Medtronic Mosaic porcine bioprosthesis: midterm investigational trial results.

BACKGROUND: The midterm clinical and hemodynamic performance of the Medtronic Mosaic porcine bioprosthesis was evaluated in a regulatory trial. METHODS: In nine Canadian centers, 802 bioprostheses (560 aortic and 242 mitral) were implanted between September 1994 and April 1999 in patients with a mean age of 70 years. RESULTS: Survival for aortic valve replacement at 4 years was 84.4%+/-3.1%. Freedom from valve-related or unexplained death was 95.6%+/-1.9%; structural valve deterioration, 100.0%; reoperation, 96.2%+/-1.7%; major thromboembolism, 96.1%+/-1.8%; and major antithrombotic-related hemorrhage, 96.4%+/-1.7%. Echocardiographic derived mean systolic gradient was 13.4 mm Hg at 4 years with an indexed effective orifice area of 0.7 to 0.8 cm2/m2. A significant decrease in left ventricular mass was shown over time in all valve sizes. Survival for mitral valve replacement at 4 years was 79.2%+/-6.8%. Freedom from valve-related or unexplained death was 96.5%+/-3.4%; structural valve deterioration, 100%; reoperation, 97.0%+/-3.2%; major thromboembolism, 95.7%+/-3.8%; and major antithrombotic-related hemorrhage, 95.0%+/-4.2%. Echocardiographically measured averaged mean diastolic gradient was 4.5 mm Hg. CONCLUSIONS: The Medtronic Mosaic bioprosthesis is safe and effective in both the aortic and mitral positions. The valve has low gradients in both positions and excellent left ventricular mass regression in the patients with aortic valve replacement.

Stroke during coronary bypass surgery: principal role of cerebral macroemboli.

OBJECTIVE: The purpose of this study was to gain insight into the etiology of stroke during coronary bypass surgery. METHODS: Retrospective review of prospectively gathered data on 6682 consecutive coronary bypass patients. Patients undergoing simultaneous procedures, including carotid endarterectomy, were excluded. We performed a systematic chart review of all patients who suffered a perioperative stroke. Predictors of stroke were determined with stepwise logistic regression analysis. RESULTS: The prevalence of stroke was 1.5% (n=98). Stroke patients had significantly increased intensive care unit and hospital length of stays, as well as increased mortality when compared to patients without stroke (all P< 0.001). Independent predictors of stroke were (in decreasing order of magnitude): age >70 years, left ventricular ejection fraction <40%, previous stroke or transient ischemic attack, normothermic cardiopulmonary bypass, diabetes, and peripheral vascular disease. Chart review revealed that the probable cause of stroke was macroemboli, likely from ascending aorta atherosclerosis, in 37% of patients and unknown in 38% of patients. Computerized tomography (CT) scans were obtained in 79 patients (81%). Lesions detected by CT were consistent with a macroembolic etiology: nearly all lesions were ischemic in nature and located in the distribution of major cerebral arteries, particularly the middle cerebral artery. CONCLUSIONS: Stroke is a devastating complication of coronary bypass surgery. Our multivariable risk factors for stroke, chart review, and CT findings all suggest that macroemboli, presumably from the ascending aorta, are the predominant cause of stroke during coronary bypass surgery. Future studies should be directed at minimizing the risk of embolization during cardiac surgery.

Neuropsychologic impairment after coronary bypass surgery: effect of gaseous microemboli during perfusionist interventions.

OBJECTIVE: Neuropsychologic impairment is a common complication of coronary bypass surgery. Cerebral microemboli during cardiopulmonary bypass are the principal cause of cognitive deficits after coronary bypass grafting. We have previously demonstrated that the majority of cerebral emboli occur during perfusionist interventions (ie, during the injection of air into the venous side of the cardiopulmonary bypass circuit). The purpose of this study was to determine whether an increase in perfusionist interventions is associated with an increased risk of postoperative cognitive impairment. METHODS: Patients undergoing elective coronary artery bypass grafting (n = 83) underwent a battery of neuropsychologic tests preoperatively and 3 months postoperatively. Patients were divided into 2 groups according to the median value of perfusionist interventions during cardiopulmonary bypass. Group 1 patients (n = 42) had fewer than 10 perfusionist interventions, and group 2 patients (n = 41) had 10 or more interventions. RESULTS: The 2 groups of patients were similar for all preoperative, intraoperative, and postoperative variables, with the exception of longer cardiopulmonary bypass times in group 2 patients (P <.001). Group 2 patients had lower mean scores on 9 of 10 neuropsychologic tests, with 3 (Rey Auditory Verbal Learning, Digit Span, and Visual Span) being statistically significant. Group 2 patients had worse cognitive test scores, even when controlling for increased bypass times. Group 2 patients had a nonsignificant trend toward an increased prevalence of neuropsychologic impairment 3 months postoperatively. CONCLUSIONS: Introduction of air into the cardiopulmonary bypass circuit by perfusionists, resulting in cerebral microembolization, may contribute to postoperative cognitive impairment.

