A model that predicts morbidity and mortality after coronary artery bypass graft surgery.

OBJECTIVES: This study was performed to develop a method for identifying patients at increased risk for morbidity or mortality after coronary artery bypass graft surgery. BACKGROUND: Postoperative morbidity is more common than mortality and is important because of its relation to cost. METHODS: Univariate and forward stepwise logistic regression analysis was used to retrospectively analyze a group of 1,567 consecutive patients who underwent bypass surgery between July 1991 and December 1992. We developed a model that predicted postoperative morbidity or mortality, or both, which was then prospectively validated in a group of 1,235 consecutive patients operated on between January 1993 and April 1994. A clinical risk score was derived from the model to simplify utilization of the data. RESULTS: The following factors, listed in decreasing order of significance, were found to be significant independent predictors: cardiogenic shock, emergency operation, catheterization-induced coronary artery closure, severe left ventricular dysfunction, increasing age, cardiomegaly, peripheral vascular disease, chronic renal insufficiency, diabetes mellitus, low body mass index, female gender, reoperation, anemia, cerebrovascular disease, chronic obstructive pulmonary disease, renal dysfunction, low albumin, elevated blood urea nitrogen, congestive heart failure and atrial arrhythmias. Observed morbidity and mortality for the validation group fell within the 95% confidence interval of that predicted by the model. Costs were closely related to the incidence of postoperative morbidity. CONCLUSIONS: Analysis of preoperative patient variables can predict patients at increased risk for morbidity or mortality, or both, after bypass surgery. Increased morbidity results in higher costs. Different strategies for high and low risk patients should be used in cost reduction efforts.

A model for predicting transfusion after coronary artery bypass grafting.

BACKGROUND: Blood conservation has become an important issue in cardiac surgery. This study was undertaken to determine if the need of blood transfusion could be predicted from preoperative patient variables. METHODS: From January 1, 1992, to December 31, 1993, 2,033 patients having isolated coronary artery bypass grafting procedures were studied; 1,446 (71%) were male and 587 (29%), female. The mean age was 65.1 +/- 9.9 years (range, 31 to 88 years). Emergency operation, urgent operation, and reoperations were done in 78 (4%), 188 (9%), and 189 (9%) patients, respectively. In the entire group, 1,245 (61%) received transfusion during hospitalization, and 788 (39%) did not. Logistic regression analysis was used to construct a model that predicted the need of transfusion of packed red blood cells after coronary artery bypass grafting. A transfusion risk score was constructed by assigning points to independent predictive factors on the basis of the logistic regression coefficient and the odds ratio. Preoperative predictors of transfusion were emergency operation, urgent operation, cardiogenic shock, catheterization-induced coronary occlusion, low body mass index, left ventricular ejection fraction lower than 0.30, age greater than 74 years, female sex, low red cell mass, peripheral vascular disease, insulin-dependent diabetes, creatinine level greater than 1.8 mg/dL, albumin value lower than 4 g/dL, and redo operation. RESULTS: The mean transfusion risk score for patients receiving 0, 1 to 4, and greater than 4 units of packed red blood cells was 2.3 +/- 0.9, 5.2 +/- 3.0, and 9.6 +/- 3.5, respectively (p = 0.001). Patients with a score higher than 6 had a 95% transfusion incidence. The predictive model was validated on 422 patients having coronary artery bypass grafting from January 1 to May 31, 1994. The observed rates of the validation group fell within the 95% confidence intervals of the predicted rates. CONCLUSIONS: These data demonstrate that readily available patient variables can predict patients at risk for transfusion. Routine use of aprotinin and other adjustments of cardiopulmonary bypass should be considered to reduce transfusion in high-risk patients.

Recent experience with major sternal wound complications.

During a recent 1-year period, 31 patients sustained a major sternal wound infection and sternal dehiscence developed in 6 patients. Multiple potential risk factors were tabulated in these patients and in a control group selected from 1,521 patients undergoing sternotomy during the same time period. The overall infection rate was 2.1%, and the mortality rate in the patients with sternal infection or dehiscence was 16.2%. Chronic obstructive pulmonary disease, prolonged intensive care unit stay, respiratory failure, connective tissue disease, and male sex were significantly higher in the group with sternal infection or dehiscence (p less than 0.05). Advanced age and low cardiac output episodes were more frequent in this group, but only approached statistical significance. Although several risk factors may have been interrelated, male sex and the presence of pulmonary disease were statistically independent predictors of sternal wound infection. Risk factors may be helpful in identifying high-risk patients for additional prophylactic measures.

Twenty-five-year review of the Magovern-Cromie sutureless aortic valve.

