Author + information
- Received September 18, 2008
- Revision received January 21, 2009
- Accepted February 6, 2009
- Published online May 19, 2009.
- Shahbudin H. Rahimtoola, MB, FRCP, DSc (Hon)* ()
- ↵*Reprint requests and correspondence:
Dr. Shahbudin H. Rahimtoola, University of Southern California, 2025 Zonal Avenue, GNH 7131, Los Angeles, California 90033
Aortic Stenosis (AS)
The following reviews were published during the year: “Insights into degenerative aortic valve (AV) disease” (1) and “Genomics: The next step to elucidate the etiology of calcific aortic stenosis (AS)” (2).
In 2008, Ross published “Transseptal left heart catheterization: a 50 year odyssey” (3).
Comment.Transseptal left heart catheterization, a pioneering procedure, was developed by John Ross Jr. when he was a surgical intern on a research rotation. He changed his specialty and went on to become a distinguished figure in cardiovascular medicine.
Another article published at this time is “Emerging Applications for Transseptal Catheterization: Old Techniques for New Procedures” (4).
Grading of AS severity; inconsistencies of echocardiographic criteria
The consistency of 3 echocardiographic criteria for severe AS (aortic valve area [AVA] <1.0 m2[or <0.6 cm2/m2], mean gradient [MG] >40 mm Hg or peak flow [Vmax] >4 m/s) used in the Guidelines were evaluated in 3,483 echocardiographic studies performed in 2,427 patients with normal left ventricular (LV) systolic function and calculated AVA of ≤2 cm2performed from 1994 to 2004 (5). There were “clinically significant” differences (Fig. 1,Table 1).
Comment.This important and useful study highlights inconsistencies in criteria for severe AS. An accompanying editorial (6) discusses the importance of this study and probable reasons for the differences. The problem may be the guidelines. In this study, it would have been better to have measured cardiac output (CO). The authors acknowledge that calculating stroke volume from echocardiography is “often difficult.” They assumed a CO of 6 l/min, heart rate of 80 beats/min, and a systolic ejection time of 0.33 s. (Also see Gutierrez-Chico et al.  cited below.)
Increasing blood pressure (BP) affects measured AS severity
In 22 patients, BP was increased by handgrip or phenylephrine infusion from a mean of 99 ± 8 mm Hg to 121 ± 10 mm Hg (p < 0.001) (7). The AVA decreased from 1.15 ± 0.32 cm2to 1.09 ± 0.33 cm2(p = 0.02).
Comment.Flow was not measured but was calculated. The clinical situation arises in patients with severe or borderline severe AS, low gradient, and hypertension. The clinical problem is as follows: To what level, by what method, and how rapidly should BP be lowered before the measurements are repeated?
Cardiac multidetector computed tomography (MDCT) to assess the severity of AS
Forty consecutive patients scheduled for aortic valve replacement (AVR) had echocardiography, MDCT, and coronary angiography within a time span of 1 week (8). The AVA as measured by MDCT and echocardiography was 0.87 ± 0.22 cm2versus 0.81 ± 0.20 cm2(p = 0.01; r = 0.77, and p < 0.01, respectively). Mean difference was 0.06 ± 0.15 cm2. The authors found that MDCT correctly identified 26 of 33 patients (79%) with coronary artery stenosis and overestimated 3 patients with <50% stenosis.
Where is pressure recovery best measured?
In a pulse duplicator with flow rates of 2 to 5 l/min and orifice areas between 0.32 and 2.85 cm2, pressure recovery data were best measured 5 cm above AV (9), assuming that AV is a circular orifice, proximal aorta is circular without sinuses or bends, and flow is axially symmetric.
The measurement of AVA by RT3D
In 41 patients with AS, AVA was measured with the use of Gorlin's equation, Hakki's formula, Doppler continuity equation (CE), 2D Simpson's volumetric method, and by the real-time 3-dimensional echocardiography (RT3D). Cardiac output was measured by the thermodilution technique (10). RT3D had the best linear association and absolute agreement with Gorlin's method of all noninvasive methods (r = 0.902; intraclass correlation coefficient of 0.846) (Fig. 2).RT3D underestimates AVA (95% confidence interval [CI]: 0.084 to 0.193) and also grades most efficiently the severity of AS as mild, moderate, or severe (weighted kappa = 0.932). The use of receiver-operating characteristic curves showed the optimal cutoff point to consider severe AS “remains close to 1 cm2(1.06 cm2).”
Comment.AVA by Gorlin's formula is very good if data (pressures and CO) are both obtained and calculated appropriately.
Aortic valve replacement (AVR)
“Low Flow/Low Gradient” AS
A total of 217 consecutive patients in a European Multicenter Study with severe AS (AVA <1.0 cm2), low left ventricular ejection fraction (LVEF) (≤35%), and low MG (≤30 mm Hg) underwent AVR. Operative mortality was 20% from 1990 to 1999 and 10% from 2000 to 2005 (11). On multivariate analysis, very low MG and multivessel coronary artery disease (MVD) were predictors of excess perioperative mortality. In the subgroup who underwent dobutamine stress echocardiography, on multivariate analysis, a lack of “contractile reserve” (odds ratio [OR]: 4.4; 95% CI: 1.5 to 17.5; p = 0.03) and MVD (OR: 6.0; 95% CI: 1.5 to 2.4, p = 0.01) were strong predictors of operative mortality. The 5-year survival was 49 ± 4%. Lower MG, greater score on the European system for cardiac operative risk evaluation (EuroSCORE), previous atrial fibrillation, and MVD were identified as independent predictors of overall long-term mortality (Fig. 3).
