This study sought to identify determinants of albuminuria in people with Ty
pe 2 diabetes. In 335 primary care patients, we assessed albumin creatinine
ratio (ACR) in two 24-h urine samples, and its cross-sectional association
s with protein and alcohol intake, cigarette smoking, body weight and heigh
t, glycosuria, blood pressure, hypoglycaemic and antihypertensive treatment
, gender, age, age at diagnosis. diabetes duration, family history of diabe
tes and cardiovascular diseases, ethnic origin, and education. The prevalen
ce of micro- or macro-albuminuria (ACR greater than or equal to 2.0 mg/mmol
) was 33%. Among these patients, compared to those with normo-albuminuria,
there were more men, protein intake (g/kg) estimated fl om urinary urea as
well as systolic blood pressure and glycosuria were higher, there were more
smokers, men were shorter, and a family history of diabetes was less preva
lent (all P < 0.05). In linear and logistic regression (n = 270) albuminuri
a ia was independently associated (P <less than> 0.05 unless indicated othe
rwise) with systolic blood pressure (OR10 (mmHg) = 1.32). smoking (ORex/nev
er = 2.36, ORcurrent/never = 4.89), glycosuria (ORgreater than or equal to
7t <t g/l = 2.41). gender (ORmen/women = 2.50), age in men (OR10 year = 1.6
0) (P < 0.10) and, inversely, in women (OR10 year = 0.63) (P <greater than>
0.10). On aggregation, the modifiable determinants systolic blood pressure
, smoking and glycosuria explained 12% of the variation in albuminuria. The
se factors thus are, although to a moderate extent only, potential determin
ants of albuminuria. We also observed an independent, inverse association w
ith body height (OR0.10 m = 0.47). This is in line with the hypothesis that
development in utero or during early life influences kidney function in la
ter life. (C) 2001 Elsevier Science Ireland Ltd. All rights reserved.