Short stature and diabetic nephropathyBMJ 1995; 310 doi: http://dx.doi.org/10.1136/bmj.310.6975.296 (Published 04 February 1995) Cite this as: BMJ 1995;310:296
- P Rossing, research fellowa,
- L Tarnow, research fellowa,
- F S Nielsen, research fellowa,
- S Boelskifte, statisticianb,
- B M Brenner, Samuel A Levine professor of medicinec,
- H-H Parving, chief physiciana
- a Steno Diabetes Center, DK-2820 Gentofte, Denmark
- b UNI-C, Danish Computing Centre for Research and Education, DK-8200 Aarhus N, Denmark
- c Harvard Medical School, Brigham and Women's Hospital, Boston, Massachusetts 02115, USA
- Correspondence to: Dr Rossing.
- Accepted 16 December 1994
Recent studies have suggested that low birth weight is associated with a reduced number of nephrons and hypertension in later life,1 2 both well known risk factors for renal disease. An inverse correlation between microalbuminuria and height, the latter known to vary directly with birth weight,3 has been shown in non-diabetic men.4 We investigated the association between adult height and diabetic nephropathy in a cross sectional study of a cohort of insulin dependent diabetic patients attending the Steno Diabetes Center in 1984.5
Patients, methods, and results
We selected patients according to the following criteria: age >/=18 years, a duration of diabetes >/=5 years, and age at onset of diabetes </=40 years. In all, 951 patients (500 men; mean age 40 (SD 13) years; mean age at onset 18 (10) years) were enrolled (97% of those eligible). We examined urinary albumin excretion rate (24 hour urine collections, radioimmunoassay) and height (to the nearest centimetre and given as integers). On the basis of urinary albumin excretion rate the patients were stratified into three groups: those with normal rate of excretion (normoalbuminuria) (</=30 mg/24 h, n=563); with microalbuminuria (>30 and <300 mg/24 h, n=214); and with macroalbuminuria (indicating diabetic nephropathy) (>/=300 mg/24 h in two out of three consecutive samples, n=174). Men and women were analysed separately owing to differences in height. Results were analysed with the programs of the SAS Institute (Cary, North Carolina).
Men with macroalbuminuria were significantly shorter (mean height 175.3 (SD 7.3) cm, n=98) than patients with microalbuminuria (177.0 (6.1) cm, n=113) or with normoalbuminuria (177.9 (6.9) cm, n=289) (P=0.004). As height and age at onset of diabetes and current age were associated a multiple logistic regression analysis was performed with patient group (ordered categorical variable normoalbuminuria, microalbuminuria, macroalbuminuria) as dependent variable and height, age, and age at onset of diabetes as independent variables. Height was included in the final model, with a reduction in the risk of developing diabetic nephropathy of 3.3% (95% confidence interval 0.77% to 5.8%) per centimetre increase in height.
Poor glycaemic control is a risk factor both for the development of incipient diabetic nephropathy (indicated by microalbuminuria) and overt diabetic nephropathy and for growth retardation during childhood. To adjust for this confounding effect on height we evaluated patients with onset of diabetes after the age of 19 (n=387); the difference in height between the patients with normoalbuminuria and those with microalbuminuria and macroalbuminuria was still significant (P=0.03, with the t test) (table). In women no significant relation existed between height and albuminuria group, but the trend was the same as in men (mean height 163.6 (SD 7.1) cm, n=76; 165.4 (7.0) cm, n=101; and 165.0 (6.2) cm, n=274) for patients with macroalbuminuria, microalbuminuria, and normoalbuminuria respectively (P=0.18).
Our data support the hypothesis that short stature is related to development of diabetic nephropathy in men. The difference between the sexes was also shown in a study of non-diabetic subjects.4 Moreover, poor glycaemic control in the patients with onset of diabetes in early childhood leading to impaired growth and increased risk for developing nephropathy was not the sole explanation of the relation between height and nephropathy as this relation was also shown in patients with onset of diabetes after the age of 19.
Factors that lead to impaired intrauterine somatic growth also give rise to an inborn reduction in the number of nephrons both in animal models and in humans.1 Fewer nephrons results in a diminished glomerular filtration surface area, leading to systemic and glomerular capillary hypertension. Furthermore, evidence from experimental studies of rats born to mothers exposed either to a restriction of dietary protein or to gentamicin during pregnancy suggests that fewer nephrons at birth initiates a postnatal process of progressive glomerular injury.1 Thus low birth weight related to short statute indicates congenital oligonephropathy and thereby could indicate an increased risk for developing diabetic nephropathy. In conclusion, our study supports the hypothesis that genetic predisposition or factors operating in utero or early childhood, or both, contribute to the development of diabetic nephropathy in men.