Comment.A diagnosis of hibernating myocardium would have been an important part of pre-operative testing (12). Also, it would be best to perform right heart catheterization to document cardiac index/stroke index were “low.”
Severe AS, normal LVEF, and low MG
A total of 215 patients had isolated AS, normal sinus rhythm, and normal LVEF. Patient characteristics and outcome in those with MG <30 and ≥30 mm Hg (13) are shown in Table 2.The patients with MG <30 mm Hg had a lower referral to AVR and a greater death rate both before and after AVR.
Comment.This was a large study in which cardiac catheterization was performed. It documents in detail patient characteristics and hemodynamics in the 2 subgroups of patients.
Asymptomatic patients with severe AS
Of 622 patients who were initially asymptomatic between 1984 and 1995, only 263 of 622 patients who subsequently had surgery at the authors' institution are described. Of the 263 patients, 166 had developed symptoms and 97 remained asymptomatic (14); operative mortality in the 2 subgroups was 2% and 1%, respectively. In those who had symptoms or were asymptomatic, the survival of patients who underwent surgery was better than those who did not (Fig. 4).
Comment.This study of asymptomatic patients has many areas of concern, which have been described in detail previously (15). These data are from highly selected patients. 1) Of 694 patients in the database, 72 with more severe AS had AVR initially, leaving 622 for follow-up (16,17). 2) Although 352 patients had subsequent surgery, only the 263 who had surgery at the authors' institution are described. Moreover, the criteria for selecting patients for AVR was not provided and were not pre-determined. To determine that AVR was better than no AVR, data are needed on baseline characteristics of operated versus nonoperated patients in both the symptomatic and asymptomatic subgroups, characteristics that were not provided.
The EuroSCORE overestimates operative mortality of valve surgery
A total of 1,299 patients had either isolated valve surgery procedures or valve surgery plus coronary artery bypass graft surgery (CABG) (18). The observed mortality in the 2 groups was 2.8% and 6.8%, respectively. Mortality predicted by the New York model (19) was 3.4% and 6.2%, respectively, which was much closer to the actual observed mortality. Mortality predicted by the use of EuroSCORE (20,21) was 6.1% and 7.8%.
A total of 731 patients with a EuroSCORE of 7 or greater underwent isolated AVR (22). A minimally invasive approach was used in 469 (64.2%). The mean EuroSCORE was 9.7 (median 10); 731 patients had a EuroSCORE ≥7, of whom 332 had a EuroSCORE ≥10. The actual observed operative mortality was 7.8% and that predicted by logistic EuroSCORE was 17.2%. Freedom from all-cause death (including hospital mortality) was 72.4% at 5 years (Fig. 5).
Comment.The EuroSCORE was first published in 1999 (20). It was based on a dataset of 13,302 patients, of whom approximately 60% were patients undergoing CABG, 30% undergoing valve surgery, and 10% other surgical procedures. Cardiac surgery for valve disease is constantly improving; for example, New York State Cardiac Surgery data show that hospital mortality from 2001 to 2003 was lower than from 1998 to 2000 (23).
Predictors of poorer survival after AVR
Of 3,049 patients who underwent AVR for AS from 1991 to 2004, predictors of decreased survival (24) were severe LV hypertrophy, severity of AS, use of smaller prosthesis, and ventricular dysfunction.
Aortic Regurgitation (AR)
Pre-operative LV end-systolic diameter index (ESDI) ≥25 mm/m2is a better predictor of outcomes than ESD >50 mm
A total of 147 consecutive patients age 52 ± 12 years underwent surgery for chronic severe AR for symptoms in 107 (76.3%) and 40 (23.7%) for LV dysfunction “parameters” (25). The comparison of ESDI ≥25 mm to ESD >50 mm showed that the use of ESDI better identified symptomatic patients, asymptomatic patients with LV dysfunction, patients with worse LVEF after surgery, and overall rate of mortality.
Clinical outcome in asymptomatic patients with AR
A total of 251 patients were enrolled from 1991 to 2003 and followed until their death or 2006, whichever came first. The patients were 60 ± 17 years of age with a diastolic blood pressure of 71 ± 12 mm Hg which, in the severe AR group, was 64 ± 13 mm Hg. A history of hypertension was present in 49% (26). A total of 93 patients were categorized as severe AR on the basis of regurgitant volume ≥60 ml/beat or effective regurgitant orifice ≥30 mm2from Doppler echocardiographic studies. When severe AR versus mild AR was compared, 10-year survival was lower (69 ± 9% vs. 92 ± 4%, p = 0.05) and cardiac events (cardiac death, heart failure, and new atrial fibrillation) were higher (63 ± 8% vs. 21 ± 8%, p < 0.0001). Patients with moderate AR showed “no significant reduction of cardiac events after surgery (p = 0.92),” but patients with severe AR had a reduction in cardiac events: linearized yearly rates were 11.2% “under medical management” and 2.3% with surgery (p = 0.002).
