Long-term changes in albuminuria: underlying causes and future mortality risk in a 20-year prospective cohort: the Nord-Trøndelag Health (HUNT) Study.

OBJECTIVE

Knowledge on how changing risk factors influence the progression of albuminuria over time is still limited. Furthermore, large population-based cohorts are needed to study the association between albuminuria change and mortality risk in nondiabetic study participants.

METHODS

We evaluated changes of albuminuria in 6282 nondiabetic individuals from the Norwegian population-based Nord-Trøndelag Health study. Using three albumin/creatinine ratios (ACR), we studied the influence of cardiovascular risk factors on ACR change from baseline to follow-up 11 years later. We evaluated the next 8-year mortality risk by using flexible parametric methods to identify nonlinear main effects and their two-way interactions.

RESULTS

Mean albuminuria increased significantly over 11 years (1.82-3.02 mg/mmol, P < 0.0001), but two-thirds of individuals had stable levels (ΔACR -1.40 to 1.40 mg/mmol). Higher age, ACR, and SBP as well as smoking and lower glomerular filtration rate at baseline were associated with increasing albuminuria. Study participants in the upper quartile of the increasing group had mean adjusted hazard ratio 1.31 (P = 0.004) for all-cause mortality compared with those with stable ACR. Those with decreasing ACR also had increased mortality, but the risk was strongly attenuated when adjusting for comorbidity. It also decreased the first 3 years before increasing. There was a strong interaction between baseline ACR and ΔACR. Increasing albuminuria had strongest effect on mortality in study participants with moderately increased baseline values.

CONCLUSION

Both increasing and decreasing albuminuria are significant independent predictors of mortality in nondiabetic individuals, but must be interpreted in light of baseline values. Cutoffs and clinical usefulness in nondiabetic study participants should be further investigated.