Abstract: FR-PO827
An Inverse Association of Proteinuria with Mortality in Incident Hemodialysis Patients
Session Information
- Dialysis: Hospitalization and Mortality
October 26, 2018 | Location: Exhibit Hall, San Diego Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 701 Dialysis: Hemodialysis and Frequent Dialysis
Authors
- Hishida, Manabu, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, United States
- Shafi, Tariq, Johns Hopkins University School of Medicine, Baltimore, Maryland, United States
- Appel, Lawrence J., Johns Hopkins Medical Institutions, Baltimore, Maryland, United States
- Inaguma, Daijo, Fujita Health University, Nagoya, Aichi, Japan
- Matsushita, Kunihiro, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, United States
Background
Proteinuria is a potent predictor of mortality. However, in patients with severely reduced kidney function, a few studies from the CKD Prognosis Consortium and Veterans Affairs showed a J-shaped association between proteinuria and mortality. To our knowledge, no studies have explored this association in incident dialysis patients.
Methods
We examined data from 1380 Japanese incident dialysis patients (mean age 67 years). Baseline data were collected just before or during the hospitalization at which dialysis was initiated. The associations of dipstick proteinuria (negative/trace, 1+, 2+ and ≥3+) with all-cause mortality and cardiovascular disease (CVD) mortality were quantified in Cox models after accounting for potential confounders such as age, cause of CKD, and history of CVD.
Results
Proteinuria ≥3+ was the most prevalent category (n=765 [55.4%]), followed by 2+ (430 [31.2%]), 1+ (137 [9.9%]), and negative/trace (48 [3.5%]). Patients with lower proteinuria were likely to be older and have a history of CVD, compared to the patients with higher proteinuria. During a mean follow-up of 3.3 years, there were 352 deaths (129 due to CVD). Patients with lower proteinuria had a higher risk of mortality, with unadjusted hazard ratios 3.86 (2.55-5.85) for negative/trace, 1.62 (1.16-2.26) for 1+, and 1.35 (1.07-1.71) for 2+ (Model 1 in Table). Although attenuated, this pattern remained significant after accounting for potential confounders (p for trend ≤0.001 and adjusted hazard ratio 2.60 [1.62-4.17] for negative/trace in Model 4 in Table). This association was consistent for CVD and no-CVD mortality, even when restricting to adults aged ≥70 years, or further adjusting for cardiac ejection fraction.
Conclusion
We documented a progressive inverse trend of proteinuria with mortality among incident dialysis patients, with a strikingly high risk of mortality in persons with negative/trace proteinuria. Our study highlights the prognostic value of predialysis data and suggests that absence of proteinuria as a potential indicator of the highest postdialysis mortality risk.
Funding
- Private Foundation Support