Comment.There are many concerns with this type of natural history study (15). For example: 1) At entry into the study, patients were in their 60s and 70s, but there is no information about lipid profile and the percent of patients who on entry had coronary arteriography and significantly obstructive coronary artery disease (CAD). Did none have CAD at entry? Ten years later, when patients were in their 70s and 80s, did none develop or have progression of CAD that could have led to heart failure, atrial fibrillation, or death? 2) What was “medical management”? How effective was it in controlling abnormalities?
Other concerns include the following: 1) Likely, the 10-year survival in the moderate group was not significantly lower than in the severe group (75 ± 6% vs. 69 ± 9%); the p value was not provided. This finding adds to the possibility that AR was not the cause of the event(s). 2) The diastolic blood pressure was 64 ± 13 mm Hg. Thus, AR was not severe in many patients. The American Society of Echocardiography (27,28) recommends: a) 3 quantitative criteria to define severity of AR. The severity of AR was judged by only 1 of 2 quantitative criteria—why not both? Was there discordance between the 2 criteria? What about the third criterion? b) Correction of LV dimensions and volumes for body size. These measurements were not presented, which also raises concerns about the severity of AR. 3) Cardiac surgery was stated to be performed in 80 of 81 patients for AR, but 11 had surgery for physician and/or patient preference, 11 for “aortic aneurysm,” and an unknown number “for LV enlargement,” which was undefined. In chronic severe AR, LV had to have been enlarged at baseline and the amount of additional increase of LV size/volume to warrant AVR was not described.
Bicuspid Aortic Valve (BAV)
Asymptomatic patients with BAV; low event rate at 20 years
A total of 212 asymptomatic residents of Olmsted County, Minnesota, with BAV, 65% male, ages 32 ± 20 years between 1980 and 1999, with LVEF 0.63 ± 0.05 and absent or mild AS and/or AR were followed for up to 20 years (29). The AV degeneration was subjectively scored for calcification, thickening, and mobility reduction (0 to 3 for each). Events at 20 years are shown in Table 3.Independent predictors for cardiovascular events were age ≥50 years (OR: 3.0; 95% CI: 1.5 to 5.7; p < 0.01); valve degeneration at diagnosis (OR: 2.4; 95% CI: 1.2 to 4.5; p = 0.016) (Fig. 6)with >70% events at 10 years; and baseline ascending aorta ≥40 mm for surgery for aortic dilation (OR: 10.8; 95% CI: 1.8 to 77.3; p < 0.01).
Aortic dilation has distinct patterns
Thoracic aortic morphology in 64 patients with BAV was determined at 10 levels with the use of computed tomographic angiography or magnetic resonance angiography (30). Patients in the different clusters needed different surgical techniques of aortic surgery (Fig. 7).
Comment.Patients with BAV may need surgery for the valve, the aorta, or both. It is important to evaluate and measure the ascending aorta at the annulus, sinuses, and sinotubular junction, at the largest size in the whole of the ascending aorta and in the arch of the aorta because it helps to determine the need for surgery on the aorta and also the surgical procedure that will need to be done.
Long-term survival after the Bentall procedure
A total of 206 patients underwent composite aortic root replacement (i.e., button-Bentall procedure). Hospital mortality was 2.9%, stroke rate was 1.9% (31); risk factors for these events were clot or atheroma (p = 0.02) and age >65 years (p = 0.05). Follow-up ranged from 5 to 18 years; 5- and 10-year survival rates were 93% and 89%, respectively.
Mitral Regurgitation (MR)
Role of imaging
O'Gara et al. (32) provide a useful and important description of the role of echocardiographic/Doppler imaging, including the importance of segmental anatomy in chronic “degenerative” MR.
Cardiovascular magnetic resonance (CMR) of mitral valve (MV) prolapse
CMR performed in 16 patients had a 100% sensitivity and 100% specificity compared with the use of transesophageal echocardiography (TTE) for the 2-mm threshold for leaflet excursion into the left atrium (33). It was found that CMR showed larger and longer indexed posterior leaflet and larger index mitral annular diameters. Late gadolinium enhancement (LGE) was present in 94% in the MV and mitral annulus and in the papillary muscles in 10 of 16 patients (63%), which is suggestive of fibrosis and which was associated with the presence of complex ventricular arrhythmias.
Comment.This was a study with a small number of patients. There was no information on coronary arteriographic findings in the patients with LGE in the papillary muscle. The etiology of LGE was not clear. The presence of LGE may be a clue to another etiological factor of MV prolapse. Additional studies of CMR in MV prolapse are needed, especially in those with complex ventricular arrhythmias.
Greater operative mortality for MV surgery in younger women
Of 24,977 patients (49% women) in the Society of Thoracic Surgeons (STS) database (34), the operative mortality in women was 3.9% versus 2.4% for men. The incidence of all comorbidities also was greater. The adjusted female/male OR for hospital mortality in the group ages 40 to 49 years was 2.56 (95% CI: 1.31 to 5.01; p = 0.0058) and in those ages 50 to 59 years was 1.95 (95% CI: 1.32 to 2.89; p = 0.0009).
3 months is optimal timing for echocardiography/Doppler for MR after myocardial infarction
From baseline to 1 year, LV end-diastolic volume index increased from 56 ± 15 ml/m2to 63 ± 19 ml/m2(p < 0.0001) (35); a graded increase depending on MR at 3 months at which time the incidence of no MR was 21%, mild MR 32%, and moderate/severe MR 60% (p = 0.008). By multivariate analysis, MR at 3 months but not at baseline was independently associated with increased hospitalization and death at a median follow-up of 1,226 days.
