QJM Advance Access originally published online on January 17, 2005
QJM 2005 98(2):119-126; doi:10.1093/qjmed/hci017
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
QJM vol. 98 no. 2 © Association of Physicians 2005; all rights reserved.
Poor prognosis in proteinuric type 2 diabetic patients with retinopathy: insights from the RENAAL study
From the 1Steno Diabetes Center, Copenhagen, Denmark, 2Faculty of Health Science, Aarhus University, Denmark, 3Medical Department M, Aarhus Kommunehospital, Aarhus, Denmark, 4Department of Medicine, Renal Division, Brigham and Women's Hospital, Boston, USA, and 5Baker Medical Research Institute, Melbourne, Australia
Received 30 June 2004 and in revised form 26 October 2004
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: Retinopathy is the clinical hallmark of generalized microangiopathy in diabetes.
Aim: To examine the relation of this abnormality to end-stage renal disease (ESRD) and death in type 2 diabetes.
Design: Retrospective analysis.
Methods: Of 1513 type 2 diabetic patients with nephropathy participating in the Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan (RENAAL) study, 1456 (96.5%) were assessed at baseline by ophthalmoscopy or fundus photography. RENAAL was a multinational double masked, randomized, placebo-controlled intervention study, whose primary end-point was the composite of a doubling of the baseline serum creatinine concentration, end-stage renal disease (ESRD) or death.
Results: Of those assessed at baseline, 65% had diabetic retinopathy. Patients with retinopathy had higher systolic blood pressure, albuminuria and lower glomerular filtration rate (GFR), haemoglobin and serum albumin values than those without. In univariate analyses, the presence of retinopathy was associated with a 44% increase in the primary composite end-point (hazard ratio 1.44, 95%CI 1.22–1.70, p < 0.001). Patients with retinopathy had a 52% increase in doubling of serum creatinine (p < 0.001), a 47% increased risk of ESRD (p = 0.002) and a 33% increase in risk of death (p = 0.026) compared to those without. In multivariate analyses, the presence of retinopathy was associated with a 23% increase (p = 0.015) in the primary composite end-point and a 22% increase in ESRD or death (p = 0.038).
Discussion: The presence of diabetic retinopathy at baseline is associated with more proteinuria, lower GFR, and a higher risk for ESRD and death in type 2 diabetic patients.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Retinopathy is a clinical hallmark of generalized microangiopathy in diabetes. Proteinuric type 2 diabetic patients who also have retinopathy have a glomerulopathy index above the normal range,1 and the presence of diabetic retinopathy strongly suggests that diabetic glomerulopathy is the cause of albuminuria in type 2 diabetic patients.2–4 A positive association between the degree of retinopathy and diabetic glomerulopathy has been found in type 2 diabetes.1,4 There is also evidence that proteinuric diabetic patients with diabetic glomerulopathy tend towards a faster rate of decline in GFR, compared to those without diabetic glomerulopathy.5–8 Approximately 30% of albuminuric type 2 diabetic patients without retinopathy have normal glomerular structure or non-diabetic kidney diseases in renal biopsy studies,2,9 and non-diabetic retinal microvascular abnormalities are associated with an increased risk for macrovascular events.10,11
In this study, based on data from the Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan (RENAAL) study,12 we analysed the association between diabetic retinopathy and renal and cardiovascular events, including end-stage renal disease (ESRD) and death.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The study design and description of the patient population has been published previously.13 In brief, the RENAAL study was a multinational, double blind, randomized, placebo-controlled study which assessed the renal and cardiovascular effects of the AT1 receptor antagonist losartan in type 2 diabetic subjects with overt nephropathy. As previously described in the initial report of the major findings of the RENAAL study,12 the primary endpoint was a composite of three events: a doubling of serum creatinine, ESRD, or death, whichever occurred first. In addition, there were a number of secondary endpoints, including cardiovascular morbidity, cardiovascular mortality and progression of renal disease (assessed by the slope of the reciprocal of serum creatinine).13
This report focuses on the relationship between both the presence and severity of retinopathy and the various renal and cardiovascular endpoints of the RENAAL study. Retinopathy was not a predefined endpoint of the study, but was evaluated at baseline in >96% of subjects. It was also not formally evaluated as part of the RENAAL protocol, since the investigators simply were asked to record the degree of diabetic retinopathy at a time as close as possible to baseline. Consequently, methods for detecting retinopathy, assessment of retinopathy degree and indication for laser photocoagulation were not identical. We have no information on the type of laser photocoagulation, i.e. scatter or focal, and no centralized grading process was applied.
