Renal disease progression in autosomal dominant polycystic kidney disease

Abstract

Background: Autosomal dominant polycystic kidney disease is a lifelong progressive disorder. However, how age, blood pressure, and stage of chronic kidney disease (CKD) affect the rate of kidney function deterioration is not clearly understood.

Methods: In this long-term observational case study up to 13.9 years (median observation period for slope was 3.3 years), serum creatinine was serially measured in 255 mostly adult patients. The glomerular filtration rate was estimated (eGFR) using a modified Modification of Diet in Renal Disease Study method. The total kidney volume (TKV) has been measured in 86 patients at one center since 2006.

Results: As age increased, eGFR declined significantly (P < 0.0001), but the annual rate of decline of eGFR did not correlate with age or initially measured eGFR. In patients with CKD stage 1, eGFR declined at a rate which was not significantly different from other advanced CKD stages. Hypertensive patients had lower eGFR and larger TKV than normotensive patients at a young adult age. The slopes of regression lines of eGFR and TKV in relation to age were not different between high and normal blood pressure groups.

Conclusion: The declining rate of eGFR was relatively constant and did not correlate with age or eGFR after adolescence. eGFR was already low in young adult patients with hypertension. As age increased after adolescence, eGFR declined and TKV increased similarly between normal and high blood pressure groups. eGFR starts to decline in patients with normal eGFR, suggesting that the decline starts earlier than previously thought.

Fig. 1

Initially measured eGFR distribution in relation to age (n = 255). y = −1.757x + 141.28, r = −0.6871, P < 0.0001

Fig. 2

Relationship of eGFR slope to age (a) and initial eGFR (b) (n = 196). a Spearman’s rank correlation coefficient (r) = 0.0728, P = 0.3094. b Spearman’s rank correlation coefficient (r) = −0.0412, P = 0.5654. No significant relationship is seen between eGFR slope and age, or between eGFR slope and initially measured eGFR. Mean observation time of eGFR was 4.2 ± 3.0 years

Fig. 3

1/Cr is plotted against age in 106 patients who had been followed up for more than 3 years. 1/Cr declines in most patients at an individually variable rate. Pattern of decline appears not to be age-dependent

Fig. 4

eGFR changes in patients followed for more than 5 years (n = 36). In 5 patients shown by a red line, the declining curve changed from moderate to rapid during follow up. The change points varied in relation to age or eGFR level. Other patients are shown in blue for easy identification

Fig. 5

a Initially measured eGFRs are plotted against age in normotensive (blue) and hypertensive (red) patients. Regression analysis for normal blood pressure group: y = 151.08 − 1.546x (where y = eGFR and x = age, r = −0.7791, P < 0.0001, n = 70) and that for hypertensive group: y = 132.30 − 1.666x (r = −0.6587, P < 0.0001, n = 158). b The relationship between age and log-transformed TKV in normotensive (blue) and hypertensive (red) patients. Regression analysis for normal blood pressure group; y = 2.7003 + 0.006275x (where y = log TKV and x = age, r = 0.57859, P = 0.0075, n = 20) and that for hypertensive group; y = 3.0339 + 0.004452x (r = 0.23144, P = 0.0615, n = 66). In both a and b, the regression lines for normotensive and hypertensive patients were not considered to be identical, with different y-intercepts, since there was a significant difference (P < 0.01, F test) in the y-intercept of the two regression lines under the null hypothesis that the y-intercept of two lines was identical. There was no significant difference (P = 0.6061 in a or P = 0.6079 in b, F test) in the slope of the two lines under the null hypothesis that the slope of the two lines was identical