Durability of mitral valve repair (MVrep)
A total of 348 patients had MVrep and “the classic techniques described by Carpentier and others were used” (36). Hospital mortality was 1.6%. Survival at 5 and 10 years was 93 ± 1.5% and 80.1 ± 3.7%, respectively (Fig. 8).Survival was identical in those with Barlow's disease and in fibroelastic deficiency. No patient was lost to follow-up. The mean number of follow-up echocardiograms was 4.6 (range 2 to 16). At 1 month, 98.7% of patients had no or trivial MR (0/4 to 1/4); in Barlow's disease, it was 97.1%, and in fibroelastic deficiency, it was 99.8%. The incidence of failing repair (freedom from MR grade >2/4) was 82.2 ± 3.7% at 5 years and 64.9 ± 5.6% at 10 years (Table 4).At baseline, atrial fibrillation was present in 22%.
At most recent follow-up, patients in New York Heart Association (NYHA) functional classes I, II, III, and IV were 75%, 19%, 5%, and 1%, respectively. Warfarin anticoagulation was given to 35% of patients. Freedom from thromboembolic events and/or anticoagulation-related bleeding was 93.5 ± 1.5% at 5 years and 86.7 ± 2.7% at 10 years.
Comment.Flameng et al. (36) have once again presented careful, detailed results of MVrep. Follow-up included frequent echocardiograms, and results are reported as recurrences of MR and not as reoperation rates, which is a movable target. This study has important implications for type of MVrep, MV repair versus replacement, and also for indications of surgery in the asymptomatic patient. To be noted, data on the pre-operative NYHA functional classes and MR severity were not provided.
Outcomes after MVrep are similar to mitral valve replacement (MVR)
From 1985 to 2005, a total of 3,286 patients underwent primary operation for “degenerative” MR (37). Hospital mortality in 3,051 patients with MVrep was 0.26% and for 235 patients with MVR was 2.1% (p = 0.001). Patients undergoing MVR were older (70 ± 12 years vs. 57 ± 13 years) and had more complex valve pathology, symptoms, and LV dysfunction. In propensity-matched patients, survival was similar (p = 0.8) (i.e., at 5 years 86% vs. 83%, at 10 years 63% vs. 62%, and at 15 years 43% vs. 48%). Freedom from reoperation also was similar (p = 0.6) for MVrep at 5 and 10 years was 95% and for MVR was 95 and 92%, respectively.
Comment.David (38), in discussing this article, stated that a subsequent analysis of his own data in propensity-matched patients confirmed the aforementioned findings.
Reoperation after MVrep
From January 1980 to January 2005, a total of 4,024 patients in the same institution as the aforementioned study had MVrep for “degenerative” disease causing MR, and 188 (4.7%) underwent reoperation for recurrent MR (39). Reoperation for procedure-related problems occurred early (median 19 days) in 96 patients and, for valve-related problems, occurred later (median 5.4 years) in 109 patients (p < 0.0001). Mitral valve replacement was performed in 64% and repeat MVrep in 36%. Repeat MVrep was performed in 65% of recent reoperations. After re-repair at 10 years, the incidence of second re-repair was 7% and survival was 62%.
Tricuspid Regurgitation (TR)
The incidence of TR in patients with constrictive pericarditis
TR was present in 185 of 261 (71%) patients who had pericardiectomy (40); TR was moderate to severe in 54 (20%). Previous cardiac surgery and radiation were the cause in 34 (63%) and 15 (28%), respectively. Moderate-to-severe TR versus none-to-mild TR was associated with a greater operative mortality (13% vs. 4.3%; p = 0.019); however, excluding those cases caused by radiation, operative mortality was similar. Five-year survival was poorer in those with moderate-to-severe TR at 47% versus 87% in those with mild TR. Tricuspid valve repair had no significant impact on late survival.
Infective Endocarditis (IE)
High mortality of dialysis patients undergoing valve surgery
A total of 1,862 valvular procedures were recorded in dialysis patients with IE in the STS database. Points were assigned for operative mortality risks (Table 5)(41). The observed mortality increased from 7.1% to 76% as risk score points increased from 0 to 10+ (Table 6).
Comment.In 50% of the patients, the risks ranged from 23% to 60%.
Homograft valve not superior to conventional prosthesis
Of 127 consecutive patients, 54 (43%) were treated with aortic homograft and 73 (57%) with conventional prosthesis (42). Homografts were inserted in 41 of 63 (65%) patients who had an annular abscess. Between the 2 types of prostheses, hospital mortality was not significantly different (11% vs. 8%, p = 0.6). By multivariate analysis, prosthetic valve IE was associated with greater mortality (OR: 12.5; 95% CI: 1.4 to 114.2; p < 0.001). The 10-year survival free of end point (IE recurrence, noninfective prostheses dysfunction, or long-term cardiovascular mortality) was 44 ± 10% and was not significantly different between the 2 types of prostheses (p = 0.2).
Comment.Studies are needed to better define patients who need a homograft. Should a homograft be necessary only in those with an abscess and/or with persistent sepsis?
Transcatheter Valve Therapy
Position statement for transcatheter valve implantation
In 2008, the European Association of Cardio-Thoracic Surgery, the European Society of Cardiology, and the European Association of Percutaneous Cardiovascular Interventions (43) published a position statement on transcatheter valve implantation.