Statistical analysis
All participants (n = 1513) were included from randomization through to the study termination date, regardless of adherence to study medication. The presence of diabetic retinopathy was initially assessed at baseline by ophthalmoscopy, fundus photography and fluorescein angiography. First, patients were divided into those with and those without retinopathy. Retinopathy patients were further divided into those with background retinopathy (non-proliferative) and those with severe retinopathy, defined as current or previous treatment with laser photocoagulation due to maculopathy or proliferative retinopathy. Baseline assessment was performed in 1456 patients (96.5%). In the remaining 57, data from medical records were available, and indicated that 36 had retinopathy and 21 did not have retinopathy. Nevertheless, the major analyses used only those with baseline retinopathy assessments (n = 1456).
Dichotomous variables (Table 1) were compared using the 
2 test. A two-sample t-test was used to explore the impact of the presence of retinopathy on the various renal and cardiovascular endpoints. When the impact of severity of retinopathy was assessed, a one way ANOVA was initially used. To further assess the impact of retinopathy, a Cox regression model was used with categorical variables as covariates to determine the hazard ratio of each category over a selected reference and 95%CI for hazard ratio (HR). The relationship between baseline retinopathy and endpoints of interest was explored with the group without retinopathy being treated as the reference for the HR. To explore retinopathy as an independent risk factor for both renal and cardiovascular events, a multivariate Cox model was used. The model incorporated non-renal risk factors, including blood pressure, prior type of antihypertensive drugs, HbA1c, lipids, uric acid and additional baseline covariates (age, gender, race, BMI, smoking and insulin use). The HRs for the presence and severity of retinopathy, adjusting for these risk factors, were then calculated. Analyses were performed with and without proteinuria and estimated level of GFR being included. The assumption of proportionality for the Cox analysis was tested and the criteria were met. The backward selection method was used to eliminate covariates which had p values 
0.05. The impact of retinopathy on the effect of losartan on various renal endpoints was also analysed using a Cox regression model. SAS version 8 was used for the analysis. A p value of <0.05 was considered to indicate statistical significance. All statistical tests were two-sided.
|
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The demographic, clinical and laboratory data in patients with and without diabetic retinopathy at baseline are shown in Table 1. The demographic and clinical data at baseline were similar in the two groups, except for the following variables, which were higher or more frequent in the group with retinopathy: systolic blood pressure; use of certain antihypertensive drugs, including dihydropyridine CCBs; diuretics and agents that interrupt the renin-angiotensin system; insulin use; neuropathy; and amputation. Patients with baseline retinopathy also had more severe nephropathy, as characterized by greater degree of proteinuria, higher serum creatinine concentration and lower serum albumin and haemoglobin levels. No differences in prevalence of smoking, serum cholesterol or glycaemic control were observed between the two groups. Comparing patients with severe (17%) vs. background retinopathy (83%), no significant differences at baseline in demographic, clinical or laboratory variables were seen, except for a higher prevalence of neuropathy in the severe retinopathy group (data not shown). The following regional differences in prevalence of diabetic retinopathy were noted: North America, 54.7%; Europe, 61.0%; Asia, 71.2%; and Latin America, 83.2% (p < 0.001).
By univariate analysis, the presence of diabetic retinopathy was associated with a 44% increase in the primary composite end-point (HR 1.44, 95%CI 1.22–1.70, p < 0.001, Figure 1a). A similar pattern was observed for the individual components of the composite end-point: doubling of serum creatinine concentration (52%, p < 0.001), ESRD (47%, p = 0.002) and death 33% (p = 0.026). A similar increase in the hazard risk was observed for the important end point of renal death (ESRD) or death (42%, p < 0.001, Figure 1b). Severe and background retinopathy were both associated with a steeper rates of decline in renal function (assessed by the reciprocal of the serum creatinine concentration) by 35% and 23%, respectively (severe retinopathy, mean slope 0.35 ± 0.25 mg/dl/year; background retinopathy, 0.32 ± 0.25; no retinopathy, 0.26 ± 0.26; p < 0.001). The difference between the two retinopathy groups did not reach statistical significance (severe vs. background p = 0.087).
|
No significant association with retinopathy status (background vs. severe) was observed for the major cardiovascular end-points, including the composite cardiovascular end-point of the first event of myocardial infarction, stroke, hospitalization for unstable angina, hospitalization for heart failure, coronary or peripheral revascularization and death from cardiovascular causes (whichever occurred first). The impact of retinopathy on both renal and cardiovascular HRs was similar in the losartan and placebo groups. The beneficial effect of losartan on the various renal end-points was not significantly influenced by the presence of retinopathy (data not shown).
