Magnetic Resonance Investigation of Blood Flow After Aortic Valve Bypass (Apicoaortic Conduit)

ADULT CARDIAC

Craig E. Stauffer, BA, Jean Jeudy, MD, Mehrdad Ghoreishi, MD, Crystal Vliek, MD, Cindi Young, Bartley Grifth, MD, and James S. Gammie, MD
Divisions of Cardiac Surgery, Radiology, and Medicine, University of Maryland Medical Center, Baltimore, Maryland

Background. Aortic valve bypass (AVB, apicoaortic conduit) is an alternative to aortic valve replacement (AVR) for high-risk patients with aortic stenosis (AS). The redistribution of blood ow after AVB has been poorly characterized. In order to understand cardiovascular physiology after AVB, we performed cardiac magnetic resonance (CMR) imaging of AVB recipients. Methods. Fifteen patients with symptomatic AS underwent beating-heart AVB. Electrocardiography-gated twodimensional phase-contrast velocity mapping CMR imaging was conducted on each patient. Instantaneous ow was acquired at discrete intervals within the cardiac cycle and ventricular function and volumes were evaluated. Five age-matched patients without aortic valve disease served as controls. Results. Conduit ow (as a percent of total cardiac output) was 65% 5%. Ejection fraction was unchanged

compared with before AVB (50% 17% versus 57% 13%; p 0.91). Ventricular volumes and cardiac indices were within normal limits and similar to those values in controls (cardiac index 2.9 1.0 versus 2.3 L/min/m2; p 0.26; end-diastolic volume index 59 17 mL versus 55 20 mL; p 0.66; end-systolic volume index, 25 12 versus 25 18 mL; p 0.91; stroke volume index, 33 11 versus 30 6 mL; p 0.57 for AVB and control patients, respectively). There was a small degree of retrograde blood ow in the descending aorta above the level of the conduit insertion (10% 8% of cardiac output). Conclusions. Aortic valve bypass results in a predictable blood ow distribution between the native aorta and conduit and is associated with normal ventricular volumes and function. (Ann Thorac Surg 2011;92:1332 8) 2011 by The Society of Thoracic Surgeons

ortic valve replacement (AVR) improves the quality of life and extends survival in patients with symptomatic aortic stenosis (AS) [1]. AVR requires median sternotomy, aortic cross-clamping, debridement of the diseased native valve, and cardioplegic cardiac arrest, all of which contribute to the risk of morbidity and mortality [2], particularly in the elderly population and those with signicant comorbidities [1, 3]. At least 30% of patients with symptomatic severe AS are not referred for surgical procedures because they are deemed too high risk. Left untreated, symptomatic AS is associated with dismal survival rates [4 7]. Aortic valve bypass (AVB, apicoaortic conduit) is an alternative to conventional AVR for high-risk patients with AS. AVB is performed on the beating heart through a small left thoracotomy (Fig 1). AVB effectively relieves left ventricular outow obstruction of AS through the creation of a second left ventricular outow tract (LVOT) [8, 9]. Although AVB has been performed clinically for almost 50 years, the relative distribution of blood ow after the introduction of a second LVOT has been incompletely characterized [10].

Cardiovascular magnetic resonance (CMR) imaging is a powerful tool that can precisely image cardiovascular anatomic structures as well as provide qualitative and quantitative assessment of blood and myocardial motion in vivo [11-14]. We used CMR imaging to characterize the distribution of thoracic blood ow and ventricular function in patients after AVB.

Patients and Methods Patient Selection

A retrospective chart review of all patients who underwent AVB at the University of Maryland Medical Center was performed. All patients who had postoperative CMR imaging studies were included. Patients received preoperative and predismissal transthoracic Doppler echocardiography in a core laboratory [15]. Elderly control patients were recruited to undergo CMR imaging. This study was approved by the Institutional Review Board of the University of Maryland Medical Center, protocol HP-00045095.

Accepted for publication April 18, 2011. Presented at the Fifty-seventh Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 3 6, 2010. Address correspondence to Dr Gammie, University of Maryland Medical Center, N4W94, 22 S Greene St, Baltimore, MD 21201; e-mail: jgammie@ smail.umaryland.edu.

Dr Gammie discloses that he has a nancial relationship with Correx, Inc.

