Dong Seop Jeong

Midterm angiographic follow-up after off-pump coronary artery bypass: Serial comparison using early, 1-year, and 5-year postoperative angiograms

OBJECTIVE We analyzed the angiographic changes of the anastomotic sites at three time points for 5 years after off-pump coronary artery bypass surgery. METHODS Of the 402 patients who underwent off-pump coronary artery bypass surgery between January 1998 and December 2001, 240 patients who received the early, 1-year, and 5-year follow-up coronary angiograms regardless of the patients anginal symptoms were studied. Morphologic changes of grafts were traced by the FitzGibbon grading system. RESULTS Overall graft patency rates (FitzGibbon grade A+B) at early, 1-year, and 5-year angiography were 98.6%, 91.9%, and 88.3%, respectively. Graft patency rates in the left anterior descending artery, left circumflex artery, and right coronary artery territories were similar at early angiograms (P = .162). However, graft patency rate in the left anterior descending artery territory was higher than that in the left circumflex artery and right coronary artery territories at both the 1-year (P < .001) and 5-year (P < .001) angiograms. Of the 31 FitzGibbon grade B arterial grafts (internal thoracic artery and right gastroepiploic artery) at early angiography, 10 became occluded and 19 became grade A at 5-year angiography. In the saphenous vein grafts, grade B lesions gradually increased during the 5 postoperative years (2.6% vs 6.5% vs 13.3%). CONCLUSIONS Midterm angiographic follow-up demonstrated acceptable patency rates of grafts after off-pump coronary artery bypass surgery. Approximately half of the FitzGibbon grade B arterial grafts in the early angiograms became grade A at 5 years after surgery, but the proportion of grade B saphenous vein grafts gradually increased over the 5 postoperative years.

Long-term experience of surgical treatment for aortic regurgitation attributable to Behçet's disease.

BACKGROUND Cardiac involvement in Behçets disease is a rare but severe complication and presents challenges to cardiac surgeons as a result of late valve detachment or pseudoaneurysms of the aortic root after valve surgery. Few reports have been published on this topic. In this article, clinical data and surgical outcomes in patients with aortic regurgitation attributable to Behçets disease were analyzed. METHODS Nineteen patients with aortic regurgitation attributable to Behçets disease were surgically treated between March 1986 and June 2008. There were 15 men and 4 women with ages ranging from 24 to 55 years (mean, 39 +/- 7 years). Mean follow-up duration from index operations was 77.4 +/- 68.1 months (range, 9 to 271 months). RESULTS Overall mortality was 47.3% (9 of 19 patients), but no early deaths occurred at index operations. All deaths occurred after second operations, and the causes of death were low cardiac output (n = 6) and sudden aggravation of aortic regurgitation (n = 3). Erythrocyte sedimentation rates and C-reactive protein concentrations were negatively correlated with event-free period. Event-free survival at 13 years was 39.2% +/- 14.1% in patients who underwent aortic root replacement, but this was 4% +/- 3.9% in patients who underwent valve replacement (p = 0.001). Event-free survival at 13 years in patients who were administered immunosuppressive therapies was 33.7% +/- 11.0% and 0% in patients not administered immunosuppressive therapy (p = 0.001). CONCLUSIONS The mortality in this condition was very high and was found to depend on levels of postoperative inflammatory markers. Aortic root replacement and postoperative immunosuppressive therapy may be helpful.