We reviewed 25 years (4,798 patient-years) of aortic valve replacement with the Magovern-Cromie sutureless valve. Operative mortality was 11% for isolated aortic valve replacement and 15% for aortic valve replacement with concomitant cardiac procedures. Since 1981, operative mortality has declined to 4.9%. Valve-related morbidity was in the lower expected ranges for prosthetic aortic valves: ball variance, 0.3%/patient-year; paraprosthetic leak, 0.41%/patient-year; valve endocarditis, 0.43%/patient-year; valve thrombosis, 0.04%/patient-year; and embolic events, 3.95%/patient-year. The incidence of aortic valve reoperation was 0.76%/patient-year. The 5-year, 10-year, and 20-year probability of survival corrected for normal mortality was 77%, 64%, and 52% for all discharged patients. This review confirms the Magovern-Cromie valve to be a safe, durable, and efficient prosthetic valve.

Traumatic aortic transection: repair using left atrial to femoral bypass.

Traumatic aortic transection is a life-threatening surgical emergency. Therapy must be directed at rapid repair and prevention of postoperative complications, the most serious being paraplegia. Controversy over the optimal method of repair exists-specifically whether the use of a shunt modifies the outcome. Our series of 17 patients using left atrial to femoral bypass with the Biomedicus pump will be discussed. Preoperative preparation and operative technique will be outlined. Mortality in this series was 18%, the incidence of paraplegia was 0.

Paced skeletal muscle for dynamic cardiomyoplasty.

Four patients, each with a history of myocardial infarction and diffuse coronary artery disease, underwent application of left latissimus dorsi (LD) muscle with intact neurovascular bundle to the anterolateral wall of the left ventricle. The muscle was conditioned over a six-week period subsequent to operation in 3 patients and was conditioned preoperatively with a burst stimulus in the fourth. Biopsy specimens confirm the experimental data that human skeletal muscle can be electrically conditioned over a six- to ten-week period to contain mainly fatigue-resistant type I fibers. All patients survived the procedure, and 3 showed improvement secondary to aneurysmectomy. In Patient 1, a modified resection was performed, and at 28 months after operation, at the 75-W level of exercise, the ejection fraction was 54% paced versus 45% nonpaced. In Patient 2, at 12 months, the ejection fraction at rest was 44% paced versus 30% nonpaced. Doppler echo studies confirmed the presence of the flap and its function in the paced and nonpaced mode. The third patient died of a sudden ventricular arrhythmia 2 months following operation. An infected, nonfunctioning, degenerated flap was found at autopsy. Patient 4 did not have an aneurysm. She received a bypass graft to the right coronary artery and underwent cardiomyopexy in an attempt to relieve medically refractory incapacitating chronic congestive heart failure. Ten months postoperatively, ejection fraction at rest was 33% paced versus 25% nonpaced. Constrictive myopathy has not been encountered in any of these patients.

Recovery of the failing canine heart with biventricular support in a previously fatal experimental model.

Prolonged normothermic ischemia in the canine model is generally fatal with standard resuscitative techniques. To determine whether such myocardial injury is recoverable with biventricular support, we subjected 10 dogs to 45 minutes of ischemia at 37 degrees C. After ischemia, the animals were supported for 24 hours with biventricular assist with the centrifugal pump. During early reperfusion, none of the hearts could sustain a stable rhythm or blood pressure. Myocardial adenosine triphosphate concentration, expressed as micromoles per gram of heart protein, was dramatically reduced from a control of 31.5 +/- 2.4 to 14.6 +/- 2.9 (p less than 0.01 versus control), a 54% reduction. Ultrastructural analysis did not reveal the explosive cell swelling of irreversible cell injury. After 12 hours of biventricular assist, developed pressure partially recovered to 60.0 +/- 10 mm Hg (p less than 0.01 versus control) and maximal positive dP/dt measured 2,649 +/- 412 mm Hg/sec (p less than 0.01 versus control). Adenosine triphosphate concentration increased to 25.2 +/- 5.5 (p less than 0.01 versus control). Electron microscopic examination showed less chromatin clumping, no further mitochondrial distortion, and more abundant glycogen. After 24 hours of biventricular assist, cardiac output in the seven dogs successfully weaned from biventricular assist measured 3.6 +/- 0.6 L/min, developed pressure recovered to 76.3 +/- 8.9 mm Hg, and its first derivative recovered to 4,282 +/- 585 mm Hg/sec (all measurements not significant compared with control). Examination by an electron microscope revealed no severe mitochondrial injury.

Paced latissimus dorsi used for dynamic cardiomyoplasty of left ventricular aneurysms.