Results of the BRAT study--a pilot study investigating the possible significance of ASA nonresponsiveness on the benefits and risks of ASA on thrombosis in patients undergoing coronary artery bypass surgery.

BACKGROUND: Several studies suggest that acetylsalicylic acid (ASA) is less effective in preventing thrombotic events in ASA nonresponder patients. If so, the thrombotic event rate in ASA nonresponders should be higher than in ASA responders. OBJECTIVE: To conduct a prospective, multicentre observational pilot study to determine the thrombotic event rates in ASA responders and nonresponders. PATIENTS AND METHODS: Patients undergoing nonurgent coronary artery bypass grafting (CABG) who were prescribed 325 mg ASA/day were recruited. Patients were classified as an ASA responder or nonresponder based on the ASA effect (or lack thereof) on their bleeding times. All thrombotic events that occurred in the two years following CABG were recorded. These data were stored in a blinded fashion until the last patient follow-up, and then adjudicated by a validation committee. RESULTS: A total of 289 patients recruited at three sites completed the two-year follow-up. Of these patients, 45.3% were classified as ASA responders and 54.7% were classified as ASA nonresponders. Of ASA responders, 6.9% had thrombotic events compared with 9.5% of the ASA nonresponders, but this difference was not significant (P=0.526). CONCLUSIONS: While ASA responder or nonresponder status did not appear to affect the thrombotic event rate in patients undergoing nonurgent CABG, the possibility that ASA responder or nonresponder status affects the thrombotic event rate in more acutely ill CABG patients cannot be excluded.

Ruptured sinus of valsalva aneurysm: early recurrence and fate of the aortic valve.

BACKGROUND: We reviewed our experience with congenital ruptured sinus of Valsalva aneurysms (RSVA) to determine patterns of early recurrence and the fate of the aortic valve (AV). METHODS: Over a 28-year period, RSVA was identified in 34 patients, (mean age 31.6 years). Primary closure of the RSVA was performed in 10 patients, and a patch employed in 24. Aortic insufficiency was present in 24 patients. AV replacement (AVR) was performed in 5 patients; AV repair in 6. RESULTS: Follow-up of 9.2 +/- 8.3 years (6 months to 24 years) was complete in all but 2 patients. Five early fistula recurrences (in 4 patients) correlated with primary rather than patch closure (p < 0.03). Kaplan-Meier survival at 10 years is 90 +/- 7%. Freedom from reoperative AVR at 10 years is 83 +/- 9%. Late AVR was performed in 6 patients for progressive aortic insufficiency due to bicuspid valve (n = 3), cusp disease of affected sinus (n = 2), or aortic root dilatation (n = 2). CONCLUSIONS: Patch closure of the RSVA should be routinely employed. A bicuspid valve may be associated with the late need for AVR.

Decreasing incidence of stroke during valvular surgery.

