Gregory L. Kay

Inappropriate Use of Bivariable Analysis to Screen Risk Factors for Use in Multivariable Analysis

The use of bivariable selection (BVS) for selecting variables to be used in multivariable analysis is inappropriate despite its common usage in medical sciences. In BVS, if the statistical p value of a risk factor in bivariable analysis is greater than an arbitrary value (often p = 0.05), then this factor will not be allowed to compete for inclusion in multivariable analysis. This type of variable selection is inappropriate because the BVS method wrongly rejects potentially important variables when the relationship between an outcome and a risk factor is confounded by any confounder and when this confounder is not properly controlled. This article uses both hypothetical and actual data to show how a nonsignificant risk factor in bivariable analysis may actually be a significant risk factor in multivariable analysis if confounding is properly controlled. Furthermore, problems resulting from the automated forward and stepwise modeling with or without the presence of confounding are also addressed. To avoid these improper procedures and deficiencies, alternatives in performing multivariable analysis, including advantages and disadvantages of the BVS method and automated stepwise modeling, are reviewed and discussed.

Influence of ejection fraction on hospital mortality, morbidity, and costs for CABG patients

BACKGROUNDnPreoperative ejection fraction (EF) has been shown to adversely affect postoperative hospital mortality and morbidity for patients undergoing isolated coronary artery bypass grafting.nnnMETHODSnTo investigate influence of EF on isolated coronary artery bypass grafting outcomes (overall hospital mortality, hospital cardiac mortality, hospital morbidity, and hospital costs), data were reviewed from 1,354 consecutive patients who underwent isolated coronary artery bypass grafting between January 1, 1990, and April 30, 1992, at a single nonprofit hospital. Overall hospital mortality was 4.06% (cardiac, 2.36%). Hospital morbidity was 14.25% (including mortality). Hospital costs (not charges) averaged

Tricuspid Regurgitation Associated With Mitral Valve Disease: Repair and Replacement

16,673 per patient. To explore the impact of preoperative EF, EF was stratified into regular intervals. Each interval was then compared with regard to hospital mortality, morbidity, and average costs. A new statistical tool, discharge analysis, was developed to analyze the cost data. This was necessary because previous efforts at cost analysis have used tools inappropriate for real world cost data.nnnRESULTSnThe statistical analysis showed that patients with EF of 0.40 or greater had the best outcomes (lowest mortality, morbidity, and cost). Once the EF is 0.40 or greater the EF does not carry further predictive value. At EF less than 0.40, patients with EF less than 0.30 have a poorer outcome than patients with EF of 0.30 to 0.39.nnnCONCLUSIONSn(1) Ejection fraction is a valid predictor of mortality, morbidity and resource utilization based on statistical analysis. (2) Patients can be broadly grouped as having EF greater than 0.40, less than 0.30, or from 0.30 to 0.39 with regard to clinical and cost outcomes. (3) Postoperative length of stay is not predicted by risk-adjusted EF. (4) A new tool, discharge analysis, is presented to facilitate cost analysis.

Effect of mild renal dysfunction (s-crea 1.2–2.2 mg/dl) on presentation characteristics and short- and long-term outcomes of on-pump cardiac surgery patients

Between January 1975 and June 1988, 156 patients with combined mitral and tricuspid valve disease underwent mitral and tricuspid valve repair or replacement. There were 127 (81%) patients with tricuspid valve repair and 29 (19%) patients with tricuspid valve replacement. Hospital mortality was 14% and was strongly influenced by preoperative pulmonary hypertension (systolic pressure greater than 65 mm Hg) and poor left ventricular function (ejection fraction less than 0.4). Five-year survival for the entire series was 57% +/- 5%; 12-year survival was 44% +/- 9%. Ejection fraction was the only age-adjusted risk factor for long-term survival. Of the patients who underwent tricuspid annuloplasty, 91% +/- 4% were free from reoperation after 10 years, indistinguishable from valve replacement (90% +/- 7%). Our tricuspid annuloplasty is simple and effective, and exhibits excellent long-term durability as well as immediate hemodynamic improvement.