Two patients are described, each with a large left ventricular aneurysm and severe coronary artery disease, and each with an ejection fraction lower than 30% and in congestive heart failure. In both, the left latissimus dorsi (LD) muscle was used in the repair of the ventricular aneurysm because preoperative studies demonstrated that there was concomitant coronary artery disease, and there was a strong suggestion that resection of the entire aneurysm would seriously compromise the residual ventricular capacity. One patient had an 18-year history of coronary occlusion with two infarctions. A large, calcified ventricular aneurysm developed, and despite vigorous medical treatment, intractable congestive heart failure and angina persisted. The diffuse coronary artery disease made this patient a poor candidate for bypass grafting. The other patient sustained an acute myocardial infarction 5 months prior to operation. The left anterior descending coronary artery was totally occluded, and a large apical aneurysm developed along with an akinetic anterior wall and septum. After his heart attack, the patient had progressive dyspnea on exertion. Following operation in both patients, the transpositioned LD, then a component in the repair of the left ventricular wall, was electrically trained to synchronously contract with each systole, driven by a standard dual-chamber cardiac pacemaker. Steady improvement and a return to normal activities were observed in both patients. There was an indication of improved ejection fraction with synchronous contraction of the skeletal muscle.

Mechanical support of the failing heart.

Mechanical ventricular assist with a centrifugal pump with or without anticoagulation for an extended period has been used in 41 patients with postcardiotomy ventricular failure. Left ventricular, right ventricular, and biventricular assist were required. The efficacy and safety of mechanical ventricular assist have been documented. Marked improvement in survival has been observed in the more recent part of this series, and is attributed to earlier employment of the assist device, maintenance of better flow rates near physiological levels, and use of biventricular assist to provide effective circulatory support. Mechanical ventricular assist is easy to use, and the conversion from ordinary cardiopulmonary bypass is also easy. Therefore, mechanical assist provides a very effective means of temporary circulatory assist.

Is percutaneous insertion of the intra-aortic balloon pump through the femoral artery the safest technique?

Between Jan. 1, 1979, and Sept. 30, 1983, 423 intra-aortic balloon pumps (IABPs) were successfully inserted into 400 patients; IABPs could not be inserted in 36 patients (success rate 91.7%). Before 1980 all balloons were inserted surgically through a graft and by 1983 virtually all IABPs were inserted by percutaneous techniques. Infectious complications occurred in 12.3% of IABPs inserted by the open technique and 1.5% of IABPs inserted percutaneously (p less than 0.001). Major ischemic complications were not significantly different between the two groups when all of the patients were analyzed together (p greater than 0.5) and when only the surviving patients were analyzed (p greater than 0.75). However, ischemic complications occurred in 32.9% of women in the study and only 19.7% of the men (p less than 0.005). The percutaneous insertion of the IABP is recommended as the technique of choice because of the ease of insertion, lack of infectious complications, and similar rate of major ischemic complications when compared with IABPs inserted by open surgical means through a graft anastomosed to the common femoral artery.

Infectious complications and cost-effectiveness of open resuscitation in the surgical intensive care unit after cardiac surgery.

From July, 1982, to May, 1984, 2,412 patients underwent cardiac surgery. Open resuscitation through a midline sternotomy was performed in the surgical intensive care unit (SICU) 88 times in 64 patients one minute to 10 days after admission. There were 49 initial survivors; 31 patients had primary closure in the SICU (Group 1), and 18 patients had delayed closure (Group 2). In Group 1 there was 1 death. Wound infection developed in 2 of the 30 survivors--Escherichia coli in 1 and Staphylococcus epidermidis in 1. The latter required subsequent debridement. In Group 2 there were 8 survivors and no wound infections. Fifteen patients could not be resuscitated because of ventricular arrhythmia (13%), asystole (33%), cardiogenic shock (47%), and tamponade (7%). Only 2 of 38 patients, or 5%, experienced wound infections. This study demonstrates that open resuscitation in the SICU is a safe, rapid, and cost-effective procedure that will allow earlier intervention, diagnosis, and treatment. In no instance was death attributed to wound infection, and at our institution, this method resulted in cost savings of more than $1,000 per patient.

Use of the BioMedicus centrifugal pump in traumatic tears of the thoracic aorta.