BACKGROUND: The predictors and causes of stroke after valvular surgery are incompletely defined. We examined the incidence, predictors, and mechanisms of stroke during valvular procedures over a 15-year time period. METHODS AND RESULTS: We retrospectively reviewed prospectively gathered data on 5954 consecutive patients undergoing valvular procedures at our institution from 1982 to 1996. Stroke was defined as persistent central nervous system deficit, usually with confirmatory CT imaging. Patients were divided into 3 groups according to date of operation: group 1, 1982 to 1986 (n = 1819); group 2, 1987 to 1991 (n = 2022); and group 3, 1992 to 1996 (n = 2113). Chart review was undertaken of all patients who developed stroke (n = 189). Stroke occurred in 3.8% of group 1 patients, 3.3% of group 2, and 2.6% of group 3 (P = 0.120). The decreasing incidence of stroke over time was confirmed by multivariable logistic regression analysis, in which earlier date of operation was an independent risk factor for stroke (P < 0.001). Predictors of stroke identified by multivariable logistic regression were (listed in decreasing order): (1) endocarditis (OR, 3.0; 95% CI, 1.8 to 5.0); (2) age > 74 years (OR, 2.3; 95% CI, 1.5 to 3.7); (3) earlier time period of operation (1982 to 1986: OR, 2.2; 95% CI, 1.5 to 3.2; 1987 to 1991: OR, 1.5; 95% CI, 1.0 to 2.2); (4) urgent timing (OR, 2.0; 95% CI, 1.4 to 2.8); (5) concomitant coronary bypass (OR, 2.0; 95% CI, 1.4 to 2.8); and (6) reoperation (OR, 1.7; 95% CI, 1.2 to 2.4). In more recent years of operation, we found an increasing prevalence of age > 74 years (7.4% in group 1, 9.5% in group 2, and 15.3% in group 3; P < 0.001), urgent timing (11%, 26%, and 34%, P < 0.001), and concomitant coronary bypass surgery (25%, 27%, and 33%; P < 0.001). CONCLUSIONS: The incidence of stroke during valvular surgery has decreased with time, despite an increased prevalence of risk factors. Predictors of stroke suggest 3 major causes (multivariable predictors in parentheses): atherosclerotic emboli (elderly age, concomitant coronary bypass), shock (urgent timing, reoperation), and septic emboli (endocarditis).

Medtronic mosaic porcine bioprosthesis satisfactory early clinical performance.

BACKGROUND: The Medtronic (Minneapolis, MN) Mosaic porcine bioprosthesis is an investigational prosthesis which incorporates zero-pressure fixation, aortic root predilation, low profile stent, and alpha oleic acid antimineralization treatment. METHODS: From September 1994 to August 1996, 289 patients (mean age 70 years, range, 28 to 88 years) had 227 (78.5%) aortic valve replacements and 62 (21.5%) mitral valve replacements. Concomitant procedures were performed in 61.2% (139) of aortic valve replacements and 54.8% (34) of mitral valve replacements. Of the aortic valve replacement group 70 (30.8%) were in the 61 to 70 age group and 134 (59.0%) were 71 years or older. Of the mitral valve replacements, 23 (37.1%) were 61 to 70 years and 30 (48.4%) 71 years or older. RESULTS: The early mortality, overall, was 4.2% (12 of 289); for aortic valve replacement it was 4.0% (9) and for mitral valve replacement it was 4.8% (3). The late mortality for aortic valve replacement was 2.6% per patient-year (3 events, 1.3% of total) and for mitral valve replacement it was 3.3% per patient-year (one event, 1.6% of total). The reoperative rate for aortic valve replacement was 3.0% per patient-year (4), while there were no mitral valve replacement reoperations. The freedom from major thromboembolism was 97.3%+/-1.6% for aortic valve replacement and 94.7%+/-3.0% for mitral valve replacement at 1 to 1.5 years. The freedom from reoperation was 96.7%+/-1.7% for aortic valve replacement; there was no reoperation for mitral valve replacement. There were no cases of structural valve deterioration. In the aortic position the mean systolic gradient was low, approximately 11 mm Hg, across all sizes (range 8 to 12 mm Hg at 3 months and 10 to 13 mm Hg at 12 months). In the mitral position the mean diastolic gradient was approximately 5 mm Hg (range, 2 to 6 mm Hg) for all sizes 25 to 31 mm at the early and 1 year follow-up echocardiographic assessment. CONCLUSIONS: The early clinical performance and in vivo hemodynamics are encouraging.

Risk factors for stroke following coronary bypass surgery.

Improvements in surgical technique and advances in myocardial protection have resulted in low rates of morbidity and mortality despite a greater incidence of high-risk patients. Noncardiac morbidity prolongs hospital stays and increases the costs of cardiac surgery. This study examines the preoperative predictors of stroke following isolated coronary bypass surgery. The clinical records of 3910 consecutive patients who underwent isolated coronary bypass surgery at the University of Toronto were reviewed. Stepwise logistic regression identified six independent predictors of stroke following CABG (percent in parentheses) and calculated factor adjusted odds ratios (OR) for each risk factor. Triple vessel coronary artery disease was the most important predictor (1.9%, OR 5.71), followed by normothermic systemic perfusion (3.8%, OR 4.85), age > 70 years (3.2%, OR 3.88), a previous history of transient ischemic attacks or stroke prior to surgery (6.1%, OR 3.7), peripheral vascular disease (4.7%, OR 2.77), and diabetes mellitus (2.6%, OR 2.01). The mechanism of stroke is likely different between these high-risk groups and strategies to prevent postoperative stroke should focus on the mechanisms responsible in high-risk patients.