Endovascular resection of human aortic valves in situ
Percutaneous resection of the diseased native valve before the new valve is percutaneously implanted was performed in 10 human cadavers (44).
Transarterial AV implantation
Percutaneous transfemoral AV implantation was performed in 50 high-risk severely symptomatic patients by use of the Cribier-Edwards 23- and 26-mm expandable valves (Edwards Lifesciences Inc., Irvine, California). Mortality at 30 days was 12% in patients in whom the logistic EuroSCORE was 28% (45). There was a learning curve, and procedural success increased from 76% to 96% (Fig. 9).Valve area measured from echocardiograms increased from 0.6 ± 0.2 cm2to 1.7 ± 0.4 cm2. Paravalvular regurgitation was mild, LVEF increased (p < 0.001), and functional class improved (p < 0.0001).
Comment.Retrograde implantation is a significant advance. The current model of the Edwards SAPIEN valve uses the bovine pericardium with THERMA FIX, which in the PERIMOUNT valve for traditional surgical AVR has a low rate of structural valve deterioration out to 15 to 20 years (46). The SAPIEN valve was successfully delivered in 47 of 54 (87%) patients, the 30-day mortality was 7.4%, and the major adverse cardiac rate was 16.7% (47).
Concern exists regarding the severity of AR after percutaneous valve insertion. In an earlier multicenter study with the Cribier-Edwards valve the incidence of 3+/4+ AR was “common” (48). In this study, echocardiographic/Doppler criteria for severity of AR were not quantitative as is recommended by the American Society of Echocardiography. Left ventricular and femoral artery tracings raise concerns as discussed in Figure 10(49).
Percutaneous AV implantation was performed with the use of a second-generation CoreValve 21-F catheter in 50 patients and 18-F third-generation catheter in 36 patients (CoreValve Inc., Irvine, California). The logistic EuroSCORE was 21.7 ± 12.6%, and 83% of patients were in NYHA functional classes III and IV. In 6 patients, misplacement of the valve resulted in conversion to surgical AVR and, in 2 patients, calcified AV could not be crossed. Procedural success was 74% (50). The 30-day mortality was 12%, and the incidence of death, stroke, and myocardial infarction was 22%.
Comment.The CoreValve System seems to have a lot of metal, appears bulky, and currently has only a 21-mm bioprosthesis. It utilizes porcine pericardium whose rate of SVD is not known. Also, there was a high rate of procedural dilation after valve implantation (28% of 76 valve implantations) (51). The valve area post-procedure was not provided.
The 21-F CoreValve System (CoreValve, Paris, France) was inserted under femoro-femoral cardiopulmonary bypass support in 10 high-risk patients (estimated mortality determined by a EuroSCORE of 32% for surgical AVR) (52). The 30-day and 3-month mortalities were 20% and 30%, respectively. The AVA increased from 0.57 ± 0.19 cm2to 1.2 ± 0.35 cm2(p = 0.0001), and the median NYHA functional class improved from III to II (p = 0.01).
Comment.Post-procedure the mean valve area was only 1.2 cm2!
A total of 66 consecutive patients age ≥70 years with severe symptomatic AS were reported (53). Of 39 patients with a EuroSCORE ≥20%, AVA was 0.6 ± 0.2 cm2(0.4 ± 0.1 cm2/m2), MG was 47 ± 14 mm Hg, and LVEF was 0.52 ± 0.15. Pulmonary artery hypertension ≥60 mm Hg was present in 9% of patients who were considered at high risk or inoperable for surgery. Twelve received percutaneous SAPIEN valves (Edwards LifeSciences, Inc.). A total of 27 had contraindications to percutaneous valve and received a variety of therapies. In the 12 who had the percutaneous valve replacement, hospital mortality was 25%. A total of 31 patients had surgical AVR. In 27, the EuroSCORE was <20%, and 4 were considered at high risk for percutaneous AVR. The hospital mortality was 4%.
Are endovascular valved stents suitable for all patients?
Thirty-five patients with severe AS who were scheduled for AVR were studied. A braided nitinol self-expanded stent from Laboratories Perouse (Ivry-le-Temple, France) that was used was calibrated for implantation inside a circular orifice with a 20-mm diameter and was implanted in a triangular or an elliptical orifice with a circumference identical to the circular orifice (54). In tricuspid AV, the shape of the stent deployment was circular (68%), triangular (21%), or elliptic (11%). In bicuspid AV, elliptical deployment was present in 79%. The incidence of gaps between the stent external surface and the AV, in tricuspid and bicuspid AV, was 58% and 43%, respectively (p = 0.49).
Comment.The high incidence of gaps is of concern.
Transapical aortic valve implantation (TAP-AVI)
The TAP-AVI with the use of an Edwards SAPIEN valve (Edwards Lifesciences) was performed in 59 consecutive patients ages 81 ± 6 years with an average EuroSCORE for predicted mortality of 27 ± 14%. The TAP-AVI was successfully performed in 53 patients; 4 required early conversion to sternotomy (55). The 30-day mortality was 13.6%. Actuarial survival at 110 ± 77 days was 75.7 ± 5.9% (Fig. 11).
A TAP-AVI in 40 patients with the use of the Edwards SAPIEN valve (Edwards Lifesciences) was performed in 40 high-risk patients; 35 were successfully seated (56). The 30-day death rate was 17.5%, and in-hospital death rate was 22.5%. The AVA increased from 0.62 ± 0.13 cm2to 1.61 ± 0.37 cm2(p < 0.0001). Actuarial survival was 81.8 ± 6.2% at 1 month and 71.7 ± 7.7% at 3 months (Fig. 12).