Tricuspid Reoperation After Left-Sided Rheumatic Valve Operations

BACKGROUND The management of late tricuspid regurgitation after left-sided valve operations in rheumatic patients remains controversial. The aim of this study was to analyze clinical and echocardiographic outcomes of tricuspid valve procedures after left-sided valve operations in rheumatic patients. METHODS This study enrolled 106 rheumatic patients with a history of left-sided valve operations who were undergoing tricuspid valve procedures (53 replacements, 53 repairs). Follow-up was 97% complete, with a mean follow-up of 62 ± 42 months. Clinical and echocardiographic data were analyzed. RESULTS The early mortality rate was 1.9% (2 of 106 patients). There was no significant difference in cumulative survival at 10 years between patients who underwent tricuspid valve replacement (63.1% ± 13.2%) or repair (80.7% ± 0.8%, p = 0.317). Multivariable Cox regression analysis revealed that old age (hazard ratio [HR], 6.5; p = 0.007), anemia (HR, 10.9; p = 0.004), and left ventricular ejection fraction of less than 0.4 (HR, 10.3; p = 0.001) were predictors of major adverse cardiac events. Among patients who underwent tricuspid valve repair, multivariate analysis revealed that the aortic transprosthetic mean pressure gradient at late follow-up was an independent predictor of late tricuspid regurgitation. CONCLUSIONS Tricuspid valve procedures after left-sided valve operations in rheumatic patients can be performed at low risk with good clinical outcomes. For improved clinical outcomes, early surgical intervention should be considered before the development of anemia and left ventricular dysfunction. A lower aortic transprosthetic mean pressure gradient may help prevent late progression of tricuspid regurgitation in a clinical setting.

Fate of functional tricuspid regurgitation in aortic stenosis after aortic valve replacement.

OBJECTIVE Functional tricuspid regurgitation (TR) is found not infrequently in conjunction with aortic stenosis. The aim of the present study was to evaluate the changes in TR and to identify the predictors of late progression after aortic valve replacement. METHODS We evaluated 354 patients who had undergone aortic valve replacement for aortic stenosis from January 1995 to December 2009. Patients with mitral regurgitation were excluded. Of the 354 patients, 54 had TR greater than mild. The mean follow-up duration was 4.4±4.3 years (maximum, 15). The serial echocardiographic and clinical data were analyzed. RESULTS No early mortality occurred, and the late cardiac mortality rate was 3.9% (14 of 354). In the 48 patients with TR greater than mild, TR did not improve in 23 (49.1%) during the follow-up period. Freedom from cardiac mortality at 10 years was lower in those with TR greater than mild than in patients without TR (61.6%±16.7% vs 93.0%±2.9%, P=.008). Left ventricular diastolic function correlated with right ventricular systolic pressure (P<.001) and the degree of TR during follow-up (P=.001). Multivariate analysis showed that postoperative atrial fibrillation (odds ratio, 6.8; P=.001) and the aortic transprosthetic mean pressure gradient (odds ratio, 1.1; P=.028) predicted late TR greater than mild. CONCLUSIONS Not only did TR in patients with aortic stenosis frequently persist after aortic valve replacement, it was progressive in some. This finding was associated with left ventricular diastolic dysfunction. A concomitant tricuspid valve procedure could be considered in selected patients with aortic stenosis to avoid late TR.

Iatrogenic type A aortic dissection during cardiac surgery.

We reviewed our experience of intraoperative type A aortic dissection during cardiovascular surgery. From January 1998 to May 2009, intraoperative aortic dissection occurred in 10 of 3421 cardiac surgical patients (M:F=4:6, 62.4+/-8.0 years). Preoperative diagnoses were valvular heart disease (n=6), ischemic heart disease (n=2), combined disease (n=1) and aortic aneurysm (n=1). All underwent total circulatory arrest (TCA) with retrograde cerebral perfusion and the torn aorta was replaced (n=8) or repaired (n=2). Iatrogenic type A dissection occurred in 0.29% of patients. It was related with cannulation of ascending aorta (n=4), axillary artery (n=2), aortic root (n=2), and femoral artery (n=1) and aortotomy repair (n=1). Mortality rate was 40% (4/10). After adoption of routine intraoperative transesophageal echocardiography, mortality rate decreased from 75% (3/4) to 17% (1/6) (P=0.190). We initiated TCA before achieving deep hypothermia in three of four non-survivors. There was a trend of increased mortality when the disease extended beyond aortic arch (67%, 4/6 vs. 0%, 0/4; P=0.076). Although intraoperative aortic dissection occurred in <0.3% of our patient population, mortality was high, especially when it extended beyond the arch vessels. Better results were expected when early recognition and proper treatment under deep hypothermic circulatory arrest could be performed.

