). No interaction between diabetes status and progression of retinopathy was observed for any CVD outcome in multivariate analysis (p=0.95).
With regards to individual CVD events, progression of retinopathy by two or more steps was also associated with increased risk of concurrent MI in both univariate analysis (OR 4.4 (2.10, 9.23), p<0.001), and multivariate analysis (OR 4.28 (1.67, 10.95), p=0.002, Table 2).
When we considered stroke, MI and PAD together, as atherosclerotic CVD events, retinopathy progression by 2 or more steps was associated with increased risk in both univariate analysis (OR 3.09 (1.56, 6.12), p=0.001), and multivariate analysis (OR 2.44 (1.05, 5.67), p=0.04, Table 2).
Finally, retinopathy progression by two or more steps was associated with increased risk of concurrent HF in univariate analysis (OR 3.47 (1.69, 7.12), p<0.001), and multivariate analysis (OR 2.43 (1.00, 5.90), p=0.049, Table 2). However, no significant associations were observed for CVA and PAD (Table 2).
When participants without follow up photographs were included in the analysis using multiple imputation, the odds ratio decreased for any CVD to 2.28 (1.33, 3.90), p=0.003, in the univariate analysis and 1.66 (0.87, 3.16), p=0.12 in the multivariate analysis, Table 3.
Table 3:
Association between 2-step between retinal level step change and incidence of each cardiovascular outcome using additional imputed retinal level data.
| Retinal Level Step Change | Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|---|
| <2 step worsening | ≥2 step worsening | OR (95% CI) | p | OR (95% CI) | p | ||
| Any cardiovascular event ** | No | (91.9%) | (83.3%) | Ref | 0.003* | Ref | 0.12* |
| Yes | (8.1%) | (16.7%) | 2.28 (1.33, 3.90) | 1.66 (0.87, 3.16) | |||
| Atherosclerotic event ** | No | (94.8%) | (89.2%) | Ref | 0.02 | Ref | 0.27 |
| Yes | (5.2%) | (10.8%) | 2.20 (1.16, 4.16) | 1.54 (0.71, 3.33) | |||
| Stroke | No | (98.6%) | (97.8%) | Ref | 0.58 | Ref | 0.89 |
| Yes | (1.4%) | (2.2%) | 1.47 (0.37, 5.82) | 1.11 (0.24, 5.13) | |||
| Acute myocardial infarction | No | (97.0%) | (91.6%) | Ref | 0.002 | Ref | 0.07 |
| Yes | (3.0%) | (8.4%) | 2.95 (1.50, 5.81) | 2.15 (0.94, 4.92) | |||
| Peripheral artery disease | No | (98.7%) | (97.1%) | Ref | 0.18 | Ref | 0.43 |
| Yes | (1.3%) | (2.9%) | 2.14 (0.70, 6.62) | 1.71 (0.45, 6.46) | |||
| Heart failure | No | (95.2%) | (89.7%) | Ref | 0.01 | Ref | 0.29 |
| Yes | (4.8%) | (10.3%) | 2.24 (1.18, 4.27) | 1.52 (0.70, 3.28) | |||
Adjusted for retinal level at first photo, age, sex, smoking, LDL, HDL, triglyceride level, systolic BP, hypertension, diabetes, hemoglobin A1C, eGFR, and 24-hour urine protein
The p-values for the interaction between step change and diabetes status are 0.65 unadjusted, and 0.99 adjusted
Atherosclerotic events include stroke, acute myocardial infarction, and peripheral artery disease.
Any cardiovascular events include all atherosclerotic events and heart failure.
No statistically significant associations were detected between changes in arteriolar diameter and concurrent incidence of any CVD event in both univariate analysis (p=0.95) or multivariate analysis (p=0.76, Table 4).
