Examining Short-Term Within-Person Variability in Clinical Markers of Kidney Function

Chronic kidney disease (CKD) is defined by abnormalities in glomerular filtration rate (GFR) and/or albuminuria; CKD affects approximately 10% of the US population. Endogenous plasma or serum filtration markers, most commonly creatinine, are used to estimate GFR; albuminuria is quantified by measuring urinary albumin concentration (UAC) or urinary albumin-creatinine ratio (UACR). There are well established reference ranges for estimated GFR (eGFR) and albuminuria in healthy individuals.

For patients with CKD, determining the clinical significance of a change in eGFR or albuminuria requires knowledge of the expected variability in the absence of underlying clinical changes. However, there are relatively few data on the inherent biologic variability of eGFR and UACR in the setting of CKD.

Sushrut S. Waikar, MD, MPH, and colleagues, on behalf of the Chronic Kidney Disease Biomarkers Consortium Investigators, recently conducted a cross-sectional study to provide estimates of the short-term within-person biologic variability in measures of kidney function, including albuminuria (UAC and UACR) and plasma eGFR markers (creatinine and cystatin C, ß₂-microglobulin [B2M], and beta trace protein [BTP]). Study results were reported in the American Journal of Kidney Diseases [2018;72(4):538-546].

Study participants were clinically stable outpatients with CKD (n=50) attending a nephrology subspecialty practice at Brigham and Women’s Hospital, Boston, Massachusetts. Patients were recruited as a convenience sample during a visit to the nephrology clinic between February 2011 and February 2014.

The study exposure was repeat measurements from serially collected samples across three study visits. Urine and blood samples were collected shortly following a clinic visit (8 am to 5 pm) and then at two requested follow-up study visits within an approximate 2-week period (range, 1-4 weeks). At each of the two subsequent visits, participants were asked to bring a refrigerated first-void morning urine sample and then to provide a fresh urine sample during the study visit. Up to three blood specimens and up to five urine samples were collected.

Outcomes of interest were measurements of urine albumin concentration (UAC), urinary albumin-creatinine ratio (IUACR), and plasma creatinine, cystatin C, B2M, and BTP. Using log-transformed measurements, the researchers calculated within-person coefficients of variation (CV_w) values and corresponding reference change positive and negative (RCV_pos and RCV_neg) values.

The 50 participants provided a total of 139 plasma samples and 227 urine samples during the 4-week study period. Mean age of the cohort was 56.8 years, 44% were female, 32% were African American, and 32% had diabetes mellitus. At the time of study enrollment, 72% (n=36) were taking an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and 42% were taking a diuretic. Values from samples obtained at the first timepoint were median estimated glomerular filtration rate (eGFR) using the CKD-Epidemiology Collaboration creatinine equation, 33 mL/min/1.73 m²; and median random spot UACR, 143 mg/g.

Median CV_w (RCV_pos:RCV_neg) values of filtration markers were serum creatinine, 5.4%; cystatin C, 4.1%;  BTP, 7.4%; and B2M, 5.6%.

Results for albuminuria were 33.2% (95% confidence interval [CI], 14.4%-52.0%) for first-morning UAC, 50.6% (95% CI, 39.0%-62.3%) for random spot UAC, 32.5% (95% CI, 23.2%-41.9%) for first-morning UACR, and 29.7% (95% CI, 19.5%-40.0%) for random spot UACR. There were no significant differences in median CV_w values in first-morning versus random-void specimens for UACR or UAC. Variability in random UACRs was sufficient to qualify as a change in albuminuria categorization in only six of 50 participants, all between moderately increased (30-300 mg/g) and severely increased (>300 mg/g) categories.

CV_w values for filtration markers were comparable across the range of baseline eGFRs. CV_w values for UAC and UACR were comparable across the range of baseline albuminuria values. There were no statistically significant differences in CVw values for each of the markers in individuals with versus without a given characteristic (eg, diabetes vs no diabetes, African American vs non-African American, or eGFR or UACR at median or less vs more than median).

The researchers cited some limitations to the study, including the small sample size with numerous causes of CKD, the inability to collect and measure albuminuria in 24-hour samples, not measuring urine total protein, and enrolling participants in a convenience sample from individuals attending a CKD clinic rather than in a research setting where clinical variables could be better controlled.

In conclusion, the researchers said, “Assessment of the clinical significance of changes in kidney disease markers such as eGFR and albuminuria needs to take into account the expected within-person variability of these measurements in CKD. For filtration markers for GFR estimation, within-person variability appears to be low (CV_w <10%), but sufficient to cause potential false-positive designations of acute kidney injury in individuals with more advanced CKD. For albuminuria, within-person variability may be substantial, and increases or decreases of 50% in an individual patient should not be taken as sole evidence of disease progression or remission without repeat testing and consideration of the clinical context.”

Takeaway Points

1. A cross-sectional study in clinically stable patients with chronic kidney disease (CKD) sought to provide estimates of the short-term within-person biologic variability in measures of kidney function and plasma estimated glomerular filtration rate (eGFR) markers.
2. Markers of eGFR appear to have relatively low short-term within-person variability.
3. Albuminuria appears to have relatively high short-term within-person variability, making it difficult to distinguish random variability form meaningful biologic changes.