Isolated tricuspid valve surgery for severe tricuspid regurgitation following prior left heart valve surgery: analysis of outcome in 34 patients.

BACKGROUND AND AIMS OF THE STUDY: Patients with symptoms of right heart failure due to severe tricuspid regurgitation following a prior operation on left heart valves present a difficult problem. The outcome of tricuspid surgery in this setting is not well defined. We describe a single-center experience of isolated tricuspid valve surgery after prior left heart valve surgery, and analyze potential risk factors for a poor outcome. METHODS: Thirty-four patients who underwent isolated tricuspid valve operation for severe tricuspid regurgitation following prior valvular surgery for left-sided valve disease between 1980 and 1997 were identified. Charts were reviewed for clinical, echocardiographic, catheterization and surgical data. Follow up of survivors was conducted by telephone to ascertain functional status. RESULTS: Three patients died in hospital (early mortality rate, 8.8%). At a follow up of 71 +/- 39 months, 13 patients were alive and 21 reached an end-point (three cardiac reoperations, 18 deaths). Event-free actuarial survival at five years was 41.6 +/- 9.2%. Patients who were alive at follow up had a mean NYHA functional class of 2.1 +/- 0.6 compared with 3.4 +/- 0.5 preoperatively; 85% of survivors were symptomatically improved. Predictors of poor outcome were: increased age at the time of tricuspid surgery (p = 0.0007) and higher number of prior cardiac operations (one versus two or three, p-value 0.01, relative risk 3.4). Pulmonary artery systolic pressure, left ventricular ejection fraction, right ventricular function and size, annulus diameter, tricuspid valve pathology, and valve replacement versus repair were not predictive of outcome. CONCLUSIONS: Isolated tricuspid valve surgery for severe tricuspid regurgitation following prior surgery for left-sided heart valve disease can be performed with acceptable early mortality. There remains a high late mortality that is predicted only by age and the number of previous cardiac operations. However, in this selected group of severely symptomatic patients, significant improvement in symptoms are achieved in the survivors.

Aortic valve homografts in adults: a clinical perspective.

Aortic valve replacement is a lifesaving measure in patients with severe aortic valve disease. In the United States, the most commonly used prostheses are the mechanical and bioprosthetic valves. With mechanical valves, long-term anticoagulation is necessary because of high thrombogenic potential. Bioprosthetic valves have a relatively high incidence of structural failure, especially in younger patients. Aortic valve homografts, derived from human heart donors or autopsy material, provide an alternative to mechanical or animal valves. The advantages of the homograft in comparison with the mechanical prostheses are the low incidence of thromboembolism without anticoagulation and lower valvular gradients in smaller sizes. Homografts are relatively resistant to endocarditis and are the valve of choice during active endocarditis. Their major mode of failure has been aortic regurgitation; however, recent advances in preservation and operative techniques have decreased this problem. Whether implantation of an aortic valve homograft should be the procedure of choice in subsets of patients remains controversial. Herein we review the history, techniques, results, complications, and current indications for aortic valve homografts.

Enhanced angiogenesis and unfavorable remodeling in injured porcine coronary artery lesions: effects of local basic fibroblast growth factor delivery.

There is interest in the role of growth factors in the genesis of arterial remodeling. We studied local administration of basic fibroblast growth factor (bFGF) to coronary lesions to determine whether there is a difference in remodeling and whether neovascularization could be induced in such stenoses and distal myocardium. Pigs were randomized to balloon infusion of either saline or bFGF at each thermally injured arterial site. After the animals were killed, their internal elastic lamina, neointima, and lumen areas were measured. Capillaries were counted in the arteries and myocardium. There was a greater loss of lumen and internal elastic in the bFGF group. The neointima, media, and myocardium in the bFGF treated arteries had statistically more capillaries. This study showed that local intracoronary bFGF, at a dose that results in arterial luminal revascularization in injured segments, adversely affects arterial remodeling. Thus, the angiogenic response to exogenous bFGF may be offset by concomitant shrinkage of injured arterial segments.

Arterial remodeling after experimental percutaneous injury is highly dependent on adventitial injury and histopathology.

