Description

Groeger et al developed a model for predicting mortality in cancer patients who require mechanical ventilation for respiratory failure. In general these patients have a poor prognosis, and having a reasonable estimate of the prognosis can aid physicians and family members in making management decisions. This was a multicenter study, with the authors from Memorial Sloan-Kettering, Mount Sinai Medical Center, MD Anderson, City of Hope National Medical Center, and Johns Hopkins Hospital.


Patient population: age >= 16 years with malignant disease

 

Parameters:

(1) intubation relative to ICU admission

(2) tumor type

(3) disease status

(4) history of allogenic bone marrow transplant (BMT)

(5) history of tumor surgery with curative intent

(6) cardiac rhythm: life threatening if (a) ventricular tachycardia or fibrillation; (b) symptomatic supraventricular dysrhythmia; (c) new heart block

(7) DIC

(8) vasopressor therapy: (a) any dose of norepinephrine, epinephrine, phenylephrine, dobutamine, amrinone, milrinone, nitroprusside, nicardipine; (b) dopamine > 3 µg/kg/min for >= 1 hour

 

Parameter

Finding

Points

intubation relative to ICU admission

<= 24 hours

0

 

> 24 hours

1

tumor type

not leukemia

0

 

leukemia

1

disease status

neither progression nor recurrence

0

 

progression or recurrence

1

history of allogenic BMT

no

0

 

yes

1

history of previous surgery with curative intent

no (no surgery, palliative surgery)

0

 

yes

1

cardiac rhythm

normal

0

 

non-life threatening arrhythmias

0

 

life threatening arrhythmias

1

DIC

no

0

 

yes

1

vasopressor therapy

no

0

 

yes

1

 

where:

• A history of surgery with the intent to cure was the only protective factor.

 

X =

= (0.7331 * (points for intubation)) + (0.6462 * (points for tumor type)) + (0.7042 * (points for disease status)) + (0.8202 * (points for BMT)) - (0.7093 * (points for surgery)) + (0.7924 * (points for cardiac rhythm)) + (1.3564 * (points for DIC)) + (0.6721 * (points for vasopressor therapy)) – 0.1466

 

probability of hospital mortality =

= EXP(X) / (1 + EXP(X))


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