METHODS

Patients

The IDNT was an investigator-initiated, prospective, three-arm, randomized, double-blinded study of patients with diabetic nephropathy and hypertension to determine the relative efficacy of antihypertensive treatment based on irbesartan or amlodipine or placebo plus other antihypertensive agents. A description of the protocol and details of the baseline characteristics and clinical outcomes of this study have been previously published⁷. In addition to coded medication, all patients also received open-label hypertensive treatment, excluding angiotensin-converting enzyme inhibitors (ACEi), angiotensin receptor blockers (ARBs), or calcium channel blockers (CCBs) as needed to reduce blood pressure⁷. Briefly, the inclusion criteria were patients aged between 30 to 70 years having type 2 diabetes mellitus complicated by overt nephropathy (24-hour protein excretion rate 900 mg/day), serum creatinine between 1.0 and 3.0 mg/dL if female, or between 1.2 and 3.0 mg/dL if male, and either baseline seated blood pressures >135/85 mm Hg or having presented receiving medication for hypertension⁷. Exclusion criteria included the presence of a kidney disease other than diabetic nephropathy, absolute requirement for systemic immunosuppressive therapy, or a serum potassium concentration outside the normal range⁷.

Treatment

Eligible patients were randomly assigned in a 1:1:1 fashion to treatment with irbesartan 300 mg/day, amlodipine 10 mg/day, or matched placebo.

Outcomes

The primary outcome of the main trial was time to the composite of doubling of baseline serum creatinine concentration, end-stage renal disease (defined as a serum creatinine 6 mg/dL, renal dialysis, or transplantation), or all-cause mortality⁷. A prespecified CV outcome was the composite of cardiovascular death, nonfatal myocardial infarction (MI), heart failure (HF) resulting in hospitalization, stroke, unplanned coronary revascularization procedures, unplanned peripheral revascularization, and/or below knee amputation⁷; this CV composite was used to evaluate risk for experiencing a CV event. CV events were not ascertained after end-stage renal disease was reached.

Statistical analysis

The cardiovascular composite end point was used as the dependent outcome variable. As a marker of renal function, eGFR using the four-component Modification of Diet in Renal Disease (MDRD) formula was used, defined as: 186 (serum creatinine in mg/dL)^-1.154 (age in years)^-0.203; for women and non-Caucasians, the product was multiplied by a correction factor of 0.742 and 1.21, respectively⁸.

Analysis for CV predictors

All patients were followed for the cardiovascular composite (time to first event) until either end-stage renal disease was reached or they were lost to follow-up. Forty-one potential predictors (listed in the Appendix) were identified a priori among the demographic, history, and laboratory variables. Univariate analyses were performed using proportional hazards regression models (Cox regressions) for each of the 41 candidate predictors. To assess the roles of the various candidate predictors further, analysis was performed with multivariable Cox models allowing all 41 potential predictors to compete for entry. A stepwise procedure was used in which variables were added or removed from the model, depending on whether their contribution to the model was statistically significant, and on the relative degree of significance. A final multivariable model was evolved, all of whose terms contributed to the most parsimonious set of predictive variables.

The initial stepwise processing was performed on the subset of 1032 of the 1715 randomized patients who had complete data on all 41 candidate variables. Thereafter, at each stage, analysis was carried out on the larger number of patients who had complete data just for those predictors already in the model, or, in the case of adding variables, for those predictors plus each candidate variable not in the model individually in its own turn. In this way, the later processing used the maximum possible number of patients at each stage. The final multivariate model used data from 1484 patients.

The criterion for statistical significance was the 0.05 probability level. All models, univariate as well as multivariate, had terms forced in for randomized study medication so that all analyses were automatically adjusted for any effects study medication might have had, independently of whether those effects attained significance. All analyses were carried out with the Statistical Analysis System (SAS), versions 6 or 8 (Cary, NC, USA). In keeping with the intent-to-treat principle, all randomized patients were included in all analyses where availability of data made it possible, regardless of subsequent adherence to coded medication.

RESULTS

The baseline characteristics of the cohort are shown in Table 1. The average age was 58.9 7.7 years; almost 30% had evidence of prior cardiovascular disease at baseline. Mean eGFR was 48.1 18.3 mL/min, with a range of 11.4 mL/min to 155.4 mL/min. Mean 24-hour urinary albumin:creatinine ratio was 1416.2 1261.1 mg/g, with range between 16.7 mg/g to 12074.1 mg/g.

Table 1 - Baseline characteristics.