When HRs were calculated comparing the three groups (severe, background and no retinopathy), no impact of severity of diabetic retinopathy could be detected on any of the renal or cardiovascular end points (Table 2).
|
Multivariate analysis was performed to determine whether retinopathy per se influenced any of the end points, after accounting for potential confounding non-renal factors such as blood pressure, HbA1c, lipids, smoking or use of agents such as dihydropyridine CCBs and insulin (Table 3). After accounting for these factors in a Cox model, the presence of retinopathy remained associated with a 23% increase (p = 0.015) in the primary composite end point and a 22% increase (p = 0.038) in ESRD or death. However, when renal factors such as proteinuria and estimated GFR were included in the multivariate analysis, retinopathy was eliminated as one of the predictors of the composite end-point or ESRD (data not shown). No effect of retinopathy was observed on all-cause death, which was primarily influenced by age, HbA1c, Hispanic ethnicity, LDL cholesterol, blood pressure and uric acid.
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Based on prospective observational data from the RENAAL study, our analysis demonstrates that the presence of any degree of diabetic retinopathy is associated with more proteinuria and a higher risk for ESRD and death in type 2 diabetic patients. Furthermore, each of the individual components of the composite end point was increased in patients with diabetic retinopathy, compared to patients without diabetic retinopathy. The inverse of serum creatinine, a continuous variable for measuring progression of kidney function was also significantly lower in patients with retinopathy consistent with more rapid progression of renal disease. The predictive power of diabetic retinopathy for subsequent ESRD and death in the RENAAL study was independent of a number of variables, including blood pressure, glycaemic control and lipids. The effects of retinopathy were eliminated after adjusting for baseline proteinuria or estimated GFR. Both these variables are closely linked to the presence of diabetic retinopathy (see Table 1). Indeed, renal and retinal disease in diabetes may be considered manifestations of a generalized diabetes-related disorder, microangiopathy.14 Since diabetic retinopathy and markers of diabetic nephropathy (proteinuria and GFR) are strongly associated, it may not be appropriate both from either physiological or a statistical point of view to include retinopathy and nephropathy in the same multiple regression analyses. From a clinical standpoint, the presence of diabetic retinopathy marks a patient for a high risk of nephropathy and ultimately progression to ESRD.
Ophthalmoscopy was commonly applied for retinopathy categorization at baseline, and this method cannot be considered a highly sensitive approach. Consequently, it is highly likely that a substantial proportion of those patients classified as having no retinopathy at baseline did in fact have disease. This bias would serve to dilute the hazard ratios when comparing no vs. background retinopathy. Furthermore, it is also likely that cases of severe retinopathy were missed, again diluting the risk estimates when comparing severe vs. background retinopathy. Therefore, we are probably underestimating the significance of the importance and link between diabetic retinopathy and progression of renal disease. Retinopathy should be assessed systematically in future renal studies. The regional differences in the prevalence of retinopathy suggest that genetic predisposition or protection, in addition to well known non-genetic risk factors for development of diabetic retinopathy, may vary considerably.
We found no relationship between cardiovascular end-points and retinopathy. This may relate to the way retinopathy was assessed in our study, in addition to the lack of power due to the relative small number of cardiovascular events. However, the HRs were very close to 1, with no apparent trend for any of the different cardiovascular end points.
The beneficial effects of losartan on the various renal end points were not significantly influenced by the presence or absence of retinopathy. The worsening impact of diabetic retinopathy on GFR decline has previously been demonstrated in studies of small numbers of proteinuric type 2 diabetic patients with rather well-preserved kidney function.5,8 In both studies, differences in the surrogate end point, rate of decline in GFR, were demonstrated, but neither had the power to report on the clinically relevant renal end-point of doubling of serum creatinine or ESRD. In recent preliminary analysis from the Irbesartan in type 2 Diabetic Nephropathy Trial (IDNT) (1715 patients), the presence of retinopathy increased the risk of progression of renal disease (ESRD) as well as cardiovascular events and death.15 Interestingly, a large population–based cardiovascular study in middle-aged people has revealed that retinal microvascular abnormalities are related to the development of coronary heart disease and stroke.10,11 Long-term prospective observational studies have documented that diabetic retinopathy is a risk factor for development of diabetic kidney disease both in type 116 and type 217 diabetes.
