Surgery for primary cardiac tumors. Clinical experience and surgical results in 60 patients.

BACKGROUND: We reviewed our clinical experience with primary cardiac tumors, attempting to clarify the surgical management of these rare entities. METHODS: Between October 1978 and November 1999, we experienced 60 surgical cases of primary cardiac tumors. There were 23 male and 37 female patients (age range, 7 months to 84 years). Tumors included the following 3 groups: myxomas (n=49), nonmyxoma benign tumors (n=3), and malignant tumors (n=8). We reviewed the presenting symptoms, diagnostic data, anatomical findings, and surgical techniques, and evaluated the surgical RESULTS. Late follow-up was 95% complete (mean follow-up, 7.7+/-7.1 years). RESULTS: Tumors produced obstructive, embolic, and/or constitutional symptoms in most cases. Generally, echocardiography alone gave sufficient information for operation. Full-thickness excision was performed in 42 patients with myxoma. Complete excision was achieved in all of the nonmyxoma benign tumors and in none of the malignancies. Early mortalities in the 3 groups were 8.2% (4/49), 0% (0/3), and 12.5% (1/8), respectively. Late mortalities were 9.5% (4/42), 0% (0/3), and 100% (7/7), respectively. One patient with myxoma had recurrence, the cause of which was likely to be inadequate resection. The late deaths in patients with malignancies were due to metastasis or local recurrence. CONCLUSIONS: Benign tumors are generally curable if surgically excised. Preoperative refractory cardiac dysfunction or embolism should be avoided by the accurate evaluation on echocardiography. The prognosis of malignant tumors is poor if they are only debulked. However, aggressive surgery that can palliate obstruction and allow time for adjuvant therapy should be carried out.

Obstruction of St Jude medical valves in the aortic position: a consideration for pathogenic mechanism of prosthetic valve obstruction.

Between 1995 and 2000, 8 patients with St. Jude Medical (SJM) valves in the aortic position required 9 redo valve replacement for prosthetic valve obstruction. Obstruction of the prosthetic valve was diagnosed by simultaneous echocardiography and cineradiography, and process of restricted leaflet movement that progressed to hemodynamic impairment was observed by serial studies in three recent patients. An oral anticoagulation was considered to be adequate in all patients except one patient who had withdrawal of warfrain. Pannus was the sole cause of valve obstruction in seven events in 6 patients, and both thrombus and pannus in 2 patients. Pannus overgrowth was found on the inflow aspect of the SJM valve, and involved the ends of the straight edge of the leaflets over pivot guards. These results suggest that pannus might play the primary role in development of obstruction of aortic SJM valves in patients on adequate oral anticoagulation.

Influence of renal function on serum and urinary heart fatty acid-binding protein levels.

BACKGROUND: Serum heart fatty acid-binding protein (H-FABP) has been reported to be a sensitive and early indicator of myocardial damage. However, circulating H-FABP may be cleared considerably from kidney, similar to that found for myoglobin. Therefore, the possibility exists that any change in renal function affects serum H-FABP concentration, and thus leads to erroneous interpretation. To evaluate the influence of renal function on H-FABP levels, we conducted a prospective study. METHODS: Nineteen patients undergoing isolated primary coronary artery bypass grafting were enrolled in this study. The patients were classified by the preoperative creatinine clearance into two groups: the control group (n=12); patients with creatinine clearance of 40 mL/min or greater, and the renal dysfunction group (n=7); patients with creatinine clearance of less than 40 mL/min. Serum H-FABP, CK-MB, troponin-T and urinary H-FABP levels were measured perioperatively. RESULTS: None of the patients had perioperative myocardial infarction. No significant differences were found in CK-MB and troponin-T levels between the groups. The renal dysfunction group resulted in significantly (p<0.05) higher serum H-FABP levels and lower urinary H-FABP levels than those in the control group, postoperatively. The creatinine clearance correlated inversely with the peak levels of serum H-FABP (r=-0.75, p=0.0001) and correlated with the peak levels of urinary H-FABP (r=0.64, p=0.003). CONCLUSIONS: The results indicate that the kidneys play an important role in the clearance of serum H-FABP. Thus, caution must be taken in interpreting this marker for myocardial damage during cardiac surgery in patients with renal dysfunction.

Repair of left ventricular rupture following mitral valve replacement concomitant with left atrial reduction procedure--intracardiac patch and extracardiac buttress suture.