Table 4:
Association between change in retinal vessel calibers and incidence of any cardiovascular event.
| Any Cardiovascular Event | Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|---|
| No 860 (94%) | Yes 59 (6%) | OR (95% CI) | P* | OR (95% CI) | P* | ||
| ΔArteriole diameter | ≤-10 μm | 210 (93.3%) | 15 (6.7%) | 1.04 (0.56, 1.92) | 0.95 | 0.87 (0.45, 1.69) | 0.76 |
| >-10 to <10 μm | 582 (93.6%) | 40 (6.4%) | Ref | Ref | |||
| ≥10 μm | 68 (94.4%) | 4 (5.6%) | 0.86 (0.30, 2.47) | 0.67 (0.21, 2.11) | |||
| ΔVenule diameter | ≤-10 μm | 242 (91.7%) | 22 (8.3%) | 1.90 (1.04, 3.47) | 0.04 | 1.66 (0.86, 3.22) | 0.14 |
| >-10 to <10 μm | 480 (95.4%) | 23 (4.6%) | Ref | Ref | |||
| ≥10 μm | 138 (90.8%) | 14 (9.2%) | 2.12 (1.06, 4.22) | 1.94 (0.92, 4.08) | |||
Adjusted for retinal level at first photo, age, sex, smoking, LDL, HDL, triglycerides, systolic BP, diabetes, hypertension, hemoglobin A1C, eGFR, and 24-hour urine protein
The p-values for interaction between change in vessel calibers and diabetes status are 0.67 and 0.31 for arteriole diameter unadjusted and adjusted, and 0.06 and 0.11 for venule diameter unadjusted and adjusted.
Venular diameter change, on the other hand, showed a trend of an association with increased risk of any CVD in univariate analysis (p=0.04). Decrease of diameter was associated with OR 1.9 (1.04, 3.47) and increase of diameter was associated with OR 2.12 (1.06, 4.22). No association was detected on multivariate analysis (p=0.14, Table 4). The imputation analysis showed that the risk estimates were further reduced for all outcomes (data not shown).
Finally, there were no significant interactions between diabetes and arteriolar or venous vascular diameters changes and any CVD outcome (p=0.31 for arterioles and p=0.11 for venules in the adjusted analysis, Table 4).
Discussion
Abnormalities of the retinal vasculature are often present in diabetic and non-diabetic populations. These abnormalities have been associated with increased risks of progression of systemic diseases.(17) Previous studies have shown an association between retinopathy and CKD and CVD.(4,18)
In this study we examine the association of retinopathy worsening and concurrent incidence of CVD in a large cohort of patients with CKD. Although most of the participants in our study did not show progression of retinopathy by two or more steps, we detected a strong association between progression of retinopathy and the concurrent development of any CVD, MI and CHF. This association remained significant even after adjustment for well known risk factors for progression of CVD and CKD, suggesting that progression of retinopathy offers information of concurrent CVD development beyond that provided by established risk factors. In addition, these findings strongly suggest that vascular changes in the retina have commonalities with vascular pathology influencing adverse clinical CVD events.
The secondary analysis with imputed data shows attenuation of the associations between progression of retinopathy and incidence of CVD and this could be due to differences between the participants with and without follow up photography, or imprecisions of the imputation model.
Wong at all (17) and Liew et al (18) have reported in population based studies and Grunwald et al (4) found in participants with CKD that baseline retinopathy is associated with future increased risk of heart disease in persons with and without diabetes. Our current study materially expands on previous studies by suggesting that progression of retinopathy may provide additional information on risk for concurrent CVD.
We have previously reported a significant 9-fold increase in risk for CVA development in patients with advanced retinopathy.(4) These patients with advanced retinopathy at baseline were excluded from the current analysis because they could not progress by 2 or more steps in retinopathy. This may have contributed to the lack of a significant association between progression of retinopathy and concurrent stroke. In addition, we have only 13 CVA events in this study and therefore, our power is limited.
We did not find any association between retinal arteriolar diameters and development of cardiovascular disease. There was a trend of increased risk of any CVD with both increase and decrease for ≥10 μm in venular diameter in the univariate analysis, p=0.04. However, after adjustment for risk factors for CVD and CKD, these trends were not statistically significant. Venular dilatation has been associated in previous studies with poor glycemic control (19), obesity inflammation, endothelium dysfunction (20), progression of diabetic retinopathy (21) and genetic markers (22), all factors that probably have a role in the development of CVD. Our current results are also consistent with other population studies showing associations between larger venular diameter and CVD (23, 24), although not all studies have reported this type of association. (25)
Strengths of our study are the prospective collection of photographic material with detailed evaluation of retinopathy features, and extensive collection of clinical data with adjudication of CVD events in a well described population of CKD participants.
