Welch Center Uniting Medicine Public Health Proteinuria as

Welch Center Uniting Medicine & Public Health Proteinuria as a Surrogate Outcome in Chronic Kidney Disease A workshop co-sponsored by the National Kidney Foundation and U. S. Food and Drug Administration Prevalence of Albuminuria, and its Relationship to Decreased GFR and Outcomes Josef Coresh, MD, Ph. D Director, Cardiovascular Epidemiology Program Welch Center Departments of Epidemiology, Medicine & Biostatistics Johns Hopkins University Disclosures: None

Outline • CKD prevalence – Albuminuria by stage & its persistence • CVD risk in relation to: – Albuminuria and e. GFR • Creatinine (e. GFRMDRD) • Cystatin C (e. GFRCys. C) • Conclusions

Prevalence of CKD: NHANES Surveys Representing the US Adult Population • Study Population: stratified random sample – NHANES III (1988 to 1994): n=15, 488* – NHANES 1999 -2000: n=4, 101* – NHANES 2001 -2002: n=4, 684 – NHANES 2003 -2004: n=4, 448 • Serum creatinine: calibrated to be comparable to the MDRD creatinine assay using frozen serum (Am J Kidney Dis. 2007; 50: 918 -26) • GFR Estimate: MDRD Study 4 -variable equation • Extrapolation to US population: NCHS published survey weights adjusting for age, sex, race and non-response *Coresh et al. JAMA. 2007; 298(17): 2038 -47

Distribution of Albumin to Creatinine Ratio: NHANES 1988 -2004 micro ”normal” Albumin to Creatinine Ration, mg/g macro

Prevalence of Diagnosed Diabetes and Hypertension by Albuminuria: NHANES 1988 -2004 Proportion ACR mg/g Prevalence <5 41% (9, 920) 5 -29 30 -299 50% (13, 000) 7. 1% (2648) 300+ 1. 2% (573) Age, y 42 48 60 56

CKD Prevalence Estimates Require GFR Estimation (e. GFR) • MDRD Study Equation (GFR ml/min/1. 73 m 2) e. GFR = 186 x (SCr)-1. 154 x (age)-0. 203 x (0. 742 if female) x (1. 210 if African American) e. GFR = 175 x (Standardized SCr)-1. 154 x (age)-0. 203 x (0. 742 if female) x (1. 210 if African American) • Cockcroft-Gault (CG ml/min) Ccr = (140 -age) x weight x 0. 85 (if female)/(SCr) BSA Adjusted = CG * 1. 73 / BSA formula • Equations in other populations – Children, Chinese, Japanese, Kidney Donors N Engl J Med. 2006; 354(23): 2473 -83. Clin Chem 2007; 53(4): 766 -72

Estimated GFR Distribution Estimated GFR, ml/min/1. 73 m 2 The conservative trends analysis eliminated the difference in mean GFR between surveys. The vertical line demarcates an estimated GFR of 60 ml/min/1. 73 m 2 which defines decreased GFR.

Relationship of e. GFR to Albuminuria & Hypertension: NHANES III Albuminuria only High BP only Albuminuria & High BP Normal Am J Kidney Dis 2002; 39: (2) S 49

Chronic Kidney Disease (CKD) Definition • Kidney damage for 3+ months as defined by structural of functional abnormalities of the kidney, with or without decreased GFR manifest by either: – Pathological abnormalities, or – Markers of kidney damage including abnormalities in the composition of the blood or urine, or abnormalities in imaging tests • GFR < 60 ml/min/1. 73 m 2 for 3+ months with or without kidney damage Am J Kidney Dis 2002; 39: (2) S 1 -S 266 Ann Intern Med 2003; 139(2): 137 -147

Persistence of Albuminuria – Spot Urine ACR in 2 visits a median of 17 days apart: NHANES III Log(ACR) – Second Visit 10 Macro >300 mg/g 5 Micro 30 -299 mg/g 0 Micro 30 -299 mg/g “Normal” -5 -5 0 Macro >300 mg/g 5 Log(ACR) – First Visit 10

Persistence of Albuminuria: NHANES III Albuminuria at Albuminuria on a Repeat First Visit, mg/g Visit (median 17 days later) Micro, 30 -299* Macro 300+ e. GFR 90+ e. GFR <90 50. 9% (n=57) 100% 75. 0% (n=36) 100% * 53. 9% and 72. 7% for gender specific cutoffs for microalbuminuria 17 -250 mg/g for men and 25 -355 mg/g for women

US Trends in the Prevalence of CKD by Age and Stage: NHANES 1988 -2004 50% Prevalence, % CKD Stage 40% Stage 4 Stage 3 30% Stage 2 Persistent Stage 1 albuminuria >30 mg/g e. GFRMDRD 15 -29 e. GFRMDRD 30 -59 20% 10% 20 -39 40 -59 60 -69 99 -0 4 4 -9 88 4 -0 99 88 Survey years: Age Group: -9 4 0% 70+ Coresh et al. JAMA. 2007; 298: 2038 -2047