Rupture of the posterior wall of the left ventricle after mitral valve replacement is a dire complication associated with a very high mortality. This study reports a successful repair of type I left ventricular rupture, which occurred after mitral valve replacement concomitant with a left atrial reduction procedure, by combination of an intracardiac patch and an extracardiac buttress suture. In a case such as this, in which hemostasis is quite difficult to establish, this combination technique is particularly effective.

Antiinflammatory effects of colforsin daropate hydrochloride, a novel water-soluble forskolin derivative.

BACKGROUND: To evaluate the effects of colforsin daropate hydrochloride (colforsin), a water-soluble forskolin derivative, on hemodynamics and systemic inflammatory response after cardiopulmonary bypass, we conducted a prospective randomized study. METHODS: Twenty-nine patients undergoing coronary artery bypass grafting were randomized to receive either colforsin treatment (colforsin; n = 14) or no colforsin treatment (control; n = 15). Administration of colforsin (0.5 microg.kg(-1).min(-1)) was started after induction of anesthesia and was continued for 6 hours. Perioperative cytokine and cyclic adenosine monophosphate levels, hemodynamics, and respiratory function were measured serially. RESULTS: Marked positive inotropic and vasodilatory effects were observed in patients receiving colforsin. Interleukin 1beta, interleukin 6, and interleukin 8 levels after cardiopulmonary bypass were significantly (p < 0.05) lower in the colforsin group. Plasma levels of cyclic adenosine monophosphate increased significantly (p < 0.05) in the colforsin group, and the levels correlated inversely (r = -0.56, p = 0.002) with the respiratory index after cardiopulmonary bypass. CONCLUSIONS: Intraoperative administration of colforsin daropate hydrochloride had potent inotropic and vasodilatory activity and attenuated cytokine production and respiratory dysfunction after cardiopulmonary bypass. The results indicate that the technique can be a novel therapeutic strategy for the systemic inflammatory response associated with cardiopulmonary bypass.

[Effects of colforsin daropate hydrochloride in patients undergoing coronary artery bypass surgery].

The effects of Adehl, colforsin daropate hydrochloride, on hemodynamics were studied in patients undergoing cardiac surgery. Twenty-six patients who underwent coronary artery bypass grafting were divided into two groups according to the intraoperative administration of Adehl. The control group (n = 14) received no Adehl treatment and the Adehl group (n = 12) received Adehl infusion immediately after anesthesia induction (0.5 microgram.kg-1.min-1) for 6 hours. Hemodynamic measurements and clinical results were accessed perioperatively. The Adehl group resulted in significantly (p < 0.05) lower pulmonary capillary wedge pressure and systemic vascular resistance, and significantly (p < 0.05) greater cardiac indices and left ventricular stroke work indices than those in the control group. No significant difference was found in the rate pressure product between the groups. The Adehl group resulted in significantly shorter duration of intubation and ICU stay. Adehl was not associated with a significant increase in the prevalence of adverse effects. The results suggest that Adehl has positive inotropic and vasodilator effects without increasing myocardial oxygen consumption. Thus, it is suggested that Adehl can be a useful agent for the perioperative management in patients undergoing cardiac surgery.

Surgical treatment of prosthetic valve endocarditis with left ventricular-aortic discontinuity: reconstruction of the left ventricular outflow tract with a xenopericardial conduit.

BACKGROUND AND AIM OF THE STUDY: Aortic prosthetic valve endocarditis (PVE) with annular destruction presents a challenge that requires techniques to eradicate the infection and correct the hemodynamic abnormality. METHODS: Between July 1, 1996 and March 31, 2000, six patients with native or PVE of the aortic valve and aortic annular destruction underwent surgical treatment. Of these patients, three (two men, one woman; mean age 71.0 years) had circumferential annular destruction of the aortic annulus, and formed the basis of this study. The microorganisms responsible for the infection were Streptococcus spp. in two patients and Staphylococcus aureus in one patient. In addition to aggressive debridement of the infected tissue, repair was achieved by reconstruction of the left ventricular outflow tract with a xenopericardial conduit and fixation of the new prosthetic valve to the conduit. RESULTS: One patient with ventricular septal perforation, multiple systemic embolism and sepsis died of low cardiac output syndrome soon after surgery. Two operative survivors were followed up for 9 and 51 months, with no late deaths. No patient has experienced recurrent infection, pericardial patch aneurysm, or prosthetic valve detachment. CONCLUSION: These operative procedures provide easy and secure fixation of the pericardial patch to the healthy tissue under excellent operative view, as well as a sturdy structure for the fixation of the new prosthesis, and complete exclusion of the abscess cavity from the blood stream.

