Korean Journal of Nephrology 1991;10(1):1-7.
만성 신부전증 진행 예측 지표 ( Prediction Error ) 에 관한 연구
김교순
Abstract
The progression of chronic renal failure was deter- mined in 61 patients who proceeded to end stage failure. Derivatives of Scr, including Scr, 1/Scr and log Scr, were defined for the period Scr was between 1.5 and 5 mg/dl. Regression equations were used to predict the time, in months(mos), to Scr =>10 mg/dl. The prediction error(PE) was determined as the predicted time minus the actual time for each Scr transformation. The follow- ing results were obtained. 1) The PE for 1/Scr was lower than the PE for either log Scr or Scr (median: 2.1, 15,7, 47.0 mos respectively; p < 0. 001). 2) The PE for 1/Scr of male and female patients was 2.1 and 1.2 mos respectively. 3) The PE for 1/Scr of less than 45 years, 45-65 years, more than 65 years was 4.5, 1.2 and 1.4 mos respectively. 4) The PE for 1/Scr according to underlying disease was chronic glomerulonephritis 5.5 mos, diabetic ne- phropathy 1.2 mos and hypertensive nephrosclerosis -0. 2 mos, respectively. 5) The PE for 1/Scr in the patients with a mean diastolic blood pressure of more than 90 mmHg and less than 90 mmHg was 4.5 and 1.2 mos. Several factors such as sex, age, underlying diseases and presence or absence of diastolic hypertension had no significant effect on the PE. The logarithmic transfor- mation tended to predict a slower progression of chronic renal failure than actually occurred. The Scr without transformation was not adequate for prediction because of large PE. In conclusion, reciprocal Scr appeared to be a useful model to predict the rate of progression of chronic renal failure, especially in the early stage of renal failure.
TOOLS
METRICS Graph View
  • 366 View
  • 11 Download
Related articles


ABOUT
BROWSE ARTICLES
EDITORIAL POLICY
FOR CONTRIBUTORS
Editorial Office
#301, (Miseung Bldg.) 23, Apgujenog-ro 30-gil, Gangnam-gu, Seoul 06022, Korea
Tel: +82-2-3486-8736    Fax: +82-2-3486-8737    E-mail: registry@ksn.or.kr                

Copyright © 2024 by The Korean Society of Nephrology.

Developed in M2PI

Close layer