Yu et al reported a model for predicting acute kidney injury in a patient undergoing partial hepatectomy. This can help to identify a patient who may require more aggressive management of renal function. The authors are from the Medical School of Chinese PLA, the First Medical Center of Chinese PLA General Hospital, and Nankai University in China.
Patient selection: age >= 65 years, status post partial hepatectomy
Parameters:
(1) age in years from 64 to 84 years
(2) chronic kidney disease
(3) NSAID use
(4) hepatic inflow occlusion
(5) estimated blood loss, from 0 to 3,000 mL
(6) blood transfusion
points for age =
= (3.29 * (age)) - 210.56
points for blood loss =
= 0.0333 * (loss in mL)
|
Parameter
|
Finding
|
Points
|
|
chronic kidney disease
|
no
|
0
|
|
|
yes
|
48.08
|
|
NSAID use
|
no
|
0
|
|
|
yes
|
65.83
|
|
hepatic inflow occlusion
|
no
|
0
|
|
|
yes
|
25.19
|
|
blood transfusion
|
no
|
0
|
|
|
yes
|
41.21
|
total score =
= SUM(points for all of the parameters)
Interpretation:
• minimum score: 0
• maximum score: 346.13
value of X =
= (0.01926 * (score)) - 3.994
probability of acute kidney injury =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is 0.73 in the training and 0.71 in the validation cohorts.