Full table

Of the 1715 randomized patients, 518 (30.2%) experienced the CV composite end point. Proportions of the CV composite experienced as the first event were: 4.1% for CV death, 5.5% for nonfatal MI, 11.8% for HF, 3.6% for stroke, 3.4% for unplanned coronary revascularization, and 1.8% for unplanned peripheral revascularization and/or amputation.

Univariate analysis for predictors

Demographic and clinical risk factors for CV events

Older age at randomization, longer duration of diabetes mellitus, and history of CVD were associated with cardiovascular events (Table 2). Other associations included history of congestive heart failure, history of myocardial infarction, prior stroke, ever smoking, and history of angina at baseline. Use of aspirin/antiplatelet agents, calcium channel blockers, diuretics, and insulin at presentation associated with CV events. Individuals of Hispanic, Asian, or Pacific Islander background were less likely to experience CV events than non-Hispanic Caucasians or blacks.

Table 2 - Significant univariate predictors.

Full table

Physical examination findings of neuropathy and retinopathy at baseline correlated with increased CV risk (Table 2). Higher-seated systolic and lower diastolic blood pressures at baseline appeared predictive for CV events.

Baseline laboratory characteristics for CV events

With respect to serum chemistries, individuals who experienced a CV event tended to have lower serum phosphorous, calcium, albumin, hemoglobin, and higher LDL cholesterol and blood urea nitrogen (BUN) at baseline (Table 2). Lower eGFR was associated with CV events. Baseline serum parameters that did not associate with CV events were HbA₁c, glucose, and HDL cholesterol. Regarding baseline urinary chemistries, high 24-hour urinary albumin:creatinine excretion ratio was predictive for CV events.

The proportion of individuals experiencing the CV composite end point progressively increased with increasing quartiles of baseline urine albumin:creatinine ratio Figure 1a, P < 0.01); similarly, the proportion experiencing CV events progressively increased with decreasing quartiles of serum albumin Figure 1b, P < 0.01).

Figure 1.

Proportion of individuals experiencing the CV composite end point by quartiles of (A) baseline urinary albumin:creatinine ratio (mg/g) and (B) baseline serum albumin (g/dL).

Full figure and legend (22K)

Multivariate analysis for predictors

The baseline demographic, clinical characteristics, and laboratory parameters were analyzed simultaneously to determine which measurements were independent risk factors of a CV event adjusted for all the others entering the model (Table 3). Eight independent risk factors were identified and included: older age, male gender, history of CVD, history of congestive heart failure, longer duration of diabetes, low serum albumin, and high urine albumin:creatinine excretion ratios. Also, relative to white or black race, those of Hispanic, Asian/Pacific Islander, Indian/Pakistani, and all others were significantly less likely to experience CV events.

Table 3 - Significant baseline predictors for CV events from multivariate analysis.

Full table

CONCLUSION

Compared with the general population, individuals with diabetic nephropathy and hypertension are at high risk for adverse CV outcomes; despite blood pressure–lowering therapy, over 30% of the IDNT cohort experienced at least one CV event. This analysis has identified baseline patient characteristics, beyond conventional risk factors, which are predictive for adverse cardiac outcomes in an already high CV risk population, having both diabetic nephropathy and hypertension.

As anticipated, a history of CVD and/or HF identified a very high-risk group for future events; this can be explained by the integration of a number of factors. Having either history at baseline signifies an individual with complex medical history (hypertension, smoking, dyslipidemia, insulin use, and CV medications), greater atherosclerotic burden, underlying cardiac dysfunction, and more complicated diabetes, all of which are established predictors for cardiac events.

Older age and male gender predicted CV events. Independent of age, a longer duration of diabetes at baseline also was predictive for CV events. Diabetes is associated with accelerated atherosclerosis; the age-adjusted relative risk for CV death is three-fold higher among diabetics than the general population⁹. A longer duration of diabetes reflects prolonged exposure to the hyperglycemic state associated with development of both micro- and macrovascular complications¹⁰. Baseline Hb_A1C and/or blood glucose levels were not associated with events in our study. Previous studies documenting association of Hb_A1C and/or fasting blood glucose levels to cardiac events, however, did not only rely upon baseline values, and followed their cohorts serially over time^11,12. Isolated Hb_A1C and blood glucose levels at baseline only represent glycemic control for a brief period at most, and may not appropriately reflect many years of hyperglycemia exposure.