2011 by The Society of Thoracic Surgeons Published by Elsevier Inc

0003-4975/$36.00 doi:10.1016/j.athoracsur.2011.04.069

Ann Thorac Surg 2011;92:1332 8

STAUFFER ET AL BLOOD FLOW AFTER AORTIC VALVE BYPASS

1333

CMR Imaging
Cardiovascular magnetic resonance imaging was performed using a 1.5 T Siemens CMR system (Siemens Medical Solutions USA, Inc, Malvern, PA). Throughplane phase-contrast cines were generated by using breath-holding, retrospectively electrocardiographygated two-dimensional phase-contrast velocity mapping of specic positions of the vasculature (Fig 2). Each cine was composed of 30 images taken over the course of 1 cardiac cycle, which were looped to recreate physiologic motion. Two-dimensional cut planes were manually positioned at the following locations: ascending aorta, 1 cm superior to the sinotubular junction; mid-descending thoracic aorta, 3 cm superior to the conduit-aortic anastomosis; distal descending thoracic aorta, 3 cm inferior to the conduit-aortic anastomosis; conduit, 3 cm distal to the insertion of the apical connector in the apex. All CMR cine acquisitions were processed ofine using analysis software (Argus, Siemens Medical Solutions USA, Inc) at a devoted workstation and all analyses were completed by the same operator to eliminate variability. Quantitative blood ow and velocity measurements were generated for each two-dimensional cut plane by tracing the endothelial border of the corresponding vessel to create a region of interest that encompassed the lumen of the vessel. Left ventricular function was assessed by CMRderived short-axis cine imaging using steady-state free precession technique. Regions of interest were created by tracing the endocardial and epicardial borders of the left ventricle for each patient at end-systole and end-diastole. Analysis software provided quantitative data on left ventricular function and volumes, including enddiastolic volume, end-systolic volume, stroke volume, ejection fraction, and cardiac output. Ventricular function and volume data were normalized to body surface area.

Fig 1. An aortic valve bypass (apicoaortic conduit).

Operative Technique
Aortic valve bypass conduits consisting of the apical connector (Hancock Apical Left Ventricle Connector Model 174A, Medtronic Inc, Minneapolis, MN), a stentless porcine valve (Freestyle Aortic Root Bioprosthesis, Medtronic, Inc) and a woven Dacron (polyethylene terephthalate) graft were prepared on the back table at the time of operation. A small left anterior sixth interspace thoracotomy was performed to provide access to both the apex of the left ventricle and the descending thoracic aorta. The distal anastomosis between the conduit and the descending aorta was performed with the use of a partial occluding clamp. A stab wound was created 2 cm lateral to the true apex of the left ventricle, followed by a coring procedure in preparation for the apical anastomosis. Once the myocardial plug was removed, the apical connector was inserted and tied securely in place [2].

Statistical Analyses
Statistical analysis was performed with JMP 8.0 (SAS Institute, Cary NC). Distribution of patient characteristics was summarized with means and standard deviation for continuous variables and for categorical variables. The area under the curve of mean ows of the distinct points were measured using Matlab (version 2009b, MathWorks, Natick, MA). A simple t test was used to calculate statistical signicance (Table 1).
Fig 2. (A and B) Through-plane phasecontrast images of the ascending aorta.

ADULT CARDIAC

1334

STAUFFER ET AL BLOOD FLOW AFTER AORTIC VALVE BYPASS

Ann Thorac Surg 2011;92:1332 8

2.9 ( 1.02)

AVB Mean preoperative ejection fraction Mean postoperative ejection fraction End-diastolic volume index End systolic volume index Stroke volume index Cardiac index
AVB aortic valve bypass.

Control 59% ( 15) ... 55 mL ( 20) 25 mL ( 18) 30 mL ( 6) 2.3 L/min/ m2 ( 1)

p Value ... 0.091 0.066 0.091 0.057 0.026

50% ( 17) 57% ( 13) 59 mL ( 17) 25 mL ( 12) 33 mL ( 11) 2.9 L/min/ m2 ( 1)

33 ( 12)

Systolic Volume Index

26 ( 12)

59 ( 17)