BACKGROUND: The extent and nature of unfavorable geometric remodeling, especially related to the adventitia, has not been studied previously. The purpose of this study was to examine two methods of experimental arterial injury, characterize the extent of remodeling, and determine if remodeling is injury-specific. METHODS: Two methods for producing coronary stenoses in pigs were used: heat injury using thermal balloon angioplasty (resulting in adventitial fibrosis), and copper stent implantation (resulting in intense inflammation). Histomorphometric parameters included changes in neointimal thickness (delta neointima) from uninjured to injured sections, and differences in area circumscribed by the internal and external elastic laminas (delta internal elastic lamina area and delta external elastic lamina area, respectively). Remodeling was calculated for each lesion as the enlargement of the external elastic lamina area or internal elastic lamina area for incremental neointimal thickening, expressed as the slopes delta external elastic area/delta neointima and delta internal elastic lamina area/delta neointima. RESULTS: Remodeling indices for the heat lesions for the heat lesions were negative (delta internal elastic lamina area/delta neointima = 0.15, delta external elastic lamina area/delta neointima = 0.64) and indicated little remodeling in contrast to copper stent injury (delta internal elastic lamina area/delta neointima = 0.95, delta external elastic lamina area/delta neointima = 1.20). CONCLUSIONS: Remodeling in fibrotic compared to inflammatory lesions differs markedly, and may explain increased restenosis rates observed in thermal balloon angioplasty in patients. This formulation may be useful to study remodeling and restenosis following interventional technologies.

Systolic aortic valve compression from partial dehiscence of an aortic valve homograft.

Implantation of valve prostheses provide improvement of symptoms and prolongation of life in selected patients with valvular heart disease. Meticulous follow-up of patients after valve surgery is essential as complications of valve failure, valve dehiscence, valve thrombosis, and infection may occur. The major mode of failure of aortic valve homografts is valve regurgitation, which is readily detected by physical examination. We report a case of left ventricular outflow obstruction after implantation of an aortic valve homograft.

Porcine model of stent thrombosis: platelets are the primary component of acute stent closure.

Acute stent thrombosis remains a major concern of coronary stent implantation. Animal studies using stents do not adequately mimic this clinical problem, since stent placement is rarely associated with acute closure. The purpose of this study was to develop and characterize a porcine model of stent thrombosis. Improved understanding through such a model may be useful toward preventing and treating acute stent closure. Whole blood was drawn from domestic crossbred swine one day before study. Platelets were isolated, labeled with 111-In tropolone, and reinjected within 18 hr of the study. Bilateral carotid arteries were exposed, and severe injury induced by a series of mechanical crushes. This method produced histologic injury similar to human coronary angioplasty, with medial disruption and large dissections protruding into the lumen. Stenting was performed in standard fashion with 3.5-mm JJIS stents. Local platelet deposition was measured and recorded as 111-In radioactivity using a miniaturized scintillation detector (Dosimeter Corp.) mounted directly at the artery injury site. This measurement was made in real time at 1-min intervals. Similarly, volumetric blood flow was measured in real time by Doppler flowmeter. Eighteen arteries of nine pigs were studied. In nine arteries from nine pigs, crush injury only was performed and monitored. In the contralateral artery, crush injury was followed immediately by placement of a 3.5-mm Palmaz-Schatz (coronary) stent. Blood flow decreased rapidly following injury in both groups and followed a cyclic pattern. Eight arteries of the crush alone and two arteries of the crush plus stent groups were totally occluded 1 hr after crush. 111-In counts normalized to baseline were significantly higher at 1 hr in both groups compared to baseline; in the stented group, counts were higher than in the unstented group. Blood flow was higher in the stented group than in unstented group for 1 hr. Histopathologic observation of the thrombi forming in both crush-only and crush-stent injuries showed severe medial dissections with obstructing medial flap formation. The thrombi forming in both groups were highly platelet rich. This model of stent and arterial thrombosis showed rapid formation of platelet-rich thrombus, cyclic blood flow variations, and acute occlusion in 20% of cases. Stent placement at arterial injury sites is associated with thrombus that is predominantly platelet rich. Stent placement at injury sites enhances platelet deposition over crush injury alone. Despite greater numbers of platelets, as shown by increased 111-In counts, stenting improved vessel patency. These were likely due to higher volumetric blood flow, continuous deposition, and embolization of labeled platelets.