The degree of albuminuria, assessed by urinary albumin:creatinine ratio, contributed significantly to increased risk; this despite all individuals already having clinically significant baseline proteinuria (900 mg/day). Every natural log unit increase in urinary albumin:creatinine ratio was associated with a 29% increase in the risk for CV events, the other predictors being assumed fixed. Albuminuria is a powerful predictor for progression to end-stage renal disease^6,13. It not only reflects glomerular damage, but is also a sensitive indicator of generalized atherosclerotic-mediated vasculopathy¹⁴ and is considered to be a marker of significant vascular risk^15,16,17. Several retrospective and prospective studies have consistently demonstrated that worsening albuminuria is paralleled by increasing cardiovascular complications^4,17. Albuminuria clusters with a number of vascular risk factors inclusive of hypertension, renal dysfunction, dyslipidemia, hyperhomocysteinemia, and several inflammatory markers¹⁷; this may also explain why traditional risk factors such as hypertension and dyslipidemia lost significance following multivariable analysis. The precise mechanism by which albuminuria portends increased CV risk is currently unknown; several derangements in the fibrinolysis and coagulation systems have been noted, favoring a thrombogenic state¹⁸. Low serum albumin was also found to be predictive for events; this may not only reflect the severity of albuminuria, but has also been shown to signify excessive catabolism¹⁹. Low serum albumin levels appear to signify a heightened state of inflammation, which in itself is a strong predictor for CV events²⁰.

In this cohort, although lower baseline eGFR demonstrated univariate association, it was not an independent predictor for CV events. Many studies have demonstrated significant influence of renal dysfunction upon adverse cardiovascular outcomes in community and high-risk cardiac populations. Particular studies have included analyses involving the Framingham cohort, NHANES I, Atherosclerotic Risk in the Community (ARIC), Survival and Ventricular Enlargement (SAVE), Studies of Left Ventricular Dysfunction (SOLVD), Thrombolysis in Myocardial Infarction (TIMI), and Valsartan in Acute Myocardial Infarction (VALIANT) trials^2,21,22,23. In most studies, reduced renal function, whether measured by eGFR or serum creatinine, was associated with higher mortality and adverse CV event rates. However, these studies consisted of mixed populations of chronic kidney disease (CKD) and normal renal function patients; the prevalence of CKD and/or diabetes in these studies was relatively low, yet these individuals were found to be at high risk for CV events. These studies did not examine the influence of albuminuria. Among diabetics with nephropathy, the degree of albuminuria has been shown to demonstrate an inverse relationship with eGFR²⁴. In the context of diabetic nephropathy, in which CKD and high risk for CV events is already established, low eGFR and/or high serum creatinine may already be accounted for by the degree of albuminuria, which in itself is a surrogate of renal dysfunction severity²⁵.

Lower hemoglobin concentrations conferred increased risk in the univariate analysis; however, it was not an independent predictor for CV events. Chronic anemia has been demonstrated to be an independent risk factor for adverse CV outcomes, particularly in patients with kidney disease and/or heart failure^26,27. Anemia is also an independent predictor for renal disease progression²⁸ because it reflects progressive erythropoietin deficiency and the negative influence of increasing uremic factors upon erythropoiesis²⁹. The mechanism for the association between anemia and CV events is unclear; in addition to reduced oxygen supply to already ischemic tissue, it is postulated that the presence of anemia may also be signifying systemic inflammatory activity³⁰. Based on such hypotheses, the loss of its significance following multivariable analysis may have been accounted for by other proinflammatory factors such as albuminuria.

Individuals of Hispanic, Asian, or Pacific Islander extraction, or those classed as 'other' were less likely to experience CV events. This is in contrast to what has been reported in the literature. A number of studies involving registries have demonstrated that individuals of non-Caucasian extraction are at higher risk for both cardiovascular and renal disease; these same populations also appear to be experiencing a rising prevalence and incidence of type 2 diabetes mellitus³¹. Genetic predisposition, socioeconomic status, reduced awareness, and cultural factors play important roles.

Given so many potential risk factors were analyzed raises the probability of false positives (i.e., of incorrectly identifying one or more covariates as predictors of risk). In addition, given the IDNT study was not specifically designed for risk factor analysis, chances for false negatives (i.e., of failing to recognize one or more legitimate predictors) may be high also; of note, we did have not complete electrocardiographic and/or chest x-ray data, which may have altered our model.

Patients with overt diabetic nephropathy and hypertension are at significant risk for adverse cardiac events. Among this cohort, individuals at higher risk for CV events are more likely to be male, older in age, have a longer duration of diabetes, prior history of cardiovascular disease and/or CHF, marked albuminuria, and hypoalbuminemia. In addition to traditional coronary risk factors, the degree of albuminuria and low serum albumin levels are important independent prognostic variables for cardiovascular events in this high-risk cohort.

