Prediction of Severe Perioperative Adverse Outcomes Using the Data of the Multicenter Study of General Anesthesia

Description

The Multicenter Study of General Anesthesia was conducted at 15 university-affiliated hospitals and provided an analysis of adverse outcomes in 17,201 patients.

Parameters studied:

(1) concurrent diseases: 7 included

(2) type of surgery: 8 subtypes

(3) age in years, by decade, from 20 to 80

(4) ASA physical status

(5) anesthetic agent used

(6) miscellaneous: gender, obesity, and smoking

Anesthetic used:

(1) The 3 anesthetic agents studied were fentanyl, halothane and isoflurane, with patients receiving enflurane as the "reference" group.

(2) Severe ventricular arrhythmias were more common with halothane.

(3) Severe hypertension and severe bronchospasm were more common with fentanyl.

(4) Severe tachycardia was more common with isoflurane.

Obesity was defined as present if the person was > 1.15 ideal weight

(1) for males: ideal weight in pounds = 106 + (6 * ((height in inches) - 60))

(2) for females: ideal weight in pounds = 100 + (5 * ((height in inches) - 60))

Relatively Frequent Adverse Outcomes

Parameter	Finding	any severe outcome	any severe CV outcome	hypotension
disease	myocardial infarction <= 1 year	-0.80	-0.70
	cardiac failure	-0.75	-0.64
type of surgery	gynecologic	0.55	0.57	0.83
	abdominal	-0.66	-0.62
	neurologic	-0.45	-0.49
	cardiovascular	-1.56	-1.96	-1.90
	thoracic	-0.80	-0.75
age	20 years	0.09	0.05	0.93
	30 years	0.12	0.01	0.42
	40 years	0.22	0.14	-0.13
	50 years	-0.06	-0.26	-0.33
	60 years	-0.30	-0.60	-0.41
	70 years	-0.41	-0.76	-0.72
	80 years	-0.78	-0.91	-0.80
ASA physical status	ASA 2	-0.10	-0.13	-0.66
	ASA 3	-0.47	-0.75	-1.52
	ASA 4	-0.95	-1.21	-2.55
	obesity	0.15
	smoking	0.24
study anesthetic	fentanyl	-0.49	-0.47
	halothane	-0.41	-0.51
	isoflurane	-0.36	-0.38
logistic constant		3.48	4.36	5.44

from Table 1, page 6, Forrest et al (1992)

where:

• any severe outcome includes death

• CV = cardiovascular

Parameter	Finding	hyper-tension	any severe respiratory outcome	tachycardia
disease	hypertension	-0.75		0.62
	COPD		-0.81
	myocardial infarction > 1 year			-1.18
	cardiac failure		-1.32
type of surgery	musculoskeletal	1.25
	gynecologic	1.08		0.44
	abdominal		-0.56	-1.06
	EENT			-0.55
	endocrinologic			-0.55
	cardiovascular	-1.18		-1.74
male gender			0.59
ASA physical status	ASA 2	-0.90
	ASA 3	-1.54
	ASA 4	-0.75
obesity			0.67
smoking		0.25	0.43	0.54
study anesthetic	fentanyl	-1.44		-0.56
	halothane	-0.37		-0.09
	isoflurane	-0.58		-0.87
logistic constant		5.77	3.97	5.22

from Table 1, page 6, Forrest et al (1992)

Parameter	Finding	ventricular arrhythmia	bradycardia	atrial arrhythmia
disease	hypertension		-0.67
	myocardial ischemia	1.19	-0.97
	myocardial infarction > 1 year	-1.05
	ventricular arrhythmia	-1.68
type of surgery	abdominal	-0.60
	cardiovascular	-1.38		-2.26
age	20 years			1.28
	30 years			0.57
	40 years			0.80
	50 years			1.55
	60 years			0.54
	70 years			0.05
	80 years			-1.30
ASA physical status	ASA 2			-0.64
	ASA 3			-1.74
	ASA 4			-3.15
study anesthetic	fentanyl	-0.55
	halothane	-2.27
	isoflurane	-0.02
logistic constant		6.74	5.66	6.71

from Table 1, page 6, Forrest et al (1992)

Infrequent Complications

Parameter	Finding	cardiac failure	broncho-spasm	myocardial ischemia
disease	myocardial ischemia	-1.29	-1.60
	COPD		-1.60
age	20 years			1.00
	30 years			1.00
	40 years			1.00
	50 years			-7.33
	60 years			-8.21
	70 years			-7.52
	80 years			-9.34
obesity			1.14
smoking			1.30
study anesthetic	fentanyl		-1.22
	halothane		0.26
	isoflurane		0.32
logistic constant		15.86	6.74	14.20

from Table A4, page 13, Forrest et al (1992)

Parameter	Finding	myocardial infarction	respiratory failure	secretions
ASA physical status	ASA 2	-6.61	-8.24
	ASA 3	-9.98	-8.96
	ASA 4	-13.36	-11.18
smoking				2.48
logistic constant		15.20	15.20	6.89

from Table A4, page 13, Forrest et al (1992)

Calculating the Probability of an Adverse Outcome

probability of outcome =

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

X =

= SUM(coefficients for risk factors present) + (logistic constant for outcome)

Implementation notes:

• I am not certain of how best to handle the situation where a different agent is used from the 4 listed (fentanyl, halothane, isoflurane, enflurance). In the implementation these will not receive a score, which may introduce an error into the probability estimate.

• For age: (1) pediatric patients excluded, (2) if > 85, will use data for 80, (3) will round age to nearest decade (viz 26 to 30).

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

Description

Specialty

ICD-10