Traumatic blunt thoracic aortic injury is a clinical entity of increasing incidence. After the diagnosis of traumatic tear of the aorta is made, there is some controversy over whether the aorta should be repaired using cardiopulmonary bypass, a heparinized shunt, or cross-clamping and graft interposition without a shunt or bypass. At Allegheny General Hospital, 19 patients were treated for traumatic tears of the thoracic aorta between July 1, 1977, and June 30, 1983. They can be divided into two groups: Group 1 (July 1, 1977, through October 31, 1981), in which no shunt or bypass or only a heparinized shunt was used, and Group 2 (November 1, 1981, through June 30, 1983), in which left atrium-femoral artery bypass was performed using a BioMedicus heparinless pump and tubing. Among the 10 patients in Group 1, 4 died and 2 had paraplegia postoperatively. Among the 9 patients in Group 2, 1 died and none experienced paraplegia following operation. We believe that the BioMedicus centrifugal pump is a simple, safe means of perfusing the lower body, kidneys, and spinal column without necessitating heparinization in a patient with multiple injuries or the placement of a cumbersome heparinized shunt. Because of the simplicity and the reliability demonstrated, this pump should be considered for use in all patients with traumatic tears of the thoracic aorta.

The surgical resection and chemotherapy of metastatic osteogenic sarcoma of the right ventricle.

The case report of a patient with metastatic osteogenic sarcoma of the right ventricle who had had an interscapulothoracic amputation in October, 1970, and a left lower lobectomy in June, 1973, for pulmonary metastasis is presented. The patient was hospitalized in October, 1974, with signs and symptoms of right ventricular outflow obstruction and arrhythmia, and cardiac workup established the presence of a right ventricular tumor. The lesion was successfully resected using cardiopulmonary bypass, and the diagnosis of metastatic osteogenic sarcoma was confirmed. The patient did well after the operation and returned to normal activity. She was placed on adjuvant Adriamycin (doxorubicin) chemotherapy, but 6 months later died of Adriamycin toxicity.

Immunoprofile studies for patients with bronchogenic carcinoma. I. Correlation of pretherapy studies with survival.

The general immune competence of 146 patients with bronchogenic carcinoma was measured, prior to irradiation therapy, by determining dinitrochlorobenzene (DNCB) reactivity, delayed cutaneous hypersensitivity (DCH) response to microbial antigens, peripheral lymphocyte counts, peripheral T and B lymphocyte counts, and the response of patient's lymphocytes to stimulation by phytohemagglutinin (PHA), concanavallin A (Con A) and pokeweed mitogen (PWM). Analyses were performed by the life-table method to determine the correlation of the immune status of these patients with survival rates. Statistically significant differences in survival were noted between the groups of patients with normal values when compared with the patients with abnormal values for the majority of the tests of general immunity. A stage of disease correlation with survival rate was noted for all groups of patients with abnormal immune measurements, but it was absent for many of the immune parameters when patients with normal values were compared. The effects of histology, age, and sex did not appear to influence the survival data as significantly as did the immune status of the patient. These data indicate that measurements of general immune competence may be of significant prognostic value for the management of patients with bronchogenic carcinoma. The measurement of DNCB reactivity shows the strongest correlation with survival rate.

A thirteen-year review of the Magovern-Cromie aortic valve.

A review of 375 cases of aortic valve replacement with the Magovern-Cromie sutureless prosthesis demonstrated a 43 per cent survival rate 10 to 13 years following the operation. Of the 308 patients surviving zero to 30 days, 238 patients are alive after one to 14 years. A total of 70 patients (22 per cent died during the 14 year follow-up period. The actuarial survival rate of all patients at risk was 64 per cent at the fifth year and 51 per cent at the tenth year. For those patients who lived at least one year after surgery, the probability of survival was 84 per cent and 70 per cent at 5 and 10 years, respectively. These results compare favorably with reports of up to only 8 years with the Smeloff-Cutter and Starr-Edwards prostheses. There was only a 1.5 per cent incidence of perivalvular leak with this method of fixation, and no valve has migrated. Ball variance has not occurred with any valve inserted since 1964, and the incidence of thromboembolism with the Model A4 prosthesis is 2.4 per cent. Improvement in results depend on reduction of thromboembolism, lessening of late and operative morality rates, and by earlier operative intervention to reduce the late deaths from cardiogenic factors.

Familial myxomas in four siblings.

Familial myxomas of the heart are very rare, with only two previous reports noted in the literature. This report documents four siblings with myxomas, and three of these patients had two or more myxomas. One sibling had four myxomas excised. The clinical findings, history, examination, and surgical treatment are described.

Carotid sinus syndrome: new surgical considerations.

1. Direct surgical denervation of the carotid sinus would appear to offer the best long-term results for carotid sinus syncope. 2. Carotid sinus denervation should be done under temporary cardiac pacing for safety and for completeness of denervation. 3. Arteriography is indicated to eliminate those cases of cerebral vascular insufficiency that might mimic carotid sinus syndrome. Credit must be given to Dr. R. Hughes, Clearfield, Pa., who recognized this diagnosis and referred the patient for therapy.