Several factors may play a role in the poor outcome for proteinuric type 2 diabetic patients with retinopathy. Retinopathy is the clinical hallmark of generalized microangiopathy in diabetes. Proteinuric type 2 diabetic patients with retinopathy have been reported to have abnormal glomerulopathy indices, based on quantitative morphometry,1,4 and the most severe degree of diabetic glomerulopathy was observed in proteinuric type 2 diabetic patients with the most severe eye disease, i.e. proliferative retinopathy.1,4 A close renal structural-functional relationship has been demonstrated in proteinuric type 2 patients. Specifically, patients with more severe glomerular lesions also had enhanced albuminuria and an accelerated rate of decline in GFR.1 Importantly, approximately 30% of an unbiased, consecutive cohort of proteinuric type 2 diabetic patients without diabetic retinopathy had either normal renal glomerular structure or non-diabetic nephropathies.9 Fioretto et al.3 reported normal or near-normal renal structure in 29%, and ‘atypical’ patterns of injury (absent or only mild diabetic glomerular changes, associated with mainly tubulointerstitial and arteriolar abnormalities) in 41% of type 2 diabetic patients with microalbuminuria. Diabetic retinopathy was present in all microalbuminuric type 2 diabetic patients with typical diabetic glomerulopathy. In a biopsy study of an unbiased proteinuric type 2 diabetic patient population closely resembling the IDNT and the RENAAL patients, there were two discrete patterns of glomerular pathology: retinopathy was associated with Kimmelstiel-Wilson nodules but not with mesangial sclerosis lesions.4 Marked proteinuria was associated with both lesions, although those with Kimmelstiel-Wilson lesions had lost a larger proportion of their kidney function. Recently, we found that in proteinuric type 2 diabetic patients without retinopathy, those with diabetic glomerulopathies tended towards a faster rate of decline in GFR, compared to those with non-diabetic glomerulopathies.7 Diabetic retinopathy is a requirement for the clinical diagnosis of diabetic nephropathy, which can be made in patients with diabetes who have persistent albuminuria without clinical or laboratory evidence of other kidney or renal tract disease.18
Apart from the association between diabetic retinopathy and renal structure and function, retinal abnormalities are related to markers of inflammation (e.g. white-cell count, fibrinogen concentration) and endothelial dysfunction (e.g. von Willebrand factor), consistent with microvascular pathology.19 In addition, abnormal autoregulation of blood flow to various tissues and organs has been demonstrated in patients with diabetic retinopathy.20
An important clinical message from our study is that more aggressive vasculoprotective treatment should be applied in proteinuric type 2 diabetic patients with diabetic retinopathy in order to improve their prognosis. The main strengths of our study include a large, multiracial, well-defined cohort of proteinuric type 2 diabetic subjects, with prolonged observation and validated identification of predefined primary and secondary end-points. However, several limitations should be noted. Firstly, validated fundus photos of predefined retinal fields were not applied. Secondly, standard grading of retinal microvascular abnormalities was not feasible. Thirdly, there was a lack of systematic eye examination during the trial. Finally, selection may have been biased according to the inclusion and exclusion criteria of the RENAAL study.
In conclusion, in our cohort of proteinuric type 2 diabetic patients, the presence of diabetic retinopathy was associated with a higher risk for ESRD and death. This damaging impact appears to be independent of arterial blood pressure, glycaemic control, hypercholesterolemia and ethnicity. To increase the power of renoprotective trials in the future, patients with diabetic retinopathy should be studied, since this is a group with a higher rate of progression of diabetic renal disease..
|
Acknowledgments |
|---|
The authors wish to express their appreciation to Denise Ramjit for assistance with the manuscript, and Mr Zhongxin Zhang for assistance with statistical analyses. The RENAAL study was an investigator-initiated study supported by Merck Research Laboratories.
|
Footnotes |
|---|
Address correspondence to Dr H.-H. Parving, Steno Diabetes Center, Niels Steensensvej 2, DK-2820 Gentofte, Denmark. e-mail: hhparving{at}dadlnet.dk
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Østerby R, Gall M-A, Schmitz A, Nielsen FS, Nyberg G, Parving H-H. Glomerular structure and function in proteinuric Type 2 (non-insulin-dependent) diabetic patients. Diabetologia 1993; 36:1064–70.[CrossRef][Web of Science][Medline]
2. Parving H-H, Gall M-A, Skøtt P, Jørgensen HE, Løkkegaard H, Jørgensen F, et al. Prevalence and causes of albuminuria in non-insulin-dependent diabetic patients. Kidney Int 1992; 41:758–62.[Web of Science][Medline]
3. Fioretto P, Mauer SM, Brocco E, Velussi M, Frigato F, Muollo B, et al. Patterns of renal injury in NIDDM patients with microalbuminuria. Diabetologia 1996; 39:1569–76.[CrossRef][Web of Science][Medline]
4. Schwartz MM, Lewis EJ, Leonard-Martin T, Breyer Lewis J, Battle D, The Collaborative Study Group. Renal pathology patterns in type II diabetes mellitus: relationship with retinopathy. Nephrol Dial Transplant 1998; 13:2547–52.