Visceral and limb perfusion during thoracoabdominal aortic aneurysm repair.

Patients undergoing thoracoabdominal aortic aneurysm repair are at high risk of operative morbidity and death. Aortic clamping and unclamping stresses the myocardium, interrupts visceral and limb perfusion, and leads to metabolic acidosis. Use of a simple technique to preserve distal perfusion during the period of aortic clamping may reduce perioperative morbidity. We describe a technique of visceral and limb perfusion that may reduce surgical risk in high-risk patients.

Right ventricular failure associated with left ventricular failure.

RV and LV failure frequently coexist. Experimental evidence suggests that RV failure results from a primary insult to the right ventricle and an increase in pulmonary vascular resistance. LV failure results in an elevation of the left atrial pressure and thereby a reduction in the transpulmonary hydrostatic gradient. Because RV function depends on the contraction of the left ventricle, this contribution is reduced during LV failure. Treatment should be aimed first at restoring LV function. If RV failure persists and is not due to a simple mechanical problem, treatment of RV failure should be commenced. If volume loading, inotropes, and pH adjustments do not result in adequate RV output, a mechanical assist device should be considered. RHBP is the most powerful device, and it is capable of restoring systemic perfusion even when there is no residual RV function. Clinical results with this device have been most encouraging--most patients demonstrate improved RV function and a decrease in pulmonary vascular resistance after several days of treatment, allowing them to be weaned from assist. However, this technique is very invasive and requires constant close attention. PABC is a simple but less powerful assist device. Experimental studies suggest that if RV function is not extremely depressed--cardiac output is greater than 50 percent of normal--PABC may be effective in restoring systemic perfusion to normal levels. Current results suggest that RHBP remains the gold standard for severe RV failure, but further clinical experience with PABC may more clearly define its role in the management of RV failure.

A physiological approach to surgery for acute rupture of the papillary muscle.

There is controversy regarding the optimal management of patients in whom acute papillary muscle rupture develops. This study evaluates the effect of division of the anterolateral papillary muscle on left ventricular (LV) function and compares two methods of treatment--mitral valve replacement (MVR) and mitral valve repair. Thirteen pigs were placed on cardiopulmonary bypass, and interventions were performed in an isolated beating heart preparation. LV function was assessed with a compliant intraventricular balloon at baseline, after division of the anterolateral papillary muscle (Divided), after repair of the divided papillary muscle (Repair), and finally after MVR. Division of the anterolateral papillary muscle caused a significant deterioration in LV function. Function was maintained at this level after mitral valve repair but deteriorated with MVR. Developed pressure measured at baseline was 179 +/- 13 mm Hg; Divided, 148 +/- 11 mm Hg (p less than 0.05 versus baseline); Repair, 149 +/- 15 mm Hg; and MVR, 95 +/- 8 mm Hg (p less than 0.05 versus Divided) at a balloon volume of 20 ml. These results suggest that LV function is impaired by papillary muscle rupture. Repair of the ruptured papillary muscle is associated with better LV function than is MVR.

A rational approach to the selection of an assist device for the failing right ventricle.

Assist devices have become available for the treatment of right ventricular (RV) failure. The present study assesses the efficacy of pulmonary artery balloon counterpulsation (PABC) in restoring RV output to normal levels. In a porcine model of RV failure during left heart bypass, PABC restored RV output to normal, provided that the RV output was not depressed to less than 50% of baseline values. When RV failure was more severe, PABC was not effective in restoring RV output to base line. This study suggests that a rational approach should be developed for the use of right heart assist devices. In moderately severe RV failure, PABC may be successful in restoring cardiac output to normal. However, in extreme failure a right heart bypass is necessary.

The effects of cardioplegic arrest on right atrial function.