References

1. DRIES, DL, EXNER, DV, DOMANSKI, MJ, GREENBERG, B, STEVENSON, LW: The prognostic implications of renal insufficiency in asymptomatic and symptomatic patients with left ventricular systolic dysfunction. J Am Coll Cardiol 2000 35: 681–689,  | Article | PubMed | ISI | ChemPort |
2. CULLETON, BF, LARSON, MG, EVANS, JC, et al: Prevalence and correlates of elevated serum creatinine levels: The Framingham Heart Study. Arch Intern Med 1999 159: 1785–1790,  | Article | PubMed | ISI | ChemPort |
3. FOLEY, RN, HERZOG, CA, COLLINS, AJ: Blood pressure and long-term mortality in United States hemodialysis patients: USRDS Waves 3 and 4 Study. Kidney Int 2002 62: 1784–1790,  | Article | PubMed |
4. DINNEEN, SF, GERSTEIN, HC: The association of microalbuminuria and mortality in non–insulin-dependent diabetes mellitus. A systematic overview of the literature. Arch Intern Med 1997 157: 1413–1418,  | Article | PubMed | ISI | ChemPort |
5. KANNEL, WB: Lipids, diabetes, and coronary heart disease: Insights from the Framingham Study. Am Heart J 1985 110: 1100–1107,  | PubMed |
6. UUSITUPA, MI, NISKANEN, LK, SIITONEN, O, et al: Ten-year cardiovascular mortality in relation to risk factors and abnormalities in lipoprotein composition in type 2 (non–insulin-dependent) diabetic and non-diabetic subjects. Diabetologia 1993 36: 1175–1184,  | Article | PubMed | ISI | ChemPort |
7. LEWIS, EJ, HUNSICKER, LG, CLARKE, WR, et al: Renoprotective effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med 2001 345: 851–860,  | Article | PubMed | ISI | ChemPort |
8. LEVEY, AS, BOSCH, JP, LEWIS, JB, et al: A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999 130: 461–470,  | PubMed | ISI | ChemPort |
9. SOWERS, JR, EPSTEIN, M, FROHLICH, ED: Diabetes, hypertension, and cardiovascular disease: An update. Hypertension 2001 37: 1053–1059,  | PubMed | ISI | ChemPort |
10. KLEIN, R: Hyperglycemia and microvascular and macrovascular disease in diabetes. Diabetes Care 1995 18: 258–268,  | PubMed | ChemPort |
11. LAAKSO, M: Long-term influence of glycaemic control on cardiovascular risk in NIDDM. Diabetes Res Clin Pract 1996 31(Suppl):S61–S65,  | PubMed |
12. MUGGEO, M, VERLATO, G, BONORA, E, et al: Long-term instability of fasting plasma glucose predicts mortality in elderly NIDDM patients: The Verona Diabetes Study. Diabetologia 1995 38: 672–679,  | PubMed |
13. DAMSGAARD, EM, FROLAND, A, JORGENSEN, OD, MORGENSEN, CE: Prognostic value of urinary albumin excretion rate and other risk factors in elderly diabetic patients and non-diabetic control subjects surviving the first 5 years after assessment. Diabetologia 1993 36: 1030–1036,  | PubMed |
14. DECKERT, T, KOFOED-ENEVOLDSEN, A, NORGAARD, K, et al: Microalbuminuria. Implications for micro- and macrovascular disease. Diabetes Care 1992 15: 1181–1191,  | PubMed | ISI | ChemPort |
15. BORCH-JOHNSEN, K, FELDT-RASMUSSEN, B, STRANDGAARD, S, et al: Urinary albumin excretion. An independent predictor of ischemic heart disease. Arterioscler Thromb Vasc Biol 1999 19: 1992–1997,  | PubMed | ISI | ChemPort |
16. YUDKIN, JS, FORREST, RD, JACKSON, CA: Microalbuminuria as predictor of vascular disease in non-diabetic subjects. Islington Diabetes Survey. Lancet 1988 2: 530–533,  | PubMed | ISI | ChemPort |
17. GERSTEIN, HC, MANN, JF, YI, Q, et al: Albuminuria and risk of cardiovascular events, death, and heart failure in diabetic and nondiabetic individuals. JAMA 2001 286: 421–426,  | Article | PubMed | ISI | ChemPort |
18. KARIO, K, MATSUO, T, KOBAYASHI, H, et al: Activation of tissue factor-induced coagulation and endothelial cell dysfunction in non–insulin-dependent diabetic patients with microalbuminuria. Arterioscler Thromb Vasc Biol 1995 15: 1114–1120,  | PubMed |
19. MOSHAGE, HJ, JANSSEN, JA, FRANSSEN, JH, et al: Study of the molecular mechanism of decreased liver synthesis of albumin in inflammation. J Clin Invest 1987 79: 1635–1641,  | PubMed | ISI | ChemPort |