Management of acute massive pulmonary embolism: Is surgical embolectomy inferior to thrombolysis?☆

BACKGROUND Although current guidelines for pulmonary embolism (PE) treatment recommend surgical embolectomy when thrombolysis is contraindicated or has failed, their clinical outcomes rarely have been compared directly. METHODS After excluding patients aged under 18 years and those with submassive or non-massive PE, 45 consecutive patients (median age, 68 years; 62% female; 31% experienced cardiac arrest before PE treatment onset; 33% had cancer diagnosis history; and 29% received extracorporeal membrane oxygenation [ECMO]) who underwent only thrombolysis (TL group; n=19) or surgical embolectomy (SE group; n=26, including 4 who had failed thrombolysis) for acute massive PE from 2000 to 2013 at Samsung Medical Center were enrolled to assess cardiac mortality as primary outcome. RESULTS Median follow-up duration was 17.2 months. In the SE group, significantly higher proportions of patients had recent surgery and ECMO. Overall 30-day all-cause mortality rate was 24% (n=11), without significant difference between the SE (15%) and TL (37%) groups (P=0.098); however, cardiac mortality rate was significantly higher in the TL than SE group (Log rank P=0.023). TL was an independent multivariate predictor of cardiac death (P=0.03). CONCLUSION In this small retrospective single center experience, surgical embolectomy is associated with lower cardiac mortality risk than thrombolysis, which might render it first-line treatment option for acute massive PE for patients without life-limiting comorbidities.

Comparison of Right Internal Thoracic Artery and Right Gastroepiploic Artery Y Grafts Anastomosed to the Left Internal Thoracic Artery

BACKGROUND Early and 1-year results of arterial Y composite grafts anastomosed to the in situ left internal thoracic artery were studied. METHODS Three hundred twelve patients who underwent off-pump coronary artery bypass using arterial Y composite grafts for revascularization of the left coronary artery territory were analyzed. A skeletonized right internal thoracic artery (RITA) or right gastroepiploic artery (RGEA) was anastomosed to the side of the left internal thoracic artery to construct a Y composite graft. Propensity-matched analysis was used to match patients using RITA (RITA group, n = 102) with patients using RGEA (RGEA group, n = 102). Postoperative coronary angiographies were performed early (200 of 204; 1.8 +/- 1.7 days) and 1 year (171 of 204, 11.3 +/- 2.5 months) postoperatively. RESULTS There were no differences in postoperative mortalities (1 of 102 versus 2 of 102; p = 1.000) and morbidities including atrial fibrillation, mediastinitis, and perioperative myocardial infarction between the RITA and RGEA groups (not significant). Early and 1-year postoperative angiographies showed that there were no significant differences in patency rate between the two groups (early, 99.4% versus 99.3%; p = 1.000; 1-year, 95.4% versus 97.4%; p = 0.251). When the early and 1-year patency rates were compared based on the side-arm graft used, there were no differences in patency rates of RITA versus RGEA grafts between the two groups (early, 99.4% versus 100%; p = 1.000; 1-year, 96.5% versus 97.7%; p = 0.724). CONCLUSIONS Construction of Y composite grafts using the RITA or RGEA showed comparable results including patency rates early and 1 year postoperatively.

Long-Term Results of the Leaflet Extension Technique in Aortic Regurgitation: Thirteen Years of Experience in a Single Center

BACKGROUND We evaluated the effectiveness and durability of the leaflet extension technique for correction of aortic regurgitation (AR) and the long-term clinical results. METHODS Between March 1995 and August 2004, 41 consecutive patients were included. The mean age was 32.2 +/- 13.9 years. The causes of AR were rheumatic in 31 patients (75.5%), degenerative in 2 patients (4.9%), bicuspid aortic valve in 4 patients (9.8%), infective endocarditis in 1 patient (2.4%), and congenital in 3 patients (7.3%). Leaflet extensions were performed in three leaflets for 32 patients, two leaflets for 3 patients, and only one leaflet for 6 patients. The mean follow-up duration was 92.9 +/- 48.4 months. RESULTS There were no early deaths and 2 late deaths. One patient died of cancer and the other patient died of infective endocarditis. The cardiac-related mortality was 2.4% (1 of 41 patients). During a mean follow-up of 7 years, severe AR was detected in 1 patient and moderate AR in 6 patients (17.0%; 7 of 41 patients). The causes of recurrent AR were infective endocarditis in 3 patients, disease progression in 3 patients, and Behçets diseases in 1 patient. We performed 6 reoperations (14.6%), 3 in patients owing to infective endocarditis, 2 in patients owing to disease progression, and 1 in a patient owing to the suture dehiscence associated with Behçets disease. The cumulative survival was 92.6% at 13 years. Freedom from recurrent AR was 97.5% at 5 years, 81.7% at 10 years, and 68.1% at 13 years. CONCLUSIONS The long-term durability of the leaflet extension technique was acceptable. The reoperations increased with time, but pericardial leaflet dysfunction was not the cause.

