Weiss et al developed models for predicting mortality in older adults with severe kidney disease. These can help to identify a patient who may benefit from more aggressive management. The authors are from Oregon Health and Sciences University, McGill University, Kaiser Permanente Northwest and Veterans Administration Puget Sound Healthcare.
Patient selection: kidney disease, age >= 80 years, GFR <= 30 mL per min per 1.73 square meter
Parameters:
(1) index GFR in mL per min per 1.73 square meters, from 2 to 30
(2) second GFR in mL per min per 1.73 square meters, from 5 to 30
(3) baseline systolic blood pressure in mm Hg, from 80 to 240
(4) age in years
(5) body mass index (BMI) in kg per square meter, from 10 to 60
(6) gender
(7) history of renal failure
(8) history of cardiovascular disease
(9) tobacco use
(10) history of hypertension
(11) therapy with a statin
(12) number of non-disease specific measures (weight loss >= 10 lbs in 12 months, orthostatic hypotension, hip fracture in the 2 years prior to entry, dementia diagnosis in the 3 months prior to entry, fall in 2 years prior to entry, >= 2 ED or hospital admissions in 2 years prior to entry)
(3) number of antihypertensive medications
points for initial GFR =
= (-3.169 * (GFR)) + 95.12
points for second GFR from 20 to 30 mL per min per 1.73 square meter = 2
points for second GFR from 5 to 20 mL per min per 1.73 square meter =
= (0.1286 * (GFR)) - 0.5
points for systolic blood pressure from 80 to 120 =
= 102.6 - (0.721 * (SBP))
points for systolic blood pressure from 120 to 140 =
= 72 - (0.5 * (SBP))
points for systolic blood pressure from 140 to 240 =
= 4.684 - (0.01974 * (SBP))
points for age =
= (3.633 * (years)) - 291.9
points for BMI from 25 to 60 =
= (0.01613 * ((BMI)^2)) - (2.235 * (BMI)) + 76.26
points for BMI from 10 to 25 =
= (0.2333 * ((BMI)^2)) - (12.93 * (BMI)) + 202.5
|
Parameter
|
Finding
|
Points
|
|
gender
|
male
|
30
|
|
|
female
|
0
|
|
history of renal failure
|
absent
|
0
|
|
|
present
|
24
|
|
history of cardiovascular disease
|
absent
|
0
|
|
|
present
|
11
|
|
tobacco use
|
absent
|
0
|
|
|
present
|
11
|
|
history of hypertension
|
absent
|
7
|
|
|
present
|
0
|
|
therapy with a statin
|
yes
|
0
|
|
|
no
|
15
|
|
number of non-disease specific measures
|
0
|
0
|
|
|
1 or 2
|
46
|
|
|
3 to 6
|
62
|
|
number of antihypertensive medications
|
0
|
0
|
|
|
1 or 2
|
5
|
|
|
>= 3
|
9
|
total score =
= SUM(points for all 13 parameters)
Interpretation:
• minimum score: 0
• maximum score: 465
• The higher the score the greater the risk of mortality.
|
Score
|
2-Year Mortality
|
|
< 45
|
< 10%
|
|
45 to 186
|
(0.00146 * ((score)^2)) - (0.18457 * (score)) + 7.9773
|
|
186 tp 278
|
(-0.0017197 * ((score)^2)) + (1.2367 * (score)) - 115.47
|
|
> 278
|
> 95%
|
Performance:
• The area under the ROC curve is 0.68.