Triple valve replacement with bileaflet mechanical valves: is the mechanical valve the proper choice for the tricuspid position?

A more durable mechanical valve may be a better choice for the tricuspid position than a bioprosthesis when the patient already has mechanical prosthesis in the left side of the heart. Eleven cases of triple valve replacement (total follow-up period, 49.5 patient years), all with mechanical valves, are reviewed to assess optimal valve selection. Nine patients had undergone a total of 12 previous cardiac surgeries. Three patients died in hospital (27.3%), but there were no late deaths among the survivors. Two cases of valve thrombosis in the tricuspid position occurred (linearized incidence: 4.04%/patient years) and 1 of these required reoperation. Because of this high incidence of valve thrombosis, the bileaflet mechanical valve is not considered to be the best choice. Even if mechanical valves are implanted in the left side of the heart, a bioprosthesis may be a better choice at the tricuspid position.

Cardio-aortic operation in octogenarians.

OBJECTIVE: Aging of the population is a current phenomenon in Japan, and life expectancy at 80 years old is getting longer. So we reviewed cardio-aortic operations on octogenarians at our institution. SUBJECTS AND METHODS: Thirty-three consecutive octogenarian patients who had undergone cardio-aortic operations from 1992 to 1998 were studied. There were 14 men and 19 women. The mean age was 81.9 years. Of the 33, 19 patients (58%) were in New York Heart Association class IV, and 21 patients (64%) were operated on urgently or in emergency. The procedures undergone were operation for coronary artery disease in 17 patients, operation for valvular disease in 7 patients, operation for thoracic-aorta in 7 patients, and others in 2 patients. RESULTS: The hospital mortality rate was 27% (9 patients). However, 89% of patients experiencing hospital death were in New York Heart Association class IV preoperatively and had required an emergency/urgent operation. On the other hand, there was only one hospital death (1/12, 8.3%) among the elective patients. The statistically significant risk factors for hospital death were renal insufficiency, shock, New York Heart Association class IV, intra-aortic balloon pumping, and longer cardiopulmonary bypass time. The one-, three-, and five-year-survival rate was 73%, 68%, and 55%, respectively. Of the survivors, 77% were in class I or II. CONCLUSION: Although octogenarians' hospital mortality was still very high, the mid-term results were acceptable and the survivors' quality of life was satisfactory. These data suggested that we should operate on cardio-aortic patients before they reach a very serious state, especially in octogenarians.

Coronary artery bypass grafting in patients with mild renal insufficiency.

UNLABELLED: It is well known that dialysis-dependent renal failure increases the likelihood of a poor outcome following cardiac surgery. However, it is not known whether non-dialysis-dependent mild renal insufficiency also influences clinical outcome. Fifty-five patients with non-dialysis-dependent renal insufficiency undergoing coronary artery bypass grafting (CABG) (Renal group: serum creatinine level >1.5 mg/dl) were enrolled. These patients were then matched on prognostic variables to 148 patients with normal renal function ( CONTROL GROUP: serum creatinine level <1.5 mg/dl). The early postoperative clinical results showed that patients in the Renal group were more likely to develop postoperative renal failure (18% vs 1%: p=0.0002) and hemorrhage requiring re-exploration (11% vs 2%; p=0.01). Total morbidity was significantly higher in the Renal group (40% vs 22%; p=0.01). Multivariate analysis revealed that the Renal group was the second most important predictor of morbidity (odds ratio (OR) =2.2) behind left ventricular dysfunction (OR=2.9). The Renal group was also the second most important predictor of postoperative renal failure (OR=12.5). Therefore, non-dialysis-dependent mild renal insufficiency also increases the risk of morbidity following CABG.

High-dose cimetidine reduces proinflammatory reaction after cardiac surgery with cardiopulmonary bypass.