20. BERGSTROM, J, LINDHOLM, B: Malnutrition, cardiac disease, and mortality: An integrated point of view. Am J Kidney Dis 1998 32: 834–841,  | PubMed | ISI | ChemPort |
21. GARG, AX, CLARK, WF, HAYNES, RB, HOUSE, AA: Moderate renal insufficiency and the risk of cardiovascular mortality: Results from the NHANES I. Kidney Int 2002 61: 1486–1494,  | Article | PubMed | ISI |
22. AL AHMAD, A, RAND, WM, MANJUNATH, G, et al: Reduced kidney function and anemia as risk factors for mortality in patients with left ventricular dysfunction. J Am Coll Cardiol 2001 38: 955–962,  | Article | PubMed | ISI | ChemPort |
23. GIBSON, CM, PINTO, DS, MURPHY, SA, et al: Association of creatinine and creatinine clearance on presentation in acute myocardial infarction with subsequent mortality. J Am Coll Cardiol 2003 42: 1535–1543,  | PubMed |
24. MACISAAC, RJ, TSALAMANDRIS, C, PANAGIOTOPOULOS, S, et al: Nonalbuminuric renal insufficiency in type 2 diabetes. Diabetes Care 2004 27: 195–200,  | PubMed |
25. GARG, AX, KIBERD, BA, CLARK, WF, et al: Albuminuria and renal insufficiency prevalence guides population screening: Results from the NHANES III. Kidney Int 2002 61: 2165–2175,  | Article | PubMed | ISI |
26. SARNAK, MJ, TIGHIOUART, H, MANJUNATH, G, et al: Anemia as a risk factor for cardiovascular disease in The Atherosclerosis Risk in Communities (ARIC) study. J Am Coll Cardiol 2002 40: 27–33,  | Article | PubMed | ISI |
27. JURKOVITZ, CT, ABRAMSON, JL, VACCARINO, LV, et al: Association of high serum creatinine and anemia increases the risk of coronary events: Results from the prospective community-based atherosclerosis risk in communities (ARIC) study. J Am Soc Nephrol 2003 14: 2919–2925,  | Article | PubMed | ISI | ChemPort |
28. BREYER, JA, BAIN, RP, EVANS, JK, et al: Predictors of the progression of renal insufficiency in patients with insulin-dependent diabetes and overt diabetic nephropathy. The Collaborative Study Group. Kidney Int 1996 50: 1651–1658,  | PubMed | ISI | ChemPort |
29. INOMATA, S, ITOH, M, IMAI, H, SATO, T: Serum levels of erythropoietin as a novel marker reflecting the severity of diabetic nephropathy. Nephron 1997 75: 426–430,  | PubMed |
30. HEGARTY, J, FOLEY, RN: Anaemia, renal insufficiency and cardiovascular outcome. Nephrol Dial Transplant 2001 16(Suppl 1):102–104,  | PubMed | ISI | ChemPort |
31. TIERNEY, WM, MCDONALD, CJ, LUFT, FC: Renal disease in hypertensive adults: Effect of race and type II diabetes mellitus. Am J Kidney Dis 1989 13: 485–493,  | PubMed | ISI | ChemPort |

Appendices

APPENDIX

The following 41 potential predictors were used in the CV events analysis:

Continuous variables: age (years), serum albumin (g/dL), body mass index, BUN, serum calcium (mg/dL), LDL (mg/dL), HDL (mg/dL), LDL:HDL ratio, duration of diabetes (years), blood glucose whether or not fasted, hemoglobin, hemoglobin A1_c, serum potassium (mg/dL), MDRD estimate of GFR (mL/min), serum sodium (mg/dL), serum phosphorus (mg/dL), 24-hr urinary albumin:creatinine ratio, seated diastolic blood pressure (mm Hg), seated systolic blood pressure (mm Hg).

Categorical variables (yes/no): male gender, black race, other race, family history of diabetes, family history of hypertension, history of neuropathy, history of retinopathy, smoking at entry, history of ever smoking, history of CV disease, history of MI, history of CHF, history of stroke, history of angina, use at entry of ACEi, aspirin/antiplatelet drugs, -blockers, calcium channel blockers, diuretics, insulin, metformin/phenformin, sulfonylureas.

Note: Serum calcium was adjusted for serum albumin. Natural log of 24-hr urinary albumin:creatinine ratio was used. Other race denotes Hispanic, Asian, and/or Pacific Islander, or those classed simply as 'other'.