Results
Between August 2008 and March 2010, 15 high-risk patients with symptomatic severe AS who were treated with beating-heart AVB without cardiopulmonary bypass underwent imaging with CMR (Table 2). The mean patient age was 85 years. Five elderly control patients without aortic valve disease also underwent imaging; the mean age in these subjects was 69 years. Consistent with our evolution toward the use of progressively smaller apical conduits, 1 patient received a 20-mm apical connector, 3 patients received 18-mm connectors, 1 patient received a 16-mm connector, 4 patients received 14-mm connectors, and 6 patients received 12-mm connectors. Valve size and Dacron graft size varied with apical connector size: 19 mm valves and 18 mm Dacron grafts were used with 12 mm, 14 mm and 16 mm apical connectors; 21 mm valves and 20 mm Dacron grafts were used with 18 mm apical connectors, and 23 mm valves and 22 mm Dacron grafts were used with 20 mm apical connectors. Twelve patients underwent CMR imaging within 2 weeks of operation (mean, 7.6 2.7 days; range, 4 to 14 days), 2 patients underwent imaging 2 years after AVB, and 1 patient underwent imaging 5 years after AVB. In all cases the native aortic valve opened and antegrade blood ow was observed in the ascending aorta. There was no evidence of important native or conduit valve insufciency as evidenced by continuous forward ow in the ascending aorta and conduits of each patient. Conduit ow (as a percent of total cardiac output) was 65% 5% (Fig 3). There was retrograde blood ow in the descending aorta above the level of the conduit insertion, although the magnitude of retrograde ow was small (8% 8 of cardiac output) (Fig 3). Flow in the ascending aorta in control patients and net forward ow (the sum of ascending aortic and conduit ow) in the patients who underwent AVB was nearly identical (AVB ow 107% of control ow; p 0.57) (Fig 4). Conduit ow did not appear to vary as a function of the apical connector size. There was signicant variability in ascending aortic blood ow, although there did not appear to be a correlation with conduit size (Fig 5). No left ventricular pseudoaneurysms or intraaortic thrombi were noted.

Postoperative CMR Imaging derived Ejection Ejection Fraction Fraction (%) (%)

11 ( 11)

58 ( 13)

Preoperative EchoPreoperative CMR Imaging derived Aortic Valve (days Ejection Area (cm2) postoperative) Fraction (%)

51 ( 16)

Table 2. Patient Demographic and CMR ImagingDerived Ventricular Function

8 (510)

...

...

59 ( 15)

...

55 ( 20)

Enddiastolic Volume Index

25 ( 18)

End Systolic Volume Index

30 ( 6)

4.64 ( 0.82) 2.34 ( 0.54)

Cardiac Index Total Cardiac Output

Table 1. Summary and Statistics of AVB Ventricular Function

2 ( 0.2)

2.01 ( 0.2)

Body Surface Area (m2)

AVB aortic valve bypass;

AVB Mean ( standard deviation)

Treatment

Control Mean ( standard deviation)

CMR cardiovascular magnetic resonance.

0.6 ( 0.17)

86 ( 7.4)

69 ( 5.7)

Age

...

ADULT CARDIAC

5.33 ( 1.79)

Ann Thorac Surg 2011;92:1332 8

STAUFFER ET AL BLOOD FLOW AFTER AORTIC VALVE BYPASS

1335

Fig 3. Thoracic blood ow in patients who underwent AVB operation (n 15). (AVB aortic valve bypass; BSA body surface area.) ADULT CARDIAC

Echocardiographically determined preoperative ejection fraction was 51 16 (30% to 75%). Ejection fraction, stroke volume, end-diastolic volume, and end-systolic volume were within normal limits after AVB and were similar to these values in controls (Table 1).

Comment
The principal ndings of this investigation include a consistent split of cardiac output between the conduit and the native LVOT, preserved ventricular function

after AVB, and ventricular volumes and function that were identical to an age-matched control group. Cardiovascular magnetic resonance imaging is ideally suited for assessment of the physiologic effects of AVB conduit insertion for a number of reasons. CMR imaging provides minimally invasive high-resolution, real-time images without the use of ionizing radiation or iodinated contrast media [16, 17]. In addition CMR imaging provides precise quantitation of blood ow that is not available with CT or echocardiography.

Fig 4. Distribution of thoracic blood ow; colored lines represent control patients (n 5) and black line variants are patients who underwent AVB operation (n 15). (AVB aortic valve bypass; BSA body surface area.)