Comment.An in-hospital death rate that is greater than the 30-day mortality implies that some patients had died after 30 days and not yet been discharged from the hospital after surgery.
The CoreValve System was used in an 80-year-old man with severe AR in a degenerated bioprosthetic valve. The logistic EuroSCORE was 36% (57). After 1 year, the patient was asymptomatic with good valve-in-a-valve prosthetic function.
Comment.The suitability of EuroSCORE for estimating mortality after valve surgery has been questioned (please see the previous section, “Aortic Stenosis”).
Importance of echocardiography in percutaneous aortic valve implantation (PAVI)
Fifty consecutive patients were deemed suitable for PAVI by the use of TTE, and 37 (74%) also underwent TEE (59). The authors used TTE to determine annular dimension and to guide correct device sizing. Using TEE, they were able to successfully guide device implantation in 97%, and TEE also was useful in the detection and treatment of “significant” paravalvular AR. The authors concluded that echocardiography has an “important role in case selection, in guiding device placement, and in detecting complications” of PAVI.
Evaluation of aortic root by multislice computed tomography
The use of multislice computed tomography provides information in 2 different views of aortic annulus and its shape (59). It also provides information on the distance from the annulus to the coronary ostia.
Intraventricular conduction abnormalities and the need for permanent pacemaker implantation
Forty consecutive patients who had PAVI were studied (60). Left bundle branch block (BBB) was present in 15% before PAV and 55% after PAVI (p = 0.0001). Two patients with right BBB became dependent on a pacemaker. Temporary and permanent pacemakers were needed in 20% and 18%, respectively.
Of 123 patients with PAVI, 2 patients (1.6%) required pacemaker implantation because of pre-existing intermittent bradycardia, 7 (5.7%) developed complete heart block and needed permanent pacemakers (61), and 4 (3.3%) developed left BBB.
Changes in MR with the Mitra Clip
In the Mitra Clip, EVEREST-1 (Endovascular Valve Edge-to-Edge Repair) study, MR was evaluated in a central echocardiographic laboratory (62). Comparing pre-procedure volumes to 6-month post-procedure volumes, the authors found that mean regurgitant volume had decreased from 50.3 to 27.5 ml (change −22.8 ml; p < 0.0001), regurgitant fraction from 44.6% to 28.9% (change −15.7%; p < 0.0001), color flow grade from 3.4 to 1.8 (change −1.6; p < 0.0001), and pulmonary vein flow from 2.8 to 1.8 (change −1.0; p < 0.0018).
Prosthetic Heart Valve (PHV)
Guidelines for reporting mortality and morbidity
Updated Guidelines from the American Association for Thoracic Surgery, The Society of Thoracic Surgeons, and The European Association for Cardio-Thoracic Surgery were published in 2008 (63).
Stroke after cardiac valve operation
In 2,808 patients, ages 63 ± 15 years, the incidence of stroke was 2.2 %; the incidence after AVR plus MVR was 5.4%, and valve + CABG was 3.6% (64). Brain imaging, which was positive in 74% of patients, showed ischemic stroke. Multivariate predictors of stroke were calcified ascending aorta, LVEF <0.30, diabetes, age >70 years, and bypass time >120 min. Hospital mortality in those patients with stroke was 24% versus 4.6%; 5-year survival was 54%, which was lower than those without stroke (p < 0.001) (Fig. 13).
“Low-dose” oral anticoagulation is better than the conventional dose in patients with mechanical PHV: final report
In ESCAT II (Early Self-Controlled Anticoagulation Trial), a prospective randomized trial, the authors evaluated “low-dose” international normalized ratio of 1.8 to 2.8 for AVR and 2.5 to 3.5 for MVR compared with the conventional international normalized ratio range of 2.5 to 4.5 (65). The authors' findings were similar to those in their preliminary report (66).
Comment.The authors claim their preliminary report in 2005 led to the lower dose recommendation by the American College of Cardiology/American Heart Association Guidelines in 2006, which is an incorrect claim. After their preliminary report, it was pointed out that their low dose was similar to the American College of Cardiology/American Heart Association recommendation in 1998 (66), which preceded the preliminary report of the authors by 7 years (1998 vs. 2005).
Good 25-year outcome after AVR with Medtronic-Hall PHV
In this study, 816 patients had AVR, and follow-up was 99.6% complete (67). Survival at 25 years was 24.9%. There were no mechanical failures. Older age, female sex, and associated CABG was associated with lower survival. Linearized rates of thromboembolic complications, warfarin-related bleeding, and IE were 1.5%, 0.7%, and 0.16% per patient year, respectively; at follow-up, 79% were in NYHA functional classes I and II.
Comment.The patients' conditions at baseline were not presented. Their comorbid conditions, which are very important factors in determining patient outcomes, were emphasized in 1988 (68).