However, our study has limitations. About one third of our participants did not have follow up photographs. To address this issue we performed additional analysis using imputed data that showed some attenuation of our associations. Our average follow-up of 3.5 years was relatively short which contributed to reduced statistical power. Additionally, our findings should be considered in light of the fact that some participants had photographs of poor quality that did not allow for definitive evaluation. Finally, exclusions of pre-existing CVD at entry into the RCRIC study was based only on participant self-report.
In summary, our results show a strong association between retinopathy progression and incident CVD in both univariate and multivariate analysis suggesting that progression of retinopathy offers additional information on CVD beyond that provided by well-known risks for CVD progression. In addition, our results from this study and our previous reports showing prevalence and progression of retinopathy in participants with CKD (1, 4) justify vigilance about eye disease in these patients. Finally, progression of retinopathy is associated with progression of CVD, and therefore, patients that have retinopathy progression should be assessed for CVD. A larger study is needed to further probe the pathophysiologic mechanisms that underpin the progression of retinopathy and heart disease.
Acknowledgements:
All authors have provided substantial contributions to the conception or design of the work. All authors have provided input into the drafting of the manuscript. All authors have approved the final version of the manuscript. All authors have agreed to be accountable for the accuracy and integrity of the manuscript.
We would like to acknowledge the contributions of the CRIC Study investigators: Jiang He, MD PhD and Akinlolu Ojo, MD, PhD.
Funding:
This study was supported by NIH grant RO1 DK 74151. CRIC was funded by NIDDK cooperative agreements (U01DK060990, U01DK060984, U01DK061022, U01DK061021, U01DK061028, U01DK060980, U01DK060963, and U01DK060902), University of Pennsylvania Clinical and Translational Science Award NIH/NCATS UL1TR000003, Johns Hopkins University UL1 TR-000424, University of Maryland GCRC M01 RR-16500, Clinical and Translational Science Collaborative of Cleveland, UL1TR000439, Michigan Institute for Clinical and Health Research (MICHR) UL1TR000433, University of Illinois at Chicago CTSA UL1RR029879, Tulane COBRE for Clinical and Translational Research in Cardiometabolic Diseases P20 GM109036, Kaiser Permanente NIH/NCRR UCSF-CTSI UL1 RR-024131, Vivian S. Lasko Research Fund, Nina C. Mackall Trust, and Research to Prevent Blindness.
None of the funding sources had any role in study design, writing of the report and decision to submit this article for publication.
Financial Disclosures:
Dr. Raymond Townsend is a consultant for Medtronic, ROX Medical, and receives royalties from UpToDate. Dr. Go has received research funding from Novartis, GlaxoSmithKline and Sanofi. All other coauthors have no financial conflict of interest regarding the contents of this manuscript.
References
- 1.Grunwald JE, Ying GS, Maguire MG, Pistilli M, Daniel E, Alexander J, Whittock-Martin R, Parker CR, Mohler E, Lo JC, Townsend R, Gadegbeku CA, Lash JP, Fink JC, Rahman M, Feldman H, Kusek J, Xie D, Coleman M, Keane MJ The CRIC Study Group. Association Between Retinopathy and Cardiovascular Disease in Patients with Chronic Kidney Disease (From the Chronic Renal Insufficiency Cohort [CRIC] Study). Am J Cardiology 2012;110:246–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Feldman HI, Appel LJ, Chertow GM, Cifelli D, Cizman B, Daugirdas J, Fink JC, Franklin-Becker ED, Go AS, Hamm LL, He J, Hostetter T, Hsu CY, Jamerson K, Joffe M, Kusek JW, Landis JR, Lash JP, Miller ER, Rahman M, Townsend RR, Wright JT., Chronic Renal Insufficiency Cohort (CRIC) Study Investigators: The Chronic Renal Insufficiency Cohort (CRIC) study: Design and methods. J Am Soc Nephrol 2003;14:S148–S153. [DOI] [PubMed] [Google Scholar]