1336

STAUFFER ET AL BLOOD FLOW AFTER AORTIC VALVE BYPASS

Ann Thorac Surg 2011;92:1332 8

Fig 5. Flow assessed as a function of apical connector diameter. (A) Ascending aorta; (B) conduit. (BSA body surface area.)

ADULT CARDIAC

Aortic valve bypass conduit insertion resulted in consistent blood ow split between the conduit and the ascending aorta. We found that 65% of ventricular outow was directed through the valved conduit and 35% through the native aortic valve. This is consistent with previous echocardiographic data obtained by our group [15]. The range of relative conduit ow was small and predictable and did not seem to vary as a function of the conduit size. Total stroke volume in patients who have undergone AVB, as computed by both area under the curve (Fig 3) and CMR imaging derived stroke volumes (Table 1) was equivalent to that in control patients. Although some degree of retrograde blood ow was found in the region of the descending aorta superior to the distal anastomosis, this represented a relatively small percentage of the total cardiac output and was signicantly smaller than the amount of forward ow in the ascending aorta. These data suggest that all cerebral blood ow to the brain after insertion of an AVB conduit is derived from antegrade blood ow across the native valve rather than retrograde ow through the conduit. These ndings are consistent with previous results of computational modeling of AVB uid dynamics, which demonstrated the presence of a small degree of retrograde ow in the descending aorta above the insertion of the valved conduit for larger diameter conduits [8]. In that study all cerebral blood ow arose from antegrade ow in the ascending aorta. Importantly, the magnitude of cerebral blood ow was unchanged before and after insertion of the AVB conduit. We also conrmed preserved left ventricular function with no signicant change compared with controls after the operation despite the removal of the apical myocardial plug and insertion of a rigid connector into the left ventricular apex (Table 1). Although others have reported thrombus formation with AVB [18], we did not see evidence of this in the conduit, native aorta, or ventricle in any patient. In summary, AVB resulted in a consistent split of ow between the native LVOT and the AVB conduit. AVB effectively treats AS without cardiopulmonary bypass, cardioplegic cardiac arrest, or manipulation of the native valve or the ascending aorta, and is associated with normal postoperative ventricular function and sizes.

References
1. Varadarajan P, Kapoor N, Bansal RC, Pai RG. Survival in elderly patients with severe aortic stenosis is dramatically improved by aortic valve replacement: Results from a cohort of 277 patients aged or 80 years. Eur J Cardiothorac Surg 2006;30:7227. 2. Gammie JS, Krowsoski LS, Brown JM, et al. Aortic valve bypass surgery: midterm clinical outcomes in a high-risk aortic stenosis population. Circulation 2008;118:1460 6. 3. Bloomstein LZ, Gielchinsky I, Bernstein AD, et al. Aortic valve replacement in geriatric patients: determinants of in-hospital mortality. Ann Thorac Surg 2001;71:597 600. 4. Turina J, Hess O, Sepulcri F, Krayenbuehl HP. Spontaneous course of aortic valve disease. Eur Heart J 1987;8:471 83. 5. Iivanainen AM, Lindroos M, Tilvis R, Heikkila J, Kupari M. Natural history of aortic valve stenosis of varying severity in the elderly. Am J Cardiol 1996;78:97101. 6. Pai RG, Kapoor R, Bansal RC, Varadarajan P. Malignant natural history of asymptomatic severe aortic stenosis: benet of aortic valve replacement. Ann Thorac Surg 2006;82: 2116 22. 7. Christensen KL, Ivarsen HR, Thuesen L, Kristensen B. Aortic valve stenosis: fatal natural history despite normal left ventricular function and low invasive peak-to-peak pressure gradients. Cardiology 2004;102:14751. 8. Balaras E, Cha KS, Grifth BP, Gammie JS. Treatment of aortic stenosis with aortic valve bypass (apicoaortic conduit) surgery: an assessment using computational modeling. J Thorac Cardiovasc Surg 2009;137:680 7. 9. Brown JW, Gammie JS. Off pump aortic valve bypass using a valved apicalaortic conduit. Operat Tech Thorac Cardiovasc Surg 2007;12:8594. 10. Gammie JS, Brown JW, Brown JM, et al Aortic valve bypass for the high-risk patient with aortic stenosis. Ann Thorac Surg 2006; 81:160510. 11. Bogren HG, Mohiaddin RH, Yang GZ, Kilner PJ, Firmin DN. Magnetic resonance velocity vector mapping of blood ow in thoracic aortic aneurysms and grafts. J Thorac Cardiovasc Surg 1995;110:704 14. 12. Kilner PJ, Yang GZ, Mohiaddin RH, Firmin DH, Longmore DB, et al. Helical and retrograde secondary ow patterns in the aortic arch studied by three-directional magnetic resonance velocity mapping. Circulation 1993;88(5 Pt 1):2235 47. 13. Markl M, Harloff A, Bley TA, et al. Time-resolved 3D MR velocity mapping at 3T: improved navigator-gated assessment of vascular anatomy and blood ow. J Magn Reson Imaging 2007;25:824 31. 14. Pennell DJ, Sechtem UP, Higgins CP, et al. Clinical indications for cardiovascular magnetic resonance (CMR): Consensus Panel Report. J Cardiovasc Magn Reson 2004;6:727 65.