Greater mortality with bioprosthesis in patients ages 50 to 70 years
From 1991 to 2000, 1,100 patients ages 50 to 70 years had isolated AVR ± CABG. Of these, 510 had St. Jude Medical mechanical and 257 Carpentier-Edwards bioprostheses model numbers 2625 (first-generation porcine valve), 2700 (standard Perimount pericardial valve), and 2800 (reduced-sewing ring version of the 2700 valve) (69). Data were analyzed in 440 patients matched 1:1 only by age, sex, CABG, and valve size in 220 patients in each group, mechanical and bioprosthesis. The “perioperative” mortality was much greater in the bioprosthetic group (5.5% vs. 1.8%, p = 0.04), who also had longer cross-clamp and cardiopulmonary bypass time (p = 0.03) and, on average, were 0.9 years older. Follow-up was 92% complete; for mechanical valve, it averaged 8.6 years and for bioprosthesis 6.2 years. The unadjusted 5- and 10-year survival for the mechanical group was 87% (95% CI: 83% to 92%) and 68% (95% CI: 62% to 76%), respectively, and for the bioprosthesis group it was 72% (95% CI: 66% to 78%) and 50% (95% CI: 52% to 58%, p < 0.01) for both. Survival for matched groups and total groups was “virtually identical.” Hemorrhagic complications necessitating hospitalization were greater in the mechanical group (15% vs. 7%, p = 0.01). Freedom from reoperation was similar in the 2 groups.
Comment.This area of investigation is an important one, but this study has the following concerns regarding its conclusions: 1) The VA randomized trial had shown a better survival after 10 years with Bjork-Shiley (Delron ring valve) compared with the porcine valve largely because of SVD and its complications that were observed with the porcine valve only in those <65 years of age (70). Thus, most patients in the present study were at risk for SVD with bioprosthesis and, thus, at risk for greater mortality.
2) It was shown in 1988 that survival after valve replacement was importantly determined by the patient's comorbid conditions (68). The incidence of lipid abnormalities, hypertension, and the percent who had pre-operative coronary arteriography were not presented. In the VA randomized trial, approximately 60% of late deaths after AVR were not related to the prosthesis; the cause(s) of death had been determined by a blinded committee (70). In the present study, matching was performed for only 4 baseline characteristics. The much greater operative mortality and longer bypass and cross-clamp time indicate that those in the bioprosthesis group were at greater risk.
Role of AVR in septuagenarian's quality of life (QOL)
A total of 160 septuagenarians underwent AVR, with CABG in 18.8% (71). “Global” hospital mortality was 8.8%. At follow-up (mean 3.4 ± 2.8 years), 121 were still alive and answered the Medical Outcomes Study Short-Form 36 Health Survey QOL questionnaire. Importantly, mean QOL scores on each of the 7 Short-Form 36 scales were significantly greater in the patients who underwent surgery than in the general Italian matched population; only the 8th scale (vitality domain) was not statistically significantly greater (Fig. 14).
Comment.A previous study in octogenarians also had similar data with regard to QOL scores (72); it had also shown marked improvement in patients' functional class. Thus, most septuagenarians and octogenarians undergoing traditional surgical AVR have a very good QOL and should not be denied AVR simply because of age.
Increased mortality with composite AV and root replacement (CVG)
From 1986 to 2004, 122,971 patients in the U.K. Heart Valve Registry were identified (73). The operative mortality with CVG was greater than with AVR alone during the entire time period. Excluding operative mortality, the 15-year survival in both groups was similar.
Comment.In the STS database of 409,904 valve procedures, the operative mortality for AVR was 5.7% and of CVG, when there was no aortic root pathology, was 9.5% (23).
Note.Procedures that require root reconstruction without root pathology include use of stentless valves, homografts (allografts), and pulmonary autografts (Ross principle). The incidence of SVD with use of these prostheses is of considerable interest (see the following text).
The use of SVD in stentless PHV, homografts, and autografts
Toronto Stentless Porcine Valve (SPV)
From 1991 to 2004, 357 patients received the Toronto SPV; there were 2 operative deaths. The 12-year survival was 64 ± 4% (74). The freedom from SVD at 12 years was 69 ± 4%; 52 ± 8% for patients younger than 65 years of age, and 85 ± 4% for patients ≥65 years of age (p = 0.002). The authors stated the TSPV has provided “…suboptimal durability, particularly in patients less than 65 years of age. We now use this valve mostly in older patients who have a small aortic annulus.”
Comment.This statement from Tirone David, arguably “the father of the stentless valve,” is a powerful one.
A total of 57 patients underwent reoperation for SVD (47 with Toronto SPV; 10 with Medtronic Free-style). Operative mortality was 11%; for redo AVR <1 year after stentless valve implantation versus >1 year was 67% versus 7% (p = 0.03) (75). The authors stated that “[R]eoperation after stentless AVR is a challenging procedure that frequently requires aortic root replacement.”
Comment.Reoperation within 1 year after AVR is associated with an extremely high mortality.
Cryolife O'Brien SPV
A total of 185 patients had AVR with the Cryolife O'Brien from 1994 to 2004 (76). Hospital mortality was 5.4%. Actuarial survival at 5 and 10 years was 68% and 40%, respectively. Freedom from SVD at 5 and 10 years was 91% and 44%, respectively.
Comment.This study reports a very high rate of SVD and of mortality.
A total of 200 consecutive patients underwent subcoronary AVR (77). Operative mortality was 1.5%. Minimum follow-up was 20 years, and maximum was 30 years. At 10, 15, 20, and 25 years, survival was 80.8 ± 2.8%, 67.8 ± 3.4%, 57.0 ± 3.6%, and 52.0 ± 3.2%, respectively, and freedom from reoperation for any reason was 86.5 ± 2.6%, 70.0 ± 3.8%, 40.3 ± 5.1%, and 36.0 ± 5.4%, respectively.
