Description

D'Journo et al reported a model for predicting 90-day mortality for a patient undergoing esophagectomy for cancer. The authors are from multiple institutions from around the world and members of the International Esodata Study Group.


Patient selection: esophagectomy for cancer, not ECOG 4

 

Parameters:

(1) age in years

(2) body mass index (BMI) in kg per square meter

(3) sex

(4) ECOG performance status

(5) myocardial infarction

(6) connective tissue disease (SLE, RA, etc)

(7) peripheral vascular disease

(8) liver disease, moderate or severe

(9) neoadjuvant therapy

(10) hospital volume for esophagectomies per year

 

Parameter

Finding

Points

age

<= 40 years

0

 

41 to 60 years

1

 

61 to 80 years

0

 

> 80 years

-3

BMI

< 18.5

-3

 

18.5 to 24.9

0

 

>= 25

1

sex

male

0

 

female

2

ECOG PS

0

0

 

1

-1

 

2

-2

 

3

-4

myocardial infarction

no

0

 

yes

-2

connective tissue disease

no

0

 

yes

-3

peripheral vascular disease

no

0

 

yes

-2

liver disease, moderate or severe

no

0

 

yes

-5

neoadjuvant therapy

none

0

 

definitive chemoradiation

-2

 

chemotherapy only

0

 

chemoradiotherapy

-1

hospital volume

<= 108.6

0

 

> 108.6

1

 

total score =

= SUM(points for all of the parameters)

 

Interpretation:

• minimum score: -24

• maximum score: 5

 

value of X =

= (0.311 * (score)) + 3.382

 

probability of 90-day mortality =

= 1 / (1 + EXP(X))

 

Total Score

Risk of 90-Day Mortality

90-Day Mortality

>= 1

very low

1.8%

0

low

3%

-1 or -2

medium

5.8%

-3 or -4

high

8.9%

<= -5

very high

18.2%

 

Performance:

• The area under the ROC curve was 0.68 in the derivation and 0.64 in the validation cohort.


To read more or access our algorithms and calculators, please log in or register.