Sun et al reported a model and nomogram for predicting the risk of end-stage renal disease (ESRD) for a patient with diabetic nephropathy. The authors are from Zhengzhou University in China.
Patient selection: diabetic with biopsy-confirmed nephropathy
Parameters:
(1) serum cystatin C in mg/L, from 0 to 7.5
(2) eGFR in mL per min per 1.73 square meters
(3) serum BNP in 10^3pg/mL
(4) log10 (urine albumin to creatinine ratio in mg/g), or LOG10(ACR)
(5) pathologic grade, from 1 to 5 (maps to Class I, Class IIa, Class IIb, III or IV)
points for cystatin C =
= (14.2857 * (cystatin C)) - 7.14286
points for eGFR =
= 41.5 - (0.31923 * (GFR))
points for serum BNP =
= 6.3 - (0.063 * (BNP in 10^3))
points for log10ACR =
= (5.8 * (LOG10ACR)) - 2.32
points for pathologic grade =
= (4.625 * (grade)) - 4.625
total score =
= SUM(points for all of the parameters)
Interpretation:
• minimum score: 0
• maximum score: 183.7
value of X =
= (0.1018 * (score)) - 7.185
probability of ESRD =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is 0.86 in the derivation and 0.87 in the validation cohorts.