Early experience of off-pump coronary artery bypass using Octopus Tissue Stabilizer

OBJECTIVESnThe objective of this study is to evaluate differences in patient presentation and short- and long-term outcomes between patients dichotomized by the level of preoperative s-creatinine (s-crea) without renal failure and to use EuroSCORE (ES) risk stratification for validating differences and for predictive purposes.nnnMETHODSnA thousand consecutive cardiac surgery patients from January 1999 through May 2000 were analyzed. Patients with off-pump surgery or s-crea >200 micromol/l (>2.2 mg/dl) were excluded leaving 885 patients for analysis. Group 1 (n=703) had s-crea 0.5-1.2 mg/dl and Group 2 (n=182) had elevated s-crea 1.3-2.2 mg/dl but no renal insufficiency.nnnRESULTSnGroup 2 patients were older (P<0.0001), had a higher percentage of males (P=0.008), had lower left ventricular ejection fraction (LVEF) (P=0.001), had higher New York Heart Association (NYHA) classification (P<0.0001), had more diabetics (P=0.001) and had more patients with a history of congestive heart failure (CHF) (P<0.0001). Both additive ES (AES) and logistic ES (LES) variables were higher in Group 2 patients, AES 8.45+/-4.28% vs. 6.05+/-3.80% (P<0.0001) and LES 17.7+/-19.1% vs. 9.57+/-13.3% (P<0.0001). Proportions of emergency operations and use of intra-aortic balloon pulsation (IABP) support did not differ. There were more coronary artery bypass grafting (CABG) with or without concomitant procedures in Group 1 but otherwise the procedures performed were similar. Cardiopulmonary bypass (CPB) times did not differ (P=0.1). Operative mortality was similar (P=0.06) but hospital mortality was higher in Group 2: 19/10.4% vs. 25/3.6% (P<0.0001), odds ratio (OR) 3.16. Total length of stay (LOS) and length of stay in the postoperative intensive care unit (ICU) did not differ. Postoperative renal failure (PORF) (s-crea increase to >2.25 mg/dl or >200 micromol/l) developed in 38/4.5% patients in Group 1 and in 41/22.5% patients in Group 2 (P<0.0001), OR=5.08. Follow-up all-cause mortality was higher in Group 2: 68/37.4% vs. 167/23.8% (P<0.0001), OR=1.91. Both ES definitions predicted hospital mortality, LOS, ICU, PORF and long-term mortality well, while increased s-crea predicted PORF and long-term mortality in both groups.nnnCONCLUSIONSnMild increase in s-crea is a marker for patients with increased cardiac risk factors and the risk for poor outcomes. Both ES definitions are highly predictive of the outcomes.

Prediction of long-term survival by preoperative exercise testing in patients with depressed ejection fraction undergoing myocardial revascularization

OBJECTIVEnThe Octopus Tissue Stabilizer stabilizes segments of the beating heart using a series of suction cups that do not damage myocardial tissue or compromise hemodynamics. This allows the heart to be positioned with the arterial target on a stable platform. In this study we present our early experience of off-pump coronary artery bypass using the Octopus Tissue Stabilizer.nnnMETHODSnBetween October 1997 and June 1998, 50 patients underwent off-pump coronary artery bypass using the Octopus Tissue Stabilizer (7.5% of all coronary artery bypass cases; average age: 67.0 +/- 12.2). Preoperative ejection fraction was 15-70% (43.0 +/- 14.0%). Median sternotomy was used for all operations.nnnRESULTSnThe average number of grafts per patient was 2.5 +/- 0.9. The average operating time was 191 +/- 47 minutes. No patients returned to the operating room for postoperative bleeding or tamponade. Average intraoperative blood loss was 297 +/- 190 ml. No patient experienced perioperative myocardial infarction. There were five hospital deaths and one late death. All patients were free of angina at discharge. No patients returned with angina or required clinical restudy.nnnCONCLUSIONSnOff-pump coronary artery bypass offers the surgeon another tool to apply for appropriate patients. The Octopus Tissue Stabilizer allows the surgeon access to all parts of coronary circulation, extending the range of off-pump coronary artery bypass procedures, when the anatomy is appropriate.

Measured posterior annuloplasty for repair of non-ischemic mitral regurgitation. A single unit follow-up.

Ejection fraction is a major determinant of morbidity and mortality for patients with ischemic heart disease. Patients with an ejection fraction of 0.40 or less are generally recognized as having a poorer prognosis than those patients with an ejection fraction of 0.50 or better and remain a heterogeneous group. It would be useful if patients with a favorable surgical prognosis could be identified preoperatively. Fifty-five patients who underwent coronary artery bypass grafting and had an ejection fraction less than 0.40 (mean of 0.23 +/- 0.07 standard deviation) were studied by catheter measurement of pulmonary arterial pressure and radionuclide left ventriculography. Heart rate, systemic blood pressure, pulmonary artery pressures, cardiac output, and ejection fraction were measured, at rest, after nitroglycerin was given intravenously and with supine bicycle exercise. Forty-seven patients who had follow-up longer than 4 years were divided into two groups according to their life status (alive or dead) 4 years after operation. Measured variables of exercise stress tests and clinical characteristics were entered into factor analysis to obtain a cardiac function factor score for predicting the life status after 4 years. The cardiac function factor score was highly loaded by stroke index (rest, nitroglycerin), cardiac index (exercise), systemic vascular resistance index (exercise), and history of congestive heart failure. The cardiac function factor provided a predictive value superior to that of any individual variable. By dividing the patients into two groups by cardiac function factor score, the actuarial 5-year survival was 72% versus 17% for good and poor prognosis groups, respectively (p < 0.0001). Preoperative exercise stress testing data integrated by factor analysis provide a predictive tool for patients with a low ejection fraction.