BACKGROUND: Cimetidine, which is usually used for gastric ulcer, enhances cellular immunity. The effect of cimetidine on perioperative proinflammatory response after cardiac surgery with cardiopulmonary bypass was investigated. METHODS: Elective coronary artery bypass graft cases in which CPB was performed were placed randomly in a cimetidine (C) group (n = 20) or a no-treatment (N) group (n = 20). The time course of plasma levels of neutrophil elastase, interleukin (IL)-6 and IL-8, leukocyte counts, lymphocyte recovery ratio, C-reactive protein, creatine-kinase-MB, and oxygenation index were analyzed. RESULTS: The plasma levels of neutrophil elastase and IL-8 were inhibited in the C groups at 2 hours after CPB termination. In a comparison of the two groups, the C group demonstrated higher lymphocyte recovery ratio and lower C-reactive protein on postoperative day 5 and shorter intubation time. No intergroup differences were observed in IL-6, leukocyte counts, creatine-kinase-MB levels, or oxygenation index. CONCLUSIONS: Cimetidine may reduce surgical stress and augment the immune system after cardiac surgery with cardiopulmonary bypass.

Aortic dissections complicating open cardiac surgery: report of three cases.

Between June 1991 and February 1999, three patients suffered ascending aortic dissection as a complication of cardiopulmonary bypass operations with aortic cannulation at our hospital. The dissection occurred during the operation in two of the three patients and several months after the operation in one. Among a total of 2207 cardiac operations performed during this period, the incidence of perioperative ascending aortic dissection was 0.14%. In addition to visual inspection and palpation, either epicardial or transesophageal echocardiography proved extremely useful for establishing an intraoperative diagnosis of ascending aortic dissection as a complication of open cardiac operation. One of the three patients underwent closed plication but subsequently died of vital organ ischemia. In this case, failure of reapproximation of the injured intima by closed plication might have led to extension of the dissection. Despite prolonged cardiopulmonary bypass and myocardial ischemic time, graft replacement of the ascending aorta was successfully carried out in the other two patients. Thus, we believe that graft replacement of the ascending aorta should be performed for patients with extensive aortic dissection complicating an open cardiac operation.

Biological activity of endogenous atrial natriuretic peptide during cardiopulmonary bypass.

To evaluate the effect of cardiopulmonary bypass (CPB) on atrial natriuretic peptide (ANP) biological activity in patients undergoing cardiac operations, we conducted a prospective study. Ten patients undergoing mitral valve surgery were enrolled. Plasma levels of ANP and cyclic guanosine monophosphate (cGMP), hemodynamic variables, and renal function parameters were assessed perioperatively. The molar ratio of cGMP to ANP (as a marker for ANP biological activity) decreased significantly (p < 0.05) during CPB despite similar plasma ANP levels. The ratio correlated inversely with the duration of CPB (r = -0.85, p = 0.002). The ratio also correlated with fractional sodium excretion (r = 0.65, p = 0.04) and correlated inversely with pulmonary vascular resistance (r = -0.79, p = 0.009) and atrial filling pressure (r = -0.84, p= 0.003) postoperatively. CPB decreased the molar ratio of cGMP to ANP, which may represent ANP biological activity, such as vasodilation and natriuresis. The phenomenon may contribute to water-sodium retention and pulmonary hypertension after cardiac surgery.

Effects of intraoperative administration of atrial natriuretic peptide.

BACKGROUND: Biological activity of endogenous atrial natriuretic peptide (ANP) may decrease during cardiopulmonary bypass. To evaluate the effects of intraoperative administration of exogenous ANP in patients undergoing cardiopulmonary bypass, we conducted a prospective randomized study. METHODS: Eighteen patients undergoing mitral valve surgery were randomized to receive either ANP treatment (ANP group; n = 9) or no ANP treatment (control group; n = 9). Atrial natriuretic peptide was given immediately after initiation of cardiopulmonary bypass for 6 hours (0.05 microg x kg(-1) x min(-1)). Plasma ANP, brain natriuretic peptide and cyclic guanosine monophosphate (cGMP) levels, hemodynamic variables and renal function were assessed perioperatively. RESULTS: Administration of ANP increased plasma cyclic guanosine monophosphate levels, urine output and fractional sodium excretion, and decreased preload, afterload and plasma brain natriuretic peptide levels significantly (p < 0.05). Plasma cyclic guanosine monophosphate levels correlated with plasma ANP levels (r = 0.95, p = 0.0001), correlated with fractional sodium excretion (r = 0.53, p = 0.02), and correlated inversely with systemic vascular resistance (r = -0.54, p = 0.02). CONCLUSIONS: Intraoperative administration of ANP had potent effects on natriuresis and systemic vasodilation by elevating cyclic guanosine monophosphate levels. The results suggest that the technique is useful for the management of hemodynamics and water-sodium retention after cardiopulmonary bypass.

