Systematic review on urine albumin testing for early detection of diabetic complications

OBJECTIVES: To determine whether microalbuminuria is an independent prognostic factor for the development of diabetic complications and whether improved glycaemic or blood pressure control has a greater influence on the development of diabetic complications in those with microalbuminuria than in those with normoalbuminuria. DATA SOURCES: Electronic databases up until January 2002. REVIEW METHODS: A protocol for peer review by an external expert panel was prepared that included selection criteria for data extraction and required two independent reviewers to undertake article selection and review. Completeness was assessed using hand-searching of major journals. Random effects meta-analysis was used to obtain combined estimates of relative risk (RR). Funnel plots, trim and fill methods and meta-regression were used to assess publication bias and sources of heterogeneity. RESULTS: In patients with type 1 or type 2 DM and microalbuminuria there is a RR of all-cause mortality of 1.8 95% confidence interval (CI) 1.5 to 2.1 and 1.9 (95% CI 1.7 to 2.1) respectively. Similar RRs were found for other mortality end-points, with age of cohort being inversely related to the RR in type 2 DM. In patients with type 1 DM, there is evidence that microalbuminuria or raised albumin excretion rate has only weak, if any, independent prognostic significance for the incidence of retinopathy and no evidence that it predicts progression of retinopathy, although strong evidence exists for the independent prognostic significance of microalbuminuria or raised albumin excretion rate for the development of proliferative retinopathy (crude RR of 4.1, 95% CI 1.8 to 9.4). For type 2 DM, there is no evidence of any independent prognostic significance for the incidence of retinopathy and little, if any, prognostic relationship between microalbuminuria and the progression of retinopathy or development of proliferative retinopathy. In patients with type 1 DM and microalbuminuria there is an RR of developing end-stage renal disease (ESRD) of 4.8 (95% CI 3.0 to 7.5) and a higher RR (7.5, 95% CI 5.4 to 10.5) of developing clinical proteinuria, with a significantly greater fall in glomerular filtration rate (GFR) in patients with microalbuminuria. In patients with type 2 DM, similar RRs were observed: 3.6 (95% CI 1.6 to 8.4) for developing ESRD and 7.5 (95% CI 5.2 to 10.9) for developing clinical proteinuria, with a significantly greater decline in GFR in the microalbuminuria group of 1.7 (95% CI 0.1 to 3.2) ml per minute per year compared with those who were normoalbuminuric. In adults with type 1 or type 2 DM and microalbuminuria at baseline, the numbers progressing to clinical proteinuria (19% and 24%, respectively) and those regressing to normoalbuminuria (26% and 18%, respectively) did not differ significantly. In children with type 1 DM, regression (44%) was significantly more frequent than progression (15%). In patients with type 1 or type 2 DM and microalbuminuria, there is scarce evidence as to whether improved glycaemic control has any effect on the incidence of cardiovascular disease (CVD), the incidence or progression of retinopathy, or the development of renal complications. However, among patients not stratified by albuminuria, improved glycaemic control benefits retinal and renal complications and may benefit CVD. In the effects of angiotensin-converting enzyme (ACE) inhibitors on GFR in normotensive microalbuminuric patients with type 1 DM, there was no evidence of a consistent treatment effect. There is strong evidence from 11 trials in normotensive type 1 patients with microalbuminuria of a beneficial effect of ACE inhibitor treatment on the risk of developing clinical proteinuria and on the risk of regression to normoalbuminuria. Patients with type 2 DM and microalbuminuria, whether hypertensive or not, may obtain additional cardiovascular benefit from an ACE inhibitor and there may be a beneficial effect on the development of retinopathy in normotensive patients irrespective of albuminuria. There is limited evidence that treatment of hypertensive microalbuminuric type 2 diabetic patients with blockers of the renin--angiotensin system is associated with preserved GFR, but also evidence of no differences in GFR in comparisons with other antihypertensive agents. The data on GFR in normotensive cohorts are inconclusive. In normotensive type 2 patients with microalbuminuria there is evidence from three trials (all enalapril) of a reduction in risk of developing clinical proteinuria; in hypertensive patients there is evidence from one placebo-controlled trial (irbesartan) of a reduction in this risk. Intensive compared with moderate blood pressure control did not affect the rate of progression of microalbuminuria to clinical proteinuria in the one available study. There is inconclusive evidence from four trials of any difference in the proportions of hypertensive patients progressing from microalbuminuria to clinical proteinuria when ACE inhibitors are compared with other antihypertensive agents, and in one trial regression was two-fold higher with lisinopril than with nifedipine. CONCLUSIONS: The most pronounced benefits of glycaemic control identified in this review are on retinal and renal complications in both normoalbuminuric and microalbuminuric patients considered together, with little or no evidence of any greater benefit in those with microalbuminuria. Hence, microalbuminuric status may be a false boundary when considering the benefits of glycaemic control. Classification of a person as normoalbuminuric must not serve to suggest that they will derive less benefit from optimal glycaemic control than a person who is microalbuminuric. All hypertensive patients benefit from blood pressure lowering and there is little evidence of additional benefit in those with microalbuminuria. Antihypertensive therapy with an ACE inhibitor in normotensive patients with microalbuminuria is beneficial. Monitoring microalbuminuria does not have a proven role in modulating antihypertensive therapy while the patient remains hypertensive. Recommendations for microalbuminuria research include: determining rate and predictors of development and factors involved in regression; carrying out economic evaluations of different screening strategies; investigating the effects of screening on patients; standardising screening tests to enable use of common reference ranges; evaluating the effects of lipid-lowering therapy; and using to modulate antihypertensive therapy.