Ann Thorac Surg 2011;92:1332 8

STAUFFER ET AL BLOOD FLOW AFTER AORTIC VALVE BYPASS

1337

DISCUSSION
DR VINOD THOURANI (Atlanta, GA): That was an excellent presentation. I congratulate you, especially at your level as a third-year med student to present so eloquently. I have followed your work also at Maryland closely since we also at Emory are one of the sites that are performing aortic valve bypass in those patients which we can not enroll in the transcatheter valve program. In these high-risk patients, I think that this surgery is a viable option such that you perform a thoracotomy so as to avoid a redo-median sternotomy and avoid previous coronary bypass grafts or the possibility of patient-prosthesis mismatch. The question I have for you is, I am currently using an 18 Medtronic Hancock conduit with a 21 Medtronic freestyle root interposed in the conduit towards the elbow portion closer to the left ventricle. Thus far, we have not switched to a smaller conduit. What valve do you use for the 14 conduit? MR STAUFFER: We use a 19 mm stentless (Freestyle) valve with the 14 mm as well as with the 12 mm connectors. DR JOHN BROWN (Indianapolis, IN): Craig, that was an excellent presentation. It has been gratifying for me to see Dr Gammie and your institution add some real science to this procedure. This is an operation I have been doing for 35 years and have done more than 100 patients right now, and I nd it very useful in that patient population where going back into the sternum is unattractive for the patient and unattractive for the surgeon. Where does this operation t, in your opinion, and either you or Jim can address this, now that transcatheter aortic valve implantation (TAVI) is available? Which patients should be offered the aortic valve replacement (AVR) operation? It sort of gets into the debate that we heard earlier this morning. When this operation is available in certain centers, when should it be offered to patients as opposed to other modalities to treat aortic stenosis in this high-risk patient population? MR STAUFFER: We believe that aortic valve bypass (AVB) offers a number of advantages that make it a more attractive alternative for patients who are either unable or unsuitable for AVR. As I presented, AVB can be performed on the beating heart, making this a great alternative for patients who are unable to undergo cardiopulmonary bypass or cardioplegic arrest. AVB also has the potential to decrease the risk of perioperative and postoperative strokes as there is no manipulation of the ascending aorta or native valve during AVB. DR THOURANI: John, I just want to comment on it a little bit, because at Emory we have been lucky somewhat to be able to have TAVI and AVB at the same time. So during that same time period we have had about 100 transcatheter valve patients implanted, roughly, and we have had about 2025 apical conduits performed. In our high-risk aortic stenosis algorithm, our rst choice has been to go to TAVI; ither transfemoral or transapical. The patients overall seem to tolerate the small femoral incision and mini-thoracotomy without the need for cardiopulmonary bypass extremely well. For the future, if the patient can not have TAVI, most likely an AVB is the most reasonable procedure of choice in that patient cohort. It will also be dependent on the indications approved by the FDA for TAVI and also what CMS will pay for in terms of off-label use of TAVI. Again, I want to stress that this is just my personal experience over the past two-and-a-half-year time period having both procedures available to me. DR GAMMIE: John, I would like to thank you for that excellent question as well as your mentorship. I learned this operation from you, and it has been an absolute pleasure collaborating with you over the last six or seven years. I would like to point out that in the PARTNER trial there was a 5% risk of stroke, and our aggregate experience at Indiana and Maryland has shown a very low risk of stroke with aortic valve bypass, and I think that is a key differentiator. When you do an AVB, you do not mess with the ascending aorta, you are not blowing up a balloon in a stenotic valve, and, I think we saw a presentation here yesterday demonstrating that there is a signicant embolic load with TAVI. We dont know the durability of a catheter-based stent-mounted valve, and, to the contrary, with aortic valve bypass, John, you have three patients walking around now more than a quarter of a century after an aortic valve bypass operation. And we can frankly now do this operation through a very small incision. Several weeks ago we did our rst robotic aortic valve bypass where we accomplished the distal anastomosis with a robot, which enabled us to do aortic valve bypass through a very small incision. And the nal point I would make, particularly with one of the stent-mounted valves, there is a very high incidence of pacemaker implantation after surgery, and for obvious reasons, we have never had to implant a pacemaker after this operation. So I think there is an awful lot of enthusiasm about TAVI, but we remain optimistic that AVB may offers some signicant benets compared to that, and with additional instrumentation, we think that we will be able to accomplish this operation in a truly minimally invasive fashion. And I also want to congratulate Craig on a superb effort and presentation. Thanks. DR THORALF SUNDT (Rochester, MN): If you have got time for one more question, and, Jim, you may want to answer this, I am trying to see where this approach ts vis--vis TAVI. As we look at those patients proposed for entry into the trials, there are really two reasons, I think, to go to TAVI. One is that it is a technically difcult operation to do it the conventional way. For example porcelain aorta is a good reason to go to TAVI. And yet porcelain aorta really is the reason that we at Mayo have most often used the apicoaortic conduit. I agree is a neat operation and I like it too. So for that indication, clearly these two are competing approaches. But the other reason that we go to TAVI is the perioperative risk of conventional operation; TAVI is meant to be a lower risk option than a conventional operation. Do