- 3.Lash JP, Go AS, Appel LG, He J, Ojo A, Rahman M, Townsend RR, Xie D, Cifelli D, Cohan J, Fink JC, Fischer MJ, Gadegbeku C, Hamm LL, Kusek JW, Landis JR, Andrew Narva A, Robinson N, Teal V, Feldman HI., Chronic Renal Insufficiency Cohort (CRIC) Study Group: Chronic Renal Insufficiency Cohort (CRIC) Study: Baseline characteristics and associations with kidney function. Clin J Am Soc Nephrol 2009; 4: 1302–1311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Grunwald JE, Pistilli M, Ying GS, Maguire M, Daniel E, Whittock-Martin R, Parker-Ostroff C, Mohler E, Lo JC,. Townsend RR, Gadegbeku CA, Lash JP, Fink JC, Rahman M, Feldman H, Kusek JW, Xie D, and the CRIC Study Investigators. Retinopathy and the Risk of Cardiovascular Disease in Patients with Chronic Kidney Disease (From the Chronic Renal Insufficiency Cohort Study [CRIC]). Am J Cardiol 116(10): 1527–1533, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Grunwald JE, Alexander J, Ying GS, Maguire M, Daniel E, Whittock-Martin R, Parker C, McWilliams K, Lo JC, Go A, Townsend R, Gadegbeku CA, Lash JP, Fink JC, Rahman M, Feldman H, Kusek J, Xie D, Jaar BG. The CRIC Study Group. Retinopathy and Chronic Kidney Disease in the Chronic Renal Insufficiency Cohort Study (CRIC). Arch Ophthalmol 130:1136–1144, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Grunwald JE, Pistilli M, Ying GY, Daniel E, Maguire M, Xie D, Whittock-Martin R, Parker-Ostroff C, Lo JC, Townsend R, Gadegbeku CA, Lash JP, Fink JC, Rahman M, Feldman FI, Kusek JW, and the Chronic Renal Insufficiency Cohort Study Investigators. Retinopathy and Progression of CKD: The Chronic Renal Insufficiency Cohort Study. Clin J Am Soc Nephrol 9,1217–12242014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hubbard LD, Brothers RJ, King WN, Clegg LX, Klein R, Cooper LS, Sharrett AR, Davis MD, Cai J, Atherosclerosis Risk in Communities Study Group. Atherosclerosis Risk in Communities Study Group: Methods for evaluation of retinal microvascular abnormalities associated with hypertension/sclerosis in the Atherosclerosis Risk in Communities Study. Ophthalmology 1999; 106:2269–2280. [DOI] [PubMed] [Google Scholar]
- 8.Early Treatment Diabetic Retinopathy Study Research Group. Grading diabetic retinopathy from stereoscopic color fundus photographs - an extension of the Modified Airlie House classification. ETDRS Report Number 10. Ophthalmology 1991; 98:786–806. [PubMed] [Google Scholar]
- 9.Klein R, Klein BE, Moss SE. How many steps of progression of retinopathy are meaningful? The Wisconsin epidemiologic study of diabetic retinopathy. Arch Ophthalmol 19:547–553, 2001. [DOI] [PubMed] [Google Scholar]
- 10.Chen J, Mohler ER, Xie D, Shlipak MJ, Townsend RR, Appel LJ, Raj DS, Ojo AO, Schreiber MJ, Strauss LF, Zhang X, Wang X, He J MD, Hamm LL, for the CRIC Investigators. Risk Factors for Peripheral Arterial Disease Among Patients with Chronic Kidney Disease Am J Cardiol 2012;110:136–141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.McKee PA, Castelli WP, McNamara PM, Kannel WB. The Natural History of Congestive Heart Failure: The Framingham Study. N Engl J Med 1971; 285:1441–1446. [DOI] [PubMed] [Google Scholar]
- 12.Thygesen K, Alpert JS, Jaffe AS, Simoons ML, Chaitman BR, White HD. Third Universal Definition of Myocardial Infarction. Circulation 2012;126:2020–2035, [DOI] [PubMed] [Google Scholar]