Tricuspid valve replacement with the St. Jude Medical valve: 19 years of experience.

OBJECTIVE: The choice of the valve substitute in the tricuspid position remains controversial. A St. Jude Medical valve is a choice of valve substitute and its lower thrombogenicity and excellent hemodynamic performance have been reported even in the tricuspid position. However, little is known of the long-term durability of the St. Jude Medical valve in the tricuspid position. Our long-term experience of tricuspid valve replacement showed the higher thrombogenicity than we had expected, therefore, this study was done to reconsider our strategy for valve choice. METHODS: This study reviewed 23 patient who underwent 25 tricuspid valve replacements with the St. Jude Medical valves from 1980 to 1997. The mean age was 40 years. Eleven patients (48%) were men. There were four in-hospital deaths (17%). The remaining 19 patients were all alive and followed from 2.2 to 19.0 years (mean 11.8 years). RESULTS: The overall survival, including hospital mortality, was 83%, 10 and 15 years after surgery. Valve thrombosis occurred in six patients. Freedom from valve thrombosis was 78 and 70%, 10 and 15 years after surgery, respectively. The linearized rate of the valve thrombosis was 2.9%/patient-years. Six patients required reoperation. The mean interval to reoperation was 9.5 years. Freedom from reoperation was 83% and 75%, 10 and 15 years after surgery, respectively. The linearized rate of the reoperation was 2.8%/patient-years. No structural valve deterioration was found. Echocardiographic study showed that the function of the St. Jude Medical valve without valve-related complications was well maintained. CONCLUSIONS: The higher thrombogenicity of the St. Jude Medical valve in the tricuspid position altered our choice of valve substitutes from the St. Jude Medical valve to a bioprosthesis which is lack of need for anticoagulant therapy except for juvenile patients who are able to maintain potent anticoagulant therapy.

Biocompatibility of heparin-coated extracorporeal bypass circuits: new heparin bonded bioline system.

Biocompatibility of a new type of heparin-coated cardiopulmonary bypass equipment, the Bioline, was evaluated in coronary artery bypass surgery cases. The heparin-coated (H) group (n = 15; Quadrox Bioline oxygenator/reservior and Carmeda BioMedicus BP-80 centrifugal pump) was compared with the nonheparin-coated (N) group (n = 12; uncoated, otherwise similar oxygenator, centrifugal pump, tubing, and filter set). Both groups used full systemic heparinization. The peak values of neutrophil elastase, C3a, IL-6, and IL-8 at 2 h after cardiopulmonary bypass (CPB), and C3a levels at the end of CPB and at 2 h after CPB were significantly reduced in the H group compared with those of the N group. However, no statistically significant intergroup differences were observed in thrombin-antithrombin complex, D-dimer, beta-thromboglobulin, or platelet factor-4. No significant differences were observed in hemostasis time, postoperative 12 h blood loss, required amount of blood transfusion, or intubation time. In conclusion, the Bioline demonstrated partially improved biocompatibility, in terms of leukocyte and complement activation, and proinflammatory cytokine production. However, it did not improve platelet activation, coagulation, or fibrinolysis cascade under full systemic heparinization. As a result, the clinical beneficial impact seemed to be the minimum.

Effects of angiotensin-converting enzyme inhibitor during warm blood cardioplegia.

BACKGROUND: Effects of captopril, an angiotensin-converting enzyme inhibitor, during warm blood cardioplegia were assessed in the blood-perfused, isolated rat heart. METHODS: The isolated hearts were arrested for 60 minutes with warm blood cardioplegia given at 20-minute intervals and were reperfused for 60 minutes. The control group (n = 10) received standard cardioplegia and the captopril group (n = 10) received cardioplegia supplemented with captopril (2 mmol/L). Cardiac function, myocardial metabolism, and cardiac release of circulating adhesion molecules were assessed before and after cardioplegic arrest. RESULTS: Left ventricular end-diastolic pressure and -dp/dt were significantly (p<0.05) lower and coronary blood flow was significantly (p<0.05) greater in the captopril group than the control group during reperfusion. The captopril group resulted in significantly (p<0.05) less cardiac release of lactate, thiobarbituric acid reactive substances during reperfusion. Cardiac release of intercellular adhesion molecule-1 was significantly (p<0.05) less in the captopril group at 60 minutes of reperfusion. CONCLUSIONS: The results suggest that supplementation of captopril during warm blood cardioplegia provides superior myocardial protection by suppressing lipid peroxidation and leukocyte-endothelial cell interaction during reperfusion.