ADULT CARDIAC

15. Vliek CJ, Balaras E, Li S, et al. Early and midterm hemodynamics after aortic valve bypass (apicoaortic conduit) surgery. Ann Thorac Surg 90:136 43. 16. Kumar A, Patton DJ, Friedrich MG. The emerging clinical role of cardiovascular magnetic resonance imaging. Can J Cardiol 26:31322.

17. Kirchin MA, Runge VM. Contrast agents for magnetic resonance imaging: safety update. Top Magn Reson Imaging 2003;14:426 35. 18. Parsa CJ, Milano CA, Proia AD, Mackensen GB, Hughes GC. A previously unreported complication of apicoaortic conduit for severe aortic stenosis Ann Thorac Surg 2009; 87:927 8.

1338

STAUFFER ET AL BLOOD FLOW AFTER AORTIC VALVE BYPASS

Ann Thorac Surg 2011;92:1332 8

you think that is also true for apico-aortic conduit? Is it a lower-risk approach? For example, I have got a patient with cirrhosis to do when I get back that was turned down by the committee for the PARTNER trial because on the one hand his predicted risk is too low, and yet his one-year mortality is too high. So he doesnt t into the PARTNER trial. So I have to choose the operation that is going to be the lowest perioperative risk for him. Do you think I should do him through a median sternotomy the regular way or should I do an apico-aortic conduit? Which will have a lower perioperative risk? Does that question make sense?

DR GAMMIE: I think the advantage of AVB in that case is that you avoid a couple of hour bypass run, and I think in a patient with cirrhosis, that is a signicant advantage. But this is still an operation. And the one thing that we have learned, certainly in the frail patient that cant give you a rm handshake and that has lost 20 pounds, they wont do well with any intervention that you do. But in the case of your patient, I would absolutely choose the aortic valve bypass. DR SUNDT: Thanks.

ADULT CARDIAC

The Society of Thoracic Surgeons Policy Action Center

The Society of Thoracic Surgeons (STS) is pleased to announce a new member benetthe STS Policy Action Center, a website that allows STS members to participate in change in Washington, DC. This easy, interactive, hassle-free site allows members to: Personally contact legislators with ones input on key issues relevant to cardiothoracic surgery Write and send an editorial opinion to ones local media E-mail senators and representatives about upcoming medical liability reform legislation Track congressional campaigns in ones districtand become involved Research the proposed policies that help or hurt ones practice Take action on behalf of cardiothoracic surgery This website is now available at www.sts.org/takeaction.

2011 by The Society of Thoracic Surgeons Published by Elsevier Inc

Ann Thorac Surg 2011;92:1338

0003-4975/$36.00