Prevalence of Elevated Cystatin C in US Women Proportion with cystatin C >1. 12 mg/L (>1. 12 mg/L = 99 th %ile for young healthy adults) 100% non-Hispanic white Women non-Hispanic black Mexican American 80% (n=7, 596) 60% 40% 20% 0% 10 20 30 40 50 60 70 80 90 Age(years) Kottgen et al. Am J Kidney Dis 2008; 51: 385 -394

Different Outcomes of CKD Outcome Importance for Different Outcomes CKD Stage Type of Kidney Proteinuria Disease (Diagnosis)** Concurrent complications* +++ + + ++ +++ Prognosis (next 10 -years) Risk of CVD or mortality Risk of kidney failure Rate of decline in GFR *Hypertension, anemia, malnutrition, bone disease, neuropathy & decreased quality of life **For example, diabetic, glomerular, vascular, tubulointerstitial, & cystic

Albuminuria and Risk of Cardiovascular Death General Population: Risk Seen at Very Low Levels Hazard Ratio 5. 0 20 mg/day (30 mg/g) 200 mg/day Microalbuminuria (300 mg/g) 4. 0 3. 0 2. 0 1. 0 Urinary Albumin Concentration mg/day PREVEND Study - Hillege HL et al, Circulation 2002; 106: 1777 -82.

Glomerular Filtration Rate, Albuminuria, and Risk of Cardiovascular and All-Cause Mortality in the US Population (Astor et al. Am J Epidemiol 2008, April e. PUB)

Cardiovascular Mortality Models adjusted to incidence rates of a 60 -year-old non-Hispanic White male.

All-Cause Mortality Models adjusted to incidence rates of a 60 -year-old non-Hispanic White male.

Cardiovascular mortality in NHANES III F/U Predicted incidence rates adjusted to the mortality rate of a 60 year‑old, non-Hispanic white male Astor et al. Am J Epidemiol 2008; April

All-Cause mortality in NHANES III F/U Predicted incidence rates adjusted to the mortality rate of a 60 year‑old, non-Hispanic white male Astor et al. Am J Epidemiol 2008; April

Association of Kidney Function and Albuminuria With Cardiovascular Mortality in Older vs Younger Individuals: The HUNT II Study Age & Sex Adjusted IRR ACR – average 3 spot urines • Optimal < median • Men: < 5 • Women: < 7 • High normal • Men: 5 to 19 • Women: 7 to 29 • Microalbuminuria • Men: 20 to 199 • Women: 30 to 299 e. GFR ml/min/1. 73 m 2 5 20 30 ACR, mg/g 50 Hallan et al. Arch Intern Med. 2007; 167(22): 2490 -2496

Cardiovascular Mortality by e. GFR and Albuminuria: HUNT II Study ACR • Optimal < median • Men: < 5 • Women: < 7 • High normal • Men: 5 to 19 • Women: 7 to 29 • Microalbuminuria • Men: 20 to 199 • Women: 30 to 299 *P. 05. †P. 01. ‡P. 001. Albuminuria, mg/g (average of 3 spot urines) Hallan et al. Arch Intern Med. 2007; 167(22): 2490 -2496

3 2. 5 2 1. 5 e. GFRMDRD Age ≤ 65 e. GFRCys. C . 5 1 Age > 65 High e. GFRMDRD low muscle mass (BAD) 0 Adjusted Annual Cardiovascular Mortality (%) Adjusted* Cardiovascular Mortality Risk in NHANES III Mortality Follow-Up Study 30 40 50 60 70 80 90 100 110 120 Estimated GFR (m. L/min/1. 73 m 2) *Adjusted for 13 covariates Astor et al. JASN 2007 abstract

Distorted Associations (Baseline Disease RF CVD) Adjusted* 3 -year all-cause mortality in Dialysis Patients Presence of Inflammation/Malnutrition Overall Absence of Inflammation/Malnutrition *Adjusted to the age of 60 years, female, Whites, HD and non-smokers.

Conclusions • Albuminuria is common in the population – Spot ACR provides a reasonable measure – Cutoffs are somewhat arbitrary: sex dependent cutoffs are more accurate but non-sex dependent cutoff are useful & less complicated – Microalbuminuria varies within an individual – persistence is a useful indicator – Much more common among diabetics; but a substantial proportion of the individuals with microalbuminuria have neither hypertension nor diabetes • Albuminuria and e. GFR are associated but confer independent risk (mortality, CVD mortality)

Thank you! CKD-Epi CHOICE Study CVD-Epi Stein Hallan ARIC Staff