Revascularization for the Right Coronary Artery Territory in Off-Pump Coronary Artery Bypass Surgery

BACKGROUND Graft selection for the right coronary artery territory remains controversial. The objective of this study was to analyze outcomes from revascularization of the right coronary artery territory using the right internal thoracic artery (RITA), the right gastroepiploic artery (RGEA), or a saphenous vein graft (SVG). METHODS Between January 2001 and December 2010, 1,434 patients who underwent off-pump coronary artery bypass surgery using the bilateral mammary arteries were enrolled. Propensity score analysis was used to match patients who underwent revascularization of the right coronary artery territory with the RITA in a Y-composite fashion (RITA group, n=292), the RGEA (RGEA group, n=292), and the SVG (SVG group, n=292). Clinical and angiographic data were analyzed. RESULTS There were no intergroup differences in terms of in-hospital mortality (0.3% [1 of 292], 0% [0 of 292], and 1% [3 of 292], p=0.332). Freedom from major adverse events including death, stoke, myocardial infarction, and reintervention at 10 years was similar among the three groups (87.8%±3.0% in the RITA group versus 92.4%±1.7% in the RGEA group versus 86.7%±3.4% in the SVG group; p=0.466). A stratified regression analysis showed that use of the saphenous vein was predictive of graft failure (p=0.044, hazard ratio 3.9). Proximal stenosis (<90%) was predictive of graft failure in the arterial groups (p=0.024, hazard ratio 3.1), but not in the SVG group (p=0.112). CONCLUSIONS Arterial grafts should be considered the first choice for right coronary artery territory revascularization in off-pump coronary artery bypass. However, SVG should be considered when proximal stenosis is less than 90%.

Long-Term Clinical Impact of Functional Mitral Regurgitation After Aortic Valve Replacement

BACKGROUND We evaluated the impact of functional mitral regurgitation (MR) on clinical outcomes and to identify predictors of residual MR after aortic valve replacement in aortic stenosis. METHODS Three hundred and eighty-four patients who underwent primary aortic valve replacement for aortic stenosis were enrolled. Patients were divided into the no-MR group (no or trivial MR; n = 270) and the MR group (mild to moderate MR; n = 114). In the MR group, 19 patients underwent concomitant mitral valve repairs. Mean follow-up duration was 4.5 ± 3.7 years (range, 1 to 15 years). Clinical and echocardiographic data were analyzed. RESULTS There was no operative mortality, but there were 9 late cardiac deaths (2.3%). Freedom from cardiac death at 14 years was lower in the MR group than in the no-MR group (77.8% ± 12.6% versus 97.7% ± 1.4%, respectively; p = 0.045), and freedom from heart failure events at 10 years was also lower in the MR group (60.8% ± 13.4% versus 92.6% ± 2.2%, p = 0.043). On multivariate analysis, preoperative atrial fibrillation and left ventricular ejection fraction greater than 40% were predictors for residual MR at late follow-up in the untreated MR group. Cox regression analysis demonstrated that postoperative moderate MR predicted late cardiac death (p = 0.016, hazard ratio 5.2). In the MR group, the incidence of residual MR in patients who underwent mitral valve repair was 5.6% (versus 30.7% in patients without mitral valve repair, p = 0.001). CONCLUSIONS Functional MR in aortic stenosis was related to poor clinical outcomes. The results of this study suggest that concomitant mitral valve procedures could be considered in selected patients with aortic stenosis and functional MR.