Abud ASN 2008-2
0 likes762 views
This study aimed to identify predictors of rapid progression of chronic kidney disease (CKD) using data from a large CKD clinic. The study found that lower initial systolic blood pressure, advanced age at the initial clinic visit, and longer duration of treatment at the clinic were associated with slower CKD progression. However, the study results contradicted previous findings that factors like proteinuria, ACE inhibitor use, diabetes, and initial anemia predict progression. The authors believe this may be due to survivor bias, as patients who died or started dialysis were not captured, and plan to re-examine the data to avoid this error.
Abud ASN 2008-2
- 1. Cohort Characteristics Predictors of Progressive Versus Stable Chronic Kidney Disease Gary G. Abud, Jr., Joel M. Topf, MD, Robert Provenzano, MD, FACP • St. John Hospital & Medical Center, Detroit, MI Introduction In previous studies of our CKD population, we have found a large variation in the change in GFR over time among different patients. This is consistent with data showing that only a minority of patients with chronic kidney disease progress to end-stage renal disease. Despite this, much of the focus of CKD care is on preparing for dialysis. Physicians need reliable tools to predict which patients will and will not progress to dialysis. Purpose Determine factors which predict accelerated rates of CKD progression. Methods We used a database from a large, multi-center CKD clinic to identify predictors of rapid progression of CKD. All non-transplanted CKD patients in the clinic for at least one year and with at least four visits were enrolled in the study. Each subject had a rate of progression calculated by comparing the averages of the two most recent GFRs to the first two GFRs. Patients were divided into quartiles based on the rate of progression and we compared patients in the 1st quartile (fastest progression, loss of an average of 5 ml/min/yr) to the 4th quartile (slowest rate of progression, gain of an average 1.9 mL/min/yr). Multivariate logistic regression was performed to determine risk factors for rapid progression. Discussion Our data does not agree with much of the dogma on chronic kidney disease: • Proteinuria was protective rather than harmful • Use of ACEi and ARB was not protective • Diabetes was not harmful • Anemia on the first visit not predictive of progression This puts our methodology into question. We believe the primary error in our study can be demonstrated in the finding that with multivariate analysis, patients who had higher clinic vintage had significantly more stable renal function over time. This may reflect a survivor bias, since patients who do not return to the CKD clinic (due to death or dialysis) are not captured as progressing. We are re- examining the cohort including the primary end-points of death and dialysis to avoid this error. Despite the primary weaknesses in the study design there are two important findings that can guide physicians to the future progression of renal failure: 1. Lower initial blood pressures predicted slower pro- gression of renal disease. The SBP which was associated with no GFR loss was 130.9 mmHg. 2. Age was a reliable factor that could be used as a predictor of which patients are likely to progress to ESRD. Age is one of four variables used in the stan- dard MDRD equation. The finding that patients with advanced age were less likely to have progressive CKD may indicate that decreased GFR due to ad- vanced age may not carry the same prognosis as decreased GFR due to increased creatinine. Conclusion Many variables, e.g. race, DM and presence of proteinuria, thought to be associated with rapid progression of CKD were not helpful in predicting rapid progression of CKD. However, other symptoms of renal failure, such as high BP and anemia, were predictive of rapid progression of CKD. Stage 1 7% Stage 2 13% Stage 3 43% Stage 4 30% Stage 5 7% Single Variable Analysis Association to CKD Significance Urine Protein : Urine Creatinine (PCR) < 0.3 Faster Progression p = 0.009 Greater Clinic Vintage Slower Progression p = 0.01 Development of Anemia 2° CKD Faster Progression p < 0.0005 Multivariate Analysis Association to CKD Significance Lower Systolic Blood Pressure on 1st Clinic Visit Slower Progression p < 0.0005 Advanced Age on Initial Clinic Visit Slower Progression p = 0.009 Longer Duration of Treatment in Clinic Slower Progression p = 0.016 (100.00) (80.00) (60.00) (40.00) (20.00) 0.00 20.00 40.00 60.00 80.00 100.00 100 120 140 160 180 200 220 240 ∆GFR(mL/min/yr) Average of First Two Systolic Blood Pressures (mmHg) Results Factors not associated with rate of progression included: Age Diabetes Race Use of ACEi at 1st Visit Anemia at 1st Visit Initial GFR Sex Use of Vitamin D Change in Weight LDL Total Cholesterol Weight Patients in Cohort 712 Female 69% African American 31% Diabetic 43% Average Age 62.4 ± 19.6 Age Range 18 - 95 Median Clinic Vintage 371 days Quartile ∆GFR n 1st -5mL/min/yr 176 4th 1.9mL/min/yr 180