Measures of Kidney Function as CV Risk Factors in Older Adults

At present, classification of chronic kidney disease (CKD) is based on the associations of decreased estimated glomerular filtration rate (eGFR) and/or increased urinary albumin-creatinine ratio (UACR) with outcomes such as kidney failure, cardiovascular disease, and death. However, according to Andreas Kühn, MD, and colleagues in Germany, the strength of those associations in older patients is unclear.

The researchers conducted a population-based cohort study to examine (1) the association of kidney function with cardiovascular outcomes and mortality and (2) the predictive abilities of seven eGFR equations and UACR for those outcomes within the Berlin Initiative Study (BIS), a prospective cohort of older adults. Results were reported in the American Journal of Kidney Diseases [2021;77(3):386-396].

The outcomes of interest were stroke, myocardial infarction (MI), all-cause mortality, and any of the first of those events. Hazard ratios (HRs) and 95% confidence intervals (CIs) were derived from multivariable-adjusted Cox proportional hazards models for association analyses. The category-free net reclassification improvement (NRI) for all four outcomes at 5 years was used to compare discriminative abilities beween models.

The analyses were divided into two parts: (1) kidney measures as risk factors, i.e., the association of eGFR and/or UACR with cardiovascular outcomes and mortality; and (2) kidney measures as predictors, i.e., the predictive abilities of eGFR (different equations) and/or UACR for all outcomes. For part 1, GFR was estimated using the creatinine- and cystatin C-based BIS equation (eGFRBIS2(cr-cys)), developed for individuals >70 years of age. Measured GFR (mGFR) was used for sensitivity analysis (n=436).

The BIS is a study focusing on kidney function in adults ≥70 years of age in Berlin, Germany (n=2069). The current analysis included individuals with serum creatinine, cystatin C, and UACR values available at baseline and excluded those with a previous stroke or MI by either self-report or insurance claims data. A total of 1581 participants were eligible. Of those, 904 were female and mean age was 79.7 years.

Of the 1581 eligible participants, 57.1% had eGFR <60 mL/min/1.73 m2, UACR ≥30 mg/g was detected in 17.7% of those with eGFR 60 mL/min/1.73 m2 and 28.9% of those with eGFR <60 mL/min/1.73 m2.

During a median follow-up of 8.2 years, there were 193 first strokes, 125 first MIs, and 531 deaths, resulting in 683 participants with 849 events in total. Depending on the outcome, there was a median of 8.1 to 8.3 years of person-time.

For stroke, the lowest event-free survival was seen for low eGFR and any UACR (~80% after 9 years). The combination of decreased eGFR and increased UACR showed the lowest event-free survival (~85%), as well as worst overall survival, with ~70% mortality after 9 years.

When eGFR was categorized with a cutoff of 60 mL/min/1.73 m2, the lower category was associated with HRs of 2.18 (95% CI, 1.52-3.13) for stroke, 1.15 (95% CI, 0.75-1.76) for MI, 1.26 (95% CI, 1.00-1.57) for mortality, and 1.32 (95% CI, 1.09-1.61) for any event. When eGFR was categorized as ≥60, 45 to 59, or <45 mL/min/1.73 m2, HRs for stroke were increased in the lower two categories. For the outcomes of mortality and MI, the only statistically significant findings was the association of  eGFR <45 mL/min/1.73 m2 with mortality (HR, 1.57; 95% CI, 1.20-2.06).

There were significant associations between albuminuria with UACR of 30 to 300 mg/g and MI, (HR, 1.65; 95% CI, 1.09-2.51) and mortality (HR, 1.63; 95% CI, 1.34-1.98), and any event (HR, 1.45; 95% CI, 1.21-1.73); there was no significant association with stroke (HR, 0.91; 95% CI, 0.63-1.33). There were associations between albuminuria with UACR >300 mg/g and increased HRs for all-cause mortality and any event. The associations between albuminuria with UACR >300 mg/g and stroke and MI were less and non-significantly increased, due possibly to the limited event rate in this category.

In the 436 individuals with available mGFR, after adjusting for age and sex, the stroke risk for eGFR 45 to 59 mL/min/1.73 m2 was independent of UACR, in contrast to the other outcomes.

In part 2 (predictive abilities of kidney measures), the predictive discrimination abilities of seven different eGFR equations, category-free NRI (asymmetric 92% CI) was estimated when adding the two dichotomized kidney measures to two different models; either eGFR and/or UACR to the basic model or eGFR to the basic UACR model. There was significant positive NRI for eGFR calculated using the cystatin C-based Chronic Kidney Disease Epidemiology Collaboration equation and the eGFRBIS2(cr-cys) and Full Age Spectrum equations. UACR demonstrated significant positive NRIs for MI and mortality.

Limitations to the study cited by the authors were primarily attributable to the age of the participants, multimorbidity and increased mortality may had led to bias, eGFR and UACR categorization and blood pressure measurements were based on single assessments at baseline, and the lack of data on causes of death.

In summary, the researchers said, “Our results suggest that CKD G3a with UACR <30 mg/g is strongly associated with risk for stroke but not for MI and all-cause mortality in our elderly BIS population. In contrast, UACR of 30 to 300 mg/g did not seem to be associated with stroke risk but was strongly associated with MI and all-cause mortality. This may suggest a different pathophysiologic link between CKD and stroke risk. Thus, our study supports interpreting eGFR <60 mL/min/1.73 m2 as a risk factor for cardiovascular events in older adults, which has been an ongoing debate, even if the association was found to be weaker compared with younger persons. Furthermore, eGFR based on cystatin C level improved risk predictions of stroke in our cohort of adults older than 70 years, confirming the prognostic benefit of cystatin C level in old age.”

Takeaway Points

  1. In the general population, estimated glomerular filtration rate (eGFR) and urinary albumin-creatinine ratio (UACR) are associated with risk of cardiovascular events; the strength of that association in older adults is unclear.
  2. Researchers conducted a population-based cohort study to examine associations between eGFR and UACR and stroke, myocardial infarction (MI), and all-cause mortality.
  3. There was an association between eGFR 45 to 59 mL/min/1.73 m2 and stroke but not MI or all-cause mortality; in contrast, UACR of 30 to 300 mg/g was associated with MI and all-cause death but not stroke.