Mitral valve lipomatous hamartoma infiltrating myocardium

UNLABELLEDnThe aim of this report is to evaluate short- and long-term outcomes of annuloplasty method of our choice: measured posterior annuloplasty (MPA). MPA is a piece of a Duran ring cut to the length of free-edge of anterior mitral leaflet (AML) and anchored with multiple pledgeted U-sutures from trigone to trigone into the posterior annulus.nnnMATERIAL AND METHODSnFrom 1988 to 2000, 103 consecutive patients with non-ischemic mitral regurgitation were scheduled preoperatively to be repaired by MPA.nnnRESULTSnPreoperative mitral valve regurgitation (MR) grade was 3.8+/-0.5 and decreased to 0.1+/-0.3 (P<0.0001) after repair. One patient was converted to insertion of mechanical prosthesis after grade 3 MR persisted after septal myectomy and MPA. Three patients needed instant revision of the repair one due to SAM and two due to stenosis. No patient had a stenosis or unacceptable (>1) MR after the procedure. There was one operative death (1.0%) and 3 hospital/30-day deaths (2.9%). Sixteen patients (16.3%) expired during the follow-up to 91 months (mean 57.4+/-19.5, median 60 months) none due to failure of MPA. There were no reoperations due to failure of MPA. Three patients had a reoperation, one for dehiscence of reconstruction after P2 resection and two patients due to progression of anterior leaflet degeneration and calcification with 4+ MR. New York Heart Association (NYHA) functional classification decreased from 2.3+/-0.8 to 1.4+/-0.6 (P<0.0001) and only one patient had an increase from II to III. Eighty-eight patients (96.7%) were in NYHA class I-II. Ten patients had an increase of MR from 0 to trace or 1 and one from 0 to 2. Two patients were diagnosed with mild stenosis without need of reoperation.nnnCONCLUSIONSnMPA is a durable and stable alternative for repair of non-ischemic mitral regurgitation of different etiologies. The technique gives an objective measure of the length of the band and no patient is left with a significant MR or mitral valve stenosis (MS). First-time success rate is very high and instant repairs few and minor. Freedom of MPA related reoperations is 100%.

1019-23 Use of a Cost Outcome Risk Score to Simultaneously Stratify Mortality, Morbidity, and Cost Outcomes for Cardiac Valve Patients

FIGURE 1. Levophase left ventriculogram showing the ventricular tumor. Primary cardiac neoplasms are very rare. The incidence is reported to be 0.001% to 0.03% in autopsy series. Of these, about 10% are valvular and the majority are benign. Only a few malignant valvular tumors have been described in the literature. Most valvular tumors are small and asymptomatic but may present with a variety of symptoms including arrhythmias, particulate or thrombotic embolizations, valvular stenosis or regurgitation with congestive heart failure, and even sudden death. Primary valvular lipomatous hamartomas have been described until now in only 9 patients: 5 patients with mitral hamartomas, 3 with tricuspid hamartomas, and 1 with aortic hamartoma.

920-48 Influence of Preoperative Risk Factors on Intraoperative and Postoperative Cerebrovascular Accidents in Cardiac Surgery

To investigate whether risks in postoperative outcomes (mortality, morbidity and costs [not charges]) can be simultaneously stratified preoperatively, data were reviewed for 551 consecutive patients who underwent valve surgery. Patient age 64.0 (±xa015.4) years and 43.4% females. There were 279 (50.6%) aortic valve procedures, 177 (32.1%) mitral valve procedures. 7 (1.3%) tricuspid valve procedures, 88 (16.0%) combined valve procedures, and 150 (20.7%) valve plus coronary procedures. The average total hospital stay was 13.7 (±xa014) days and postoperative stay was 11.0 (±xa012.6) days. To develop a cost outcome risk score, multivariate analysis of hospital costs via the Cox proportional hazards model was applied. This model adjusts for patients death, patientss health status at discharge and is not distorted by cost outliers. Patient age, gender, body surface area, body mass index. surgeon case-volume, procedural urgency, preoperative ejection fraction, history of diabetes, infective endocarditis, pulmonary hypertension, history of CABG, prior MI, and type of procedures were incorporated. Score N Hospital † Hasp. Total † Hasp. Cardiact † Hospital † Costs (