OBJECTIVE - To investigate the predictive value of microalbuminuria an
d the annual increase of albumin excretion as risk factors for diabeti
c nephropathy. RESEARCH DESIGN AND METHODS - A 5-year follow-up of pat
ients with microalbuminuria (urinary albumin excretion [UAE] = 30-299
mg/24 h) and matched patients with normoalbuminuria (UAE < 30 mg/24 h)
. The initial classification was based on one single 24-h urine collec
tion. The annual increase in UAE was calculated by linear regression a
nalysis of log-transformed UAE on time. This study was conducted at th
e outpatient clinic of the Steno Diabetes Center. The study subjects i
ncluded 118 insulin-dependent diabetes mellitus (IDDM) patients betwee
n 18 and 50 years of age with microalbuminuria and 112 matched control
patients with normal UAE with an age at diabetes onset of <31 years.
The main outcome measures were UAE, annual change in UAE rate (percent
age per year), and the prevalence of retinopathy. RESULTS - After 5 ye
ars, 39 (33%, 24-42 CI (95% confidence interval]) patients with microa
lbuminuria had normoalbuminuria, 57 (48%, 38-57 CI) still had microalb
uminuria, and 22 (19%, 12-27 CI) had developed diabetic nephropathy. A
mong the 112 patients with normoalbuminuria in 1985, 9 (8%, 4-15 CI) h
ad developed microalbuminuria, and 2 (2%, 0-6 CI) had developed diabet
ic nephropathy. Of the 79 patients with persistent albuminuria, only 3
6 (46%, 34-57 CD were progressors with a rate of progression of >5%/ye
ar. Progressors had significantly higher HbA1c, higher mean blood pres
sure, and a higher incidence of proliferative retinopathy compared wit
h nonprogressors. Multiple regression analysis only identified mean Hb
A1c as an independent predictor of the rate of progression. Smoking wa
s significantly more prevalent in patients with persistent albuminuria
. CONCLUSIONS - Microalbuminuria is a predictor of progression to diab
etic nephropathy; however, not as strong as suggested previously. Calc
ulation of the annual increase in UAE seems to be a more specific meth
od of identifying patients who will develop diabetic nephropathy.