5. Gall M-A, Nielsen FS, Smidt UM, Parving H-H. The course of kidney function in type 2 (non-insulin-dependent) diabetic patients with diabetic nephropathy. Diabetologia 1993; 36:1071–8.[CrossRef][Web of Science][Medline]
6. Brocco E, Fioretto P, Mauer SM, Saller A, Carraro A, Frigato F, et al. Renal structure and function in non-insulin dependent diabetic patients with microalbuminuria. Kidney Int 1997; 63:S40–4.
7. Christensen PK, Larsen S, Horn T, Olsen S, Parving H-H. Renal function and structure in albuminuric type 2 diabetic patients without retinopathy. Nephrol Dial Transplant 2001; 16:2337–47.
8. Trevisan R, Vedovato M, Mazzon C, Coracina A, Iori E, Tiengo A, et al. Concomitance of diabetic retinopathy and proteinuria accelerates the rate of decline of kidney function in type 2 diabetic patients. Diabetes Care 2002; 25:2026–31.
9. Christensen PK, Larsen S, Horn T, Olsen S, Parving H-H. The causes of albuminuria in patients with type 2 diabetes without diabetic retinopathy. Kidney Int 2000; 58:1719–31.[CrossRef][Web of Science][Medline]
10. Wong TY, Klein R, Couper DJ, Cooper LS, Shahar E, Hubbard LD et al. Retinal microvascular abnormalities and incident stroke: the Atherosclerosis Risk in Communities Study. Lancet 2001; 358:1134–40.[CrossRef][Web of Science][Medline]
11. Wong TY, Klein R, Sharrett AR, Duncan BB, Couper DJ, Tielsch JM, et al. Retinal arteriolar narrowing and risk of coronary heart disease in men and women. The Atherosclerosis Risk in Communities Study. JAMA 2002; 287:1153–9.
12. Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch WE, Parving H-H, et al. Effects of Losartan on Renal and Cardiovascular outcomes in patients with type 2 diabetes and nephropathy. N Engl J Med 2001; 345:861–9.
13. Brenner BM, Cooper ME, de Zeeuw D, Grunfeld JP, Keane WF, Kurokawa K, et al. The Losartan renal protection study—rationale, study design and baseline characteristics of RENAAL (Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan). J Renin Angiotensin Aldosterone System 2000; 1:328–35.[CrossRef]
14. Deckert T, Feldt-Rasmussen B, Borch-Johnsen K, Jensen T, Kofoed-Enevoldsen A. Albuminuria reflects widespread vascular damage. The Steno hypothesis. Diabetologia 1989; 32:219–26.[CrossRef][Web of Science][Medline]
15. Rodby I, Gilbert R. The presence of retinopathy is associated with poor renal cardiovascular outcomes in type II diabetic nephropathy. J Am Soc Nephrol 2002; 13:644–5A.
16. Rossing P, Hougaard P, Parving H-H. Risk factors for development of incipient and overt diabetic nephropathy in Type 1 diabetic patients. Diabetes Care 2002; 25:859–64.
17. Gall M-A, Hougaard P, Borch-Johnsen K, Parving H-H. Risk factors for development of incipient and overt diabetic nephropathy in patients with non-insulin dependent diabetes mellitus: prospective, observational study. Br Med J 1997; 314:783–8.
18. Parving H-H. Initiation and progression of diabetic nephropathy. N Engl J Med 1996; 335:1682–3.
19. Klein R, Sharrett AR, Klein BE, Chambless LE, Cooper LS, Hubbard LD et al. Are retinal arteriolar abnormalities related to atherosclerosis?: The Atherosclerosis Risk in Communities Study. Arterioscler Thromb Vasc Biol 2000; 20:1644–50.
20. Kastrup J. The diabetic arteriole: The impact of diabetic microangiopathy on microcirculatory control. Dan Med Bull 1988; 35:334–45.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Greenstein, M. Tavakoli, M. Mojaddidi, A. Al-Sunni, G. Matfin, and R. A Malik Review: Microvascular complications: evaluation and monitoring relevance to clinical practice, clinical trials, and drug development The British Journal of Diabetes & Vascular Disease, July 1, 2007; 7(4): 166 - 171. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||