- 13.Bansal N, Xie D, Sha D, Appel LJ, t Deo R, Feldman HI, He J, Jamerson K, Kusek JW, Messe S, Navaneethan SD, Rahman M, Ricardo AC, Soliman EZ, Townsend R, Go AS. Cardiovascular Events after New-Onset Atrial Fibrillation in Adults with CKD: Results from the Chronic Renal Insufficiency Cohort (CRIC) Study. JASN 29 (12) 2859–2869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sandsmark DK, Messé SR, Zhang X, Roy J, Nessel L, Lee Hamm L, He J, Horwitz EJ, Jaar BG, Kallem RR, Kusek JW, Mohler ER 3rd, Porter A, Seliger SL, Sozio SM, Townsend RR, Feldman HI, Kasner SE; CRIC Study Investigators. Proteinuria, but Not eGFR, Predicts Stroke Risk in Chronic Kidney Disease: Chronic Renal Insufficiency Cohort Study. Stroke 2015; 46:2075–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Anderson AH, Yang W, Hsu CY, Joffe MM, Leonard MB, Xie D, Chen J, Greene T, Jaar BG, Kao P, Kusek JW, Landis JR, Lash JP, Townsend RR, Weir MR, Feldman HI; CRIC Study Investigators: Estimating GFR among participants in the Chronic Renal Insufficiency Cohort (CRIC) Study. American Journal of Kidney Diseases 2012;60: 250–261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Rubin DB Multiple Imputation for Nonresponse in Surveys, New York: John Wiley & Sons, Inc. 1987. [Google Scholar]
- 17.Wong TY, Klein R, Sharrett AR, Manolio TA, Hubbard LD, Marino EK, Kuller L, Burke G, Tracy RP, Polak JF, Gottdiener JS, Siscovick DS. The Prevalence and Risk Factors of Retinal Microvascular Abnormalities in Older Persons. The Cardiovascular Health Study. Ophthalmology 2003; 110:658–666. [DOI] [PubMed] [Google Scholar]
- 18.Liew G, Wong TY, Mitchell P, Cheung N, Wang JJ. Retinopathy predicts coronary heart disease mortality. Heart 2009; 95:391–394. [DOI] [PubMed] [Google Scholar]
- 19.Ikram MK, Janssen JA, Roos AM, Rietveld I, Witteman JC, Breteler MM, Hofman A, van Duijn CM, de Jong PT. Retinal vessel diameters and risk of impaired fasting glucose or diabetes: the Rotterdam study. Diabetes 2006; 44:506–510. [DOI] [PubMed] [Google Scholar]
- 20.Klein BE, Knudtson MD, Tsai MY, Klein R. The relation of markers of inflammation and endothelial dysfunction to the prevalence and progression of diabetic retinopathy. Wisconsin epidemiologic study of diabetic retinopathy. Arch Ophthalmol 2009; 127:1175–1182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Roy MS, Klein R, Janal MN. Retinal venular diameter as an early indicator of progression to proliferative diabetic retinopathy with and without high-risk characteristics in African Americans with Type 1 diabetes mellitus. Arch Ophthalmol 2011; 129:8–15. [DOI] [PubMed] [Google Scholar]
- 22.Fahi SJ, Sun C, Zhu G, Healey PR, Spector TD, Martin NG, Mitchell P, Wong TY, Mackey DA, Hammond CJ, and Andrew T. The relationship between retinal arteriolar and venular calibers is genetically mediated, and each is associated with risk of cardiovascular disease. Invest Ophthalmol and Vis Sci 2011; 52:975–981. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wong TY, Kamineni A, Klein R, Sharrett AR, Klein BE, Siscovick DS, Cushman M, Duncan BB. Quantitative retinal venular caliber and risk of cardiovascular disease in older persons: the cardiovascular health study. Arch Intern Med 2006; 166:2388–2394. [DOI] [PubMed] [Google Scholar]
- 24.McClintic BR, McClintic JI, Bisogniano JD. The relationship between retinal microvascular abnormalities and coronary heart disease: A review. Am J Med 2010; 123:374e1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Sabanayagam C, Shankar A, Klein BE,Lee KE, Muntner P, Nieto FJ, Tsai MY, Cruickshanks KJ, Schubert CR, Brazy PC, Coresh J, Klein R. Bidirectional association between retinal vessel diameters and estimated GFR decline: The beaver Dam CKD Study. Am J Kidney Dis 2011;57:682–691. [DOI] [PMC free article] [PubMed] [Google Scholar]