Obstruction of St Jude Medical valves in the aortic position: significance of a combination of cineradiography and echocardiography.

BACKGROUND: Obstruction of the St Jude Medical valve (St Jude Medical, Inc, St Paul, Minn) is a rare but serious complication. METHODS: Cineradiographic and echocardiographic evaluations of aortic St Jude Medical valves were simultaneously performed on 54 patients, with no signs of prosthetic valve dysfunction late after surgery. RESULTS: Although closing angles of the leaflets corresponded closely with the manufacturer data, restricted opening of the leaflets (opening angle >/= 20 degrees ) was found in 16 (group D) of the 54 patients by means of cineradiography. The opening angles were equal to or less than 14 degrees in the other 23 patients (group N) and between 15 degrees and 19 degrees in the remaining 15 (group M). Doppler-derived transprosthetic pressure gradients were significantly higher (P =.03) and the velocity index was significantly lower (P =.003) in group D than in group N. However, no significant differences were found in those values between group N and group M. Replacement of the aortic St Jude Medical valves was performed in 5 of the 16 patients, and the remaining 11 have been followed up because of relatively low pressure gradients. The cause of restricted leaflet movement was pannus formation without thrombosis in 4 patients and valve thrombosis with pannus formation in one. CONCLUSIONS: Reduced valve orifice area and restricted opening of the leaflets resulting from excess growth of pannus probably led to obstruction of the aortic St Jude Medical valves. A combination of cineradiography and echocardiography makes it possible to provide an accurate and detailed diagnosis of obstruction of the valve.

ATS prosthetic valve motion: an in vitro analysis.

BACKGROUND AND AIM OF THE STUDY: We have reported clinical findings that normally functioning open pivot ATS valves did not open completely. In order to analyze features of the ATS valve motion more precisely, in vitro tests were conducted. METHODS: Opening angles and pressure gradients of the ATS valve were measured and compared with those of the St. Jude Medical (SJM) valve under steady flow, but with various outlet configurations. Second, opening angles of the two valves were measured under pulsatile conditions in two different outflow configurations: (i) a 'straight outlet' where the leaflets did not extend into the tapering outflow chamber, and (ii) an 'abrupt enlargement outlet' where the leaflets extended directly into the enlarged outflow space. Third, flow visualization studies were made under steady flow conditions in the straight and abrupt enlargement outlet conduits, respectively. RESULTS: Under steady flow conditions, opening of the ATS valve was restricted in most outflow configurations; only when the outlet angle was 0 degrees did the valve open fully. The SJM valve opened completely in all downstream configurations. Despite restricted opening in the ATS valve, the pressure gradient was similar in both valves. Under pulsatile conditions, both valves opened fully in the straight outlet; however, in the abrupt enlargement outlet the ATS valve opened incompletely and the SJM valve completely. Substantial turbulent flow was observed at the outside of the leaflet and corners of the conduit, notably with the ATS valve. CONCLUSION: This study showed that the ATS valve did not open fully except when the outflow was straight, and the leaflet did not extend into an enlarged downstream chamber. Structural features of the ATS, such as its axis being located close to the straight edge and its leaflets extending further downstream from the ring orifice, may cause this unique valve behavior.

Plasma exchange for hyperbilirubinemia following implantation of a left ventricle assist system: a case report.

A 49-year-old patient with end-stage dilated cardiomyopathy underwent implantation of a left ventricular assist system (LVAS). Although the systemic circulation seemed to be improved, the serum total bilirubin (Tbili) level increased sharply in the early postoperative period (preoperative Tbili, 5.7 mg/dl; postoperative day 3, 33.6 mg/dl). Plasma exchange (PE) was performed 7 times from postoperative day 4, and the Tbili level decreased to 16.3 mg/dl by postoperative day 11. Thereafter, serum Tbili normalized concomitant with improved circulatory condition. The cause of the hyperbilirubinemia was considered to be temporary right ventricular dysfunction or hepatic sinusoid endothelial dysfunction. The liver function was recoverable, so PE had been effective in this case. Unfortunately, the patient suffered a midbrain infarction and ultimately died. From this experience, PE is recommended if it is judged that liver function can be preserved and circulation is adequate, but its implementation should not be delayed. It is essential that LVAS is implanted before damage occurs to end-organ function and thus prevent hyperbilirubinemia.