Document Detail

Development and validation of a Bayesian model for perioperative cardiac risk assessment in a cohort of 1,081 vascular surgical candidates.
MedLine Citation:
PMID:  8613603     Owner:  NLM     Status:  MEDLINE    
OBJECTIVES: This study sought to develop and validate a Bayesian risk prediction model for vascular surgery candidates. BACKGROUND: Patients who require surgical treatment of peripheral vascular disease are at increased risk of perioperative cardiac morbidity and mortality. Existing prediction models tend to underestimate risk in vascular surgery candidates. METHODS: The cohort comprised 1,081 consecutive vascular surgery candidates at five medical centers. Of these, 567 patients from two centers ("training" set) were used to develop the model, and 514 patients from three centers were used to validate it ("validation" set). Risk scores were developed using logistic regression for clinical variables: advanced age (>70 years), angina, history of myocardial infarction, diabetes mellitus, history of congestive heart failure and prior coronary revascularization. A second model was developed from dipyridamole-thallium predictors of myocardial infarction (i.e., fixed and reversible myocardial defects and ST changes). Model performance was assessed by comparing observed event rates with risk estimates and by performing receiver-operating characteristic curve (ROC) analysis. RESULTS: The postoperative cardiac event rate was 8% for both sets. Prognostic accuracy (i.e., ROC area) was 74 +/- 3% (mean +/- SD) for the clinical and 81 +/- 3% for the clinical and dipyridamole-thallium models. Among the validation sets, areas were 74 +/- 9%, 72 +/- 7% and 76 +/- 5% for each center. Observed and estimated rates were comparable for both sets. By the clinical model, the observed rates were 3%, 8% and 18% for patients classified as low, moderate and high risk by clinical factors (p<0.0001). The addition of dipyridamole-thallium data reclassified >80% of the moderate risk patients into low (3%) and high (19%) risk categories (p<0.0001) but provided no stratification for patients classified as low or high risk according to the clinical model. CONCLUSIONS: Simple clinical markers, weighted according to prognostic impact, will reliably stratify risk in vascular surgery candidates referred for dipyridamole-thallium testing, thus obviating the need for the more expensive testing. Our prediction model retains its prognostic accuracy when applied to the validation sets and can reliably estimate risk in this group.
G J L'Italien; S D Paul; R C Hendel; J A Leppo; M C Cohen; L A Fleisher; K A Brown; S W Zarich; R P Cambria; B S Cutler; K A Eagle
Related Documents :
23910223 - Utility of a novel error-stepping method to improve gradient-based parameter identifica...
23953593 - Determining the impact of instrument variation and automated software algorithms on the...
20379603 - Advances in nmr-based biofluid analysis and metabolite profiling.
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  Journal of the American College of Cardiology     Volume:  27     ISSN:  0735-1097     ISO Abbreviation:  J. Am. Coll. Cardiol.     Publication Date:  1996 Mar 
Date Detail:
Created Date:  1996-06-03     Completed Date:  1996-06-03     Revised Date:  2010-03-24    
Medline Journal Info:
Nlm Unique ID:  8301365     Medline TA:  J Am Coll Cardiol     Country:  UNITED STATES    
Other Details:
Languages:  eng     Pagination:  779-86     Citation Subset:  AIM; IM    
Vascular Unit, Massachusetts General Hospital, Boston, Massachusetts 02114, USA.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Bayes Theorem
Cohort Studies
Dipyridamole / diagnostic use
Heart Diseases / epidemiology*,  radionuclide imaging
Logistic Models
Models, Statistical*
Postoperative Complications / epidemiology*,  radionuclide imaging
Predictive Value of Tests
ROC Curve
Reproducibility of Results
Risk Assessment
Risk Factors
Sensitivity and Specificity
Thallium Radioisotopes / diagnostic use
Vascular Diseases / surgery*
Vasodilator Agents / diagnostic use
Reg. No./Substance:
0/Thallium Radioisotopes; 0/Vasodilator Agents; 58-32-2/Dipyridamole
Comment In:
J Am Coll Cardiol. 1996 Mar 15;27(4):799-802   [PMID:  8613605 ]

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Previous Document:  Circadian variations of onset of acute myocardial infarction and efficacy of thrombolytic therapy.
Next Document:  Meta-analysis of intravenous dipyridamole-thallium-201 imaging (1985 to 1994) and dobutamine echocar...