Atrial electrical and mechanical activity persists during cardioplegic arrest. It has been postulated that atrial ischemia may occur and cause deterioration in atrial function. This study was designed to assess the effect of cardioplegic arrest on right atrial function. Twenty-one pigs were placed on cardiopulmonary bypass (CPB), and the right atrium was isolated from the circulation by snaring both venae cavae and incising the coronary sinus. The tricuspid valve was closed through a small right ventriculotomy, and baseline atrial function was assessed using a compliant balloon in the atrium. Fourteen pigs underwent one hour of cardioplegic arrest (7 with cardioplegia alone [CCA group] and 7 with the addition of topical hypothermia [CCA + TH group]) followed by one hour of normothermic reperfusion. Seven other pigs were placed on CPB for the same period of time (CPB group). Atrial electrical and mechanical activity persisted at 45 beats per minute in the CCA group but was virtually abolished in the CCA + TH group. Cardioplegic arrest caused considerable deterioration in right atrial function (developed pressure, 18.9 +/- 0.8 [baseline] versus 14.1 +/- 0.7 mm Hg; p less than 0.05; first derivative of atrial pressure [dP/dt], 187 +/- 19 versus 134 +/- 25 mm Hg per second; p less than 0.05; 60 minutes of reperfusion and balloon volume of 20 ml). It was not affected by topical cooling. Right atrial developed pressure was maintained, but dP/dt was significantly reduced in the CPB group. This study suggests that cardioplegic arrest does not protect the atrium.

Toward a better understanding of the etiology of left ventricular dysfunction after mitral valve replacement: an experimental study with possible clinical implications.

The relatively high morbidity and mortality of mitral valve replacement (MVR) appears to be related to an impairment in left ventricular (LV) function. This two-part investigation was designed to assess the effect of MVR on global LV function in an isolated heart preparation and to evaluate whether a mitral prosthesis with a flexible annulus would be of benefit. In Part I (14 pigs), the effects of each step in MVR were studied. Division of the chordae tendineae caused a severe deterioration in LV function (systolic pressure, 180 +/- 13 versus 120 +/- 10 mm Hg; p less than 0.05; developed pressure, 167 +/- 13 versus 108 +/- 11 mm Hg; p less than 0.05; first derivative of LV pressure [dP/dt], 2,630 +/- 300 versus 1,610 +/- 180 mm Hg/sec; p less than 0.05; balloon volume, 30 ml). Fixation of the mitral annulus prior to division of the chordae tendineae resulted in a small decrease (not significant) in LV function but had no effect after the chordae tendineae were divided. In Part II (10 pigs), two mitral annular prostheses were studied: a standard rigid prosthesis and a prosthesis of identical size but with a flexible annulus. LV function was better with the flexible than the rigid prosthesis (systolic pressure, 118 +/- 10 versus 89 +/- 5% control, p less than 0.02; developed pressure, 120 +/- 11 versus 87 +/- 5% control; p less than 0.02; dP/dt, 119 +/- 10 versus 85 +/- 4% control; p less than 0.02; balloon volume, 30 ml).(ABSTRACT TRUNCATED AT 250 WORDS)

Is atrial function affected by conventional cardioplegic arrest?

Conventional cardioplegic arrest results in persistent atrial electrical and mechanical activity. This activity has been postulated to result in atrial ischemia which can induce postoperative arrhythmias and impair the transport function of the atrium. In this study, the effects of simple cardiopulmonary bypass (CPB) (seven pigs) and conventional cardioplegic arrest (CCA) (seven pigs) on right atrial function were evaluated. Function was assessed in an isolated right atrial preparation with a compliant balloon inserted via the superior vena cava. CCA for 1 hr produced significant deterioration in right atrial function (developed pressure 14.1 +/- 0.7 vs 18.9 +/- 0.8 mm Hg, P less than 0.05, diastolic pressure 10.0 +/- 1 vs 4.5 +/- 1.4 mm Hg, P less than 0.05, dP/dt 134 +/- 25 vs 187 +/- 19 mm Hg/sec, P less than 0.05 at a balloon volume of 20 ml after 1 hr of reperfusion). CPB alone caused no alteration in pressures in the right atrium but was associated with a late decrease in dP/dt (developed pressure 19.3 +/- 1.8 vs 18.9 +/- 0.8 mm Hg, diastolic pressure 4.0 +/- 1.2 vs 4.5 +/- 1.4 mm Hg, dP/dt 148 +/- 18 vs 187 +/- 19, P less than 0.05 at a balloon volume of 20 ml at a time corresponding to 1 hr of reperfusion in the CCA group). These results are consistent with the postulate that conventional techniques of cardioplegic arrest are associated with ischemic dysfunction of the right atrium.