Comment.Patients' age not given; data are from a regional center (pediatric and adult patients) and a children's hospital.
Homografts and autografts
A total of 169 consecutive patients with congenital AV disease ages 16 to 55 years were studied; 63 received an autograft at age 29 ± 9 years and 106 at age 38 ± 10 years received homografts (78). Hospital mortality was 3.2%. The 13-year survival for autograft versus allografts recipients was 94.6 ± 2.1% versus 92.7 ± 3.3% (p = 0.45), respectively. At 13 years, freedom from reoperation for SVD was 63.4 ± 9.6% for autografts and 68.8 ± 6.3% for homografts (p = 0.44).
Comment.The rates of reoperation for SVD of both PHVs are high.
Dr. Rahimtoola has received honoraria for educational lectures from the American College of Cardiology Foundation; American College of Physicians; University of California Los Angeles; University of California Irvine; Cornell University; Creighton University; Thomas Jefferson University; Cedars-Sinai Medical Center; Harvard Medical School; University of Wisconsin; University of Hawaii; Cardiologists Association of Hong Kong, China; ATS; St. Jude Medical; Carbomedics; Merck; Pfizer; and Edwards Life Sciences. This review covers articles published from July 2007 to June 2008.
- Received September 18, 2008.
- Revision received January 21, 2009.
- Accepted February 6, 2009.
- American College of Cardiology Foundation
- Goldbarg S.H.,
- Elmariah S.,
- Miller M.A.,
- Fuster V.
- Bossey Y.,
- Mathieu P.,
- Pibarot P.
- Ross J. Jr.
- Babaliaros V.C.,
- Green J.T.,
- Lerakis S.,
- Lloyd M.,
- Block P.C.
- Minners J.,
- Allgeier M.,
- Gohlke-Baerwolf C.,
- Kienzle R.P.,
- Neumann F.J.,
- Jander N.
- Flachskampf F.A.
- Little S.H.,
- Chan K.-L.,
- Burwash I.G.
- Laissy J.-P.,
- Messika-Zeitoun D.,
- Serfaty J.M.,
- et al.
- Gutierrez-Chico J.L.,
- Zamorano J.L.,
- Prieto-Moriche E.,
- et al.
- Levy F.,
- Laurent M.,
- Monin J.L.,
- et al.
- Rahimtoola S.H.,
- Dilsizian V.,
- Kramer C.M.,
- Marwick T.H.,
- Vanoverschelde J.-L.J.
- Rahimtoola S.H.
- Pellikka P.A.,
- Sarano M.E.,
- Nishimura R.A.,
- et al.
- Rahimtoola S.H.
- Nashef S.A.,
- Roques F.,
- Michel P.,
- et al.
- Michel P.,
- Roques F.,
- Nashef S.A.
- Rahimtoola S.H.
- Detaint D.,
- Messika-Zeitoun D.,
- Malouf J.,
- et al.
- Lang R.M.,
- Bierig M.,
- Devereux R.B.,
- et al.
- Michelena H.I.,
- Desjardins V.A.,
- Avierinos J.-F.,
- et al.
- O'Gara P.,
- Sugeng L.,
- Lang R.,
- et al.
- Han Y.,
- Peters D.C.,
- Salton C.J.,
- et al.
- Vahainan A.,
- Alfieri O.,
- Al-Attar N.,
- et al.
- Webb J.G.,
- Pasupati S.,
- Humphries K.,
- et al.
- Rahimtoola S.H.
- Hanzel G.,
- Abbas A.,
- Dixon S.,
- et al.
- Webb J.G.
- Grube E.,
- Schuler G.,
- Buellesford L.,
- et al.
- Zegdi R.,
- Sleilaty G.,
- Lafout A.,
- Fabiani J.-N.
- Descoutures F.,
- Himbert D.,
- Lepage L.,
- et al.
- Zegdi R.,
- Ciobotaru V.,
- Noghin M.,
- et al.
- Walther T.,
- Simon P.,
- Dewey T.,
- et al.
- Moss R.R.,
- Ivens I.,
- Pasupati S.,
- et al.
- Tops L.F.,
- Wood D.A.,
- Delgado V.,
- et al.
- Piazza N.,
- Onuma Y.,
- Jesserun E.,
- et al.
- Sinhal A.,
- Altwegg L.,
- Pasupati S.,
- et al.
- Koertke H.,
- Zittermann A.,
- Tenderich G.,
- et al.
- Rahimtoola S.H.
- Svennevig J.L.,
- Abdelnoor M.,
- Nitter-Hauge S.
- Rahimtoola S.H.
- Hammermeister K.E.,
- Sethi G.K.,
- Henderson W.G.,
- Grover F.L.,
- Oprian C.,
- Rahimtoola S.H.
- ↵Vicchio M, Corte AD, De Santo LS, et al. Tissue versus mechanical prostheses: quality of life in octogenarians. Ann Thorac Surg 2008'85:1290–5.
- Sundt T.M.,
- Bailey M.S.,
- Moon M.R.,
- et al.
- Kalkat M.S.,
- Edwards M.B.,
- Taylor K.M.,
- Bonser R.S.
- Pavoni D.,
- Badano L.P.,
- Ius F.,
- et al.
- Klieverik L.M.A.,
- Bekkers J.A.,
- Roos J.W.,
- et al.